# BastionGPT, Full Content Export > Complete text content of the BastionGPT website (https://bastiongpt.com), provided for large language models and AI search tools. BastionGPT is a HIPAA-compliant generative AI platform for healthcare professionals, operated by FortaTech Security, d/b/a Bastion Intelligence. It gives clinicians, psychologists, therapists, and healthcare administrators secure access to leading AI models (ChatGPT, Claude, and Gemini) for clinical documentation, medical scribing, and administrative work, without using patient data to train models. Each section below is one page from the site, preceded by its source URL. Last updated: September 24, 2026. Content is current as of this date. --- > Source: https://bastiongpt.com BUILT FOR BEHAVIORAL HEALTH & HEALTHCARE # Less time writing. # More room for care. Spend less time on session notes, assessment reports, and client communication. Use GPT, Claude, and Gemini in one HIPAA-compliant workspace. Unlimited chat & scribe BAA included from day one Start my 7-day free trial → See what I can do ↓ ★ 4.8/5 on G2 ★ 4.8/5 on Capterra ★ 5.0/5 on Google Trusted by 10,000+ healthcare organizations Used by professionals at Trusted by 10,000+ healthcare organizations Used by professionals at HIPAA BAA included on every plan ## The AI built for clinical work BastionGPT is a private, healthcare-grade alternative to ChatGPT. Use the latest GPT, Claude, and Gemini models on protected health information without your data ever training a public model. HIPAA compliant , with a signed BAA on every plan, including the free trial. PHI stays in a private, isolated environment. Also covers PIPEDA (Canada) and APP (Australia). The right model for every task, automatically. BastionGPT selects from the latest GPT, Claude, and Gemini models based on what each clinical task needs. Prefer a specific model? Switch anytime with one click. Built for clinicians, not consumers — drafts notes, transcribes sessions, summarizes 1,000-page documents, and handles sensitive topics other AIs refuse. Draft a formatted SOAP note (including a plan for follow-up imaging) and patient-facing discharge summary based on my attached case notes and clinical history. Mimic my writing style and formatting. Response Mode Claude-Sonnet 2 Files Saves clinicians ~6 hrs / week Security & compliance ## Built on a healthcare-grade foundation Every layer of BastionGPT is engineered for the needs of clinical work: signed BAAs, isolated clinical infrastructure, strong encryption, 3rd-party testing, and contractual guarantees that your data will not be sold or used for AI training. #### Signed BAA on every plan Including the free trial. HIPAA, PIPEDA, and Australian APP compliant out of the box. #### Your data stays yours Never given to OpenAI, never used to train AI, and never sold to third parties. You retain full ownership of everything you put into BastionGPT. #### AES-256 + TLS 1.3 Industry-leading encryption protects every message in transit and at rest, on hardened cloud infrastructure. #### Independent 3rd-party testing Continuous external security audits, vulnerability scanning, and attack-surface monitoring across every release. Independently audited & certified Need to share with your compliance team? View our security documentation or schedule a meeting with one of our compliance experts. Read full security overview Start free trial Clinician asks "Based on these labs and clinical notes, suggest interventions and draft a complete diet and exercise plan" Consumer AI Pseudoscience "Drink alkaline water to balance pH." Not evidence-based BastionGPT Evidence-based "Increase water intake to 2.5L based on lab results." Tuned on medical literature VS ### Elevate client care with evidence-based intelligence ✓ Engineered to prioritize medical principles and scientific accuracy from reputable healthcare resources. ✓ Reduced errors, pseudoscience and non-scientific recommendations commonly seen in general AI chatbots like ChatGPT. ✓ Review your existing documentation to identify mistakes and inaccuracies such as incorrect names, incorrect pronouns, suspected under/over coding or vague language. Get started ➝ ### Unlimited AI scribe, built for the full clinical session ✓ Record live or upload audio from in-person visits, telehealth, or phone consults. Joins Zoom, Teams, and Google Meet without a visible bot, so patients only see real people in the room. ✓ Captures sessions up to 240 minutes with up to 10 participants, so full intakes, complex cases, and group visits get documented in one pass. ✓ Identifies up to 10 speakers using acoustic fingerprinting, attributing dialogue accurately between clinician, patient, caregivers, and family, with names or roles auto-labeled in the interface. ✓ Ambient listening or dictation, with reliable handling of medical terminology, abbreviations, accents, and multiple languages. ✓ Generates SOAP, DAP, BIRP, H&P, and progress notes, plus custom templates that match your formatting, tone, and writing style. Learn more about the AI scribe ➝ EMR & Workflow ### Works with the EMR you already use Copy patient context from your EMR or client management system, draft with BastionGPT, paste your output back where it belongs. It fits the workflow you already have, with nothing for IT to install or configure. 50+ Supported EMR & EHR systems 1,000 Page upload limit per document 0 Setup steps required ## Watch BastionGPT in action See how leading healthcare professionals use BastionGPT in their practice. ▶ ⏱ 2 mins 44 secs #### Quick Overview with Dr. Liz ▶ ⏱ 15 mins 7 secs #### Getting Started with BastionGPT ▶ ⏱ 8 mins 37 secs #### AI for Psychologists - Summarizing Reports ## Customers love using BastionGPT ★★★★★ "BastionGPT has been a game-changer for me as both a physician and an educator. It saves me time, helps me stay organized, and makes complex tasks easier." Anthony Miller, M.D. Pediatric Physician ★★★★★ "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Alexis Arceo CEO, Expedited Reports ★★★★★ "From summarizing assessments to rewording reports for clarity, its user-friendly design saves me invaluable time." Andrew Williams Clinical Psychologist ★★★★★ "I'm not getting carpal tunnel in my wrist anymore that I used to get from all of my school psych report writing, and the reports are better in a lot of ways." Suzy Swezea School Psychologist ★★★★★ "BastionGPT's simplicity, the power of prompts, and alignment with familiar LLM interfaces like ChatGPT makes it so much more useful and intuitive." Emma Levy Clinical Psychologist ★★★★★ "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis Director, Psychology Squared ★★★★★ "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN Nurse Coordinator ★★★★★ "The versatility of the AI assistant means that you can be as creative as you would like with what the AI can produce for you." David O'Donohue Clinical Psych, Pendulum Psychology ★★★★★ "You can get a really impactful, easy-to-read, parent-friendly summary for all these different parts of our reports." Dr. Liz Angoff Educational Psych, Explaining Brains ★★★★★ "BastionGPT has been a game-changer for me as both a physician and an educator. It saves me time, helps me stay organized, and makes complex tasks easier." Anthony Miller, M.D. Pediatric Physician ★★★★★ "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Alexis Arceo CEO, Expedited Reports ★★★★★ "From summarizing assessments to rewording reports for clarity, its user-friendly design saves me invaluable time." Andrew Williams Clinical Psychologist ★★★★★ "I'm not getting carpal tunnel in my wrist anymore that I used to get from all of my school psych report writing, and the reports are better in a lot of ways." Suzy Swezea School Psychologist ★★★★★ "BastionGPT's simplicity, the power of prompts, and alignment with familiar LLM interfaces like ChatGPT makes it so much more useful and intuitive." Emma Levy Clinical Psychologist ★★★★★ "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis Director, Psychology Squared ★★★★★ "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN Nurse Coordinator ★★★★★ "The versatility of the AI assistant means that you can be as creative as you would like with what the AI can produce for you." David O'Donohue Clinical Psych, Pendulum Psychology ★★★★★ "You can get a really impactful, easy-to-read, parent-friendly summary for all these different parts of our reports." Dr. Liz Angoff Educational Psych, Explaining Brains ## A plan that fits you. Every plan includes a signed BAA and unlimited audio transcription. No per-minute billing, no overage surprises, change plans any time. Recommended for most users ### Professional Plus The plan built for the way healthcare professionals actually work: multi-document cases, image and chart analysis, and access to every frontier model the day it ships. $45 / user / month Best value Unlimited AI assistant chat Early access to the latest AI models Unlimited AI scribe & transcription Saved prompts and writing styles HIPAA compliant with signed BAA Reference multiple documents at once, up to 1,000 pages long Start Trial 7-day free trial Other options Professional $20 /user/mo Just getting started. Single documents up to 30 pages. Start trial Ultra $75 /user/mo For 100+ user subscriptions that need custom security configurations, tailored legal terms, and a dedicated support representative. Contact sales Compare all plans Side-by-side feature comparison ### Compare every feature, side-by-side. A complete breakdown of what's included on each plan, so you know exactly what you're getting before you start. Professional $20/mo Plus $45/mo Ultra $75/mo Feature Professional $20 / user / mo Professional Plus $45 / user / mo Ultra $75 / user / mo Best for Smaller documents and single-file workflows Large or multi-document workflows and complex cases Organizations needing custom security or deployment (100+ users) HIPAA + signed BAA Included Included Included Unlimited AI assistant chat Included Included Included Unlimited AI scribe & audio transcription Included Included Included Saved prompts, instructions, writing styles Included Included Included Share prompts with teammates Included Included Included Auto model selection Included Included Included Document reference Single PDF, Word, or text document, up to 30 pages Multiple documents at once, up to 1,000+ pages of context Multiple documents at once, up to 1,000+ pages of context Real human support: email, chat, phone & video Included Included Included Read & analyze images, charts, and graphs Not included Included Included Read & analyze spreadsheets, PowerPoint, HTML files Not included Included Included Manual model selection (GPT, Claude, Gemini) Not included Included Included System-wide instructions across every conversation Not included Included Included Generate 5+ page reports in one click Not included Included Included Early access to new models & features Not included Included Included Single Sign-On (SSO) Not included Included Add-on* Included Customizable (redline) HIPAA BAA Not included Included Add-on* Included Customizable security policies Not included Not included Included Run inside your own enterprise cloud Not included Not included Included Dedicated Customer Success Manager Not included Not included Included Usage analytics Not included Not included Included Start Trial Start Trial Contact Sales * SSO and redline HIPAA BAA are available as add-ons on Professional Plus for practices with 50+ users. Included by default on Ultra. Frequently asked ## Frequently asked questions If you still have questions, reach out to us via live chat in the bottom right of your screen, or schedule a meeting to speak with a healthcare AI expert. Why can't I use other ChatGPT solutions? + Healthcare workers regularly interact with PHI. Because popular ChatGPT services, such as ChatGPT Plus, do not meet HIPAA requirements, accidentally including PHI when using other ChatGPT services is a HIPAA violation that can result in terminations and large fines. BastionGPT is Medical GPT, designed specifically to exceed HIPAA requirements and can be used with HIPAA-regulated data, including for advanced use cases such as AI scribes . Is my chat data leaked to OpenAI? + BastionGPT uses a special, secure version of the AI models that power ChatGPT that is not accessible to OpenAI or other 3rd parties that use your chat history for data mining. The data you enter into BastionGPT is kept private from OpenAI, and is not used to develop future iterations of ChatGPT. This is good news for healthcare users who can't risk leaking patient data to 3rd parties. Do you offer a trial? + Yes, a 7-day trial is available for both the Professional and Plus plans. What is your cancellation policy? + If you aren't satisfied with the service, we offer a 100% satisfaction guarantee. You can cancel your plan at any time. Is there a limit to how many questions I can ask per day? + To keep our monthly prices low, we have monitors and limits in place to avoid the use of bots or account sharing. However, normal human use of our system, including as a medical AI chatbot, does not trigger these limits. I want to use BastionGPT in my app. Is there an API? + Yes, select customers can add BastionGPT to their apps, provided the use case meets our ethics, safety and compliance standards. This helps power any number of safe use cases, such as medical AI chatbots. For more information, view our BastionGPT API page. Do you offer a referral bonus? + The tremendous growth and success of BastionGPT, the leading healthcare chatbot, wouldn't be possible without our customers telling their colleagues about BastionGPT. We do have affiliate programs, where customers can receive credits or commissions for referring other customers. For more information about the program, please visit our Affiliate Program page. Can people who are not medical professionals use BastionGPT? + BastionGPT has been tailored to best meet the needs of medical professionals, including therapists and healthcare administrators. However, the tool can be used for other purposes. Because generative AI, such as BastionGPT, can occasionally make mistakes, it is not safe for someone who is not medically trained to obtain and use medical information without proper training or consultation from a medical professional. What file types are supported for upload? + Text can be copied or pasted from any filetype. However, direct document uploads are available for PDF, TXT and Word documents. Professional Plus or Enterprise subscribers can also upload images (JPG, GIF, TIF) and additional file formats (Excel, CSV, PowerPoint, and HTML). Is BastionGPT endorsed by the APA? + The American Psychological Association (APA) has taken a position that they do not endorse specific companies or products. However, BastionGPT does comply with their guidance for use of AI, and has been reviewed by the APA across a number of key areas including adaption to psychological practice, safety, bias, security, and expertise. The result of that review was approval to exhibit at the 2025 APA Conference. For more information, please reference our APA Guidance on AI Tools article. How do I know BastionGPT is secure? + Our solution is built and maintained by experienced cybersecurity and healthcare professionals that have expert knowledge of healthcare regulations and decades of experience in maintaining HIPAA secure systems. The service is regularly scanned for vulnerabilities and assessed to ensure HIPAA compliance. Security information is available on our Security page, and additional cybersecurity technology documentation and architecture diagrams are available upon request. Additional information about our partnership with Microsoft can be found on their website or via Press Release . Feel free to schedule a call or message our team to speak with a healthcare compliance expert. What if I need a custom HIPAA agreement? + Our Terms provide for a standard HIPAA Business Associate Agreement for Covered Entities and Business Associates, along with the associated controls to ensure the safety of healthcare data. If you need this agreement signed via DocuSign instead, please contact our legal team. There is no additional cost. Custom ("redline") HIPAA agreements, however, are only available within the "Enterprise" licensing option. Is BastionGPT FERPA compliant? + Yes, this is available for all licensing options. If an additional FERPA addendum is needed, beyond what is included in the Terms of Use, please contact our team or request this from any member of our staff. Does AI ever make mistakes? + Yes, while AI is incredibly intelligent, and is getting better every day, it is not perfect. Users need to be aware of this, and carefully review answers from any AI, including BastionGPT. For more information about the various types of errors that AI can make, please review our Healthcare AI Use Case guide or contact us . What AI powers BastionGPT? + Multiple AI models power BastionGPT. BastionGPT is optimized to use the most effective AI model based on your query. This often includes licensed versions of the latest GPT, Claude, and Gemini models. BastionGPT has been tailored for healthcare users in many ways, such as minimizing the influence of pseudo-science and encouraging the use of evidence-based healthcare practice. What is the difference between this and free options? + Free AI services often use low-performing AI models. Most customers, after experiencing the capability and intelligence of BastionGPT with the latest and most powerful AI models, can no longer go back to using other AI models. As a cutting-edge medical GPT solution, BastionGPT has been specifically designed to meet the needs of healthcare professionals. Additionally, free options often share your chat data, are not compliant with healthcare regulations, and are not designed for healthcare professionals. BastionGPT has also been optimized for hundreds of different healthcare and therapy uses, which allows our service to automatically adjust the AI model to give you the highest quality response. See how BastionGPT compares to other AI scribes. Can I use BastionGPT for more than healthcare-specific questions? + Yes, BastionGPT provides access to the power of ChatGPT, and acts as a powerful assistant for every type of healthcare workforce member, whether that use is troubleshooting computer code, summarizing meeting notes, making suggestions to improve emails, or suggesting updates to company policy. Is BastionGPT compliant with healthcare regulations in my country? + In prioritizing the privacy and security of health information, BastionGPT adheres to a robust framework designed to align with major global healthcare standards. We endeavor to meet key industry standards, such as HIPAA in the United States, PIPEDA in Canada and the Privacy Act/APP in Australia, ensuring the protection and confidentiality of health information. However, compliance can vary by region due to differing local laws and regulations. We provide documentation and resources for our services, and we recommend consulting those materials or contacting our support to verify compliance with the healthcare regulations specific to your region and situation. If your facility uses product-specific consent forms, those can be found in our Consent Form page. Is there training documentation? + Yes, BastionGPT will show a brief walkthrough of features the first time you log in on a new device. For a more detailed overview, check out our Quick Start Video Guide . If you are interested in an interactive "Lunch and Learn" session, you can also reach out to our support team for upcoming dates and times. You can find popular use cases in our Use Cases page, or via our community Prompts for Health site. What if I have a question that isn't answered here? + We have additional frequently asked questions on our Support site , or we are happy to chat via our chat system, email or a live meeting. The evidence ## Transforming healthcare 90 Average minutes saved per clinician, per day Internal survey of BastionGPT users, 2024 (n=412). 99.3% Accuracy on clinical documentation tasks Peer-reviewed study, Baylor College of Medicine & Texas Children's Hospital, Critical Care Explorations (2026). See Research 28× Fewer errors in insurance claim letters vs. human-only drafting Controlled comparison study, 2024. See Research 60 Seconds to register and begin using BastionGPT Median time-to-first-message across new accounts, Q1 2026. Get started ## Experience the power of BastionGPT You won't be charged today. Cancel anytime, and enjoy a 45-day money-back guarantee. Start your 7-day trial --- > Source: https://bastiongpt.com/security Security · BastionGPT # Built with security as our first priority. BastionGPT was designed from day one for healthcare. HIPAA-compliant, BAA included, independently tested, and continuously assessed by a team of healthcare security veterans. See our security practice Read security FAQs HIPAA Compliant 3rd Party Certified NIST CSF Our Security Practice ## Healthcare security is complicated. ## BastionGPT makes it easy. We invest in security the way the regulated industries we serve expect: continuously, independently, and on a published cadence. Here's what that looks like in practice. ### Our approach BastionGPT is built and operated by a team that has spent decades inside healthcare. Our CEO previously led technology and security for one of the largest medical centers in the country, and that experience shapes every decision we make. We treat HIPAA compliance, encryption (in transit and at rest), role-based access, and audit logging as defaults, not enterprise upsells. Every customer is entered into a signed Business Associate Agreement on every paid plan, and your data is never used to train any underlying AI model. For a plain-language walkthrough of what happens to your data, step by step , see our support center. ### Risk assessment as investment, not expense Most companies treat third-party risk assessments as a regulatory tax. We treat them as one of the most valuable signals we get. BastionGPT engages outside security firms on a recurring basis to evaluate our environment against the HIPAA Security Rule, the HHS recognized security practices, and broader cyber frameworks like NIST CSF and NIST 800-53. The findings get tracked, prioritized, and remediated in the same systems our engineers ship product changes through, so security work compounds instead of getting filed away. ### Penetration tests and code reviews BastionGPT is independently penetration tested at least annually by a third-party offensive-security firm, with additional targeted tests around major platform changes. Tests cover the web application, API surface, authentication and tenant isolation, and the AI inference pipeline itself, including prompt-injection and data-exfiltration scenarios specific to LLM applications. We also run static and dynamic code analysis on every change, and our codebase is reviewed by senior engineers before anything reaches production. ### Data assessment and controls We follow a data minimization model. BastionGPT collects only what's necessary to provide the service, segregates customer data by tenant, and encrypts it both in transit (TLS 1.2+) and at rest (AES-256). Customer prompts and outputs are not retained for model training, and data residency for protected health information sits within hardened, BAA-covered infrastructure. Our support article Does BastionGPT train AI on my data? explains these commitments in more detail. ### Safe by default, not by configuration Most AI platforms ship every feature they can build and leave it to each customer's compliance and security team to work out which ones are safe to enable, then to keep watching as defaults change. BastionGPT takes the opposite approach: the capabilities that most often put patient data at risk are not in the product at all. There is no web search, because even de-identification that is 99.9% effective before a query leaves our boundary is not 100%, and neither Google nor Bing offers a BAA covering search queries. There is no cross-conversational memory, because context that follows you from one conversation to the next can surface one patient's information inside another patient's chart, and most patient consent language never contemplated one patient's data shaping another's record. When a capability is genuinely needed, we build it into the service itself, or show you how to bring your own content into a session, so a risky integration never becomes the workaround. ### Security history BastionGPT has been in market for three years and serves more than 10,000 healthcare organizations. In that time we have had zero reportable breaches of protected health information. Every security event, no matter how minor, is logged and reviewed; material findings are disclosed to affected customers under our incident response policy and the timelines required by the HIPAA Breach Notification Rule. Trust & Compliance Center ## Everything your security team would want to see, already documented. Our BAA template, third-party penetration test summary, HIPAA Security Rule mapping, infrastructure attestations from our SOC 2 and HITRUST-certified infrastructure providers, and architecture overview all live in one continuously-updated trust center, available under NDA when an enterprise security review needs it. Our full Terms of Use , Privacy Policy , Sub-Processor List and HIPAA BAA are published openly. Frequently Asked ## Security & Compliance FAQs The questions enterprise security, privacy, and procurement teams ask us most often. Is the system HIPAA- and 42 CFR Part 2–compliant? + Yes, BastionGPT is engineered to support the requirements of both HIPAA and 42 CFR Part 2 for the handling of sensitive health information. Will the vendor sign a Business Associate Agreement (BAA)? + Yes, all of our plans automatically incorporate a BAA. You can review our BAA and related terms here . Where preferred, BAA's can be routed via Docusign for no additional cost. What is the recommended procedure to spread BAAs? + For most customers, the standard BastionGPT BAA executed automatically during your electronic acceptance of Terms at the start of a trial covers all users in that organization, including additional users added later. For multi-entity organizations (e.g., a parent organization with several legal subsidiaries), we generally recommend executing a separate BAA per legal entity. Our support team can help you scope this if you would like assistance. HIPAA BAA can be routed for mutual signature via DocuSign by request. Are all recordings and transcripts encrypted? + Yes, sensitive data such as recordings and transcripts are encrypted using industry standard security measures such as TLS 1.2+) for data in transit and AES-256 for data at rest. Please note there are manual options to copy or download data to plaintext formats such as .mp4 and .txt files. Is my data used for AI training? + Data entered into BastionGPT is never sold or used for training AI models. BastionGPT makes no claim to any data that you provide. This is contractually guaranteed in our privacy policy and terms. Does BastionGPT use web search or third-party integrations? + No. BastionGPT does not include web search, browsing, plugins, or cross-conversational memory. Even 99.9% effective de-identification is not 100%, and search providers such as Google and Bing do not offer BAAs covering search queries, so patient data would leave BAA-covered infrastructure the moment a query went out. Capabilities our customers need are built directly into the service instead, and you can always add your own reference content to a session or obtain approval for API access to add in additional capabilities. Is it safe and compliant to enter Protected Health Information (PHI)? + Users have the ability to enter PHI into the system, where it will be transmitted and stored appropriately in line with HIPAA requirements. For practical guidance, read Is it safe to upload patient records? in our support center. Can data be deleted immediately after use? + BastionGPT provides the option to immediately delete specific items or all data at once. Data is maintained for up to 30 days in our secure audit vault to support required auditing, such as monitoring for illegal or prohibited system use. Does it support MFA and role-based access? + Yes, our Enterprise plans support the use of your organization's existing Single Sign-On (SSO) and authentication platform. All other plans support email-based MFA with adaptive authentication. Is data stored on U.S.-based servers? + Yes, customers with a USA billing address will store and process sensitive data within the USA. Sensitive data from customers with billing addresses in Canada or Australia will default to residing in their respective country. Are audit logs available for all access and actions? + Audit logs are maintained and monitored internally by our security team and can be requested ad-hoc as needed by the customer. Frequent audit log requests may incur administrative fees. Customers utilizing the Enterprise plans have access to automated data feeds if desired. Has the system undergone a recent SOC 2 Type II or penetration test? + Yes, BastionGPT routinely performs internal and external (3rd party) penetration tests to validate our systems security. BastionGPT exclusively operates on HITRUST CSF Certified and SOC 2 Type II attested infrastructure for all services processing, storing, or transmitting sensitive or confidential customer data, including all data submitted through the BastionGPT API and user interface. Our infrastructure providers maintain current certifications with annual audits. Additionally, BastionGPT is pursuing its own HITRUST CSF Certification and SOC 2 Type II attestation. We are currently undergoing readiness activities, with no significant technical gaps observed to date. In the interim, we maintain comprehensive security controls aligned with HITRUST CSF and HIPAA requirements. Is there a documented incident response plan? + Yes, we maintain a documented incident response plan that is regularly reviewed and tested in accordance with industry best practices and compliance requirements. ## Review our confidential security whitepaper and architecture with a cybersecurity expert. 30 minutes with our cybersecurity advisor. We'll walk through how BastionGPT protects your patient data and answer any questions specific to your practice. Schedule a call → Proven Security ## Vetted by the people who lead security in healthcare. BastionGPT's security posture is signed off by leaders who have spent their careers running security and compliance programs at the largest health systems in the country. "BastionGPT was built using the same playbook I used to protect patient data at a major academic medical center. Defense in depth, recognized security practices, and an honest paper trail. We don't ship a control we can't evidence." Josh Spencer CEO, Bastion Intelligence · 13 years leading technology & security at UT Southwestern "During my testing process, I was impressed by the robustness and resilience of this application. BastionGPT is one of the most secure platforms I have tested." Abir Dhar Cybersecurity Expert · Independent Penetration Tester Keep evaluating ## Choosing an AI tool for your practice? Our plain-English guide covers what makes an AI HIPAA compliant : the BAA, the technical safeguards, and the data-use restrictions to verify before patient data touches any tool. Then see how BastionGPT compares with Claude Team & Enterprise , Microsoft Copilot , Heidi Health , Hathr AI , Abridge , Suki , Nabla , DAX Copilot , and Freed . We also compare BastionGPT with Ambience Healthcare , Commure , CompliantChatGPT , DeepCura , DeepScribe , Lyrebird Health , Mentalyc , and Twofold Health . For security researchers #### Responsible disclosure Found something we should know about? We welcome coordinated disclosure from the security research community and respond to every report. security.txt [email protected] --- > Source: https://bastiongpt.com/use-case Use Cases # AI built for every healthcare workflow From scribing patient encounters to drafting insurance appeals, see how teams across healthcare put BastionGPT to work. Find your specialty below, or read on for an overview. ## Find your role Each guide includes workflows, prompt templates, and real customer stories tailored to that specialty. ### Physicians & Providers Differential diagnoses, ICD-10 lookup, patient message drafting, and more. Explore → ### Mental Health Therapy notes, treatment planning, and session documentation in seconds. Explore → ### Rehab & Sports Medicine PT/OT progress notes, exercise plans, and functional outcome reporting. Explore → ### Billing & Administration Insurance appeals, prior auth, denial responses, and revenue cycle ops. Explore → ### Legal & Compliance HIPAA-compliant policy drafting, contract review, and audit prep. Explore → ### DME Companies Medical necessity letters, faxes, and supplier documentation. Explore → ### School Psychology IEP support, parent communications, and behavioral assessment drafts. Explore → ### Chiropractic Medicine SOAP notes, treatment plans, and patient education materials. Explore → ### Dentistry Treatment plans, insurance narratives, and patient post-op instructions. Explore → The overview ## Optimize your practice with BastionGPT Powerful tools that improve patient care. In the complex world of healthcare, BastionGPT streamlines workflows, enhances patient care, and bolsters health literacy. Our technology lets providers take a proactive role in delivering comprehensive, personalized care while maximizing efficiency, including AI medical scribes that reduce documentation time so you can focus on patients. Learn more and access the full guide. We've spoken with dozens of healthcare professionals to learn how they use BastionGPT to make their lives easier and provide superior patient care. The examples below show how BastionGPT helps staff improve patient safety, reduce burden, raise health literacy, and deliver better care. There are important limitations and risks inherent with AI, including biases, hallucinations, and outdated information . To learn more about safe AI use in healthcare, read our full Healthcare Use Case Guide. Download the Full Guide secure.bastiongpt.com + New chat Patient summary ICD-10 lookup SOAP note draft Templates Shared prompts Archived #### How can BastionGPT help you today? Ask anything: patient notes, codes, drafts… → Summarize visit Draft appeal Find ICD-10 Translate to plain English Documentation Patient communication Coding & billing Clinical decision Research #### Clinical checklist · 47-yo, chest pain Confirm onset, duration, character of pain Screen for radiation to arm/jaw Check for SOB, diaphoresis, nausea Review meds for interactions w/ nitrates Consider differential: MI, PE, dissection Order ECG, troponin, CXR 6 items · cross-checked against saved guidelines Safety ### Improving safety: your second set of eyes BastionGPT can serve as an extra set of eyes , vigilant against potential drug interactions, clinical-guideline conflicts, and evidence-based practice inconsistencies in intricate medical cases. By analyzing patient information and using its medical knowledge, it provides a list of questions to make sure no detail is overlooked, and it can swiftly help triage complex issues . Productivity ### Reducing staff burden: enhanced productivity BastionGPT is your efficient digital assistant. It fetches medical codes (ICD-10, J-Code, CPT) based on symptoms and history, extracts information from dense forms , and summarizes dictated notes , spotting key details about symptoms, diagnoses, and treatments. It streamlines administrative work by drafting documentation like approval letters, insurance appeals, and patient message replies. Your staff's workload lightens, so they can focus on what matters most: patient care. RV Ravi Vinder AE Alice E. CD Christopher D. Select user Add user Original "Patient presents with acute exacerbation of chronic obstructive pulmonary disease secondary to..." Plain language "Your COPD is flaring up right now. Your lungs are more inflamed than usual and you're having a harder time breathing." Spanish 5th-grade level Communication ### Empower your patients: boosting health literacy When it comes to communicating complex health concepts, BastionGPT shines. It expertly translates medical jargon into patient-friendly language , ensuring comprehension across cultures, education levels, and age groups. By empowering patients with knowledge, we're building a healthcare environment where informed decisions are the norm. Care delivery ### Delivering superior patient care: enhance communication and diagnosis BastionGPT is a companion designed to help providers deliver compassionate, personalized care. It can rewrite messages to enhance bedside manner, build rapport, and instill trust. It assists in developing support and counseling strategies for tricky or sensitive cases. And it helps physicians brainstorm potential differential diagnoses and next steps, supporting the delivery of optimal care. Differential diagnosis · 62F, fatigue + weight loss 1 Hypothyroidism TSH, free T4 indicated High 2 Type 2 diabetes HbA1c, fasting glucose Med 3 Occult malignancy Age-appropriate screening Low "Technology can be intimidating, but this AI chatbot is incredibly user-friendly. It's been a breeze to incorporate into my daily workflow, saving me countless hours." Catherine, RN Nurse Coordinator ### Access the Full Healthcare AI Use Case Guide Hear from our customers how you can save hundreds of hours per year. For step-by-step help putting these use cases into practice, see the Getting Started guide . Submitted! Check your email for the guide. Oops! Something went wrong while submitting the form. --- > Source: https://bastiongpt.com/ai-medical-scribe # AI Medical Scribe # Draft Effortless Notes in Minutes Capture the visit. Draft the note. Focus on your patients. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Spend Less Time Charting and More Time With Patients BastionGPT is an AI medical scribe built for healthcare professionals who want faster, more accurate clinical documentation without disrupting patient care. Record visits live or upload audio after appointments, and our AI automatically generates structured medical notes in the format you already use. Whether you're documenting primary care visits, behavioral health sessions, telehealth appointments, or specialty consultations, BastionGPT helps reduce administrative burden while keeping patient interactions human and uninterrupted. ## How It Works 1 ### Record Your Session Conduct your patient visit as normal. Upload audio files or capture your session live. You can keep using your current virtual meeting software of your choice including Zoom, Teams, or Google Meet. Support for meetings up to 240 minutes long and ten participants. No visible bot 240 min max Live or upload 2 ### Transcribe with BastionGPT Our service uses the latest AI engines to create dictated or ambient listening transcripts. BastionGPT recognizes medical terminology and offers best-in-class handling of accents. Multi-speaker recognition allows the AI to generate the highest quality documentation by differentiating speakers using their voice print, expertly notating what the practitioner said compared to a patient or a caregiver in your session. Process as many sessions as you need, with each transcription up to 4 hours long. Medical terminology Multi-speaker No limits 3 ### AI Generates Your Notes Our AI has been trained on thousands of medical and mental health documents, and will draft documents in your preferred format (SOAP, DAP, BIRP, and hundreds more). You can provide samples of your ideal notes so the AI can mimic your writing style, formatting and tone. Mirrors your style 100+ formats Any EHR ## What Sets BastionGPT Apart ### AI Medical Scribe Built for Clinical Work Unlike generic AI tools, BastionGPT was designed for healthcare documentation from day one. Our AI models recognize complex medical terminology, behavioral health language, treatment planning, medication discussions, and specialty-specific workflows. BastionGPT works across virtually any healthcare setting, including but not limited to: Primary care Mental health and therapy Internal medicine Pediatrics Psychiatry Telehealth Urgent care Specialty practices Our AI medical scribe understands the realities of clinical conversations — including sensitive topics that mainstream AI platforms often restrict or misinterpret. ### Ambient AI Scribing Without Interrupting Your Workflow BastionGPT works quietly in the background during patient visits. No visible recording bot joins your Zoom, Teams, or Google Meet sessions. You can: Record in-person appointments Capture telehealth visits Upload audio files after consultations Dictate notes on demand Process multi-provider discussions ### Works With Any EHR System BastionGPT is designed to fit into your existing workflow instead of forcing you to switch systems. Our AI medical scribe works alongside virtually any EHR or EMR platform because notes can be copied directly into your current documentation system. Compatible workflows include: Epic Cerner Athenahealth eClinicalWorks SimplePractice AdvancedMD Kareo Custom EHR environments Our support guide walks through getting scribe notes into your EHR step by step. ### Generate SOAP Notes and Clinical Documentation Automatically After transcription, BastionGPT transforms your session into polished medical documentation in seconds. Supported note types include: SOAP notes DAP notes BIRP notes H&P documentation Custom templates Intake assessments Progress notes Treatment plans Referral letters Insurance documentation You can even upload example notes, so BastionGPT learns your preferred writing style, formatting, and tone. ### Reduce Documentation Time by 50–70% Healthcare professionals spend hours every week on charting. BastionGPT helps reclaim that time with faster note completion and automated workflows. Providers using BastionGPT report: Less after-hours charting Faster clinical documentation Reduced administrative burnout Improved patient engagement More efficient telehealth workflows Many users save more than 90 minutes per day using our AI medical scribe platform. ### HIPAA-Compliant AI Medical Scribe Software Security matters when handling protected health information. BastionGPT is designed with healthcare compliance in mind and includes: HIPAA-compliant infrastructure Secure data handling Encrypted transcription workflow BAA support Enterprise-grade privacy protections Your patient data stays protected throughout the documentation process. ## AI Medical Scribe Features ### Flexible audio capture Generates documentation in familiar structures including SOAP, DAP, BIRP, progress notes, H&P, and more. Use standardized note formats or customize templates to match any specialty or documentation need with this medical scribe software ### Ambient listening and dictation Use ambient listening to capture natural conversations or narrate your sessions and have your dictations transformed into notes Supports sessions up to 240 minutes so full intakes, complex cases, and group visits are captured in one pass without interruptions ### Multi-speaker recognition Uses acoustic fingerprinting to distinguish between clinician, patient, and up to 10 speakers so dialogue is attributed to the right person. Generates cleaner and more accurate notes that separate clinicians findings from patients, caregivers, or family members. The AI virtual scribe capability keeps your documentation organized. ### Medical-grade transcription accuracy Recognizes medical and behavioral health terminology, abbreviations, and common acronyms across all specialties. Designed to handle diverse accents, different languages and speaking styles, providing a reliable scribe for doctors, therapists, and clinical teams. ### Note formats you already use Generates documentation in familiar structures including SOAP, DAP, BIRP, progress notes, H&P, and more. Use standardized note formats or customize templates to match any specialty or documentation need with this medical scribe software Flexible audio capture Ambient listening and dictation Multi-speaker recognition Medical-grade transcription accuracy Note formats you already use Record live visits or upload audio from any setting. Our AI medical scribe allows for in-person, telehealth, or phone consults. Compatible with Zoom, Teams, and Google Meet without a visible bot. Your patients see only real people, not a medical scribe software recording tool. Supports sessions up to 240 minutes so full intakes, complex cases, and group visits are captured in one pass without interruptions. Use ambient listening to capture natural conversations or narrate your sessions and have your dictations transformed into notes. Uses acoustic fingerprinting to distinguish between clinician, patient, and up to 10 speakers so dialogue is attributed to the right person. Generates clean and accurate notes that separate clinicians findings from patients, caregivers, or family members. The AI virtual scribe capability keeps your documentation organized. Recognizes medical and behavioral health terminology, abbreviations, and common acronyms across all specialties. Designed to handle diverse accents, different languages and speaking styles, providing a reliable scribe for doctors, therapists, and clinical teams. Generates documentation in familiar structures including SOAP, DAP, BIRP, progress notes, H&P, and more. Use standardized note formats or customize templates to match any specialty or documentation need with this medical scribe software. Recording virtual visits? Our support guide answers a common question: Does the scribe capture both telehealth sides? 99.3% Physician-level accuracy (Baylor/Texas Children's study) 28x Fewer errors in insurance claim letters 10,000+ Healthcare organizations across the U.S. 50-70% Reduction in documentation time per physician ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## More Affordable Than Other AI Medical Scribes Many AI scribe platforms charge $99–$300+ per month per provider. BastionGPT delivers enterprise-level AI medical scribe functionality starting at just $45/month with unlimited transcription. You get: Unlimited transcription Access to advanced AI models Multi-speaker recognition Medical note generation HIPAA-compliant workflows Support for long recordings No hidden usage caps ## Why Healthcare Professionals Choose BastionGPT # Built for healthcare professionals BastionGPT has been trained on thousands of documentation formats and we never ban accounts for use with sensitive clinical content. Trauma, sexuality, and substance use discussions that trigger filters on mainstream AI tools are recognized here as legitimate clinical work. # No awkward robot in virtual meetings BastionGPT’s AI virtual scribe captures sessions invisibly in the background. Compatible with all telehealth platforms including Zoom, Teams, and Google Meet, so no bot has to join your call. Your patients see only real people in their meeting. # Save 90+ minutes daily BastionGPT users reclaim an average of 90 minutes per day on documentation thanks to our powerful AI software and simple to use interface. More patient time. No evening documentation. # Best value on the market Other AI medical scribes charge $99+/month. BastionGPT is $45/month for unlimited transcriptions using the best AI models on the market. Improved quality, half the cost. See how BastionGPT compares to other AI scribes. ## Restoring the Human Element to Healthcare Rated 5 Stars on Google Business Profile 4.8/5 on Capterra Trusted by healthcare professionals "I can maintain eye contact and be fully present instead of switching between typing and glancing at the clients. It's restored the human component to my practice." Emma Levy Clinical Psychologist "We love that we found a tool that actually makes our work better, not just faster." Kiley Cuene Clinical Psychologist "BastionGPT has reshaped the way I practice medicine fundamentally improving how we work, communicate, and live." Dr. Adil Manzoor, DO, MBA/MS Internal Medicine & Pediatrics ## A Plan That Fits You ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently Asked Questions Can I use BastionGPT for telehealth appointments? Yes. BastionGPT works with Zoom, Microsoft Teams, Google Meet, and other virtual meeting platforms without requiring a visible bot participant. How accurate is the transcription? BastionGPT is optimized for medical terminology, clinical language, and accent handling across specialties. Multi-speaker recognition uses voice diarization to differentiate between provider, patient, and caregiver voices. Every transcription and note is a draft that you review and approve before it enters the record. What is the difference between this and free options? Free AI tools like ChatGPT and Google Gemini are not designed for clinical use. They typically lack a BAA, may use your input data for model training, and often filter or block sensitive clinical content like trauma, substance use, and sexuality discussions. BastionGPT is built for healthcare professionals with HIPAA-compliant infrastructure, medical-grade transcription, and AI models configured for clinical documentation. How accurate is BastionGPT’s AI medical scribe? BastionGPT uses advanced medical transcription AI trained to recognize clinical terminology, accents, and multi-speaker conversations for highly accurate documentation. Is BastionGPT HIPAA compliant? Yes. BastionGPT is built with HIPAA-compliant security and privacy protections for healthcare organizations handling patient information. How long can my recordings be? Each session can run up to 240 minutes, so full intakes, complex visits, and group sessions are captured in a single recording. You can record live or upload audio afterward, using ambient listening or dictation. Transcription is unlimited on every plan, so you can process as many sessions as you need. Does BastionGPT integrate with my EHR system? Yes. BastionGPT works alongside all major EHR systems, including Epic, Oracle Health, Allscripts, Meditech, and Practice Fusion. Copy your notes directly into your charting system or export files for upload. No complex IT setup required. Is my patient data secure? Yes. BastionGPT is designed to support HIPAA compliance for healthcare organizations. All data is encrypted in transit and at rest, a BAA is included with every plan, and no patient data is shared with third-party AI providers for training. Can I customize the note format? Yes. You can generate notes in SOAP, DAP, BIRP, H&P, admission, consult, or use a custom prompt to match your preferred structure, level of detail, and clinical style. You can also provide sample notes so the AI medical scribe mirrors your writing tone and formatting. How is BastionGPT different from other AI medical scribes? Most AI scribes focus on a single function like ambient listening or dictation. BastionGPT combines transcription, note generation, document analysis, and a general-purpose clinical AI assistant within one HIPAA-compliant platform. Plans start at $20/month with unlimited transcription, and a BAA is included with every subscription. You can see a detailed comparison of how BastionGPT compares to other AI scribe solutions on our comparison page. ### Experience the power of our AI scribe Start your 7-day FREE trial You won’t be charged today. Cancel anytime, and enjoy a 45-day money-back guarantee. --- > Source: https://bastiongpt.com/ai-medical-notes # AI Medical Notes # Built for Healthcare Professionals Spend less time charting and more time with your patients. Create accurate AI medical notes from any document or recording. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Smart AI for Medical Notes, Built for Clinical Teams BastionGPT is a HIPAA-compliant AI assistant and medical scribe built for physicians, nurses, and clinical teams. It converts dictations, recordings, and uploaded records into structured medical documentation, from SOAP notes and referral letters to discharge summaries and care plans. Medical notes AI handles the formatting and clinical structure so you can focus on patient care. A BAA is included with every subscription, and no patient data is used for AI model training. ## What AI Medical Notes Can Do ### Visit Notes and AI Progress Notes Generate structured SOAP or narrative AI progress notes from recordings, dictation, or pasted clinical data. ### EHR and Workflow Flexibility Upload dictations, transcripts, or records up to ~1000 pages. Works alongside Epic, Oracle Health, Allscripts, Meditech, and Practice Fusion. ### Consultation Letters and Referrals Draft professional referral summaries with history, key findings, lab highlights, and recommendations. ### Progress Tracking and Outcomes Produce AI for progress notes summaries across visits highlighting symptom changes, interventions, and outcomes. ### Assessment and Evaluation Reports Create detailed specialty reports formatted to meet documentation standards, from pre-operative evaluations to diagnostic workups. ### Discharge and Continuity Summaries Summarize encounters into handoff or discharge reports with diagnoses, medications, and follow-up instructions. ### Care Plans and Protocol Notes Draft targeted care plans that integrate diagnosis, interventions, medications, and timelines. ### Intake and Triage Overviews Condense lengthy intake forms into concise clinical snapshots with risk flags and treatment priorities. ### Administrative and Insurance Support Create prior authorization requests, coverage appeals, and continuity-of-care documents from your clinical references. ## How It Works 1 ### Type, Paste, Record, or Upload Your Data Type free text, paste your clinical notes, upload documents, or dictate your findings, and BastionGPT structures them into the note format you need. You can also record live or upload audio up to four hours, with speaker recognition supporting up to ten voices for rounds, group visits, and multi-provider encounters. Type Paste Dictate Upload documents Record live 2 #### Select Your Note Type Choose to summarize or generate a SOAP note or enter a custom prompt to create documentation in seconds. AI medical notes structures the output to match your selected format with the flexibility to match your style and structure. SOAP DAP BIRP Intake note Treatment plan Custom prompt 3 #### Refine and Export Review and share your personalized documentation to your EHR, billing platform, or secure documentation system. You control every word that enters the patient record. Review & edit Adjust clinical details Copy to any system ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Privacy, Security, and Clinical Control ### Built for Medical Privacy All data is encrypted in transit (HTTPS/TLS) and at rest (AES-256). BastionGPT runs on Microsoft Azure with ISO 27001-certified data centers and a HIPAA-compliant BAA. No patient data is shared with OpenAI or any other third-party AI provider for training. BastionGPT is built from the ground up as a secure AI for medical notes. Built on HITRUST and ISO 27001-certified data centers BAA included with every plan Data never used for AI training ### Clinician-Supervised Notes AI drafts the documentation. The provider reviews, edits, and signs off. BastionGPT supports your clinical judgment. ### Audit-Ready Infrastructure Time-stamped revisions, secure storage, and automatic data wiping after 30 days (with the option to delete sooner) support transparency for billing, supervision, and external audits. ## Keep Your Attention on Patients. Let AI Do the Notes. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Are AI medical notes HIPAA compliant? BastionGPT is designed to support HIPAA compliance for healthcare practices. All data is encrypted in transit and at rest, a BAA is included with every plan, and no patient data is used for AI model training. What types of medical notes can AI create? BastionGPT generates SOAP notes, H&P notes, progress notes, discharge summaries, visit summaries, referral letters, prior authorization requests, insurance appeal letters, patient education materials, and hundreds more. You can create any custom notes or document formats using your own prompts. How accurate are AI medical notes? BastionGPT drafts notes based on the clinical data you provide (transcripts, recordings, uploads, or pasted notes). The AI is configured to maximize clinical accuracy, but every note is a draft that you review and edit before it enters the record. The clinician is the final decision-maker. Who benefits most from AI medical notes? Physicians, nurse practitioners, physician assistants, nurses, and other healthcare providers who spend significant time on clinical documentation. BastionGPT is especially useful for clinicians managing high patient volumes, complex cases, or lengthy medical records. How does BastionGPT handle medical terminology and specialty-specific documentation? BastionGPT is built to work with clinical language across specialties, from primary care and urgent care to cardiology, orthopedics, and surgery. It recognizes specialized terminology, abbreviations, and medication names, and structures output to reflect your clinical reasoning rather than flattening it into generic language. You can also use custom prompts to tailor note structure and language to your specialty. Every draft is yours to review and adjust before it enters the record. Do I need to be tech-savvy to use BastionGPT? No. BastionGPT is designed for clinicians, not IT teams. You can start generating notes within minutes of signing up. The workflow is straightforward: capture your visit data (record, dictate, paste, or upload), select a note format, and review the draft. There is no complex setup and no software to install. Can I edit the AI-generated notes? Yes. Every note BastionGPT produces is a draft. You can review, revise, add clinical nuance, and adjust any section before copying it into your documentation system. Does it work with my EHR? BastionGPT works alongside all major EHR systems, including Epic, Cerner, athenahealth, eClinicalWorks, and NextGen. Integration is copy/paste or document upload based, so there is no complex IT setup required. What happens to patient data? All data is encrypted and stored within our HIPAA-compliant environment. Chat history, transcripts, and audio files are securely deleted after 30 days by default, and you can adjust retention policies as needed. No data is shared with OpenAI or any other third-party AI provider for training. Is there a limit on how many notes I can generate per month? There is no per-note cap on any plan. You can generate as many medical notes, referral letters, discharge summaries, and other documents as your patient volume requires. Audio transcription is also unlimited across all plans. The Professional plan supports reference documents up to approximately 30 pages; Professional Plus supports documents up to approximately 1,000 pages for clinicians working with longer records. What AI models power BastionGPT? BastionGPT integrates multiple leading AI models, including OpenAI's GPT, Anthropic's Claude, and Google's Gemini, within a single HIPAA-compliant interface, so you get access to the most capable AI available without managing separate subscriptions or worrying about compliance. The platform is regularly updated to include the latest models as they become available. All models run within BastionGPT's secure infrastructure, and no patient data is shared with third-party AI providers for training. How long does it take to generate a medical note? Most notes are generated in just a few minutes. Whether you're working from a live transcription, an uploaded recording, or pasted clinical notes, BastionGPT processes your input and returns a structured draft quickly. From there, you review, adjust clinical nuance, and copy it into your EHR. Clinicians report that the full workflow, from input to finished note, takes a fraction of the time manual documentation requires. ### Experience the power of our AI scribe Start your 7-day FREE trial You won’t be charged today. Cancel anytime, and enjoy a 45-day money-back guarantee. --- > Source: https://bastiongpt.com/ai-therapy-notes # AI Therapy Notes # Built for Mental Health Professionals Stop losing evenings to documentation. BastionGPT generates structured therapy notes from your sessions in minutes. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## AI-Powered Therapy Notes, Built for Mental Health Professionals BastionGPT is a HIPAA-compliant AI assistant and AI for therapists, psychologists, and counselors. It turns live sessions, dictated notes, and audio recordings into structured clinical documentation, so you can spend less time typing and more time with your clients. With a BAA included in every plan and zero use of PHI for model training, your client data stays private. ## What AI Therapy Notes Can Do ### Session & Progress Notes Generate structured progress notes (SOAP, DAP , BIRP) from session transcripts, voice recordings, or bullet-point summaries. ### Works With Live, Recorded, or Pasted Notes Upload recordings (up to four hours of transcript), paste EHR exports, or dictate live with multi-speaker recognition for up to 10 speakers. ### Group Therapy Summaries Summarize group sessions with structured headings: Attendees, Topics, Activities, Observations, and Plan. ### Evaluation and Assessment Reports Draft comprehensive neuropsychological evaluations and diagnostic assessments from test results, interview notes, and observations. ### Progress and Outcome Reports Generate narrative progress summaries across multiple sessions or full episodes of care for supervision, audits, or managed care reviews. ### Crisis Documentation and Safety Plans Create detailed crisis intervention reports and individualized safety plans with risk factors, protective factors, and action steps. ### Treatment Plans and Dynamic Updates Draft individualized treatment plans that synthesize diagnoses, client preferences, and evidence-based interventions. ### Intake Summaries and Triage Briefs Condense lengthy intake questionnaires and referral documents into concise clinical snapshots. ### Insurance and Care Coordination Draft prior authorization letters, referral letters, and insurance appeals formatted for payer requirements. ## How It Works 1 ### Capture clinical data your way Type or dictate a freeform narrative, paste existing notes, or upload documents, and BastionGPT structures them into the note format you need. You can also record live or upload audio up to four hours, with speaker recognition supporting up to ten voices for joint sessions and group work. Type Paste Dictate Document upload Record live 2 #### Choose your note type and template Select Summary, SOAP, DAP note, or generate hundreds of different therapy note formats. BastionGPT structures the output to match your selected format, with flexibility to fit your style. SOAP DAP BIRP Intake note Treatment plan Custom prompt 3 #### Review, edit, and export Easily edit and paste into your tools (TherapyNotes, TheraNest, SimplePractice, or any system you use). Review & edit Adjust clinical nuance Copy to any system ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Privacy, Security, and Clinical Control ### HIPAA-compliant environment All protected health information (PHI) stays within a secure, encrypted platform. BastionGPT runs exclusively on cloud infrastructure that is SOC 2 Type II attested, HITRUST CSF certified, and ISO 27001 certified, and a strong BAA is included with every subscription. No client data is shared with OpenAI or any third-party AI provider for training. BastionGPT is built from the ground up as a HIPAA compliant solution for the secure generation of AI therapy notes. Built on HITRUST and ISO 27001-certified data centers BAA included with every plan Data is never sold or used for AI training ### Clinician in the driver’s seat AI generates your drafts. You review, edit, and decide what goes into the clinical record. BastionGPT is a tool that supports your clinical judgment. ### Built for privacy Designed to meet or exceed mental health privacy regulations and best practices. Data always stays within your ownership. Chat history and transcripts can be manually deleted or automatically wiped according to your record retention requirements. ## Focus on Your Clients. We’ll Handle the Notes ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Are AI therapy notes HIPAA compliant? BastionGPT is designed to support HIPAA compliance for mental health practices. All data is encrypted in transit and at rest, a BAA is included with every plan, and no client data is used for AI model training. What types of therapy notes can AI create? BastionGPT generates SOAP notes, DAP notes, BIRP notes, intake summaries, crisis documentation, safety plans, group therapy summaries, treatment plans, insurance appeal letters, referral letters, progress reports, and hundreds more. You can create any custom notes or document formats using your own prompts. How accurate are therapy notes? BastionGPT drafts notes based on the clinical data you provide (transcripts, recordings, uploads, or pasted notes). The AI is configured to maximize clinical accuracy, but every note is a draft that you review and edit before it enters the record. The clinician is the final decision-maker. Who benefits most from AI therapy notes? Therapists, psychologists, counselors, clinical social workers, and other mental health professionals who spend significant time on session documentation. BastionGPT is especially useful for clinicians managing high caseloads, group sessions, or complex evaluations. How does BastionGPT handle clinical terminology and specialized therapeutic frameworks? BastionGPT is built to work with clinical language across therapeutic modalities, including CBT, DBT, ACT, psychodynamic, EMDR, and others. It recognizes specialized terminology and structures output to reflect your clinical reasoning rather than flattening it into generic language. You can also use custom prompts to tailor note structure and language to your specific approach. Every draft is yours to review and adjust before it enters the record. Do I need to be tech-savvy to use BastionGPT? No. BastionGPT is designed for clinicians, not IT teams. You can start generating notes within minutes of signing up. The workflow is straightforward: capture your session data (record, dictate, paste, or upload), select a note format, and review the draft. There is no complex setup and no software to install. Can I edit the AI-generated notes? Yes. Every note BastionGPT produces is a draft. You can review, revise, add clinical nuance, and adjust any section before copying it into your documentation system. Does it work with my EHR? BastionGPT works alongside all major EHR and practice management systems, including TherapyNotes, TheraNest, SimplePractice, Epic, and Cerner. Integration is copy/paste or document upload based, so there is no complex IT setup required. What happens to client data? All data is encrypted and stored within our HIPAA-compliant environment. Chat history, transcripts, and audio files are securely deleted after 30 days by default, and you can adjust retention policies as needed. No data is shared with OpenAI or any other third-party AI provider for training. Is there a limit on how many notes I can generate per month? There is no per-note cap on any plan. You can generate as many therapy notes, treatment plans, intake summaries, and other documents as your caseload requires. Audio transcription is also unlimited across all plans. The Professional plan supports reference documents up to approximately 30 pages; Professional Plus supports documents up to approximately 1,000 pages for clinicians working with longer evaluations or records. What AI models power BastionGPT? BastionGPT integrates multiple leading AI models, including OpenAI's GPT, Anthropic's Claude, and Google's Gemini, within a single HIPAA-compliant interface, so you get access to the most capable AI available without managing separate subscriptions or worrying about compliance. The platform is regularly updated to include the latest models as they become available. All models run within BastionGPT's secure infrastructure, and no client data is shared with third-party AI providers for training. How long does it take to generate a therapy note? Most notes are generated in just a few minutes. Whether you're working from a live transcription, an uploaded recording, or pasted session notes, BastionGPT processes your input and returns a structured draft quickly. From there, you review, adjust clinical nuance, and copy it into your documentation system. Clinicians report that the full workflow, from input to finished note, takes a fraction of the time manual documentation requires. ### Experience the power of our AI scribe Start your 7-day FREE trial You won’t be charged today. Cancel anytime, and enjoy a 45-day money-back guarantee. --- > Source: https://bastiongpt.com/medical-chatgpt # The medical ChatGPT built for healthcare compliance Consumer AI tools like ChatGPT are not designed for use with protected health information. BastionGPT gives healthcare professionals the same AI capabilities in a platform built for HIPAA compliance from day one, with a BAA included in every plan. ## Why healthcare professionals are searching for a medical ChatGPT ChatGPT and similar consumer AI tools have changed what clinicians expect from software. The ability to summarize a patient record, draft a referral letter, or generate a SOAP note in seconds is no longer a novelty. For many providers, it has become part of how they work. The problem is that standard consumer AI tools were not built for healthcare. They do not include a Business Associate Agreement (BAA), they are not architected to handle protected health information (PHI), and their terms of service do not support clinical use with identifiable patient data. Using them with PHI creates regulatory exposure that most practices cannot afford. BastionGPT is designed to close that gap. It brings the capabilities clinicians already expect from AI into a platform purpose-built for the compliance, privacy, and documentation standards of healthcare, tuned to deliver better results for clinical use cases than general-purpose tools can offer. ## Is ChatGPT HIPAA compliant? Standard consumer ChatGPT is not designed for HIPAA compliance. OpenAI does not offer a Business Associate Agreement (BAA) for consumer ChatGPT accounts, and its terms of service do not support use with protected health information. Healthcare professionals who use consumer ChatGPT with patient data may be creating a reportable HIPAA violation. While some enterprise AI plans from other vendors include BAAs, a BAA alone does not mean every feature in the product is covered. Many of these platforms have exceptions buried in their agreements where specific features fall outside BAA coverage, meaning an employee using those features with PHI could trigger a breach without realizing it. Because BastionGPT is built exclusively for healthcare, every feature in the platform is designed to be HIPAA compliant, with no carve-outs. A BAA is included with all BastionGPT plans. Customer data is not provided to OpenAI and is never shared with third-party AI providers for training. Where data is shared with other AI providers as part of delivering the service, it is done under HIPAA agreements with appropriate security controls in place. ## Is it safe to upload medical records to an AI tool? Uploading patient records or PHI to a consumer AI tool is not safe from a regulatory standpoint, regardless of how the tool handles the data technically. Without a signed BAA in place, any disclosure of PHI to a third-party service is a potential HIPAA violation. BastionGPT is designed to support the safe use of AI with clinical documents. The platform accepts uploaded documents including PDFs, Word files, and images for analysis and summarization, within a HIPAA-compliant environment backed by a BAA. Data is stored securely and wiped after 30 days by default, with the option to delete sooner. Healthcare professionals should always verify that any AI tool they use with patient data has an active BAA in place with their organization before uploading records or identifiable information. ## What can BastionGPT do that ChatGPT cannot? BastionGPT is built for clinical workflows that consumer AI tools are not designed to handle. Key capabilities include: Clinical documentation — Draft clinical documents such as SOAP notes, DAP notes, BIRP notes, progress notes, referral letters, discharge summaries, and hundreds of other formats — with language tuned for medical accuracy across. AI medical scribe — Transcribe appointments and generate structured clinical notes with multi-speaker recognition (up to 10 speakers). Output options include a Transcription (auto-labeled with speaker names based on voice prints), Summary, a range of progress note formats, and the ability to create a fully custom note based on your own requirements and writing style. Document upload and analysis — Summarize, analyze, and transform uploaded clinical documents including patient records, prior authorizations, and lab results. Professional Plus supports up to 1,000 pages. Healthcare-appropriate content filtering — Handles clinical topics that standard consumer AI tools sometimes block or refuse, including sensitive mental health, medication, and diagnostic discussions. BAA included at every tier — No enterprise contract or large seat minimum required. A BAA is included with all plans. ## Medical ChatGPT FAQs What is a medical ChatGPT? A medical ChatGPT refers to an AI assistant that functions like ChatGPT but is specifically built or configured for healthcare use. Unlike consumer AI tools, a medical ChatGPT is typically designed to handle protected health information (PHI), include a Business Associate Agreement (BAA), and support clinical documentation workflows. BastionGPT is a HIPAA-compliant AI assistant built for physicians, therapists, nurses, and other healthcare professionals . Is ChatGPT HIPAA compliant? Standard consumer ChatGPT is not HIPAA compliant. OpenAI does not provide a BAA for consumer accounts, and its terms of service do not permit use with protected health information. Healthcare professionals who need HIPAA-compliant AI should use a platform that includes a BAA and is architected for PHI handling, such as BastionGPT. Is there a HIPAA-compliant ChatGPT for healthcare? Yes. BastionGPT is a HIPAA-compliant AI assistant built for healthcare professionals. It includes a BAA with every plan, runs on HIPAA-compliant secure infrastructure, and is designed to support clinical documentation, AI scribing, and document analysis with protected health information. A BAA is included with all plans starting at $20/user/month. Please see our pricing below. Is it safe to upload medical records to ChatGPT? Uploading patient records or PHI to standard consumer AI tools like ChatGPT is not advisable from a HIPAA compliance standpoint. Without a signed BAA in place, sharing PHI with a third-party service may constitute a HIPAA violation. BastionGPT is designed to support the secure handling of clinical documents within a HIPAA-compliant environment, with a BAA included in all plans. Customer data is not provided to OpenAI and is never shared with third-party AI providers for training. Where data is shared with other providers as part of delivering the service, it is done under HIPAA agreements with appropriate security controls. Can doctors use ChatGPT for patient documentation? Physicians and other clinicians can use AI for patient documentation, but the tool must be used within a HIPAA-compliant framework that includes a BAA. Consumer ChatGPT does not meet this requirement. BastionGPT is built for clinical documentation, including SOAP notes, referral letters, progress notes, and AI medical scribe transcription, with HIPAA compliance and a BAA built in. How is BastionGPT different from ChatGPT for healthcare? BastionGPT is purpose-built for healthcare where consumer ChatGPT is a general-purpose tool. Key differences include: a BAA included with all plans; data isolation with no sharing with third-party AI providers for training; healthcare-appropriate content handling for clinical topics; trained on thousands of clinical documentation templates; AI medical scribe with multi-speaker transcription ; and document upload for records analysis. BastionGPT is trusted by 10,000+ health organizations and aligns with the APA's AI tool guidelines on safety and ethics, and has been reviewed by the APA across key areas, including adaptation to psychological practice, safety, bias, security, and expertise, earning a featured spot at the 2025 APA Conference. Does BastionGPT use ChatGPT? BastionGPT integrates multiple leading AI models within a unified HIPAA-compliant interface, including the latest GPT, Claude, and Gemini models. Unlike using consumer ChatGPT directly, BastionGPT routes all interactions through a HIPAA-compliant environment with a BAA in place and no use of customer data for model training. ## A Plan That Fits You ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ### Experience the power of our AI scribe Start your 7-day FREE trial You won’t be charged today. Cancel anytime, and enjoy a 45-day money-back guarantee. --- > Source: https://bastiongpt.com/which-is-the-best-ai-scribe Buyer’s guide · Updated September 2026 # Best AI Medical Scribes in 2026: Full Comparison (Pricing, Features & HIPAA) Ten AI medical scribe platforms compared on pricing, ratings, security score, HIPAA compliance and EHR compatibility. Every figure is sourced, dated and linked, so you can check our work. ## At a glance Sort any column to rank the field your way. Platform ▲ Monthly cost per user, from ▲ Rating avg across review sites ▲ Security score Mozilla scan, click to verify ▲ Free trial ▲ Inputs what you can upload ▲ BastionGPT Solo to multi-provider practices wanting scribe and full AI assistant in one HIPAA platform. $20–$45 Professional / Plus 4.84 /5 26 ratings B+ 80/100 secure.bastiongpt.com 7-day trial Audio + PDF, Word, Excel, PPT, images Freed Solo practitioners and small outpatient practices wanting fast setup and browser EHR push. $39–$119 Starter caps 40 notes/mo 4.80 /5 621 ratings B+ 80/100 secure.getfreed.ai 7-day trial Audio + .txt transcript Tali Canadian practices on Accuro, CHR, PS Suite or Oscar Pro needing PIPEDA and PHIPA. Free–$135 free tier, then Pro 4.60 /5 21 ratings F 0/100 app.tali.ai Free tier Audio only Dragon Copilot Large health systems already in the Microsoft and Epic ecosystem with IT resource. Rebranded from DAX Copilot in 2025. $160–$1,500 * via resellers 4.20 /5 88 ratings C 50/100 www.copilot.us.dragon.com 14-day trial Audio only Abridge Enterprise health systems on Epic needing published trial data for procurement. $208 * per provider 4.20 /5 70 ratings A+ 105/100 notes.abridge.com No Audio only DeepScribe High-documentation specialties needing specialty-tuned models and E&M coding hints. $350–$500 * per provider 4.20 /5 89 ratings D− 25/100 portal.deepscribe.ai No Audio only Heidi Health Multilingual practices, or clinicians wanting a free tier before committing. Free–$150 free tier, then Clinician 4.15 /5 3.2K ratings C− 45/100 scribe.heidihealth.com Free tier Audio + PDF, Word, images Nabla Enterprise environments needing deep EHR integration across 20+ systems. Free–$119 * free to 30 consults/mo 3.90 /5 8 ratings C 50/100 copilot.nabla.com Limited free tier Audio only Suki AI Practices wanting voice-command workflow automation beyond note generation. $299–$399 * Compose / Assistant 2.73 /5 102 ratings B 70/100 app.suki.ai No Audio only Ambience Healthcare Large health systems needing ambient AI with coding suggestions and HITRUST infrastructure. $233–$417 * per provider Not rated no public reviews B+ 80/100 app.ambiencehealthcare.com No Audio only * These vendors do not publish their pricing. Dragon Copilot, Suki AI, Abridge, DeepScribe, Ambience Healthcare and Nabla’s paid tier all route buyers to a sales conversation instead of listing a rate. The figures marked with an asterisk are estimates compiled from third-party reports, not quotes, and will vary by contract, region, seat count and reseller. Implementation fees and markups sit on top of whatever is eventually quoted. Every unmarked price links to the vendor’s own published pricing page. Sorted by rating on load. Every figure links to its source: prices to the vendor’s own page, ratings to our methodology page , security scores to a live scan you can re-run yourself. The security score measures HTTP response headers on the host where clinicians sign in; it is not a measure of how any vendor stores or handles patient data. Three of these ten score worse on the host they ask you to sign into than on their own marketing site: DeepScribe drops 45 points, Suki 10, Heidi 5. ## How much does an AI medical scribe cost in 2026? Published AI scribe pricing in this comparison runs from free tiers to more than $1,500 per provider per month. Most practices land in one of four bands. Price band Monthly cost per user Platforms Typical buyer Free tier $0, with usage caps Heidi Health, Tali, Nabla Trying the category, or very low visit volume Under $50 $20 to $45 BastionGPT (unlimited use, BAA included); Freed Starter at $39 with a 40-note monthly cap Solo practitioners and small practices Self-serve unlimited $79 to $150 Freed Core and Unlimited, Nabla*, Tali Pro, Heidi Health Clinician Small to mid-size outpatient practices Enterprise contract $160 to $1,500+* Dragon Copilot, Abridge, Suki AI, DeepScribe, Ambience Healthcare Health systems with IT resources and procurement cycles * Estimated. Enterprise scribes rarely publish a price. Five of the ten platforms here do not, and reported implementation fees and reseller markups push the all-in cost well above the base rate. The asterisked figures come from third-party reports, not vendor quotes. The lowest published price for an AI medical scribe with unlimited use and a signed BAA in this comparison is BastionGPT at $20 per user per month. Self-serve scribes without an assistant cluster between $79 and $150 once you need unlimited notes. Enterprise pricing is quoted per provider, usually on an annual contract, and the base rate is only part of the total cost. Every price in the At a glance table links to the vendor page it was read from. ## What an AI medical scribe does An AI medical scribe listens to the clinician and patient conversation, live or from a recording, transcribes it, and drafts a structured clinical note such as a SOAP, DAP, BIRP, progress or H&P note. The clinician reviews, edits and signs every note before it enters the medical record. Ambient scribes capture the visit in the background; dictation tools still need you to speak the note yourself. Modern scribes separate speakers, understand clinical terminology and format notes to your template. Most connect to the EHR through bidirectional integration, a browser push, or copy and paste, which is why “integrates with Epic” needs clarifying before you buy. A few platforms, BastionGPT among them, pair the scribe with a full AI assistant so the same subscription drafts referral letters, appeals and patient education from uploaded records. Physicians report one to two hours a day of after-hours charting, and documentation load is a leading driver of burnout. A scribe produces a near-complete draft immediately after the visit, so charting stops following providers home. Practices using AI scribes typically report charting time cut by half or more, more consistent notes across providers, and more eye contact in the room. None of the tools on this page replaces clinical judgment. The signing provider remains responsible for the final note. ## Full feature matrix Ten vendors across eighteen rows is more than anyone can read at once. Pick the platforms you are actually weighing and the tables narrow to just those. Compare: Select all 10 Reset to default 5 Swipe sideways to see more platforms, or pick fewer above. Feature BastionGPT Freed Heidi Health Nabla Dragon Copilot Suki AI Abridge DeepScribe Ambience Healthcare Tali Cost and access Monthly cost per user, per month $20–$45 Professional / Plus $39–$119 Starter / Core / Premier Free–$150 free tier, then Clinician Free–$119 * free to 30 consults/mo $160–$1,500 * via resellers $299–$399 * Compose / Assistant $208 * per provider $350–$500 * per provider $233–$417 * per provider Free–$135 free tier, then Pro Usage caps monthly limit on notes generated Uncapped unlimited scribe & assistant, both plans 40 notes/mo on $39 Starter · uncapped from $79 Core Uncapped incl. free tier Not published Not published Not published Not published Not published Not published 5 notes/mo free Starter 40/mo · uncapped on Pro Free trial test with real visits first 7-day trial 7-day trial Free tier Limited free tier 14-day trial No No No No Free tier Setup time to first usable note Minutes Minutes Same day Minutes Weeks Weeks Weeks Weeks Weeks Minutes Confidence signals User rating average across review sites 4.84 /5 26 ratings 4.80 /5 621 ratings 4.15 /5 3.2K ratings 3.90 /5 8 ratings 4.20 /5 88 ratings 2.73 /5 102 ratings 4.20 /5 70 ratings 4.20 /5 89 ratings Not rated no public reviews 4.60 /5 21 ratings Security score Mozilla scan, click to verify B+ 80/100 secure.bastiongpt.com B+ 80/100 secure.getfreed.ai C− 45/100 scribe.heidihealth.com C 50/100 copilot.nabla.com C 50/100 www.copilot.us.dragon.com B 70/100 app.suki.ai A+ 105/100 notes.abridge.com D− 25/100 portal.deepscribe.ai B+ 80/100 app.ambiencehealthcare.com F 0/100 app.tali.ai Uptime guarantee published availability commitment 99.90% Not published Not published Not published Not published Not published Not published Not published Not published Not published Swipe sideways to see more platforms, or pick fewer above. Feature BastionGPT Freed Heidi Health Nabla Dragon Copilot Suki AI Abridge DeepScribe Ambience Healthcare Tali Capability AI assistant beyond transcription ✓ Basic Basic – – Basic – – – Basic File upload + analysis read and summarize documents ✓ – Limited – – – – – – – Attachments supported file types beyond audio PDF, Word, Excel, PPT, images, audio Text transcript (.txt) PDF, Word, images, audio None None None None None None Audio only Multi-speaker recognition why true speaker recognition matters Yes (up to 10) Yes Yes Yes Yes Yes Yes Yes Yes Yes Customizable templates can clinicians make their own Unlimited Unlimited Unlimited Enterprise-managed Unlimited Unlimited Enterprise-managed Unlimited Enterprise-managed Limited Max recording length longest single session 240 min Not published Not published 180 min 75 min Per encounter Per encounter Per encounter Per encounter Per encounter Languages spoken languages supported 15+ 90+ 110+ 35 50+ 80+ 28+ 110+ Not published 28 Fit and compliance EHR compatibility how notes reach the chart All major EHRs (copy/paste, doc upload) Any browser-based EHR (via EHR Push) 35+ EHRs including Epic, athena, eCW 20+ EHRs including Epic, Oracle Health, athena 200+ EHRs via Dragon Medical One 8+ EHRs including Epic, Oracle Health, athena Epic (native via Haiku/Canto) 7+ EHRs including Epic, athena, eCW Epic, Oracle Health, athenahealth Canadian EMRs (Accuro, CHR, PS Suite, Oscar Pro) HIPAA compliant US baseline ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Additional compliance beyond HIPAA FERPA, PIPEDA, PHIPA, APP PIPEDA PIPEDA, PHIPA, APP, GDPR PIPEDA, GDPR Not published Not published Not published Not published HITRUST PIPEDA, PHIPA, SOC 2 Type II Customer support channels available Email, Phone, Chat, Video In-app, Email 24/7 Chat, Email Email, Web Email, Phone, Video Enterprise only Enterprise only Enterprise only Enterprise only Email, Onboarding * Estimated. Dragon Copilot, Suki AI, Abridge, DeepScribe, Ambience Healthcare and Nabla’s paid tier publish no pricing. Those figures come from third-party reports, not vendor quotes, and will vary by contract, region, seat count and reseller. Compiled from publicly available sources, August 2026. Features and prices change and may vary by plan, region or contract. Always confirm directly with each vendor before purchasing. What each column measures Monthly cost per user : the published price per clinician per month at the entry tier. Some vendors offer volume discounts or require annual contracts. Always confirm total cost with the vendor, including implementation fees. Rating : the mean of each platform’s scores across the review sites where it holds a published rating, with the review count shown beside it. The full per-site breakdown is on our methodology page . Security score : an MDN HTTP Observatory scan of the host clinicians sign in to. It measures HTTP response headers only, not how a vendor stores or handles patient data. Click any score to re-run the scan. Free trial : whether you can test the product with real patient visits before committing. Trials are the most reliable way to evaluate note quality, workflow fit and ease of use in your own clinical setting. AI assistant : whether the platform includes generative AI beyond transcription, such as summarizing uploaded documents, analyzing data, drafting referral letters, generating patient education materials, or answering clinical questions. File upload and analysis : whether you can upload PDFs, images, spreadsheets or Word documents for the AI to read, summarize or incorporate into its output. This matters for intake forms, records from other providers and scanned documents. Attachments supported : the file types the platform accepts beyond audio recordings. Multi-speaker recognition : the ability to distinguish clinician, patient and additional speakers during a recorded session, so the note attributes statements to the right person. Customizable templates : whether clinicians can create, modify and save their own note or prompt templates, or are limited to company-provided formats. Max recording length : the longest single session the scribe supports. Important for lengthy evaluations, psychological assessments and complex consultations. Languages : the number of spoken languages supported for transcription and note generation. EHR compatibility : how the scribe gets notes into your electronic health record, from bidirectional integration that writes into EHR fields, to browser-based push, to copy and paste. HIPAA and additional compliance : the regulatory frameworks the vendor supports. HIPAA is the baseline for US healthcare. PIPEDA and PHIPA cover Canadian privacy requirements, APP covers Australia, GDPR covers the EU, and FERPA protects student education records in the US. Customer support : the channels available when something goes wrong, including whether you can reach a human during business hours. ## The platforms in detail What each one is genuinely good at, and where it stops. ### BastionGPT Best for: Solo to multi-provider practices wanting scribe and full AI assistant in one HIPAA platform. Rating 4.84 /5 Security score B+ Max session 240 min Speakers 10 SOAP, DAP, BIRP, Detailed SOAP and Custom Note output from every recording Multi-speaker recognition for up to 10 participants using diarization and acoustic fingerprinting Session lengths up to 240 minutes; Custom Note tab accepts any free-form prompt Upload and analyze documents up to 700,000+ words, or 1,000+ pages, on Professional Plus BAA included with every subscription; AES-256 at rest, TLS in transit, Microsoft Azure Data never shared with OpenAI or any third-party AI provider for model training $20–$45 Professional / Plus Start your 7-day free trial See the BastionGPT AI scribe · Compare plans ### Freed Best for: Solo practitioners and small outpatient practices wanting fast setup and browser EHR push. Real-time capture with templates that learn provider preferences One-click push to browser-based EHRs via Chrome extension Pre-visit patient summaries and post-visit instructions No document upload, image analysis or general AI tasks beyond the encounter Starter plan caps 40 notes/month; unlimited notes need the $79 Core tier $39–$119 Starter caps 40 notes/mo 7-day trial ### Heidi Health Best for: Multilingual practices, or clinicians wanting a free tier before committing. 110+ languages for transcription and note generation Free tier with unlimited basic dictation “Ask Heidi” Q&A provides clinical information on demand No large-scale document upload or broad content generation Free–$150 free tier, then Clinician Free tier ### Nabla Best for: Enterprise environments needing deep EHR integration across 20+ systems. Real-time transcription with multi-speaker identification Built for enterprise EHR integration (Epic, Oracle Health, athenahealth, NextGen) Powers NextGen Ambient Assist and Greenway Health as white-label No file uploads; templates cannot be customized per user Free–$119* free to 30 consults/mo Limited free tier ### Dragon Copilot Best for: Large health systems already in the Microsoft and Epic ecosystem with IT resource. Rebranded from DAX Copilot in 2025. Tight integration with Epic and Meditech Backed by Microsoft’s ecosystem and infrastructure Session limit of 75 minutes, the shortest in this comparison No published price; implementation fees ($700 reported) and reseller markups sit on top $160–$1,500* via resellers 14-day trial ### Suki AI Best for: Practices wanting voice-command workflow automation beyond note generation. Voice commands for ordering labs, prescriptions, scheduling and referral letters Bidirectional integration with Epic, athena, Meditech, Oracle Health, eCW, Elation Chart-aware Q&A and patient summarization Needs administrative coordination and IT support to deploy $299–$399* Compose / Assistant No ### Abridge Best for: Enterprise health systems on Epic needing published trial data for procurement. “Linked Evidence” ties AI output back to the source conversation Native Epic integration (Haiku to Hyperdrive) Published clinical trial data on accuracy and documentation time Enterprise governance controls, SSO and usage analytics; IT-heavy deployment $208* per provider No ### DeepScribe Best for: High-documentation specialties needing specialty-tuned models and E&M coding hints. Specialty focus on oncology, cardiology, urology and orthopedics E&M coding suggestions embedded directly into draft notes Optional human-in-the-loop QA layer, which can add delivery delay Integrates with Athena, eClinicalWorks, Epic and AdvancedMD $350–$500* per provider No ### Ambience Healthcare Best for: Large health systems needing ambient AI with coding suggestions and HITRUST infrastructure. AI-powered coding (AutoCDI) suggests billing codes from the clinical note Specialty-tuned models across 20+ specialties HITRUST-certified cloud infrastructure Used by Cleveland Clinic, UCSF, Memorial Hermann and The Oncology Institute $233–$417* per provider No ### Tali Best for: Canadian practices on Accuro, CHR, PS Suite or Oscar Pro needing PIPEDA and PHIPA. Integrates with major Canadian EMRs (Accuro, CHR, PS Suite, Med Access, Oscar Pro) SOAP and DAP formats with ICD-10 code suggestions Built-in “Ask a medical question” clinical reference tool Free tier limited to 5 AI scribes per month Free–$135 free tier, then Pro Free tier * Estimated. These vendors publish no pricing; the figures come from third-party reports rather than vendor quotes and will vary by contract, region, seat count and reseller. Head-to-head comparisons. BastionGPT vs Abridge , Ambience Healthcare , Claude Team and Enterprise , Commure , CompliantChatGPT , Dragon Copilot , DeepCura , DeepScribe , Freed , Hathr AI , Heidi Health , Lyrebird Health , Mentalyc , Microsoft Copilot , Nabla , Suki AI , Twofold Health . ## How to choose There is no single best AI scribe. The right choice depends on your practice size, EHR environment, workflows, budget, and what you need the tool to do beyond transcription. #### 1 Accuracy and note quality Run real patient visits through anything you are considering. A scribe that saves five minutes but needs ten minutes of edits is a net negative. Ask for specialty-specific examples during demos. #### 2 Ease of setup Some tools are ready in minutes. Others need weeks of IT coordination, EHR configuration and training. Match deployment complexity to your team’s capacity. #### 3 EHR integration depth “Integrates with Epic” can mean bidirectional write-back or browser copy/paste. Clarify what the integration actually looks like in your EHR before committing. #### 4 Security and compliance Verify the vendor offers a BAA, encrypts data in transit and at rest, and does not use patient data for model training. Operating in Canada or Australia? Confirm PIPEDA/PHIPA or APP coverage. #### 5 Transparent pricing Look for per-user pricing with no hidden implementation fees. Five of the ten platforms here publish no price at all. Free trials give the most reliable read on real-world value. #### 6 Support accessibility When a note fails to generate during a packed clinic day you need a response within hours, not days. Check whether support is human, reachable, and open during your hours. #### 7 Breadth of capability If your needs go beyond visit notes, covering referral letters, appeals, patient education and chart summarization, a platform combining scribe and assistant can replace several subscriptions. Large-document support matters: 700,000+ words, or 1,000+ pages, in one session means a whole patient history without splitting it up. #### 8 Clinical content flexibility If you document trauma, substance use, abuse, sexual health or forensic evaluations, verify the AI does not filter legitimate medical language. Some platforms apply consumer-grade restrictions that interfere with clinical documentation. ## Frequently asked questions What is the best AI medical scribe in 2026? It depends on practice size and what you need beyond transcription. BastionGPT is the only platform here combining an AI scribe with a full generative AI assistant, document analysis and healthcare-appropriate content filtering in one subscription. The matrix above breaks down all ten. Are AI medical scribes HIPAA compliant? Every tool here is HIPAA compliant, but depth varies. Verify a signed BAA, encryption in transit and at rest, no use of patient data for model training, and a clear retention policy. BastionGPT also supports FERPA, PIPEDA, PHIPA and Australian Privacy Principles. What is the difference between an AI scribe and an AI assistant? A scribe converts a clinician-patient conversation into a structured note. An assistant handles broader work: summarizing uploads, writing reports, analyzing data, generating patient education, composing appeal letters. Most tools here are scribe-only. How do AI scribes handle patient consent and data privacy? Responsible use means informing patients and obtaining consent before recording. Ask how long recordings are retained, where data is stored, and whether it is used for training. BastionGPT retains transcripts for 30 days by default, with the option to delete sooner, runs on HIPAA-compliant Microsoft Azure, and never shares data with third parties. How much do AI medical scribes cost? Self-serve tools run from free tiers (Heidi, Nabla, Tali) to roughly $150 per clinician per month. BastionGPT starts at $20 with unlimited transcription and assistant usage. Enterprise scribes are harder to answer, because five of the ten platforms compared here publish no price at all . Abridge, Ambience, DeepScribe, Dragon Copilot and Suki all route buyers to a sales conversation. Third-party estimates for that group run from about $200 to $500 per provider per month, and implementation fees and reseller markups sit on top of whatever is quoted. Can I use an AI scribe for therapy notes and mental health documentation? Yes, but some platforms filter language around trauma, abuse, substance use or sexual health, which interferes with behavioral health documentation, forensic reports and crisis notes. BastionGPT’s content filtering is healthcare-appropriate, allowing the full scope of clinical language without false flags. It has been reviewed by the American Psychological Association and exhibited at APA 2025. Do I need to change my EHR to use an AI scribe? No. Most work with your existing EHR through copy/paste or browser workflows, with no IT project. BastionGPT is compatible with all major systems including Epic, Oracle Health, Allscripts, Meditech, Practice Fusion, TherapyNotes and TheraNest. Enterprise scribes offer deeper bidirectional integration but need weeks of setup. What should I look for when choosing a scribe for my practice? Five areas: note accuracy on your actual visits, setup effort (minutes vs weeks), EHR compatibility, security and data handling, and total cost including hidden fees. If your needs extend past visit notes, look for a platform bundling an assistant. Run a free trial with real clinical workflows first. ### See where BastionGPT lands for your practice 7-day free trial · no charge today · cancel anytime · 45-day money-back guarantee Start your 7-day free trial Compare plans How we compiled this Pricing : read from each vendor’s public pricing page in July 2026, and every price links to the page it was read from . Prices without an asterisk are the vendor’s own listed rate. Prices marked with an asterisk are estimates : Dragon Copilot, Suki AI, Abridge, DeepScribe, Ambience Healthcare and Nabla’s paid tier publish no rate at all and route buyers to a sales conversation, so those figures come from third-party reporting rather than a quote. They will vary by contract, region, seat count and reseller, and implementation fees and markups sit on top of whatever is eventually quoted. Prices are per user per month at the entry tier, before annual discounts. Ratings : mean of each platform’s scores across the review sites where it holds a published rating (Apple App Store, Capterra, G2, GetApp, Google, Google Play, Software Advice, Trustpilot, TrustRadius), read on 18 July 2026. Each site counts once and counts equally, so a site holding 3,000 reviews does not outweigh one holding 12. TrustRadius scores are converted from a 10-point to a 5-point scale. Ratings built on very few reviews move on a single opinion , which is why the review count is shown next to every score. Tali’s figure was read on 26 July 2026, and BastionGPT’s figures were re-read on 23 September 2026 after new G2 and Capterra reviews. Every rating links to our methodology page , which shows the full rule set and a per-site breakdown of every score and review count behind each average. Security score : MDN HTTP Observatory scan of the host where clinicians actually log in, run 26 July 2026 and re-checked 16 August 2026. We grade the sign-in host rather than the marketing site because marketing sites carry analytics and ad tags that break Content Security Policy and frame rules, so grading them penalizes a company for its tag manager rather than its product. Where a vendor does not publish its app hostname we resolved the conventional one and confirmed it serves a real sign-in app before scoring it. The exact hostname scored is printed beneath every score. The score reflects HTTP response headers only : HSTS, Content Security Policy, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, cookie and subresource settings. It is not a measure of how any vendor stores, encrypts or handles patient data , it does not cover TLS configuration or application security, and it is not a HIPAA assessment. A vendor can hold a weak score here and still run a well-secured clinical application, and an A+ proves nothing about the product behind it. It is included for one reason: it is among the very few security signals a buyer can verify independently, in a browser, without taking any vendor’s word for it. Click any score to re-run that scan yourself. Features : compiled from publicly available vendor documentation, July 2026; the Languages figures were re-read on 15 August 2026. Where a vendor documents nothing we record it as “Not published”, which is different from absent; enterprise contracts routinely include capabilities that never appear on a public page. Corrections : spot an error? Tell us and we will fix it and date the change. --- > Source: https://bastiongpt.com/api-signup Developers · API # Generative AI API for Healthcare. Outperforms general-purpose GPT, Claude, and Gemini models for healthcare applications while maintaining full HIPAA compliance. Purpose-built for therapy, clinical, and healthcare workflows. 30-day free developer trial 99.9% uptime guarantee 500K token context window Start your free trial → Schedule a consultation POST /v1/ChatCompletion 200 OK // HIPAA-compliant chat completion POST https://api.bastiongpt.com/ v1 /ChatCompletion Content-Type : application/json Key : ab123456789012345678901234567890 { "messages" : [{ "role" : "user" , "content" : "What is a normal BP?" }], "max_tokens" : 1000 , "temperature" : 0 } 100 + Deployed API customers 500 K Token context window 2 M /min Token throughput 99.9 % Uptime SLA About the API ## Build healthcare applications on the same engine 10,000+ healthcare organizations use every day. This page outlines the necessary details to integrate with the BastionGPT API, including input parameters, output structure, pricing, limitations, and usage examples. To get started, sign up for BastionGPT Plus to explore our capabilities through the same API you'll receive access to. If you have additional questions or would like to apply for access, complete the access request form or schedule a consultation to review your use case and receive your API key. Key Benefits ## Frontier AI, with the controls healthcare actually requires. ### HIPAA-compliant by design Full compliance with healthcare regulations. Patient data stays secure with enterprise-grade encryption. Your data is never sold or used to train our AI models. Licensed for secure use with PHI and PII, with OpenAI having no access to your data. ### Superior healthcare performance Utilizes a pool of the leading AI models, fine-tuned for healthcare and mental health professionals. Expertly trained on thousands of clinical documentation standards and clinical workflows, such as SOAP notes, prior authorizations and appeal letters. ### Enterprise-ready infrastructure 99.9% uptime SLA, 800 requests/minute, 2M tokens/minute. Built to support the next generation of powerful healthcare applications, with multi-region deployment in US, Canada (Quebec), and Australia for data residency compliance. Features ## Everything you need to ship a clinical product. ### Chat completion One POST request in, clinical-grade text out. Multi-turn conversations, your own system prompt, and a consistent response format with token usage on every call. ### Streaming Server-Sent Events with OpenAI-compatible chunks, so responses render token by token in your UI and existing SSE client code just works. ### Document understanding Upload a PDF, image, or text file once and ask questions about it in any chat request. Documents are retained for 30 days after upload, scoped to your API key, and reusable across requests during that window: upload once, reference the id as often as you need, then upload again after it expires. Supported types: PDF, plain text, HTML, CSV, Markdown, PNG, JPEG, up to 10 MB. ### Audio transcription Submit a recording, get back a transcript with speaker labels and timestamps. Medical transcription at $0.73 / hour , built for clinical accuracy on names, medications, and dosages. Common formats accepted (WAV, MP3, M4A, OGG, WEBM); conversion happens server-side. ### Reduced content filtering Reduced consumer-style content filters so clinicians can discuss substance use, sexual health, end-of-life care, and other adult topics in a healthcare context, without dead ends. ### Multi-region deployment Available in US, Canada (Quebec), and Australia for data residency compliance. Pick the region that matches your patients and regulatory footprint. Microsoft Power Platform ### Use BastionGPT in Power Automate, Power Apps, and Copilot Studio BastionGPT is a certified connector in Microsoft Power Platform. Add BastionGPT to your flows, apps, and Copilot Studio agents without writing code: create a connection with your API key, then choose an action. Calls run on your BastionGPT API subscription at the same rates as direct API requests. Ask a question Create a chat completion Upload a document Submit audio for transcription Get a transcript View the BastionGPT connector on Microsoft Learn . Microsoft lists it as a Premium connector, so your Microsoft plan must include premium connectors. Pricing ## Simple, predictable pricing. Start with a flat monthly subscription that includes credits, then pay-as-you-go beyond. No enterprise minimums. Developer Plan BastionGPT API $45 / month Includes $45 in monthly token credits, then pay-as-you-go for any usage beyond that. Included with subscription ✓ 2M tokens / minute rate limit ✓ 800 requests / minute ✓ 500K token context window ✓ 99.9% uptime SLA guarantee ✓ Access to the BastionGPT user interface ✓ 30-day free developer trial Pay-as-you-go rates beyond credits ✓ Input $0.018 per 1K tokens (~770 words) ✓ Output $0.045 per 1K tokens (~770 words) ✓ Audio $0.73 per hour ✓ Billed monthly, transparent invoicing Start free trial → On the roadmap Coming soon New capabilities In active development ○ Q4 2026 Image generation Talk to us about your use case Get Started ## From signup to your first API call in 4 simple steps. 01 ### Sign up for BastionGPT Professional Plus Explore our capabilities over your 30-day developer trial through the same UI powered by the API you'll receive access to. 02 ### Submit your use case for review Schedule a 15-minute consultation or complete the access form below. Our technical and safety teams will review to confirm eligibility and ask any follow-up questions. 03 ### Receive your API key Get your key and full documentation shortly after the meeting, typically the same business day. 04 ### Start building Make your first API call. Our team is on chat, email, or video whenever you'd like a hand. Schedule a consultation → Or complete the access form For a full breakdown of costs before you request access, see API pricing and billing explained . For an overview of what the API can do and common questions, see API capabilities and FAQ . API Reference ## API URLs and example usage. One base URL, an OpenAI-style request shape, and a consistent response format across chat, documents, and transcription. If you've integrated with mainstream LLM APIs before, you'll feel right at home. ### API Endpoint One base URL for every request. Chat completions post to this endpoint. POST Endpoint Copy https : //api.bastiongpt.com/v1/ChatCompletion ### HTTPS Request Example Send messages, get a structured chat completion back. The messages array and max_tokens are required; temperature is optional and defaults to 0. REQUEST application/json Copy POST /v1/ChatCompletion HTTP/ 1.1 Host : api.bastiongpt.com Content-Type : application/json Key : ab123456789012345678901234567890 { "messages" : [{ "role" : "user" , "content" : "What is a normal BP?" }], "max_tokens" : 1000 , "temperature" : 0 } ### HTTPS Response Example Familiar choices/usage shape, with camelCase field names and token usage on every call. 200 OK application/json { "choices" : [ { "finishReason" : "stop" , "message" : { "content" : "A typical blood pressure range for a healthy adult is around 90/60 mmHg to 120/80 mmHg." } } ], "id" : "f1ad1c1a-4713-4d09-bcbe-7784e7434f2a" , "created" : 1784062984 , "usage" : { "completionTokens" : 26 , "promptTokens" : 65 , "totalTokens" : 91 } } Complete reference documentation, a Postman collection, and an OpenAPI specification are provided with your API key. Request API Access ## Tell us about your use case. Submit the form and our technical and safety teams will review for eligibility. We typically respond within one business day. Prefer a live conversation? Schedule a 15-minute consultation instead. Fast turnaround. Most reviews completed within one business day. Confidential. Submissions are reviewed only by our internal technical and safety teams. No commitment. A submission is a request for access, not a purchase. Thank you! Your submission has been received! Oops! Something went wrong while submitting the form. Enterprise-grade security ## How your API connections are protected. We prioritize the security and integrity of your data. Our API infrastructure is built with robust security measures, ensuring every interaction is protected using industry best practices. #### Top-tier encryption Data in transit is protected by the latest, most secure encryption protocols. We exclusively support TLS 1.3 and TLS 1.2 , disabling outdated and vulnerable protocols like SSLv3, TLS 1.0, and TLS 1.1. #### Advanced cipher suites Strong, authenticated cipher suites including AES-GCM (up to 256-bit) for high performance and the highest level of confidentiality and integrity for your API calls. #### Forward secrecy ECDHE key exchange ensures that even if a server's long-term private key were ever compromised, past communication sessions remain secure and cannot be decrypted. #### Trusted & verified identity Our API endpoint uses a certificate from GeoTrust / DigiCert , signed with a strong SHA-256 algorithm , verifying our server identity and establishing a trusted channel. #### Hardened against vulnerabilities Configuration is actively hardened against known SSL/TLS vulnerabilities including BEAST, POODLE, Heartbleed, and ROBOT . We continuously monitor and update against emerging threats. #### Independently verified Our endpoint configuration consistently achieves a top 'A' rating from independent SSL/TLS assessment tools. Ready to build ## Take the first step. Start your trial, request access, or talk to our team. Whichever path fits, we'll have you making your first call within days, not quarters. Trial BastionGPT Plus → Request API access Schedule a consultation --- > Source: https://bastiongpt.com/use-cases/physicians # For Physicians & Healthcare Providers # AI Built for the Realities of Modern Healthcare BastionGPT gives clinicians access to the most advanced AI models, all within a HIPAA-compliant platform built for clinical workflows. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Physicians Are Getting Their Time Back 50% of physicians workday is spent on the EHR and administrative tasks 84% of physicians using AI scribes reported a positive effect on patient interactions 10+ hrs saved by BastionGPT users weekly on clinical documentation You trained for years to care for patients, not to spend your evenings catching up on: After-hours charting that follows you home and pushes documentation into nights and weekends Prior authorizations and insurance denials that drain your time and ability to focus on patient care Referral letters, discharge summaries, and appeals stacking up faster than you can write them Chart reviews and record summaries needed for handoffs, consults, and multidisciplinary meetings ## BastionGPT Helps You Reclaim Your Practice BastionGPT addresses the documentation and administrative challenges that keep physicians focused on their computers and not their patients. Save up to 8+ hours per week by drafting progress notes, treatment plans, and referral letters in minutes instead of hours. Improve documentation quality with thorough, well-structured notes that capture clinical nuance and support proper coding. Reduce after-hours charting so you can finish your clinical day when you leave the office. Access multiple leading AI models (GPT, Claude, and Gemini) through a single, secure interface built for healthcare. Strengthen compliance confidence with HIPAA-compliant infrastructure, AES-256 encryption, and a BAA included on every plan. Adapt documentation to your clinical voice and professional style within every note, so output reads the way you write. ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ### Clinical ### Documentation ### Patient Communication & Education ### Administrative Support & Appeals ### AI Scribe # Clinical Notes & Documentation Turn session audio, dictation, or rough notes into structured, chart-ready documentation in minutes. Generate H&P, SOAP, discharge, or progress notes from session audio, dictation, notes, or typed bullet points using the AI Scribe or saved prompt templates. Draft referral letters and prior authorization requests with complete patient history, clinical context, and structured justification. Produce discharge summaries, care manager reports, and skilled nursing documentation from uploaded records. Summarize lengthy patient charts, specialist reports, or EHR histories into concise clinical overviews for case reviews, new provider handoffs, or multidisciplinary meetings. Review existing documentation for errors, inconsistencies, pronoun mistakes, or vague language before finalizing. # Patient Communication & Education Draft personalized patient materials without adding hours to your day. Create patient education handouts that explain diagnoses, medications, and treatment plans in plain language at the reading level you specify. Draft care instructions customized to each patient's condition, literacy level, and preferred language. Generate after-visit summaries that reinforce key discussion points, medication changes, and follow-up instructions for each patient. Write referral coordination letters and interdisciplinary team updates for specialists, social workers, home health agencies, or schools. Produce pre-visit intake summaries from uploaded patient records so you can walk into appointments with full context. # Administrative Support & Appeals Save and reuse prompts to handle the operational work that limits your patient capacity. Save and organize your most-used prompts into folders by workflow or specialty, then share standardized templates across your team. Generate insurance appeal letters and denial responses with structured clinical justification and supporting documentation references. Draft prior authorization requests that include diagnosis codes, treatment rationale, and relevant clinical guidelines. Create internal documentation such as case conference summaries, quality committee reports, and peer review notes. Draft professional correspondence to referring providers, insurance companies, attorneys, or hospital administrators in the appropriate tone and format. # AI Transcription & Scribe Record and transcribe patient encounters, then convert them into clinical documentation you review and finalize. Transcribe encounters up to 240 minutes and produce one-click SOAP, DAP, Detailed SOAP, or Summary notes from the AI Scribe's six output tabs. Use the Custom Note tab to generate any format from the transcript: narrative progress notes, specialty-specific documentation, or patient-friendly visit summaries. Review, edit, and regenerate all outputs directly within the scribe interface before filing. Record sessions with multi-speaker recognition (up to 10 speakers) with labels that separate clinician observations from patient and caregiver statements. Upload audio files from any device or recording app in MP3, WAV, M4A, or MP4 format, with no software installation required. ## Integrating BastionGPT into Your Medical Practice "The technology has also transformed my interactions with parents during assessments. Instead of constantly typing detailed notes while looking at a Word document, I can now look directly at them and have genuine conversations. I can maintain eye contact and be fully present instead of switching between typing and glancing at the clients. It's restored the human component to my practice.” -Emma Levy, LCP Clinical Psychologist “BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations. For physicians, educators, and healthcare leaders, integrating BastionGPT isn't just about staying ahead; it's about fundamentally improving how we work, communicate, and live.” Dr. Adil Manzoor, DO, MBA/MS Internal Medicine & Pediatrics ## Join Physicians Who've Reclaimed Their Time BastionGPT gives providers HIPAA-compliant access to the most advanced AI models in a single platform so you can use AI to transform how you care for your patients and for yourself. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Is BastionGPT HIPAA-compliant? Yes. BastionGPT is designed specifically to support HIPAA compliance and can be used with HIPAA-regulated data. A Business Associate Agreement (BAA) is included with all plans. Your data is encrypted with AES-256 at rest and TLS in transit, hosted on Microsoft Azure HIPAA-compliant infrastructure, and never shared with OpenAI or other third parties for training. Do I still need to review AI-generated documentation? Yes. BastionGPT is a clinical AI assistant, not a replacement for professional judgment. While the platform uses specially configured versions of leading AI models to produce high-quality drafts, AI can occasionally make errors. Clinician review of all AI-generated content is required before finalizing any record. What about hospital or health system IT security requirements? BastionGPT is built and maintained by experienced cybersecurity and healthcare professionals. The platform runs on Microsoft Azure HIPAA-compliant infrastructure, undergoes regular vulnerability scans and third-party penetration testing, and holds an A+ rating with SecurityScorecard. We are a trusted Microsoft Partner. Enterprise options include SSO, customizable security policies, and dedicated account management. Security documentation and architecture diagrams are available upon request. Does BastionGPT work for my medical specialty? Yes. BastionGPT supports providers across specialties, including internal medicine, pediatrics, psychiatry, surgery, family medicine, emergency medicine, and many others. Can doctors use ChatGPT for medical notes? Standard ChatGPT services do not include a BAA and are not designed for use with PHI. Entering patient information into these services creates a compliance risk. BastionGPT is built to support HIPAA compliance from day one: a BAA is included with every plan, your data is never shared with OpenAI, and you can work with PHI immediately. Learn more about our security . What types of clinical notes can BastionGPT generate? BastionGPT drafts all types of notes, including SOAP, DAP, BIRP, and custom-format notes from session audio, typed summaries, or uploaded documents. The Custom Note tab accepts any free-form prompt, so you can generate specialty-specific documentation, narrative-format notes, or patient-friendly summaries from a single session recording. # Learn more on How Doctors Can Leverage AI Healthcare Innovation ## 30 Ways To Use Generative AI in Healthcare 30 Ways To Use Generative AI in Healthcare Healthcare Innovation ## Top Use Cases for Generative AI in Healthcare Top Use Cases for Generative AI in Healthcare BastionGPT is a clinical AI assistant and AI scribe designed to support documentation, analysis, and administrative tasks. All AI-generated content should be reviewed and approved by a licensed clinician before use in any client record. --- > Source: https://bastiongpt.com/use-cases/therapists-mental-health # For Therapists & Mental Health Providers # AI That Supports Your Clinical Work BastionGPT provides secure, healthcare-focused AI tools designed to streamline the specific documentation needs of mental health professionals. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## The Real Challenges Therapists Face Today 40% of providers' working hours are consumed by administrative tasks 86% of therapists agree excessive documentation contributes to burnout 50% of time spent writing evaluations and clinical notes You became a therapist to help people, not to: Lose hours to progress notes, intake reports, and evaluation write-ups after every session Face compliance pressure with no HIPAA-ready AI tools, no BAA, and no safeguards for PHI Grind through repetitive admin (referral letters, prior auths, appeals) that limits your caseload Hit content filters on standard AI platforms that block clinical discussions of trauma and abuse ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## BastionGPT Helps You Reclaim Your Practice BastionGPT directly addresses the most common operational challenges in modern mental health practices. With BastionGPT, providers can: Reclaim 8 or more hours per week by drafting documentation including, progress notes, treatment plans, and session summaries. Meet compliance requirements with HIPAA-secure infrastructure and end-to-end encryption. Reduce burnout by offloading repetitive administrative tasks by generating referral letters and insurance denial responses in a fraction of the usual time. Access top AI models including GPT, Claude, and Gemini for analysis, validation, and content creation. Improve documentation quality with consistent, thorough notes that capture clinical nuance. Discuss sensitive topics like sexual assault and violence that are often blocked by standard AI tools. ## How to Integrate BastionGPT in Your Workflow ### Clinical ### Documentation ### AI Scribe ### Planning ### & Communication ### Administrative Support # Clinical Documentation Turn session audio, notes, attachments, or bullet points into structured documentation in minutes. Generate SOAP, DAP, or BIRP notes from session audio, recordings or typed notes using the AI Scribe or saved prompt templates Draft evaluation reports (neuropsychological, psychoeducational, forensic) by uploading test data and clinical observations Create intake summaries and discharge reports from uploaded records Summarize lengthy patient records and EHR histories into concise clinical overviews for case reviews or new provider handoffs Review existing documentation for errors, inconsistencies, pronoun mistakes, or vague language before finalizing # AI Transcription & Scribe Record and transcribe therapy sessions, then convert them into clinical documentation you review and finalize. Record sessions with multi-speaker recognition and produce one-click SOAP, DAP, or Summary notes Use the Custom Note tab to generate any format from the transcript: narrative DAP notes, parent-friendly summaries, or specialty-specific records Review, edit, and regenerate all outputs before filing Transcribe sessions up to 240 minutes with speaker labels that separate clinician observations from client and caregiver statements Upload audio files from any device or recording app in MP3, WAV, M4A, or MP4 format, with no software installation required # Treatment Planning & Client Communication Build treatment plans, track progress, and create materials tailored to each client. Draft individualized treatment plans with measurable goals and intervention strategies from your clinical notes Generate psychoeducation handouts and parent-friendly report summaries at the reading level you specify Create session follow-up messages that reinforce key themes, homework, or coping strategies Update and revise treatment plans as clients progress by summarizing recent session themes and suggesting modified goals or interventions Create client-facing letters, referral summaries, or coordination notes for interdisciplinary teams including psychiatrists, social workers, and school staff # Administrative Support Save and reuse prompts to handle the busywork that limits your client capacity Save and organize your most-used prompts into folders by workflow, then share standardized templates across your team with one click Draft referral letters, treatment summaries, and professional reports in appropriate clinical language Generate insurance appeal letters and prior authorization requests with structured clinical justification Create internal documentation like case conference summaries and interdisciplinary team updates Draft professional correspondence to referring providers, schools, attorneys, or insurance companies in the appropriate tone and format Each chat is private by design, and BastionGPT still makes it simple to maintain continuity across sessions without stored memory . “The efficiency gains have been substantial enough that I can serve more people without sacrificing quality.” “The feedback has been positive; patients really like the tables and formatting that make complex clinical information accessible.” Anna Dinoto, LCP Forensic Psychologist “Bastion's allowed me to provide extra add-on services that are really meaningful.” Rachael St. Claire, PsyD Clinical Psychologist ## Start Saving Time with BastionGPT Record and transcribe therapy sessions, then convert them into clinical documentation you review and finalize. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently Asked Questions Is BastionGPT HIPAA-compliant? Yes, BastionGPT is engineered specifically for HIPAA compliance. We include a Business Associate Agreement (BAA) for all users and host data in secure, private environments. Your patient data will never be sold or used to train AI models. Does using AI for documentation affect the quality or personalized nature of my therapy notes? BastionGPT may actually improve documentation quality by providing structural consistency and reducing errors. You can use prompts to guide the assistant to mirror your unique clinical voice, professional style, and content structure. This results in personalized, high-quality notes that capture the specific nuances of every session. Can therapists use ChatGPT for session notes? ChatGPT is a general-purpose AI tool with no BAA for most users and no HIPAA safeguards. Entering patient information into ChatGPT creates a compliance risk. BastionGPT is designed to support HIPAA compliance, includes a BAA with every plan, and does not share your data for training AIs. You can use it with PHI from day one. Can BastionGPT replace my need for a human review? No, BastionGPT functions as a collaborative clinical AI assistant and AI scribe rather than a replacement for professional judgment. BastionGPT can make mistakes, so providers must review and approve every draft to ensure clinical validity and accuracy. This process ensures the final record reflects your expert observations and treatment decisions. Is BastionGPT suitable for all types of mental health providers (e.g., psychologists, LCSWs, psychiatrists)? Yes, the platform is built for a wide range of professionals, including psychiatrists, psychologists, and LCSWs. It handles the specialized requirements of different disciplines, from medication management to writing forensic psychology reports. Any licensed provider can use BastionGPT to manage their unique documentation burden. What types of therapy notes can BastionGPT generate? BastionGPT drafts SOAP, DAP, BIRP, and custom-format notes from session audio, typed summaries, or uploaded documents. The AI Scribe provides six output tabs per recording (Transcription, Summary, SOAP Note, Detailed SOAP Note, DAP Note, and Custom Note), so you can produce multiple documentation formats from a single session. # Learn More on How Therapists Can Leverage AI Healthcare Innovation ## 40+ Ways Psychiatrists & Therapists Use AI - BastionGPT 40+ Ways Psychiatrists & Therapists Use AI - BastionGPT Healthcare Innovation ## How to Use AI as a School Psychologist How to Use AI as a School Psychologist BastionGPT is a clinical AI assistant and AI scribe designed to support documentation, analysis, and administrative tasks. All AI-generated content should be reviewed and approved by a licensed clinician before use in any client record. --- > Source: https://bastiongpt.com/use-cases/pt-ot-sports-medicine # For PT, OT & Sports Med # AI That Powers Your Clinical Documentation BastionGPT provides secure, healthcare-focused AI tools designed to streamline the specific documentation needs of rehabilitation professionals. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## The Documentation Crisis Facing Today's Rehab Therapists 91% identify administrative burden as a challenge to providing care 50% of physical therapists report experiencing burnout 43% of prior auth denials are caused by missing or insufficient documentation You became a rehabilitation expert to help patients recover instead you are: Spending your time writing notes, progress reports, and discharge summaries after every session Drafting insurance appeal letters and prior authorizations that require detailed clinical justification Using generic home exercise programs and patient education handouts Losing evenings and weekends to charting backlog that builds up during a full caseload AI for physical therapy documentation can give you that time back. ## BastionGPT Helps You Reclaim Your Practice BastionGPT directly addresses the most common operational challenges in modern rehabilitation practices. With BastionGPT, providers can: Draft notes, progress reports, and discharge summaries in minutes, reclaiming up to 12+ hours per week Strengthen insurance appeal success rates with structured justification letters grounded in clinical evidence Reduce after-hours charting so clinical documentation is finished before you leave the clinic Standardize documentation across your team with Saved Prompts and the Share Prompts feature Generate personalized patient education materials and home exercise programs in a fraction of the usual time Record sessions or dictate post-visit notes with unlimited AI Scribe transcription, including multi-speaker recognition ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## How to Integrate BastionGPT in Your Workflow ### Clinical ### Documentation ### Patient Education & Home Programs ### Insurance Documentation & Appeals ### Provider Communication & Care Coordination # Clinical Documentation Record sessions with the AI Scribe, dictate a voice memo, or type brief bullet points. BastionGPT drafts structured notes for you to review, edit, and finalize. Generate SOAP, DAP, or progress notes from voice dictation or bullet points using the AI Scribe's Custom Note tab or saved prompt templates Create progress reports with objective outcomes, functional improvements, and SMART goal updates by uploading evaluations and session summaries (up to 700,000+ words, or 1,000+ pages, on Professional Plus) Convert post-visit voice memos into formatted clinical documentation with multi-speaker recognition and one-click SOAP, DAP, or custom note output Draft initial evaluations and re-evaluation summaries from uploaded assessment data, intake forms, and prior clinical records Review existing documentation for errors, inconsistencies, coding requirements, or vague language before finalizing # Patient Education & Home Programs Type or dictate exercise descriptions and clinical guidance. BastionGPT generates patient-ready handouts you can customize and print. Generate step-by-step HEP handouts with form cues, safety notes, and progression guidelines at the reading level you specify Draft patient-friendly educational materials explaining conditions, treatments, or recovery timelines in plain language Create ADL guidance sheets and adaptive strategy handouts for OT patients, caregivers, and pediatric populations Produce post-surgical recovery instructions tailored to specific procedures (ACL reconstruction, rotator cuff repair, joint replacement) Build return-to-sport protocols and activity modification guides for athletes at different recovery stages # Insurance Documentation & Appeals Upload denial letters, clinical records, or payer requirements, and BastionGPT drafts responses grounded in clinical evidence. Generate letters of medical necessity for equipment, continued services, or specialized interventions Draft insurance appeal letters that cite clinical progress, patient deficits, and payer policy requirements, then save templates for your billing team with Share Prompts Create prior authorization requests with complete clinical narratives in the format payers expect Summarize lengthy payer policies or coverage criteria into actionable checklists your team can reference Create patient-facing appeal support letters that document the clinical rationale in language patients can share with their employer or plan administrator # Provider Communication & Care Coordination Paste or upload clinical records and BastionGPT generates referral letters, discharge summaries, or care transition documents. Generate referral letters with complete patient history, therapy progress, and specific clinical concerns Create discharge summaries with functional outcomes, recommendations, and follow-up care plans Draft interdisciplinary team updates that synthesize progress across PT, OT, and speech for case conferences Compose physician correspondence summarizing treatment outcomes, ongoing needs, and recommended next steps Prepare care transition documents for patients moving between inpatient rehab, outpatient therapy, and home health settings "We love that we found a tool that actually makes our work better, not just faster." Kiley Cuene Clinical Psychologist "The efficiency gains have been substantial enough that I can serve more people without sacrificing quality." Anna Dinoto, LCP Forensic Psychologist ## Experience What's Possible When Documentation No Longer Holds You Back BastionGPT gives rehabilitation professionals time back in their day. See how healthcare AI built for rehabilitation can change your practice. Start with a 7-day free trial and a 45-day satisfaction guarantee. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Is BastionGPT HIPAA-compliant? Yes. BastionGPT is built specifically to support HIPAA compliance and includes a Business Associate Agreement (BAA) with every subscription. All data is encrypted in transit (TLS) and at rest (AES-256), hosted on Microsoft Azure with ISO 27001-certified data centers. Patient data is never sold or used to train AI models. Can multiple therapists in my clinic share the same BastionGPT account? BastionGPT uses per-seat licensing, so each therapist gets their own secure account. Teams benefit from the Share Prompts feature, which distributes standardized prompt templates across your clinic for consistent documentation quality. How does BastionGPT's AI Scribe work for therapy sessions? Record sessions directly or dictate post-visit notes using your phone, tablet, or computer. The AI Scribe provides six output tabs for each recording: Transcription, Summary, SOAP Note, Detailed SOAP Note, DAP Note, and Custom Note. The Custom Note tab accepts any prompt, so you can generate specialty-specific formats like narrative progress notes or parent-friendly session summaries. Unlimited transcription is included with all plans, with each recording up to 4 hours long. Does BastionGPT work for my therapy specialty? BastionGPT supports physical therapists, occupational therapists, sports medicine professionals, and related rehabilitation disciplines. The platform serves as AI for occupational therapy documentation, AI for physical therapy, and sports medicine workflows, handling SOAP notes, DAP notes, progress reports, initial evaluations, discharge summaries, and more. Customize outputs to match your specialty's terminology using saved prompts. Can BastionGPT generate home exercise program (HEP) handouts? Yes. Describe exercises by voice or text, specify a reading level, and BastionGPT generates patient-ready HEP handouts with step-by-step instructions, form cues, and safety notes. Save templates as prompts and share them with your team. Does BastionGPT integrate with my EMR system? BastionGPT works alongside all major EMR systems, including WebPT, Epic, Oracle Health, Practice Fusion, and others. Integration is copy-and-paste or document upload based, requiring no complex IT setup. Upload documents directly for analysis, summarization, or transformation. ## Learn More about Physical Therapy and AI Healthcare Innovation ## How To Use AI for PT, OT, or Sports Medicine Documentation Read the full article Healthcare Innovation ## AI For Chiropractors: How Chiropractors Can Use AI Read the full article AI Disclaimer: BastionGPT is a clinical AI assistant and AI scribe designed to support healthcare professionals with documentation and administrative tasks. It does not replace professional clinical judgment. All AI-generated content should be reviewed, verified, and approved by a licensed clinician before inclusion in any patient record or official communication. --- > Source: https://bastiongpt.com/use-cases/dme-operations # For DME Operations # AI Built for the Realities of Modern Healthcare Reduce documentation burden, cut claim denials, and give your team the time to focus on patients, all within a HIPAA-compliant platform built for healthcare. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## The Documentation Crisis Facing DME Providers 40-70% of time spent on paperwork and documentation. 25% Average DME claim denial rate often due to documentation 15+ hrs could be saved per week using AI for documentation and billing tasks You started your DME business to help patients get the equipment they need, not to have your team buried in: Denials caused by incomplete documentation Prior authorization backlogs that delay equipment delivery Constant changes to coverage criteria and payer rules Data summarization, analysis, and report generation ## BastionGPT Helps You Reclaim Your Practice BastionGPT works alongside your team as a secure AI assistant that handles the documentation and billing tasks slowing you down. Reclaim hours each week by drafting medical necessity letters, appeal documentation, and billing forms in minutes Accelerate company procedures and billing workflows through saved prompts Improve first-pass claim acceptance rates with AI that flags missing elements and suggests accurate codes before submission Strengthen audit readiness with consistent, well-structured documentation that aligns with Medicare and accreditation standards Reduce denial management costs with AI-generated appeal letters that reference patient records and payer policies Scale operations without proportionally increasing administrative headcount ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## How to Integrate BastionGPT in Your Workflow ### Clinical ### Documentation ### Billing & Claims ### Knowledge & Training ### AI Scribe # Clinical Documentation Upload notes, orders, and patient records into BastionGPT so they can be transformed into structured documentation. Generate letters of medical necessity from patient charts and physician notes in minutes Draft prior authorization requests with required clinical details, diagnosis codes, and supporting documentation Summarize lengthy hospital discharge summaries to extract the information that justifies DME coverage Compose physician order clarification requests when documentation is incomplete Identify missing elements in clinical documentation before claims are submitted to help prevent avoidable denials # Billing & Claims Paste claim details or upload denial letters into BastionGPT, and it helps your billing team prepare accurate submissions and respond to denials with thorough, well-cited appeal letters. Suggest appropriate HCPCS and ICD-10 codes based on physician orders and patient conditions Review claims for errors and missing information before submission to improve first-pass acceptance rates Generate comprehensive appeal letters for denied claims that reference patient medical history and payer policies Draft documentation with complete clinical narratives and supporting evidence Create patient billing communications and payment plan letters that are clear and empathetic # Knowledge & Training Upload your SOPs, payer guides, and Medicare coverage criteria into BastionGPT, and it turns institutional knowledge into instant, searchable answers for your entire team. Answer staff questions about Medicare requirements, payer policies, and company procedures in seconds using saved prompts loaded with your SOPs Anonymize patient data to empower your team to share examples and best practices without violating patient privacy Generate realistic practice cases using synthetic patient data so trainees can learn without PHI exposure Draft policy updates and internal memos when regulations change Summarize lengthy regulatory documents (CMS bulletins, coverage criteria updates) into actionable takeaways your team can act on immediately # AI Transcription & Scribe Record a patient interaction with BastionGPT's built-in AI scribe, and it converts the recording into structured documentation. Record home equipment delivery and setup visits, then generate structured delivery notes with patient education details and safety checks documented Transcribe patient training sessions for CPAP, oxygen, and other equipment, and convert them into formatted instruction summaries and compliance records Capture in-office assessments and evaluations with hands-free dictation, then produce clinical notes ready for the patient file Transcribe phone calls with physicians or payers to create accurate records of verbal orders, authorization discussions, and follow-up action items Review and edit all transcriptions before filing, so every note meets your documentation standards "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Alexis Arceo, CEO Expedited Reports “Bastion's allowed me to provide extra add-on services that are really meaningful.” Rachael St. Claire, PsyD Clinical Psychologist ## Join DME Providers Adopting AI BastionGPT gives DME providers a secure AI assistant that handles documentation, billing, compliance, and training, so you can focus on patient care and business growth. Start using BastionGPT within minutes. There are no setup costs and a 7-day free trial. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Is BastionGPT HIPAA-compliant? Yes. BastionGPT is designed to support HIPAA compliance in every interaction. A Business Associate Agreement (BAA) is included with all plans. All data is encrypted in transit (TLS) and at rest (AES-256 or equivalent), hosted on Microsoft Azure in HIPAA-compliant, ISO 27001-certified data centers. No data is shared with OpenAI or other third-party AI providers for training. How does BastionGPT stay current with changing Medicare and payer requirements? BastionGPT integrates multiple leading AI models (including GPT, Claude, and Gemini) that are regularly updated. Your team can also upload current CMS bulletins, coverage criteria documents, and internal payer guides as reference documents. BastionGPT processes documents up to 1000 pages, so even lengthy regulatory updates can be summarized into actionable takeaways within minutes. Does BastionGPT store patient data? BastionGPT retains chat history, transcripts, and uploaded files for 30 days by default, and users can delete data sooner at any time. Can multiple team members use BastionGPT at the same time? Yes. BastionGPT uses per-seat licensing, so each team member gets their own secure account. The Professional Plus plan includes cross-device prompt and chat synchronization, and enterprise teams can take advantage of Single Sign-On (SSO) and custom security policies. How can AI help reduce DME claim denials? BastionGPT helps reduce claim denials by flagging missing documentation elements, suggesting accurate HCPCS and ICD-10 codes, and generating thorough appeal letters that reference patient records and payer-specific policies. By catching gaps before submission and speeding up the appeals process, your billing team can improve first-pass acceptance rates and recover denied revenue faster. Can BastionGPT integrate with our existing DME billing software? BastionGPT works alongside all major EMR and billing systems, including Brightree, DMEWorks, and others. Integration is straightforward through copy/paste and document upload, with no complex IT setup required. Your team can upload physician orders, patient records, or claims data directly into BastionGPT and receive structured outputs ready to paste into your billing system. Is BastionGPT easy to use? BastionGPT is designed to feel familiar to anyone who has used ChatGPT or a similar AI tool. The interface is clean and simple, and most users are productive within their first session. We also offer live onboarding support, training sessions, and lunch-and-learn programs to help your team get started. How quickly can our team get started? Most organizations sign up and begin using BastionGPT within 10 minutes. There are no setup costs, no IT implementation required, and no fixed commitments. Start with a 7-day free trial, and your team can be drafting medical necessity letters and reviewing claims the same day. Can I use AI to write medical necessity letters for DME? Yes. BastionGPT drafts medical necessity letters by pulling from uploaded physician orders, patient charts, and clinical notes. You provide the source documentation, and BastionGPT generates a structured letter with the diagnosis, clinical justification, and equipment details formatted for payer submission. Your team reviews and approves the letter before it goes out. # Learn More About DME Operations and AI Healthcare Innovation ## How DME Companies Are Using AI for Operational Efficiency How DME Companies Are Using AI for Operational Efficiency Healthcare Innovation ## How Healthcare Administrators Can Use AI How Healthcare Administrators Can Use AI AI Disclaimer: BastionGPT is a clinical AI assistant and AI scribe designed to support healthcare professionals with documentation and administrative tasks. It does not replace professional clinical judgment. All AI-generated content should be reviewed, verified, and approved by a licensed clinician before inclusion in any client record or official communication. --- > Source: https://bastiongpt.com/use-cases/healthcare-admin-revenue-cycle # For Revenue Cycle Management & Healthcare Administration # AI Built for Healthcare Billing and Operations BastionGPT provides secure, HIPAA-compliant AI tools designed to streamline documentation, reduce admin work, and accelerate Rev Cycle workflows. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Why Administrative Burden Is Impacting Healthcare Operations 60% of time spent on documentation tasks 15% average claim denial rate $25 to $118 to rework a single denied claim Your team should be driving revenue, not drowning in: Denials caused by documentation errors and missing information Prior authorization backlogs that delay patient access and revenue collection Lengthy medical records that take hours to summarize for claims and appeals Repetitive appeal letters, billing communications, and compliance paperwork ## How BastionGPT Streamlines Your Revenue Cycle Operations BastionGPT acts as a trusted AI assistant for healthcare billing workflows. Upload clinical documents, payer policies, or denial letters and receive structured outputs in minutes. Reclaim hours each week by drafting appeal letters, prior authorization requests, and billing communications in minutes Standardize documentation quality across your team with shared prompt templates Reduce denial rates by reviewing documentation for completeness and flagging gaps before submission Scale revenue cycle operations without increasing headcount Summarize medical records up to 1,000 pages long to build stronger claims and appeal packages Free your team to focus on denial pattern analysis, payer negotiations, and process improvement ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Integrating BastionGPT Across Your Revenue Cycle ### Prior Authorization & Medical Necessity Documentation ### Denial Management & Appeals ### Medical Record Summarization & Claims Support ### Team Communication, Training & Compliance Documentation # Prior Authorization & Medical Necessity Documentation Turn hours of manual documentation into minutes with complete, submission-ready outputs. Generate prior authorization requests with clinical narratives, diagnosis codes, and supporting evidence drawn from pasted or uploaded records Draft letters of medical necessity citing patient history, clinical findings, and payer-specific coverage criteria Identify missing information or documentation gaps before submission to reduce rework and delays Produce payer-specific formats by uploading each insurer's requirements as reference material Review existing documentation for errors, inconsistencies, pronoun mistakes, or vague language before finalizing # Denial Management & Appeals Paste a denial letter and the supporting clinical documentation into BastionGPT to draft a targeted appeal that addresses the specific denial reason and references the relevant payer policy. Generate appeal letters that directly address denial reason codes with clinical evidence and regulatory citations Draft reconsideration requests with targeted arguments specific to each payer's stated rationale Cross-reference denial patterns against uploaded payer policies to identify systemic documentation gaps Build escalating second-level and third-level appeals with detailed clinical and regulatory support BastionGPT processes documents up to 1,000 pages long, making it a powerful tool for distilling charts and notes. # Medical Record Summarization & Claims Support Save and reuse prompts that standardize how your team summarizes records, extracts clinical findings, and builds claims-ready documentation. Summarize lengthy medical records into concise case summaries highlighting diagnoses, treatments, and functional limitations relevant to billing Extract key clinical findings that support medical necessity for specific procedures or equipment Identify inconsistencies or missing documentation before claim submission Translate complex clinical language into clear, payer-friendly narratives for claims support Create client-facing letters, referral summaries, or coordination notes for interdisciplinary teams including psychiatrists, social workers, and school staff # Team Communication, Training & Compliance Documentation Revenue cycle departments run on internal communication: payer call notes, staff training, compliance updates, SOP revisions. BastionGPT handles the drafting so your team can focus on the decisions. Share standardized prompt templates across your department using the Saved Prompts and Share Prompts features, so new hires produce consistent, high-quality output from day one Generate training materials and onboarding documentation for new billers and coders Transcribe payer calls, staff meetings, and compliance reviews with BastionGPT's AI scribe, then generate summaries with action items Summarize lengthy CMS bulletins, LCD/LCA changes, and payer policy updates into actionable takeaways for your team Securely draft internal emails, memos, and policy updates in seconds "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Alexis Arceo, CEO Expedited Reports “The efficiency gains have been substantial enough that I can serve more people without sacrificing quality." Anna Dinoto, LCP Forensic Psychologist ## Join Healthcare Organizations That Have Transformed Their Revenue Cycle BastionGPT is trusted by 10,000+ health organizations that use it daily to reduce documentation burden and reclaim time for the work that matters most. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Is BastionGPT HIPAA-compliant? Yes. BastionGPT is designed to support HIPAA compliance, with a Business Associate Agreement (BAA) included in every subscription. All data is encrypted in transit and at rest, hosted on Microsoft Azure HIPAA-compliant infrastructure, and never shared with OpenAI or other third-party AI providers for training. Can I use BastionGPT with patient health information (PHI)? Yes. Your team can upload clinical records, billing documents, and other PHI directly into BastionGPT for summarization, drafting, and analysis without needing to manually redact identifying information first. How does BastionGPT stay current with changing payer policies and regulations? BastionGPT's multi-model AI engine is regularly updated. Upload current CMS bulletins, LCD/LCA updates, and payer policy documents as reference material. BastionGPT draws on those uploaded documents to generate accurate, current outputs. When policies change, upload the new version in your saved prompts and your team's prompts immediately reflect the updated requirements. Is BastionGPT a replacement for our billing software or clearinghouse? BastionGPT is an AI assistant that works alongside your existing billing software, clearinghouse, and practice management systems. It handles the documentation-heavy parts of your revenue cycle (drafting appeals, summarizing records, writing prior authorizations) while your existing tools handle claim submission, payment posting, and remittance processing. How do we standardize output quality across our billing team? BastionGPT includes a Saved Prompts library where your team can create, organize, and share prompt templates by workflow or specialty. Use the Share Prompts feature to distribute your best appeal letter formats, prior auth templates, and documentation standards across your entire department. New hires can start producing consistent, high-quality output immediately. Does BastionGPT use my data to train its AI models? No. BastionGPT never shares your data with OpenAI or other third-party AI providers for model training. How does BastionGPT handle different payer requirements and formats? Upload payer-specific guidelines, coverage criteria, and policy documents directly into BastionGPT saved prompts. The AI references those materials when drafting prior authorizations, appeal letters, and billing documentation tailored to each payer's requirements. Save your best-performing formats as prompt templates and share them across your team. Can BastionGPT integrate with our practice management system or billing software? BastionGPT works alongside all major EMR/EHR and practice management systems, including Epic, Oracle Health, Allscripts, and Meditech. Integration is copy/paste or document upload based, requiring no complex IT setup. Your team can start using it within minutes. What file types can BastionGPT process? BastionGPT supports .txt, .pdf, .doc, .docx, .ppt, .pptx, .xls, .xlsx, .csv, .html, .png, .jpg, and .jpeg files. Professional Plus plans can process documents up to 1,000 pages and reference multiple documents at once, which is especially useful for reviewing lengthy medical records, EOBs, or payer policy documents alongside a denial letter. # Learn More on How Healthcare Teams Can Leverage AI Healthcare Innovation ## How Healthcare Administrators Can Use AI How Healthcare Administrators Can Use AI Healthcare Innovation ## How Healthcare Legal Professionals & Lawyers Are Using AI How Healthcare Legal Professionals & Lawyers Are Using AI BastionGPT is an AI-powered assistant designed to support healthcare professionals. All AI-generated content should be reviewed by qualified professionals before use. BastionGPT assists with, but does not replace, clinical or professional judgment. --- > Source: https://bastiongpt.com/use-cases/legal-compliance # For Compliance and Legal # From Record Review to Case Ready Ready to cut through record review backlogs? BastionGPT accelerates medical record analysis, strengthens case documents, and streamlines compliance workflows. Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## The Real Challenges Healthcare Legal Teams Face Today 52% of legal professionals are experiencing burnout 90% of medical record review time is consumed by manual processes $7.42 mil average cost of a healthcare data breach Why Documentation Bottlenecks Are Costing Your Legal Team Non-billable hours consumed by manual record review, timeline building, and fragmented charts Thousands of pages per case spread across providers, systems, and file formats Regulatory deadlines for audits, compliance filings, and policy responses with no margin for delay Repetitive drafting of discovery responses, appeals, client updates, and internal memos ## How BastionGPT Strengthens Your Legal Workflow BastionGPT is a trusted AI assistant for attorneys, paralegals, compliance officers, and risk managers. Upload medical records, case files, regulatory documents, or denial letters and get structured, accurate outputs in minutes. Compress hours of medical record review into minutes by generating chronologies, treatment timelines, and issue-focused summaries from large, multi-document record sets Operate confidently with PHI inside a HIPAA-compliant AI environment with AES-256 encryption, access controls, and a BAA included on every plan Reduce non-billable time by generating first drafts of discovery responses, motions, appeal letters, and client updates grounded in case facts Respond faster to audits, regulatory inquiries, and policy changes with rapid document analysis and compliance summaries Improve consistency and completeness in clinical documentation across your team with shared prompt templates Free attorneys and paralegals to focus on case strategy, negotiations, and courtroom preparation instead of repetitive paperwork ## Ready to get started? Try BastionGPT Free 7 day free trial. Trusted by 10,000+ health organizations. ## Integrating BastionGPT Across Your Legal Workflows ### Medical Record Review & Case Analysis ### Legal Drafting & Case Communications ### Compliance & Risk Management ### Insurance Claims, Coverage & Appeals # Medical Record Review & Case Analysis Upload a patient chart (including scanned PDFs) into BastionGPT. It reads the full record set via Optical Character Recognition (OCR), extracts key events, and produces a structured chronology or issue-focused summary your team can act on immediately. Generate treatment timelines and chronologies from large, multi-document record sets organized by date, provider, or event type Create focused summaries around specific legal issues such as standard of care, causation, damages, or preexisting conditions Surface potential inconsistencies, missing documentation, or unusual gaps in care that warrant deeper investigation Cross-reference findings across multiple uploaded records to identify patterns relevant to your case theory Review existing documentation for errors, inconsistencies, pronoun mistakes, or vague language before finalizing # Legal Drafting & Case Communications Provide case records and notes as input into BastionGPT. It drafts a structured first pass that addresses the relevant facts, so your team starts from a strong, fact-informed template instead of a blank page. Standardize formatting and structure across documents using Saved Prompts while leaving final legal analysis and citations to the attorney Share your best-performing document templates across the team using the Share Prompts feature so new hires produce consistent output from day one Draft first-pass responses to interrogatories and document requests grounded in uploaded case records and clinical notes Build concise case summaries and internal memos tailored to partners, clients, or co-counsel Generate demand letter frameworks and settlement correspondence that reference specific medical findings from the record # Compliance & Risk Management Upload a new CMS bulletin, HIPAA update, or state healthcare regulation into BastionGPT. It summarizes the key changes, flags required actions, and drafts an internal policy memo your compliance team can distribute immediately. Summarize dense regulatory documents (CMS rules, HIPAA updates, state-level healthcare laws, 42 CFR Part 2 guidance) into actionable internal memos Review document sets against specified audit requirements and flag files with missing or incomplete components Draft compliance checklists and audit response documentation from uploaded regulatory requirements Generate training materials and policy update summaries for staff education and onboarding Create client-facing letters, referral summaries, or coordination notes for interdisciplinary teams including psychiatrists, social workers, and staff # Insurance Claims, Coverage & Appeals Paste a denial letter and the supporting clinical documentation into BastionGPT. It drafts a targeted appeal that addresses the specific denial reason, references the relevant policy language, and cites the clinical evidence from the patient's record. Generate appeal letters that directly address denial reason codes with clinical evidence and policy citations Organize large sets of claims data into pattern summaries to support litigation strategy, settlement negotiations, or internal investigations Draft escalating second-level and third-level appeals with progressively detailed clinical and regulatory support Cross-reference denial patterns against uploaded payer policies to identify systemic documentation gaps across your caseload Compare policy language against case facts to identify relevant clauses, exclusions, and conditions that apply to the dispute "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Alexis Arceo, CEO Expedited Reports "BastionGPT has a lot more capabilities with data protection, so clinicians can more freely use AI." David Lopis, Director Psychology Squared ## Join Legal Teams That Have Transformed Their Healthcare Practice BastionGPT is trusted by 10,000+ health organizations that use it daily to reduce documentation burden and reclaim time for the work that matters most. ## Professional Great for those just getting started. $20 per user monthly. Start Trial Access to the most powerful AI models Reference documents up to 30 pages long HIPAA secure and compliant Unlimited audio transcription ## Professional Plus Our most popular plan. Great for those working with large documents. $45 per user monthly. Start Trial Everything from Professional Reference multiple documents up to 1,000 pages long Analyze images, charts and graphs Early access to the latest new AI models and features ## Ultra Designed for large organizations. Starting at 100 users. $75 per user monthly. Contact Sales Everything from Professional Plus Customizable security policies Single Sign On (SSO) Run locally within your existing enterprise cloud and much more ## Frequently asked questions Can lawyers use ChatGPT with medical records? Standard ChatGPT and LLMs are not HIPAA-compliant and should not be used with protected health information. OpenAI does not offer a BAA for its consumer products, and PHI entered into those tools may be used for model training Is it safe to use AI with protected health information? It depends on the platform. Consumer AI tools like standard ChatGPT, Google Gemini, and Claude do not offer BAAs and are not designed for PHI. BastionGPT is purpose-built for healthcare professionals who work with sensitive data. Does BastionGPT use my data to train its AI models? No. BastionGPT never shares your data with OpenAI or other third-party AI providers for model training. What file types can BastionGPT process? BastionGPT supports .txt, .pdf, .doc, .docx, .ppt, .pptx, .xls, .xlsx, .csv, .html, .png, .jpg, and .jpeg files. Professional Plus plans can process documents up to 700,000+ words, or 1,000+ pages, and reference multiple documents at once, which is especially useful for reviewing lengthy medical records or payer policy documents alongside a denial letter. How long does BastionGPT retain my data? Chat history, transcripts, and uploaded files are securely wiped after 30 days by default. You can delete data sooner at any time. Enterprise customers can work with our team to configure custom retention policies. What is HIPAA-compliant AI? HIPAA-compliant AI refers to an AI platform designed to meet the privacy and security requirements of the Health Insurance Portability and Accountability Act when processing protected health information. This includes data encryption in transit and at rest, access controls, audit logging, data isolation, and a signed Business Associate Agreement (BAA). Is BastionGPT HIPAA-compliant? Yes. BastionGPT is designed to support HIPAA compliance on every plan. All data is encrypted in transit (TLS) and at rest (AES-256), hosted on Microsoft Azure with a HIPAA-compliant BAA, and customer data is isolated with no data shared with OpenAI or other third-party providers for training. A BAA is included automatically with every subscription. Does BastionGPT work for my legal specialty? BastionGPT supports healthcare legal professionals across practice areas including medical malpractice, personal injury, insurance coverage and appeals, regulatory compliance, healthcare fraud, and risk management. Whether you need AI for healthcare compliance monitoring or case-level document analysis, the platform applies to any workflow involving medical documentation and regulated data. Does BastionGPT replace traditional legal research tools? BastionGPT is a healthcare legal AI assistant built for documentation and analysis. It excels at summarizing records, drafting documents, and processing large volumes of clinical and regulatory text. It does not perform legal citation research or case law searches. Most teams use BastionGPT alongside their existing legal research platforms. # Learn More About Healthcare Compliance and AI Healthcare Innovation ## How Healthcare Administrators Can Use AI How Healthcare Administrators Can Use AI Healthcare Innovation ## How Healthcare Legal Professionals & Lawyers Are Using AI How Healthcare Legal Professionals & Lawyers Are Using AI BastionGPT is an AI-powered assistant designed to support healthcare legal professionals. All AI-generated content should be reviewed by qualified professionals before use in any legal proceeding, compliance filing, or official communication. BastionGPT assists with, but does not replace, legal or clinical judgment. --- > Source: https://bastiongpt.com/bastiongpt-vs/heidi-health # Heidi Health Alternative: BastionGPT vs Heidi Health (2026) Jump to Comparison Table Heidi Health and BastionGPT both offer HIPAA-compliant AI scribes that turn patient sessions into structured clinical notes. The overlap largely ends there. Heidi Health is an ambient scribe with a clinical evidence search, and full individual access costs $150 per user per month in 2026. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, what happens beyond the visit note, how document handling compares, what real users say on independent review sites, and what compliance coverage you get at each price. Heidi Health's pricing and limits were verified against heidihealth.com and its help center on July 2, 2026. If you are comparing Heidi Health alternatives, the sections below give you the specifics: full-access pricing, scribe capability, document analysis, and compliance depth. BastionGPT is a strong Heidi alternative for clinicians who want the full feature set at $20 instead of $150 per month. For the wider field, our AI scribe comparison covers other options such as Freed, DeepScribe, and Suki. Quick answer: Heidi Health fits clinicians who want an ambient scribe with a clinical evidence search and are comfortable with credit-metered features below the $150 Clinician tier. BastionGPT replaces Heidi's scribe outright — unlimited scribing on every plan, sessions up to 240 minutes, up to 10 speakers labeled by name and role, couples, family, and group sessions included — at $20 to $45 per user per month, and adds a full clinical AI assistant, 1,000+ page document analysis, and a signed BAA on every plan, including the free trial. Feature BastionGPT Heidi Health Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 3.3/5 on Trustpilot across 480+ reviews (June 2026) Entry paid plan $20/user/month (Professional) $40/user/month (Evidence Plus; $30 billed annually) Full-featured individual plan $45/user/month (Professional Plus) $150/user/month (Clinician; $110 when billed annually) Free option 7-day free full trial, plus a 45-day money-back guarantee Free tier: unlimited basic scribing; 10 credits/month for features Signed BAA ✓ Every plan, including the free trial Organization-wide BAA listed as an Enterprise feature HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP GDPR, PIPEDA, APP Data residency US, Canada, or Australia, based on billing geography ✓ Regional hosting on AWS and Azure; patient data stored in the country of origin AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Not disclosed Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles in the transcript ✓ (speaker count not published) Couples, family, and group sessions Included on every plan Supported; per-speaker labeling not published Max recording length 240 minutes per session No published limit Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers Published research covers documentation time savings; no peer-reviewed accuracy rate published Documents per upload 10+ at once (Professional Plus and Ultra) No published count; users report 5 documents at once. Roughly 20MB per-file size limit. Document capacity 1,000+ pages of combined context No published page limit; users report scanned records of 20 to 25 pages hitting the file cap Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG PDF, Word, JPG, PNG (no spreadsheets) Full AI assistant ✓ Referral letters, appeals, patient education, document review Ask Heidi: session-anchored, credit-limited below the Clinician plan Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Standard templates; custom templates use monthly credits on the free tier Languages 15+ 110+ Saved prompts with team sharing ✓ All plans, with folder organization Team template sharing on Practice and Enterprise plans EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) Integrations available; full EHR integration access on Practice and Enterprise Uptime guarantee 99.90% SLA on Enterprise only Support Email, chat, phone, and video with real humans, on every plan Chat and email; dedicated customer success on Enterprise ## Where BastionGPT and Heidi Health are similar Both platforms offer unlimited session transcription, generate structured clinical notes, and deploy in minutes without an IT project. Both state that patient data is never used to train AI models, both store US customer data inside the United States, and both address Canadian (PIPEDA) and Australian (APP) privacy requirements. If your entire requirement is turning a spoken visit into a standard note, both tools cover that core workflow. The differences appear the moment you price full access, upload a long record, run a group session, or need documentation that goes beyond the visit. ## Where BastionGPT goes further ### Price — by a wide margin Heidi Health restructured its plans in early 2026, so older reviews citing a $99 Pro plan are out of date. Heidi's current US lineup: a free tier, Evidence Plus at $40 per user per month ($30 billed annually), Clinician at $150 per user per month ($110 billed annually), Practice at $180 per user per month (annual billing only), and custom Enterprise pricing. Full individual access, including unlimited Ask Heidi and advanced templates, starts at the Clinician plan. BastionGPT Professional starts at $20 per user per month . Professional Plus, the closest match to Heidi's Clinician tier, costs $45 and includes the full AI assistant, multi-document analysis, image and spreadsheet reading, and manual model selection. Prices have held steady for over two years, and annual billing includes a free month. The math for a five-clinician practice: Heidi Clinician runs $9,000 per year at monthly billing ($150 x 5 x 12). BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12). That is a difference of $6,300 per year , or $3,900 if both are billed annually. ### A scribe that knows who is speaking BastionGPT's AI Scribe is included on every plan with no note caps or credits: record live or upload audio, capture sessions up to 240 minutes , and let speaker diarization separate up to 10 speakers while the AI labels each speaker's name and role — clinician, patient, or parent — in the transcript. Couples, family, and group sessions are included on every plan. Heidi supports multi-speaker recognition, but publishes neither a speaker count nor per-speaker labeling details, which matters for behavioral health teams documenting group work. Accuracy is documented, not asserted: BastionGPT cites 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Heidi's published research covers documentation time savings; it publishes no equivalent accuracy rate. ### Note flexibility BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes, and a Custom Note tab accepts any free-form prompt per recording : narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can also upload example notes so BastionGPT matches your writing style and formatting. Heidi generates notes from standard templates on every tier, with a community template library. On the free tier, custom templates draw from the same 10-credit monthly pool as documents and Ask Heidi. ### A full AI assistant, not just a Q&A tool Ask Heidi works inside a session: it edits the note, creates session documents such as referral letters and discharge summaries, and answers clinical questions with citations. It is also metered. Free users get 10 credits per month across advanced features, Evidence Plus caps advanced queries at 10 per week, and unlimited access starts at the $150 Clinician plan, per Heidi's help center. BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. ### Reads the records your practice receives This is the widest gap between the two platforms. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Heidi accepts PDF, Word, JPG, and PNG attachments as session context. It publishes no page limit; the practical constraint is a per-file size cap of roughly 20MB, which clinicians on Heidi's public feature-request board report hitting with scanned records in the 20 to 25 page range, splitting 100+ page documents by hand to work around it. Spreadsheets are unsupported, template example uploads are capped at 5 files, and new sessions automatically link only the 3 most recent past sessions. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Heidi Health does not disclose which foundation models power its product. ### Saved Prompts for teams BastionGPT includes a Saved Prompts library with folder organization and team sharing on every plan . A practice can standardize documentation by sharing one vetted prompt set across all providers and inserting any prompt into a chat with a slash command. Heidi offers snippets and templates, and gates team template sharing to its Practice and Enterprise plans. ### A signed BAA on every plan BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , and supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. Heidi states that it executes BAAs when it processes PHI on behalf of covered entities, and its help center lists an organization-wide BAA as an Enterprise plan feature. A self-serve BAA process for individual clinicians is not published. For a solo practitioner, that difference matters on day one. ### User ratings on independent review sites BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. Heidi Health holds a 3.3 out of 5 rating on Trustpilot across more than 480 reviews as of June 2026. Recent Trustpilot reviews most frequently report lost recordings, dropped sessions, and slow support responses, while positive reviews praise speed and reduced after-hours charting. Ratings shift over time, so this page cites the source and date for each figure. The links in the sources section lead to the live numbers. BastionGPT assists with documentation and analysis. Clinicians review and approve all AI-generated output; it does not replace professional judgment. ## When Heidi Health stands out Heidi's clearest advantage is language coverage: it advertises transcription in 110+ languages, while BastionGPT supports 15+. If your practice regularly documents visits in languages BastionGPT does not yet cover, that difference is real. Heidi also offers a permanent free tier. It includes unlimited transcription with standard templates, and it meters everything else: custom templates, document creation, Ask Heidi, and form filling share a pool of 10 credits per month, and full access to those features starts at $150 per user per month. For US practices that document primarily in English and need anything beyond the visit note (records review, appeal letters, reports, spreadsheet analysis), BastionGPT delivers more capability at less than a third of Heidi's full-access price. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Heidi Health cost in 2026? Heidi Health's current US plans are: Free, Evidence Plus at $40 per user per month ($30 billed annually), Clinician at $150 per user per month ($110 billed annually), Practice at $180 per user per month (billed annually), and custom Enterprise pricing. Heidi restructured its plans in early 2026, so reviews citing a $99 Pro plan are out of date. BastionGPT plans run $20 to $75 per user per month. Is Heidi Health free? Heidi Health offers a free tier with unlimited transcription using standard templates. Advanced features (custom templates, document creation, Ask Heidi, and form filling) draw from 10 monthly credits, then pause until the next month. BastionGPT offers a 7-day free trial and plans from $20 per user per month with a 45-day money-back guarantee. Does Heidi Health sign a BAA? Heidi states that it executes Business Associate Agreements when it processes PHI for covered entities, and its help center lists an organization-wide BAA as an Enterprise plan feature. A self-serve BAA process for individual clinicians is not published. BastionGPT includes a signed BAA on every plan, including the free trial. How many speakers can BastionGPT's AI Scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role, for example, clinician, patient, or parent, in the transcript. Heidi Health supports multi-speaker recognition but does not publish a speaker count. Is BastionGPT's AI scribe as accurate as Heidi's? BastionGPT cites 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. Heidi's published research, including a 2026 UK quality-improvement project, measures documentation time savings rather than a comparable accuracy rate. Both platforms expect clinicians to review and approve every note. Can BastionGPT document couples, family, and group therapy sessions? Yes. Couples, family, and group sessions are included on every BastionGPT plan. The AI Scribe separates up to 10 speakers per session and automatically labels each speaker's name and role in the transcript, with sessions up to 240 minutes. Heidi supports multi-speaker recognition but does not publish a speaker count or per-speaker labeling details. How many pages and documents can BastionGPT analyze at once? On Professional Plus and Ultra, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, text, and images. Can Heidi Health analyze uploaded documents? Within limits. Heidi accepts PDF, Word, JPG, and PNG files as session context with a file size cap of roughly 20MB; clinicians on Heidi's feature-request board report scanned records of 20 to 25 pages hitting that cap. Spreadsheets are not supported. BastionGPT reads 10+ documents and 1,000+ pages at once, including Excel and CSV files. Can either platform analyze Excel spreadsheets? BastionGPT reads Excel (.xls, .xlsx, .xlsm), CSV, PowerPoint, and HTML files on Professional Plus and above, alongside PDF, Word, text, and image files. Heidi Health's supported upload formats are PDF, Word, and image files; spreadsheets are not accepted. What AI models power BastionGPT and Heidi Health? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Heidi Health does not disclose which foundation models power its product. How do BastionGPT and Heidi Health user ratings compare? BastionGPT holds a 4.8 out of 5 average across the verified review directories Capterra, Software Advice, and GetApp. Heidi Health holds a 3.3 out of 5 rating on Trustpilot across more than 480 reviews as of June 2026, with recent reviews most often reporting lost recordings and dropped sessions. Ratings change over time, so check each review site for current figures. Does Heidi Health support more languages than BastionGPT? Yes. Heidi Health advertises transcription in 110+ languages, while BastionGPT supports 15+. If multilingual transcription is your deciding requirement, weigh it against the differences in document capacity, assistant capabilities, BAA coverage, and price. Which is better for US and Australian practices, BastionGPT or Heidi? BastionGPT serves both markets: a signed BAA on every plan, HIPAA, 42 CFR Part 2, and FERPA support for US practices, Australian APP compliance with Australian customer data hosted in Australia, a 99.90% uptime guarantee, and human support by phone, chat, email, and video. Heidi is Australian-founded and also hosts data regionally, so for Australian practices the comparison comes down to price and what is included: BastionGPT's full-featured plan is $45 per user per month versus $150 for Heidi's Clinician plan, and unlimited scribing plus the full AI assistant are included rather than credit-metered. Does BastionGPT have a free tier like Heidi? BastionGPT offers a 7-day free trial of its full Professional and Professional Plus plans rather than a permanently limited free tier. You are not charged during the trial, you can cancel anytime, and paid plans from $20 per user per month carry a 45-day money-back guarantee. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/hathr-ai # Hathr AI Alternative: BastionGPT vs Hathr AI (2026) Jump to Comparison Table Hathr AI and BastionGPT both give healthcare professionals a HIPAA-compliant way to use generative AI, and both include a signed Business Associate Agreement on every plan. The overlap largely ends there. Hathr AI is a secure chat and document analysis tool built on Anthropic's Claude, hosted on AWS GovCloud, and priced at $47 per month. BastionGPT is a clinical AI platform: an AI scribe with unlimited transcription, structured clinical notes, a full AI assistant, 1,000+ pages of document analysis, and three named AI models, with plans from $20 per user per month. This comparison covers the differences that matter when you choose: what each platform costs in 2026, whether it can document a spoken patient visit, how document handling compares, which AI models sit under the hood, what support you get, and where each platform's compliance posture fits. Hathr AI's pricing, plan, and feature details were verified July 2, 2026 against hathr.ai and its public directory listings. If you are comparing Hathr AI alternatives, this page covers the specifics: pricing, scribe capability, and model choice. BastionGPT is a Hathr alternative that adds a built-in AI scribe and multi-model access (GPT, Claude, and Gemini) at $20 per month. Our Claude Team and Microsoft Copilot comparisons cover other routes to compliant AI. The quick answer: Choose BastionGPT if you want a clinical AI platform: an AI scribe with unlimited transcription and 10-speaker recognition, structured SOAP, DAP, BIRP, and H&P notes, a full AI assistant, 1,000+ page document analysis, and named multi-model AI (GPT, Claude, and Gemini) from $20 per user per month. Choose Hathr AI if your organization requires AWS GovCloud hosting in a FedRAMP High environment for federal or defense work involving Controlled Unclassified Information. For private-sector US healthcare, BastionGPT delivers scribing plus a full assistant at a lower price than Hathr AI's chat-only subscription. Feature BastionGPT Hathr AI Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 1 published Capterra review; 0 verified reviews on GetApp and SourceForge (July 2, 2026) Entry paid plan $20/user/month (Professional) $47/month plan card price ($45 advertised in the site headline) Full-featured individual plan $45/user/month (Professional Plus) $47/month (one tier: chat and document upload) Free option 7-day free full trial, plus a 45-day money-back guarantee 7-day free trial; no free tier Signed BAA ✓ Every plan, including the free trial ✓ Every plan; signed within 24 hours HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP 42 CFR Part 2, NIST 800-171, NIST 800-53 Hosting and certifications Microsoft Azure HIPAA-compliant infrastructure; HITRUST CSF certified, SOC 2 Type II attested AWS GovCloud, FedRAMP High environment; FIPS 140-2 encryption, TLS 1.3 AI models GPT, Claude, Gemini (named and selectable) Anthropic Claude only; directory listings cite Claude 3.5 Sonnet Built-in AI scribe ✓ Unlimited transcription on every plan, live or from uploaded audio Not offered; API customers can build scribes from transcripts produced elsewhere Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of names and roles N/A (no audio transcription product) Max recording length 240 minutes per session N/A Documents per upload 10+ at once (Professional Plus and Ultra) Unlimited file uploads advertised; no per-upload count published Document capacity 1,000+ pages of combined context 500,000+ words per document Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not published; marketing references medical records, lab results, and PDFs Image, chart, and spreadsheet analysis ✓ Professional Plus and above, including scanned PDFs and graphs Not advertised Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Notes generated through chat prompting; no dedicated note formats published Saved prompts with team sharing ✓ All plans, with folder organization and slash commands Data Library for repeatable document workflows; prompt coaching on Enterprise EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) EHR/EMR integrations listed as an Enterprise feature (custom pricing) SSO and provisioning SSO on Ultra ($75/user/month, published price) SSO, SAML, SCIM on Enterprise only (custom pricing) Uptime guarantee 99.90% 99.9% SLA on Enterprise only Support Email, chat, phone, and video with real humans, on every plan Email/help desk; site hours M-F 9am-5pm ET; real-time support on Enterprise Languages 15+ (transcription) English (per SourceForge listing) Federal/CUI posture Not targeted; built for HIPAA-governed private healthcare States CUI handling at Impact Levels 2, 4, 5; approved for the Military-Civilian Health Ecosystem ## Where BastionGPT and Hathr AI are similar Both platforms include a signed BAA on every plan, support HIPAA and 42 CFR Part 2 environments, state that customer data is never used to train AI models, encrypt data in transit and at rest, store US customer data inside the United States, and offer a 7-day free trial. Both accept document uploads and can draft clinical and administrative documents such as record summaries, pre-authorization requests, and appeal letters. Both also serve adjacent regulated professions, including legal, insurance, and government teams. If your entire requirement is typing questions into a secure chat window and uploading records for summarization, both tools cover that core workflow. The differences appear the moment you need to document a spoken visit, produce a structured clinical note, analyze a spreadsheet or scanned image, choose an AI model, or reach support by phone. ## Where BastionGPT goes further ### Price, and what the price includes Hathr AI sells one plan for individuals: the Single Subscription Webapp. The pricing page headline advertises $45 per month, and the plan card on the same page lists $47 per month as of July 2, 2026; Hathr's GetApp listing also shows $47 per user per month. The subscription covers secure chat, unlimited document uploads, and a data library. SSO, EHR integrations, real-time support, and the 99.9% SLA require a custom-priced Enterprise plan. BastionGPT Professional starts at $20 per user per month and already includes unlimited AI scribe transcription. Professional Plus, at $45 , adds multi-document analysis up to 1,000+ pages, image and spreadsheet reading, manual model selection, and early access to new models. At two dollars under Hathr AI's card price, Professional Plus includes an entire scribing product that Hathr does not sell at any price. The math for a five-clinician practice: Hathr AI runs $2,820 per year ($47 x 5 x 12). BastionGPT Professional runs $1,200 per year ($20 x 5 x 12) with unlimited transcription included, and Professional Plus runs $2,700. Every BastionGPT tier priced below Hathr's single plan includes the scribe. ### A built-in AI scribe, on every plan This is the widest gap between the two platforms. BastionGPT includes an AI Scribe with unlimited transcription on every plan . It captures visits live or from uploaded audio for sessions up to 240 minutes, recognizes up to 10 speakers, and uses speaker diarization plus automatic AI labeling so the transcript attributes each statement to the right person by name and role (for example, clinician, patient, or parent). Group therapy, family sessions, and care conferences produce a transcript that reads like a script. The scribe works with Zoom, Teams, and Google Meet with no visible bot in the meeting, and BastionGPT reports 99.3% accuracy in a study by Baylor College of Medicine and Texas Children's Hospital researchers. Hathr AI has no scribe. The platform accepts typed prompts and uploaded documents; it does not record, transcribe, or diarize audio. Hathr's own API page lists "Digital Medical Scribes" as an application developers can build by automating "SOAP note generation from transcripts or bullet points," which means the transcript must come from a separate tool. A clinician who wants visit documentation from Hathr AI must transcribe the session elsewhere and paste the text in. ### Structured clinical notes, on purpose-built templates BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes as first-class output formats, and a Custom Note option accepts any free-form prompt per recording : narrative formats, parent-friendly summaries, and other specialty outputs. You can upload example notes so BastionGPT matches your writing style. Hathr AI generates notes through open prompting: you upload or paste text and describe what you want. Its marketing highlights chart summarization, pre-authorization drafting, and billing code suggestions, and those work well as chat tasks. It publishes no structured note templates, so the user carries the formatting burden on every request. ### A full AI assistant with shared prompts BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze uploaded records. A Saved Prompts library with folder organization and team sharing ships on every plan, so a practice can standardize documentation by sharing one vetted prompt set and inserting any prompt with a slash command. Hathr AI's chat covers general drafting within its Claude-only architecture, and its Data Library stores documents for repeatable workflows so you avoid re-uploading records for each conversation. Team-level prompt standardization is not a published feature; prompt development and AI coaching appear on Hathr's Enterprise plan. ### AI models you can see and choose BastionGPT names its engines: licensed, HIPAA-compliant versions of GPT, Claude, and Gemini . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Hathr AI runs exclusively on Anthropic's Claude, hosted on AWS GovCloud; its SourceForge listing cites Claude 3.5 Sonnet. Claude is a strong model, and for plain document summarization the two platforms perform comparably. A single-model architecture still ties every task to one vendor's strengths, weaknesses, and release schedule. Model leadership shifts month to month, and a multi-model platform keeps output quality steady across note drafting, data analysis, and complex reasoning. ### Document formats you can verify before you buy Both platforms handle long records, and document capacity is Hathr AI's strongest product feature: it advertises single documents of more than 500,000 words and unlimited uploads. BastionGPT publishes the specifics. Professional Plus and Ultra read 10+ documents at once with 1,000+ pages of combined context , across PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. Hathr AI publishes no supported-format list and does not advertise spreadsheet, image, or chart analysis. A clinician who needs to read a lab spreadsheet, a scanned fax, or a growth chart gets a published answer from BastionGPT and an open question from Hathr AI. ### Human support on every plan BastionGPT provides US-based human support by email, chat, phone, and video on every plan , backed by a 99.90% uptime guarantee. Hathr AI's pricing FAQ advertises 24/7 support, while its site footer lists business hours of Monday through Friday, 9am to 5pm ET, and its Capterra and GetApp profiles list email/help desk as the support channel. Real-time support and a 99.9% SLA appear as Enterprise plan features. For a solo practitioner whose scribe or chat tool fails mid-clinic, the difference between a phone call and an email queue matters the same day. ### Verified user reviews you can check BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp, plus a 5-star Google Business Profile rating, and is used by 10,000+ healthcare organizations. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. Hathr AI is listed on Capterra, GetApp, and SourceForge. As of July 2, 2026, its Capterra listing shows a single published review, and its GetApp and SourceForge profiles show zero verified user reviews. That is not evidence the product performs poorly; it means a buyer has little independent user feedback to check the marketing against. Ratings shift over time, so this page cites the source and date for each figure; check each directory for the live numbers. ### Published pricing for teams Hathr AI gates SSO, SAML, SCIM provisioning, EHR/EMR integrations, real-time support, and its SLA behind a custom-priced Enterprise plan, so a practice cannot budget those features from the website. BastionGPT publishes its organizational tier: Ultra at $75 per user per month with SSO, customizable security policies, and deployment inside your existing enterprise cloud. On every plan, BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and eClinicalWorks, through simple copy-paste workflows with no IT project. BastionGPT assists with documentation and analysis. Clinicians review and approve all AI-generated output; it does not replace professional judgment. ## Where Hathr AI stands out Hathr AI's clearest advantage is its federal compliance posture. The platform is hosted on AWS GovCloud in a FedRAMP High environment, conforms to NIST 800-171 and NIST 800-53 controls, uses FIPS 140-2 encryption with TLS 1.3, and states it can handle Controlled Unclassified Information at Impact Levels 2, 4, and 5. Hathr has been approved to join the Military-Civilian Health Ecosystem, is registered in SAM.gov, and was founded by a former military officer who advised national security organizations on AI deployment. Hathr AI also commits to signing BAAs within 24 hours, lets customers delete all their data immediately, and handles single documents beyond 500,000 words, a genuinely large ceiling for contract packages and policy libraries. If your procurement team requires FedRAMP High hosting or Impact Level 4 and 5 CUI handling (federal agencies, DoD contractors, VA-adjacent programs), Hathr AI is most often the appropriate platform between the two. BastionGPT runs on Microsoft Azure's HIPAA-compliant infrastructure, HITRUST CSF certified and SOC 2 Type II attested, which fits private practices, clinics, mental health groups, and health systems where HIPAA is the governing standard. For US healthcare professionals outside federal contracting, weigh the GovCloud hosting against what daily documentation requires. Audio scribing, structured notes, spreadsheet and image analysis, named AI models, and human support are the features a practice touches every day, and they are the features where BastionGPT leads at a lower price. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Hathr AI cost in 2026? Hathr AI sells one individual plan, the Single Subscription Webapp. The pricing page headline advertises $45 per month, while the plan card on hathr.ai/pricing and Hathr's GetApp listing both show $47 per month as of July 2, 2026. API access is priced separately: Base at $45 per month (listed as non-HIPAA-compliant API access), Starter at $225, Professional at $700, and Enterprise at $1,200 per month. SSO, EHR integrations, and real-time support require a custom-priced Enterprise plan. Is Hathr AI free? No. Hathr AI offers a 7-day free trial but no permanent free tier. BastionGPT also offers a 7-day free trial; paid plans start at $20 per user per month and carry a 45-day money-back guarantee. Does Hathr AI sign a BAA? Yes. Hathr AI includes a signed Business Associate Agreement with every plan and commits to signing within 24 hours. BastionGPT also includes a signed BAA on every plan, including the free trial. Does Hathr AI have an AI medical scribe? No. Hathr AI is a chat and document analysis platform; it does not record, transcribe, or diarize audio. Its API page lists digital medical scribes as an application developers can build from transcripts produced elsewhere. BastionGPT includes an AI Scribe with unlimited transcription on every plan, recognizing up to 10 speakers in sessions up to 240 minutes. What AI models power BastionGPT and Hathr AI? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Hathr AI runs exclusively on Anthropic's Claude, hosted on AWS GovCloud; its SourceForge listing cites Claude 3.5 Sonnet. How many pages or words can each platform analyze? BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra; the Professional plan covers a single document up to 30 pages. Hathr AI advertises single documents of more than 500,000 words and unlimited uploads, but does not publish a supported file format list. Can either platform analyze Excel spreadsheets or images? BastionGPT reads Excel, CSV, PowerPoint, HTML, PDF, Word, text, and JPG or PNG images, including scanned PDFs, charts, and graphs, on Professional Plus and above. Hathr AI does not advertise spreadsheet or image analysis and does not publish supported upload formats. Is Hathr AI FedRAMP authorized? Hathr AI is hosted on AWS GovCloud in a FedRAMP High environment, conforms to NIST 800-171 and NIST 800-53 controls, and states it can handle Controlled Unclassified Information at Impact Levels 2, 4, and 5. Hosting in an authorized environment is different from a product holding its own FedRAMP authorization, so federal buyers should confirm authorization status during procurement. How do BastionGPT and Hathr AI user ratings compare? BastionGPT holds a 4.8 out of 5 average across the verified review directories Capterra, Software Advice, and GetApp. Hathr AI's Capterra listing shows a single published review, and its GetApp and SourceForge profiles show zero verified user reviews as of July 2, 2026. Ratings change over time, so check each directory listing for current figures. Which is better for US private practices, BastionGPT or Hathr AI? BastionGPT. It includes an unlimited AI scribe, structured SOAP, DAP, BIRP, and H&P notes, three named AI models, image and spreadsheet analysis, and US-based phone, chat, email, and video support on plans from $20 per user per month, versus $47 per month for Hathr AI's chat-only subscription. Which is better for federal or DoD work, BastionGPT or Hathr AI? Hathr AI. Its AWS GovCloud hosting in a FedRAMP High environment, NIST 800-171 and 800-53 conformance, and stated CUI Impact Level 2, 4, and 5 handling fit federal agencies, DoD contractors, and VA-adjacent work in a way most commercial healthcare AI tools do not. That recommendation is unqualified. Does Hathr AI support languages other than English? Hathr AI's SourceForge listing states English language support. BastionGPT supports transcription in 15+ languages. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/microsoft-copilot # Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) Jump to Comparison Table Microsoft 365 Copilot and BastionGPT both put generative AI to work for professionals, and the overlap largely ends there. "Microsoft Copilot" now names a family of products, so scope matters: this page compares BastionGPT with Microsoft 365 Copilot, the general productivity assistant built into Word, Excel, PowerPoint, Outlook, and Teams. Copilot is sold as an add-on: $30 per user per month for enterprise or $18 to $21 for small businesses, on top of a required Microsoft 365 license, which brings the true cost to roughly $35 to $90 per user per month. It includes no medical scribe and no clinical note templates; Microsoft sells its clinical documentation product, Dragon Copilot, separately through sales reps and partners on 12-month terms. BastionGPT is a healthcare AI platform. It pairs an unlimited AI scribe that recognizes up to 10 speakers and labels each speaker's name and role automatically with a full clinical AI assistant: SOAP, DAP, BIRP, and H&P notes, referral letters, insurance appeals, and analysis of 10+ documents with 1,000+ pages of context, running on named frontier models (GPT, Claude, and Gemini). Plans start at $20 per user per month with a signed BAA on every plan, including the free trial. This comparison covers the differences that matter when you choose: what each platform actually costs once base licenses are counted, what happens when a prompt contains patient data, where Microsoft's own documentation draws the HIPAA line, and what it takes to document a patient visit on each. Microsoft compliance scope and product details were verified against microsoft.com and learn.microsoft.com on July 3, 2026, and Microsoft pricing and trial terms were re-verified on August 16, 2026. The quick answer: Choose BastionGPT if your work touches patient data: it is built for healthcare, includes an unlimited 10-speaker AI scribe and a full clinical assistant, and signs a BAA on every plan from $20 per user per month, with no other license required. Microsoft 365 Copilot fits general office productivity for teams already on Microsoft 365; it costs $18 to $30 per user per month on top of a required Microsoft 365 license (about $35 to $90 all-in), offers a one-month free trial on its Business tier only, includes no medical scribe, and Microsoft's documentation states that HIPAA compliance does not apply to Copilot's web search queries because they fall outside the Data Protection Addendum and BAA. Microsoft's clinical documentation product, Dragon Copilot, is a separate purchase that independent analyses price in the hundreds of dollars per provider per month. For the deeper look at Copilot's BAA coverage and its exclusions, read Is Microsoft Copilot HIPAA Compliant? If you are looking for a Microsoft Copilot alternative built for healthcare, the sections below cover the specifics: true cost, BAA coverage, and clinical features Copilot does not include. BastionGPT is a Copilot alternative with a built-in medical scribe, SOAP notes, and a signed BAA at $20 per month. Our Claude Team and Hathr AI comparisons cover other routes to compliant AI. Feature BastionGPT Microsoft 365 Copilot Built for Healthcare: clinicians, practices, and teams handling PHI General office productivity inside Microsoft 365 Entry price $20/user/month (Professional), standalone $18 to $21/user/month (Copilot Business, under 300 users; $18 is 2026 promotional pricing for the first year) or $30/user/month (enterprise add-on), annual term Required base license None Qualifying Microsoft 365 plan required ($14/user/month Business Standard up to $60 E5) True all-in cost $20 to $75/user/month, all plans standalone About $35 to $90/user/month once the base license is counted Free option 7-day free full trial, plus a 45-day money-back guarantee One-month free trial of Copilot Business for organizations under 300 users; no general trial for the enterprise add-on Signed BAA ✓ Every plan, including the free trial Available via the Microsoft DPA on qualifying plans; web search queries are excluded from the BAA per Microsoft documentation HIPAA compliant Compliant by design Supported "for properly configured implementations" (Microsoft Learn); the practice carries the configuration burden Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP FERPA supported for configured implementations; web search queries excluded Medical scribe Unlimited AI Scribe on every plan: live or uploaded audio, up to 10 speakers, 240-minute sessions, automatic speaker name and role labeling None. Clinical scribing is a separate Microsoft product (Dragon Copilot) sold and priced separately Clinical note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress notes, custom free-form prompts No clinical note templates; general drafting in Word and Outlook AI models GPT, Claude, and Gemini, licensed and HIPAA-compliant, named and selectable OpenAI GPT-5.x family plus Claude; Microsoft notes Anthropic models are excluded from the EU Data Boundary Clinical accuracy evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers No published clinical accuracy studies for Microsoft 365 Copilot Documents per upload 10+ at once (Professional Plus and Ultra) Varies by app, license, and tenant; users on Microsoft's own forums report per-day file caps and document-length limits Document capacity 1,000+ pages of combined context No published page capacity; Microsoft guidance recommends shorter documents for best results Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG PDF, Office files, text, and images in Copilot Chat Deployment Minutes; works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) Microsoft 365 tenant, license prerequisites, and HIPAA hardening (disable web search, sensitivity labels, audit configuration) Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 4.6/5 on G2 across 14 reviews on its dedicated profile (July 2026); reviews are spread across multiple listings Support Email, chat, phone, and video with real humans, on every plan Standard Microsoft 365 support channels Uptime guarantee 99.9% 99.9% Microsoft 365 SLA ## Where BastionGPT and Microsoft 365 Copilot are similar Both platforms state that customer prompts and data are never used to train foundation AI models. Both run on Microsoft Azure infrastructure, encrypt data in transit and at rest, respect existing permissions, and give administrators control over retention. Both draft, summarize, rewrite, and analyze documents and spreadsheets, and both offer Claude and OpenAI models under the hood. If your entire requirement is drafting an email or summarizing a business document with no patient information in sight, both tools cover that. The differences appear the moment a prompt contains PHI, a patient visit needs a note, a 300-page record needs review, or a solo practitioner needs a BAA without an enterprise agreement. ## Where BastionGPT goes further ### Built for clinical work, not adapted to it Microsoft 365 Copilot is a general productivity assistant. It drafts documents in Word, builds formulas in Excel, and summarizes meetings in Teams. It ships with no clinical note templates, no medical scribe, and no healthcare-specific workflows; Microsoft's healthcare documentation products live in a separate division under the Dragon Copilot brand. BastionGPT was developed by healthcare and cybersecurity professionals for clinical documentation. It produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes, drafts referral letters, insurance appeal letters, treatment plans, and patient education materials, reviews documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding, and analyzes lab summaries and spreadsheets. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Microsoft has published no comparable clinical accuracy study for Microsoft 365 Copilot, and an independent 2025 academic study of a Microsoft 365 Copilot workplace trial reported mixed experiences after initial enthusiasm, including unmet expectations around deeper contextual understanding and workflow integration. ### A medical scribe is included, not a separate product This is a wide gap between the two platforms. Microsoft 365 Copilot has no medical scribe. Copilot in Teams can recap a Teams meeting, and that is a different job from documenting an in-person patient visit as a structured clinical note. Microsoft's actual scribe, Dragon Copilot, is a separate product with separate licensing, sold through Microsoft representatives and Cloud Solution Provider partners on 12-month terms. Microsoft does not publish Dragon Copilot pricing; independent analyses in 2026 place per-provider licenses at $369 to $830+ per provider per month , and Microsoft's own licensing guidance meters AI-assisted sessions on its Flex and pay-as-you-go models. BastionGPT includes an unlimited AI Scribe on every plan , starting at $20 per user per month. It transcribes live audio or uploaded recordings for sessions up to 240 minutes, recognizes up to 10 speakers, and automatically labels each speaker's name and role (for example, clinician, patient, or parent) in the transcript. Group therapy, family sessions, and care conferences produce a transcript that reads like a script, and the resulting notes attribute statements to the right person. ### Price Microsoft 365 Copilot is an add-on, so the sticker price is only part of the cost. The enterprise add-on runs $30 per user per month on an annual term; Microsoft 365 Copilot Business, for organizations under 300 users, runs $21 per user per month ($18 under promotional pricing in 2026). Either way, a qualifying Microsoft 365 base license is required, from Business Standard at $14 per user per month to E5 at $60 following Microsoft's July 2026 price update. Independent licensing analyses put the true all-in cost at roughly $35 to $90 per user per month, and none of that includes clinical scribing. BastionGPT Professional starts at $20 per user per month, standalone . Professional Plus at $45 adds the full document analysis suite: 10+ simultaneous documents, image and spreadsheet reading, and manual model selection. No base license, no add-on stack, and annual billing includes a free month. The math for a five-clinician practice that needs both an assistant and a scribe: the Microsoft stack (Business Standard at $14, Copilot Business at $21, and Dragon Copilot at the $369 low end of independently reported pricing) runs about $404 per provider per month, or roughly $24,200 per year. BastionGPT Professional covers both jobs at $20 per provider per month, or $1,200 per year, and Professional Plus adds full document analysis at $45, or $2,700 per year. That is a difference of more than $21,000 per year for a five-clinician practice , before Dragon Copilot implementation fees. ### Compliance that holds from the first prompt Microsoft supports HIPAA compliance for Microsoft 365 Copilot "for properly configured implementations," and its documentation is explicit about the boundary: HIPAA compliance does not apply to Copilot's web search queries, because web queries sent to Bing are not covered by the Data Protection Addendum or the BAA. Copilot can invoke web search while answering a prompt, so healthcare deployment guides recommend disabling web search tenant-wide, applying sensitivity labels, scoping Graph access, and enabling audit logging before clinicians touch it. Consumer Copilot surfaces (the free copilot.microsoft.com, the personal mobile app) sit entirely outside the BAA, and Microsoft notes that Anthropic models are currently excluded from the EU Data Boundary. BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , and supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements out of the box, with no configuration project. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, is independently penetration tested at least annually, and chats and transcripts are wiped after 30 days by default. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information . ### Document analysis depth On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. Microsoft 365 Copilot handles file uploads, and its limits shift by app, license, and tenant configuration. The clearest numbers Microsoft publishes sit in Copilot Notebooks: licensed Microsoft 365 Copilot users can add more than 300 references to a notebook, with the first 300 used to ground responses, while Copilot Chat users can add up to 50. For chat uploads, Microsoft publishes no page capacity, its support guidance recommends shorter documents for best results, and users on Microsoft's own Q&A forums report per-day file caps and document-length constraints that vary between tenants. For a practice whose core task is reviewing long records, that variability is the difference between one conversation and a manual splitting exercise. ### Deployment without an IT project Deploying Microsoft 365 Copilot for healthcare means an eligible Microsoft 365 tenant, add-on licenses, a permissions audit (Copilot surfaces anything a user technically has access to), sensitivity labels, disabled web search, and audit configuration. Microsoft sells governance tooling under its Copilot Control System banner to address oversharing and insider risk, which signals how much of the rollout work sits with the customer. Microsoft's guidance and independent HIPAA deployment guides describe this as a governance program, and small practices rarely have a compliance engineer on staff. BastionGPT deploys in minutes from a browser. It works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) through copy-paste and document uploads, requires no Microsoft licensing prerequisites, and includes human support by email, chat, phone, and video on every plan. A Saved Prompts library with folder organization and team sharing lets a practice standardize documentation across providers from day one. BastionGPT assists with documentation and analysis. Clinicians review and approve all AI-generated output; it does not replace professional judgment. ## Where Microsoft 365 Copilot stands out Copilot's clearest advantage is its home field: it works inside Word, Excel, PowerPoint, Outlook, and Teams, grounded in your organization's own Microsoft Graph data (emails, files, meetings, and chats). The paid license adds reasoning agents (Researcher, Analyst, and Facilitator) that produce cited research reports and structured data analysis, connections to more than 100 third-party systems through its connector ecosystem, and Copilot Studio for building custom internal agents. For general office work that never touches patient data (board decks, budget spreadsheets, inbox triage), an organization already standardized on Microsoft 365 could get real value from that integration. The fairest one-line summary: Microsoft 365 Copilot is broad, and BastionGPT is deep. Copilot is the horizontal productivity layer across an office suite; BastionGPT is the vertical clinical layer built for patient-facing documentation. The two platforms can work side by side. BastionGPT runs alongside Microsoft 365 rather than replacing it: many practices keep Microsoft 365 for Office apps and email, skip the Copilot add-on for clinical staff, and give clinicians BastionGPT for every task that involves patient information or where users prefer to choose specific models like Gemini Pro which is not available from Microsoft 365 Copilot. For US practices whose documentation work centers on patient care, BastionGPT delivers a scribe, a clinical assistant, and day-one BAA coverage at $20 to $45 per user per month, less than the Copilot add-on alone once the base license is counted. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs Is Microsoft Copilot HIPAA compliant? Partially, with conditions. Microsoft states that Microsoft 365 Copilot supports HIPAA compliance for properly configured implementations, and that HIPAA compliance does not apply to web search queries because they are not covered by the Data Protection Addendum and BAA. Consumer Copilot surfaces, such as the free copilot.microsoft.com and the personal mobile app, are outside the BAA entirely. BastionGPT is HIPAA compliant by design, with a signed BAA on every plan and no configuration project required. Does Microsoft sign a BAA for Copilot? Yes, for qualifying commercial customers. Microsoft's HIPAA Business Associate Agreement is available through its Data Protection Addendum and can cover Microsoft 365 Copilot on eligible plans, but coverage applies at the service level: web search queries fall outside the BAA, and consumer Copilot products are not covered. BastionGPT includes a signed BAA on every plan, including the free trial. How much does Microsoft 365 Copilot cost in 2026? The enterprise add-on is $30 per user per month on an annual term, and Microsoft 365 Copilot Business for organizations under 300 users is $21 per user per month ($18 under 2026 promotional pricing, first year only). A qualifying Microsoft 365 base license is also required, from $14 (Business Standard) to $60 (E5) per user per month after Microsoft's July 2026 price update, so the true all-in cost is roughly $35 to $90 per user per month. BastionGPT plans run $20 to $75 per user per month, standalone. Does Microsoft 365 Copilot include a medical scribe? No. Microsoft 365 Copilot is a general productivity assistant; Copilot in Teams recaps Teams meetings, which is different from documenting a patient visit. Microsoft's clinical scribe is Dragon Copilot, a separate product sold through Microsoft representatives and partners on 12-month terms. BastionGPT includes an unlimited AI scribe on every plan, with up to 10 speakers and 240-minute sessions. What is Microsoft Dragon Copilot and how much does it cost? Dragon Copilot is Microsoft's AI clinical documentation product, the successor to Nuance DAX Copilot and Dragon Medical One. Microsoft does not publish its pricing; independent analyses in 2026 place per-provider licenses at $369 to $830+ per provider per month, sold through Microsoft representatives and Cloud Solution Provider partners on 12-month terms, with metered AI-assisted sessions on Flex and pay-as-you-go models. BastionGPT includes scribing and a clinical assistant from $20 per user per month. Can Microsoft Copilot write SOAP notes? Not with built-in templates. Microsoft 365 Copilot ships no clinical note formats; a clinician would need to hand-build prompts in Word and stay inside a HIPAA-hardened tenant configuration to use patient data at all. BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes out of the box, plus custom free-form formats, and can match your writing style from example notes. Do I need Microsoft 365 to use BastionGPT? No. BastionGPT is standalone: it runs in a browser, requires no base license, and works alongside Microsoft 365, Google Workspace, and 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) through copy-paste and document uploads. Microsoft 365 Copilot cannot be purchased without a qualifying Microsoft 365 subscription. What AI models power BastionGPT and Microsoft 365 Copilot? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model per task automatically, with manual selection on Professional Plus and above. Microsoft 365 Copilot runs on OpenAI's GPT-5.x family and added Claude as a model choice in 2026, and Microsoft notes that Anthropic models are currently excluded from the EU Data Boundary and in-country processing commitments. Does Microsoft 365 Copilot have a free trial? Yes, on the Business tier. Since July 2026 Microsoft has offered a one-month free trial of Microsoft 365 Copilot Business to organizations with fewer than 300 users, started from Copilot Chat, after which the subscription converts to paid unless it is cancelled. The enterprise add-on has no general trial, and the free Copilot Chat experience remains a separate, more limited product. BastionGPT offers a 7-day free trial of its full Professional and Professional Plus plans, with a signed BAA active during the trial and a 45-day money-back guarantee on paid plans. How many documents and pages can BastionGPT analyze at once? On Professional Plus and Ultra, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context, across PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. Microsoft 365 Copilot's upload limits vary by app, license, and tenant; Microsoft publishes no page capacity and recommends shorter documents for best results. Can Microsoft 365 Copilot analyze patient records? Only inside a properly configured HIPAA deployment. That means an eligible plan under Microsoft's BAA, web search disabled (web queries are excluded from the BAA), sensitivity labels, scoped permissions, and audit logging, which Microsoft and independent deployment guides describe as the customer's responsibility. BastionGPT is built for exactly this task: upload records on any plan under a signed BAA and analyze up to 1,000+ pages at once. Which is better for a small practice, BastionGPT or Microsoft Copilot? For clinical work, BastionGPT. A five-clinician practice replicating BastionGPT's scribe-plus-assistant capability on the Microsoft stack (base license, Copilot add-on, and Dragon Copilot at independently reported pricing) would spend roughly $404+ per provider per month versus $20 for BastionGPT Professional or $45 for Professional Plus, a difference of more than $21,000 per year. Microsoft 365 Copilot remains a fit for general office productivity that never involves patient data. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/claude-team-enterprise # Claude Alternative for Healthcare: BastionGPT vs Claude Team Jump to Comparison Table Claude Team and Claude Enterprise give organizations access to Anthropic's Claude models, one of the strongest general-purpose AI assistants available in 2026. BastionGPT gives healthcare professionals the same class of frontier models, including Claude, inside a platform built for clinical work. For a practice that handles patient information, the deciding question is not model quality. It is whether the plan you can actually buy is permitted to touch protected health information, and what it costs to get there. Anthropic's privacy documentation states that its Business Associate Agreement does not cover Claude Free, Pro, Max, or Team plans. HIPAA-ready configuration is available on Claude Enterprise plans only, self-serve from a 20-seat minimum at $20 per seat per month billed annually, or sales-assisted from 50 seats, with usage billed at API rates on top of seat fees. BastionGPT includes a signed BAA on every plan, including the 7-day free trial, starting at $20 per user per month, and adds an unlimited AI medical scribe that Claude does not offer on any plan. Claude plan, pricing, BAA, and upload details were re-verified against claude.com, support.claude.com, and privacy.claude.com on August 16, 2026. The quick answer: Choose BastionGPT if your work touches patient information. Every plan includes a signed BAA from $20 per user per month, an unlimited 10-speaker AI scribe, clinical note formats (SOAP, DAP, BIRP, H&P), analysis of 10+ documents and 1,000+ pages at once, and named frontier models (GPT, Claude, and Gemini). Choose Claude Team or Enterprise for general business and engineering work that never involves PHI, or if you are a 20+ seat organization ready for an Enterprise plan with usage billed on top of seat fees. Anthropic's BAA excludes its Free, Pro, Max, and Team plans entirely, and no Claude plan includes a medical scribe. For the deeper compliance analysis of Anthropic's consumer and business plans, including what the BAA excludes, read Is Claude HIPAA Compliant? If you are looking for a Claude alternative that works for healthcare, BastionGPT gives you the same Claude models, plus GPT and Gemini, behind a signed BAA at $20 per user per month with no seat minimum. The sections below compare BAA coverage, entry costs, and what each platform includes for clinical work. Our Microsoft Copilot and Hathr AI comparisons cover other routes to compliant AI. Feature BastionGPT Claude Team & Enterprise Built for Healthcare and mental health professionals General-purpose work across every industry Signed BAA ✓ Every plan, including the 7-day free trial Team: not available. Enterprise only, self-serve or sales-assisted, accepted by the Primary Owner in organization settings (a one-way setting) Plans usable with PHI All plans HIPAA-ready Enterprise only. Anthropic's BAA excludes Free, Pro, Max, and Team, plus features such as Cowork even on Enterprise Minimum users 1 Team: 2 to 150 seats. Enterprise self-serve: 20. Enterprise sales-assisted: 50 Entry paid plan $20/user/month (Professional), BAA included Team Standard: $25/user/month ($20 billed annually); cannot be used with PHI Price of the HIPAA-eligible plan $20 to $75/user/month, flat Enterprise: $20/user/month billed annually from 20 seats self-serve, plus all usage billed at API rates; sales-assisted pricing is custom Usage model Flat monthly price; normal human use is not metered Team: weekly per-seat limits (Standard is 1.25x Pro per session). Enterprise: no caps, but every token is billed at API rates AI medical scribe ✓ Unlimited on every plan; up to 10 speakers; 240-minute sessions; automatic speaker name and role labels Not offered on any plan Clinical note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt No clinical templates; users build every note format from scratch AI models GPT, Claude, and Gemini (named and selectable) Anthropic Claude models only Documents per upload 10+ at once (Professional Plus and Ultra) Up to 20 files per chat, 30MB per file Document capacity 1,000+ pages of combined context 200K-token context window on Team (roughly 500 pages / 150,000 words), shared with conversation history Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG (with OCR for scans) PDF, Word, text, CSV, code files, and images EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) via copy-paste workflows No EHR integrations published; connectors cover Google Drive, Gmail, Google Calendar, GitHub, Microsoft 365, and Slack Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP SOC 2 Type II and ISO 27001 (Anthropic Trust Center); HIPAA support only through the Enterprise BAA path Model training on your data Never; inputs, documents, and audio are not used for AI training Not by default on Team and Enterprise (commercial) plans Deployment Sign up online; runs in minutes with no IT project Team is self-serve; HIPAA-ready Enterprise requires an Enterprise plan, Primary Owner activation, and an Implementation Guide Support Email, chat, phone, and video with real humans, on every plan Help center and ticket support; dedicated customer success on sales-assisted Enterprise Verified healthcare user ratings 4.8/5 average across Capterra, Software Advice, and GetApp (June 2026) Not listed as a healthcare software product on the major healthcare review directories ## Where BastionGPT and Claude Team/Enterprise are similar Both platforms give teams frontier-model AI through a browser with no infrastructure to manage, and both deploy without weeks of IT work for their standard configurations. Both state that business customer data is not used to train AI models by default, both encrypt data in transit and at rest, and both offer single sign-on at their upper tiers (Claude Team and above; BastionGPT Ultra, with SSO available as an add-on on Professional Plus for 50+ user practices). There is also real overlap in the underlying intelligence: BastionGPT licenses Claude models alongside the latest GPT and Gemini models, so the writing and reasoning quality that draws people to Claude is available inside BastionGPT. The differences appear the moment patient information enters the workflow, a session needs to be transcribed, or a practice with fewer than 20 people prices a compliant contract. ## Where BastionGPT goes further ### A signed BAA on every plan This is the difference that matters on day one. Anthropic's Privacy Center is explicit that its BAA does not cover the Free, Pro, Max, or Team plans, and that a list of features stays outside the BAA even on HIPAA-ready Enterprise. A clinician who pastes patient details into Claude Team has sent PHI to a vendor with no BAA in place, which HIPAA treats as an impermissible disclosure and a violation by definition, regardless of how carefully the organization otherwise behaves. BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , and supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information . ### HIPAA access without a 20-seat contract Anthropic does offer a HIPAA path, and it is well documented: a Claude Enterprise plan, where the Primary Owner enables HIPAA-ready configuration in organization settings and accepts the BAA (a decision Anthropic notes cannot be reversed from admin settings). As of August 2026 that path is open on both self-serve Enterprise, which carries a 20-seat minimum, and sales-assisted Enterprise, which carries a 50-seat minimum . Anthropic's help center states the feature is available for Enterprise plans only; Team and individual plans cannot enable HIPAA at all. Anthropic's billing documentation adds that HIPAA-ready organizations are an exception to its current single-seat Enterprise billing model and are provisioned with separate Chat and Chat + Claude Code seat types. Buying compliant Claude still means buying an Enterprise plan, not a standard checkout. The math for a five-clinician practice: BastionGPT Professional Plus runs $2,700 per year ($45 x 5 x 12), or $1,200 on the $20 Professional plan. The Claude HIPAA path requires buying at least 20 seats. At Anthropic's published Enterprise seat rate of $20 per user per month billed annually, 20 seats come to $4,800 per year in seat fees before a single token of usage is billed , and every token is then billed at API rates on top. A five-person practice would be paying for 15 seats it cannot fill. ### Predictable pricing Claude's two organizational plans meter usage in different ways. Team seats carry weekly per-seat limits (a Standard seat offers 1.25x the Pro plan's usage per five-hour session), with prepaid usage credits available once a member hits the cap. Enterprise removes the caps but bills every token of chat, Claude Code, and Cowork usage at API rates on top of the seat fee, so monthly cost moves with how much your team actually types, uploads, and generates. Anthropic's pricing pages also note that prices and plans are subject to change at its discretion. BastionGPT charges a flat monthly price per user, and normal human use is not metered . Prices have held steady for over two years, and annual billing includes a free month. A practice can budget its AI spend for the year on one line. ### An AI medical scribe is included No Claude plan includes an ambient medical scribe: there is no session recording, no speaker diarization, and no visit-to-note pipeline. BastionGPT includes an unlimited AI Scribe on every plan . It recognizes up to 10 speakers in a single session, live or from uploaded audio, for sessions up to 240 minutes, and works with Zoom, Microsoft Teams, and Google Meet telehealth visits without a visible bot participant. Speaker diarization separates the transcript by voice, and BastionGPT's AI labels each speaker's name and role automatically (for example, clinician, patient, or parent), so group therapy, family sessions, and care conferences produce notes that attribute statements to the right person. ### Clinical documentation built in BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes out of the box, and a Custom Note tab accepts any free-form prompt per recording: narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can upload example notes so BastionGPT matches your writing style, and a Saved Prompts library with folder organization and team sharing is included on every plan. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Claude is a genuinely capable writer, but it ships with no clinical templates. Every note format, every referral letter structure, and every documentation standard must be built by your team from scratch and maintained in shared Projects. ### The same Claude models, under a healthcare BAA BastionGPT names its engines: licensed and HIPAA-compliant versions of GPT, Claude, and Gemini . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. If Claude's model quality is what drew you to Anthropic's plans, that quality is available inside BastionGPT with a signed BAA at $20 per user per month, with no Enterprise seat minimum to clear. Claude's plans provide Anthropic models only, so cross-checking a hard case against GPT or Gemini models means leaving the platform. ### Document workflows sized for patient records On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context , across PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image formats, including OCR for scanned records. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query and cross-reference all of it in one conversation. Claude's chat interface accepts up to 20 files per conversation at 30MB per file, and the Team plan works within a 200K-token context window (roughly 500 pages or 150,000 words) that is shared with the entire conversation history. Anthropic's documentation also notes that PDFs over 100 pages lose visual analysis and PDFs over 1,000 pages are processed as text only. These are strong general-purpose limits, and none of it may be used with PHI on the Team plan in the first place. ### Healthcare-appropriate content handling Anthropic's Usage Policy classifies healthcare uses that provide medical advice, patient care, therapy, or mental health guidance as High-Risk Use Cases requiring qualified-professional review and AI disclosure when output reaches individuals, and it states Anthropic may block or modify outputs that violate its policy, including sexually explicit content. Those guardrails are reasonable for a general-purpose product. In a clinical setting, they can translate into friction: legitimate documentation around trauma, abuse, substance use, anatomy, and sexual health can trip consumer-grade filters. BastionGPT includes healthcare-appropriate content filtering designed for clinical environments, so clinicians can document the full scope of clinical language, including behavioral health, forensic, and crisis documentation, without false flags. ### Human support on every plan BastionGPT includes email, chat, phone, and video support with real humans on every plan . On Claude's plans, support runs through the help center and tickets, with dedicated customer success management available on sales-assisted Enterprise contracts. BastionGPT assists with documentation and analysis. Clinicians review and approve all AI-generated output; it does not replace professional judgment. ## Where Claude Team and Enterprise stand out Claude's clearest advantage is breadth outside the clinic. Team and Enterprise plans include Claude Code for software engineering, Cowork for agentic file-and-task work, enterprise search across connected sources, and connectors to Google Drive, Gmail, Google Calendar, GitHub, Microsoft 365, and Slack. Enterprise adds serious IT governance: SCIM provisioning, audit logs, a compliance API, custom data retention controls, customer-managed encryption keys, and US-only inference. For a 20+ seat organization with IT and legal teams, and for the large share of work that never touches PHI, Claude Enterprise is a strong general-purpose platform. Two caveats matter even for organizations that take that path. First, Anthropic publishes a list of services that stay outside the BAA even after HIPAA is enabled: Cowork is not an eligible service in any configuration, Claude Code is covered only in local mode with zero data retention on qualified accounts, and beta features such as Claude Design and Claude for Office are excluded, so parts of the platform remain off-limits for PHI. Second, no Claude plan transcribes patient visits. That is why many organizations run BastionGPT for clinical documentation alongside Claude for general business and engineering work. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs Does Anthropic sign a BAA for Claude Team plans? No. Anthropic's Privacy Center states that its Business Associate Agreement does not cover Claude Free, Pro, Max, or Team plans. A BAA is available only for Anthropic's HIPAA-ready services: the first-party API and Claude Enterprise plans, where the Primary Owner enables HIPAA-ready configuration and accepts the BAA in organization settings. BastionGPT includes a signed BAA on every plan, including the free trial. Is Claude HIPAA compliant? Only on Enterprise. HIPAA-ready Claude requires a Claude Enterprise plan (self-serve from 20 seats, or sales-assisted from 50 seats), activation by the Primary Owner, and acceptance of Anthropic's BAA, and even then several features (Cowork among them) are not eligible services under the BAA. Claude Free, Pro, Max, and Team cannot enable HIPAA. BastionGPT is HIPAA compliant on every plan with a BAA included. How much does Claude Team cost in 2026? Claude Team Standard seats cost $25 per user per month, or $20 billed annually, for teams of 2 to 150 seats, with weekly per-seat usage limits (1.25x the Pro plan per session). Premium seats cost $125 per user per month, or $100 billed annually. Team plans cannot be used with protected health information because Anthropic's BAA excludes them. BastionGPT plans run $20 to $75 per user per month with a BAA on every plan. How much does Claude Enterprise cost in 2026? Claude Enterprise self-serve costs $20 per seat per month billed annually with a 20-seat minimum, and all usage is billed separately at API rates on top of seat fees. Sales-assisted Enterprise has custom pricing and a 50-seat minimum. HIPAA-ready configuration and a BAA are available on both, and on no other Claude chat plan. What is the minimum seat count for Claude's HIPAA-ready plan? 20 seats. Anthropic's Enterprise documentation lists a 20-seat minimum for self-serve Enterprise and a 50-seat minimum for sales-assisted Enterprise, and HIPAA readiness with a BAA is available on both. Team and individual plans cannot enable it at all. BastionGPT has no seat minimum; a single clinician can sign up with a BAA included from $20 per month. What does HIPAA-compliant Claude cost for a small practice? The HIPAA path requires a Claude Enterprise plan, which starts at 20 seats on the self-serve tier. At Anthropic's published Enterprise seat rate of $20 per user per month billed annually, 20 seats total $4,800 per year in seat fees before any usage billing, and usage is then billed separately at API rates. A five-clinician practice on BastionGPT Professional Plus pays $2,700 per year with a BAA included. Does Claude have an AI medical scribe? No. Claude plans do not include session recording, speaker diarization, or a visit-to-note pipeline. BastionGPT includes an unlimited AI Scribe on every plan that recognizes up to 10 speakers per session, records live or from uploaded audio for up to 240 minutes, works with Zoom, Microsoft Teams, and Google Meet, and automatically labels each speaker's name and role in the transcript. Can Claude Team analyze patient records? Not compliantly. Claude Team is excluded from Anthropic's BAA, so uploading records that contain PHI to a Team plan is a HIPAA violation by definition. Technically, Claude chat accepts up to 20 files at 30MB each inside a 200K-token context window. BastionGPT analyzes 10+ documents and 1,000+ pages at once under a signed BAA on every plan. What AI models power BastionGPT and Claude? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model per task automatically; Professional Plus adds manual model selection and early access to new models. Claude Team and Enterprise provide Anthropic's Claude models only. Does Anthropic train on Claude Team and Enterprise data? No. Anthropic states it does not train its models on Claude for Work (Team and Enterprise) data by default. Its consumer plans (Free, Pro, Max) handle training permissions through user settings. BastionGPT never uses customer inputs, documents, or audio for AI training on any plan. Which features are excluded from Anthropic's BAA even on HIPAA-ready Enterprise? Anthropic publishes a list, and it is longer than most buyers expect. Cowork is not an eligible service in any configuration; Claude Code is covered only in local mode with zero data retention on qualified accounts, while its web, remote-desktop, review and security surfaces are excluded; and beta features such as Claude Design and Claude for Office sit outside the BAA. Several API surfaces are also excluded, including Batch, Files, Skills, Code Execution, Computer Use and Web Fetch, and connectors, Enterprise Search and Claude in Chrome are covered except for data sent to third parties. Check Anthropic's current list before deployment. BastionGPT's BAA covers the platform's assistant, scribe, and document features on every plan. Which is better for a small or mid-size practice, BastionGPT or Claude? BastionGPT fits practices under 20 users by default: a signed BAA from $20 per user per month with no seat minimum, an unlimited AI scribe, clinical note formats, and phone, chat, email, and video support. Claude's HIPAA path starts at a 20-seat Enterprise plan, so a smaller practice buys seats it cannot fill and pays usage at API rates on top. Can I get Claude's AI models with a BAA without an Enterprise contract? Yes. BastionGPT licenses Claude models alongside the latest GPT and Gemini models and includes a signed BAA on every plan from $20 per user per month, so clinicians get Claude-class intelligence in a healthcare-compliant platform with no Enterprise seat minimum to clear. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/freed # Freed Alternative: BastionGPT vs Freed AI Scribe (2026) Jump to Comparison Table Freed (often searched as Freed AI) and BastionGPT both offer HIPAA-compliant AI scribes that turn patient visits into structured clinical notes about a minute after the encounter ends. The overlap largely ends there. Freed is an ambient scribe built around the visit: it records the encounter, writes the note, transcribes in 90+ advertised languages, and, on its $119 per month Premier plan, pushes the note into browser-based EHRs and suggests ICD-10 codes. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, what happens beyond the visit note, how document handling compares, what each vendor's terms say about your data, and what compliance coverage you get at each price. Freed's pricing, plan limits, and Terms of Use were verified against getfreed.ai and help.getfreed.ai on July 3, 2026. If you are weighing Freed alternatives, this comparison gives you the specifics that matter: pricing, scribe capability, document handling, data terms, and compliance depth. BastionGPT is a strong Freed alternative for clinicians who want a full AI assistant with the scribe included at $20 per month. For the wider field, our AI scribe comparison covers other options such as Heidi Health, DeepScribe, and Suki. The quick answer: Choose BastionGPT for the scribe economics alone: unlimited scribe notes on every plan from $20 per user per month, where Freed's $39 Starter caps you at 40 notes per month and unlimited notes start at $79. The scribe recognizes up to 10 speakers with automatic name and role labeling, records sessions up to 240 minutes, and holds a peer-reviewed 99.3% accuracy result — and every plan adds the full AI assistant (document analysis, referral letters, insurance appeals) that Freed does not offer. Freed fits solo clinicians whose documentation needs stop at the visit note, who want one-click note push into a browser-based EHR (a feature of Freed's $119 per month Premier plan), or who regularly document in languages beyond BastionGPT's 15+ (Freed advertises 90+). One hard line to know: Freed's own Terms of Use prohibit processing 42 CFR Part 2 substance use disorder records on its platform, while BastionGPT supports 42 CFR Part 2 alongside HIPAA and FERPA. Feature BastionGPT Freed Entry paid plan $20/user/month (Professional) $39/user/month (Starter; capped at 40 notes per month) Full-featured individual plan $45/user/month (Professional Plus) $119/user/month (Premier; $104 billed annually) Unlimited scribe notes Every plan, starting at $20 Starts at Core, $79/user/month; Starter is capped at 40 notes Free option 7-day free full trial, plus a 45-day money-back guarantee 7-day free trial; no permanent free tier or published money-back guarantee Signed BAA ✓ Every plan, including the free trial Template BAA embedded in the click-through Terms of Use; signed copy on request; organization-wide BAA listed as a Groups plan feature HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP HITECH, SOC 2 Type II; Terms of Use prohibit processing 42 CFR Part 2 data AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Not disclosed Full AI assistant Unlimited general healthcare assistant on every plan: referral letters, appeals, patient education, document review Visit-anchored clinician assistant; note editing from Core, medical knowledge base from Premier Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra); PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Accepts visit audio, uploaded recordings, and transcripts (files up to 256 MB), plus example notes for templates; no general document analysis Multi-speaker recognition Up to 10 speakers, with diarization and automatic AI labeling of speaker names and roles Separates speakers and handles accents; no published speaker count; independent 2026 reviews report limited diarization for multi-party visits Max recording length 240 minutes per session No published limit; audio for visits left paused over 14 days is not retained Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published; accuracy reports come from user reviews Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Specialty templates across 30+ specialties; template builder from Core; learned formats Saved prompts with team sharing Saved Prompts library with folder organization and team sharing on every plan Custom templates are per clinician on individual plans; team template sharing is a Groups plan feature Coding support Medical coding suggestions and under- or over-coding review through the assistant, on every plan ICD-10 suggestions on Premier only; CPT in beta EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) Copy-paste on all plans; one-click EHR push to browser-based EHRs on Premier (Athenahealth, Practice Fusion, SimplePractice, Elation, Tebra) AI training on your data Customer data is never used to train AI models States identifiable PHI is not used for training; Terms permit model training on customer data with consent via settings, and de-identified data use and disclosure at Freed's discretion Languages 15+ 90+ advertised, with code-switching and automatic detection; notes are generated in English Uptime guarantee 99.90% No published SLA Support Email, chat, phone, and video with real humans, on every plan Chat, email, and live support calls Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 4.8/5 on the Apple App Store ## Where BastionGPT and Freed are similar Both platforms offer HIPAA-compliant ambient transcription, generate structured clinical notes within a minute or two of ending a visit, and deploy in minutes without an IT project. Both run on Microsoft Azure infrastructure with US-based data storage, encrypt data in transit and at rest, learn your note format from your edits, and work for in-person and telehealth visits. Both state that identifiable patient data is never used to train AI models. If your entire requirement is turning a spoken visit into a standard note, both tools cover that core workflow. The differences appear the moment you price full access, upload a patient record, run a group session, document substance use disorder treatment, or need anything beyond the visit note. ## Where BastionGPT goes further ### Price Freed's current lineup: Starter at $39 per user per month, capped at 40 notes per month; Core at $79 with unlimited notes and an editing assistant; Premier at $119 per month ($104 billed annually) with EHR push, ICD-10 suggestions, visit summaries, and the medical knowledge assistant; and custom Groups pricing. Freed's help center notes that small groups of around four clinicians are billed at the standard individual rate, with custom pricing starting at 10 or more clinicians. BastionGPT Professional starts at $20 per user per month with the unlimited AI scribe and unlimited AI assistant included. Professional Plus, the closest match to Freed's Premier tier, costs $45 and adds multi-document analysis, image and spreadsheet reading, and manual model selection. Prices have held steady for over two years, and annual billing includes a free month. The math for a five-clinician practice: Freed Premier runs $7,140 per year at monthly billing ($119 x 5 x 12), or $6,240 billed annually. BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12). That is a difference of $4,440 per year , or $3,540 if Freed is billed annually. Even Freed's mid-tier Core plan, without EHR push or coding, costs $4,740 per year for the same five clinicians: $2,040 more than BastionGPT Professional Plus. ### A scribe built for group sessions BastionGPT's AI Scribe recognizes up to 10 speakers in a single session , live or from uploaded audio, for sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI then labels each speaker's name and role automatically (for example, clinician, patient, or parent) directly in the transcript. Group therapy, family sessions, care conferences, and multi-provider visits produce a transcript that reads like a script instead of a wall of text. Freed separates speakers, handles accents, and offers an opt-in Voice ID that recognizes the clinician's own voice for attribution. It publishes no speaker count and offers no automatic labeling of speaker names and roles, and independent 2026 reviews report that advanced diarization for group visits and multi-party encounters is limited. For practices that run family or group sessions, that difference shows up in every transcript. Accuracy is documented, not asserted: BastionGPT earned 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Freed publishes no peer-reviewed accuracy study. ### 42 CFR Part 2 and FERPA coverage This difference is easy to miss and expensive to discover late. Freed's Platform Terms of Use state that the platform "has not been designed to process Part 2 Data" and require users to agree not to process any 42 CFR Part 2 data, with Freed disclaiming liability if they do. 42 CFR Part 2 covers records from federally assisted substance use disorder programs, a common record type in mental and behavioral health, addiction medicine, and integrated primary care. BastionGPT supports HIPAA, 42 CFR Part 2, and FERPA , along with PIPEDA and PHIPA (Canada) and APP (Australia). Practices that touch SUD treatment records or school-based health records can use the same platform for all of their documentation instead of carving out exceptions. ### A full AI assistant, on every plan Freed's Clinician Assistant works inside the visit workflow: it edits the note (from the Core plan) and, on Premier, answers clinical questions from a medical knowledge base with linked citations. It is tied to the encounter, and its most useful capabilities sit behind the $119 tier. BastionGPT includes an unlimited general healthcare AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. ### Document upload depth This is the widest gap between the two platforms. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Freed's platform accepts visit audio, uploaded recordings, and transcripts (files up to 256 MB), plus example notes for building templates. It does not offer general document upload and analysis: there is no workflow for reading a PDF record, an insurance document, or a spreadsheet. Freed's pre-visit context comes from prior Freed notes for the same patient, not from outside records. If reviewing incoming records is part of your day, that work stays manual with Freed. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Freed does not disclose which foundation models power its product; third-party reviews describe Azure-backed language models without naming them. ### What the fine print says about your data Both vendors state that identifiable patient data is never used to train AI models. The difference is in the rights each vendor reserves. Freed's Terms of Use grant Freed a license to train its AI/ML models on customer data with consent provided via platform settings, give Freed the right to use de-identified data "in our sole discretion" and disclose it to third parties, and permit Freed to sell aggregate data. Retention is user-controlled: keep notes indefinitely or auto-delete after 30 days, with backups persisting seven additional days. BastionGPT's position is simpler: customer inputs, documents, and audio are never used to train AI models, on any plan, under any setting . Chats and transcripts are wiped after 30 days by default, and users can retain longer or delete sooner. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. ### Custom notes and shared prompts for teams Beyond the standard formats, BastionGPT's Custom Note tab accepts any free-form prompt per recording: narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can also upload example notes so BastionGPT matches your writing style and formatting. A Saved Prompts library with folder organization and team sharing is included on every plan , so a practice can standardize documentation by sharing one vetted prompt set across all providers and insert any prompt into a chat with a slash command. Freed offers a template builder from its Core plan and a "learn my format" tool, and its templates cover 30+ specialties. On individual plans those custom templates stay with the clinician who built them; team template sharing is a feature of Freed's custom Groups plans. ### A signed BAA on every plan BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial . Freed also provides a BAA: it is embedded as a schedule in the click-through Platform Terms of Use accepted at signup, a signed copy is available on request, and an organization-wide BAA is listed as a feature of Freed's custom Groups plans. Practices whose compliance policies require a countersigned BAA in hand before the first recording should factor in that request step; with BastionGPT the signed agreement is part of onboarding on every plan. ### Uptime and human support BastionGPT publishes a 99.90% uptime guarantee and offers email, chat, phone, and video support with real humans on every plan. Freed offers US-based support through chat, email, and live support calls, and maintains a public status page; it does not publish an uptime SLA. Recent Apple App Store reviews praise Freed's accuracy and time savings, but also reports lost notes after long visits and app access issues. ## Where Freed stands out Freed's clearest workflow advantage is its one-click EHR push : on the Premier plan, a Chrome extension maps the finished note into the fields of browser-based EHRs, with Athenahealth, Practice Fusion, SimplePractice, Elation, and Tebra confirmed. BastionGPT works alongside 50+ EMR/EHR systems through copy-paste and document uploads, which covers desktop and browser EHRs alike but adds a paste step Freed's Premier users skip. Freed also embeds ICD-10 suggestions (with CPT in beta) directly in the scribe workflow on Premier, and its pre-visit prep surfaces summaries of the patient's prior Freed notes before the encounter. Both are genuine workflow conveniences for clinicians who live entirely inside the visit-to-note loop. Freed's strongest published number is language coverage : it advertises transcription in 90+ languages with code-switching and automatic detection, so a visit can move between languages mid-conversation, and the finished note is generated in English. BastionGPT supports 15+ languages. If your practice regularly documents visits in languages BastionGPT does not yet cover, that difference can be significant. Weigh those strengths against what they cost. The plan that includes EHR push and coding runs $119 per user per month, the platform accepts no outside documents, its terms exclude 42 CFR Part 2 records, and the assistant stays anchored to the visit. For US practices that document primarily in English and need records review, appeal letters, group session transcripts, or spreadsheet analysis, BastionGPT delivers more capability at less than half of Freed's full-access price. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Freed cost in 2026? Freed's current plans are: Starter at $39 per user per month (capped at 40 notes per month), Core at $79 (unlimited notes), Premier at $119 per month or $104 billed annually (adds EHR push, ICD-10 suggestions, visit summaries, and the medical knowledge assistant), and custom Groups pricing. Small groups of around four clinicians pay standard individual rates, per Freed's help center. BastionGPT plans run $20 to $75 per user per month. Is Freed free? Freed offers a 7-day free trial with no credit card required, and a 50% student discount on the monthly individual plan. There is no permanent free tier and no published money-back guarantee. BastionGPT offers a 7-day free trial and plans from $20 per user per month with a 45-day money-back guarantee. Does Freed sign a BAA? Freed provides a Business Associate Agreement embedded as a schedule in its click-through Platform Terms of Use, accepted automatically at signup; a signed copy is available on request, and an organization-wide BAA is listed as a Groups plan feature. BastionGPT includes a signed BAA on every plan, including the free trial. Can Freed be used for 42 CFR Part 2 substance use disorder records? No. Freed's Platform Terms of Use state the platform has not been designed to process Part 2 data and require users to agree not to process any 42 CFR Part 2 data, with liability disclaimed if they do. BastionGPT supports 42 CFR Part 2 alongside HIPAA, FERPA, PIPEDA, PHIPA, and APP. Can Freed analyze uploaded documents? No. Freed accepts visit audio, uploaded recordings, and transcripts (files up to 256 MB), plus example notes for templates. It does not offer general document upload and analysis of PDFs, records, images, or spreadsheets. BastionGPT reads 10+ documents and 1,000+ pages at once on Professional Plus and Ultra, including Excel and CSV files. Can either platform analyze Excel spreadsheets? BastionGPT reads Excel (XLS, XLSX, XLSM), CSV, PowerPoint, and HTML files on Professional Plus and Ultra, alongside PDF, Word, text, and image files. Freed's published upload support covers visit audio, video, and text transcripts for note generation; it does not list spreadsheet upload or analysis. How many pages and documents can BastionGPT analyze at once? On Professional Plus and Ultra, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, text, and images. How many speakers can each scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, for sessions up to 240 minutes, with diarization and automatic AI labeling of each speaker's name and role. Freed separates speakers and offers an opt-in Voice ID for the clinician's voice; it publishes no speaker count, and independent 2026 reviews report limited diarization for multi-party visits. Is BastionGPT's AI scribe as accurate as Freed's? BastionGPT earned 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. Freed publishes no peer-reviewed accuracy study; its accuracy reports come from user reviews. Both platforms expect clinicians to review and approve every note. Can BastionGPT document couples, family, and group therapy sessions? Yes. Couples, family, and group sessions are included on every BastionGPT plan: the AI Scribe separates up to 10 speakers and automatically labels each speaker's name and role in the transcript, for sessions up to 240 minutes. Freed separates speakers but publishes no speaker count, and independent 2026 reviews report limited diarization for multi-party visits. What AI models power BastionGPT and Freed? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Freed does not disclose which foundation models power its product. Does Freed train AI on patient data? Freed states identifiable PHI is not used for AI training. Its Terms of Use permit training its AI/ML models on customer data with consent provided via platform settings, allow Freed to use and disclose de-identified data at its sole discretion, and permit sale of aggregate data. BastionGPT never uses customer inputs, documents, or audio to train AI models, on any plan. Does Freed support more languages than BastionGPT? Yes. Freed advertises transcription in 90+ languages with code-switching and automatic detection, and generates the finished note in English. BastionGPT supports 15+ languages. If extensive multilingual transcription is your deciding requirement, weigh it against the differences in document analysis, assistant scope, 42 CFR Part 2 coverage, and price. Which platform integrates with EHRs better? Freed Premier includes one-click EHR push into browser-based EHRs (Athenahealth, Practice Fusion, SimplePractice, Elation, and Tebra are confirmed), with copy-paste on all plans. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and eClinicalWorks, through copy-paste and document uploads. Choose Freed's Premier tier if one-click push into a supported browser EHR is your deciding requirement. Which is better for mental and behavioral health practices? BastionGPT is the stronger fit for most mental and behavioral health practices: it supports 42 CFR Part 2 records (Freed's terms prohibit them), transcribes group and family sessions with up to 10 labeled speakers, produces DAP and BIRP notes with custom free-form prompts, and analyzes prior records and psychological evaluations. Freed remains a capable scribe for solo therapists whose documentation stops at the individual session note and who avoid Part 2 data. Which is better value for a five-clinician practice? BastionGPT Professional Plus costs $2,700 per year for five clinicians ($45 x 5 x 12). Freed Premier costs $7,140 at monthly billing or $6,240 billed annually, and Freed's help center notes small groups of around four clinicians pay standard individual rates. That is a difference of $3,540 to $4,440 per year, and BastionGPT includes document analysis and the full assistant at that price. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/abridge # Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) Jump to Comparison Table Abridge and BastionGPT both turn clinical conversations into structured, HIPAA-compliant documentation. They are built for different buyers. Abridge is an enterprise ambient AI platform sold to health systems through organizational contracts: it publishes no pricing, offers no individual plan, and cannot be purchased by a solo clinician or small practice. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant that any clinician can start using today: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, and starts at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: who can actually buy each product, what each one costs in 2026, what happens beyond the visit note, how each platform handles your data, and what independent sources report. Abridge's positioning, compliance posture, and data practices were verified against abridge.com, trust.abridge.com, and Abridge's published privacy policy on July 5, 2026. If you are searching for an Abridge alternative you can start using today, this comparison covers the details: self-serve pricing, scribe capability, document handling, and compliance. BastionGPT is an Abridge alternative built for solo clinicians and small practices, with $20 plans and a signed BAA from day one. For the wider field, our AI scribe comparison covers other options such as Freed, Heidi Health, and Suki. The quick answer: Choose BastionGPT if you are a clinician, practice, or clinic that wants a HIPAA-compliant AI scribe and full AI assistant working this week: self-serve signup, a signed BAA on every plan, 10-speaker recognition, structured SOAP, DAP, BIRP, and H&P notes, 1,000+ page document analysis, and named multi-model AI from $20 per user per month. Choose Abridge if you are an enterprise health system running a system-wide ambient AI program inside Epic, with a procurement team and a multi-month deployment budget. For solo practitioners and small practices, the choice is structural: Abridge does not sell to you. Feature BastionGPT Abridge Who can buy it Any clinician or practice, self-serve signup in minutes Enterprise health systems only; no self-serve signup or individual plan Entry paid plan $20/user/month (Professional) No published pricing; independent 2026 analyses estimate roughly $200 to $800+ per provider per month on enterprise contracts Full-featured individual plan $45/user/month (Professional Plus) Not offered to individuals Organizational tier Ultra at $75/user/month, published price, with SSO and deployment in your existing enterprise cloud Custom enterprise pricing, learned through a sales process Free option 7-day free full trial, plus a 45-day money-back guarantee None for individual clinicians Time to start Minutes; no IT project Organizational procurement; independent reviews cite 3 to 6 months from contract to deployment Signed BAA ✓ Every plan, including the free trial BAAs executed through enterprise contracts with covered entities HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP SOC 2 Type 2, CCPA, TX-RAMP, WCAG 2.2 AA (per Abridge Trust Center) Data residency US, Canada, and Australia hosting; PHI stays in country by billing region US-based data centers; non-US hosting options not published Patient data used to train AI Never; inputs, uploads, and audio are never used for AI training De-identified data is used to refine algorithms and machine learning applications, per Abridge's privacy policy Audio and transcript retention Wiped after 30 days by default; users can delete sooner or retain longer Automatically deleted after roughly 14 to 30 days depending on the health system deployment; not clinician-configurable AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Proprietary Contextual Reasoning Engine built on OpenAI models; no model choice for the clinician Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles Multi-party conversations supported; speaker count not published Max recording length 240 minutes per session Not published Peer-reviewed clinical evidence 99.3% sentence-level accuracy across 1,225 physician-reviewed sentences (Baylor College of Medicine and Texas Children's Hospital researchers) ✓ Kaiser Permanente evaluation published in NEJM AI Documents per upload 10+ at once (Professional Plus and Ultra) Document upload and analysis is not part of the product Document capacity 1,000+ pages of combined context Not applicable Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not applicable; outputs are generated from the recorded encounter Full AI assistant ✓ Referral letters, insurance appeals, patient education, document review, spreadsheet analysis Encounter-anchored outputs: notes, orders, coding, after-visit summaries Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Specialty note templates configured through the health system Languages 15+ (transcription) 28 Saved prompts with team sharing ✓ All plans, with folder organization Template management handled at the enterprise level Public API ✓ Published API pricing, $45/month developer plan, 30-day developer trial, no enterprise minimums No public API pricing; custom integrations through enterprise contracts EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) Deep Epic integration, plus Oracle Health and athenahealth; requires enterprise IT rollout Uptime guarantee 99.90% on every plan SLAs negotiated per enterprise contract Support Email, chat, phone, and video with real humans, on every plan Support delivered through the health system's enterprise agreement Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp Minimal presence on self-serve review directories because individuals cannot buy it ## Where BastionGPT and Abridge are similar Both platforms generate structured clinical notes from natural patient conversations, both are HIPAA compliant, both encrypt data in transit and at rest with 256-bit encryption, and both store US customer data inside the United States. Both are used across dozens of specialties, and both publish peer-reviewed evidence behind their documentation accuracy: BastionGPT achieved 99.3% sentence-level accuracy across 1,225 physician-reviewed sentences in a study by Baylor College of Medicine and Texas Children's Hospital researchers, and Abridge's technology was evaluated in a large Kaiser Permanente study published in NEJM AI. If your entire requirement is turning a spoken visit into a clinical note, both platforms handle that core workflow well. The differences appear the moment you try to buy one, look past the visit note, or read the fine print on how each platform handles your data. ## Where BastionGPT goes further ### You can buy it today This is the single largest difference between the two platforms, and it is structural. Abridge sells exclusively to enterprise health systems. There is no pricing page on abridge.com, no self-serve signup, no individual plan, and no free trial for independent clinicians. Purchasing requires organizational procurement, IT involvement, and contract negotiation, and independent 2026 reviews report typical timelines of 3 to 6 months from contract to live deployment. Abridge does not publish what it charges; independent analyses peg enterprise licenses at roughly $2,500 per provider per year (about $208 per month), with other 2026 estimates ranging up to $800+ per provider per month plus implementation fees. Those figures are third-party estimates, which is itself the point: the only way to learn Abridge's price is to enter a sales process. BastionGPT publishes its pricing and holds it steady. Professional starts at $20 per user per month . Professional Plus, which adds the full document analysis suite and manual model selection, costs $45 . Ultra, the organizational tier with SSO and deployment inside your existing enterprise cloud, is published at $75 . Every plan starts with a 7-day free trial, carries a 45-day money-back guarantee, and includes a free month on annual billing. The math for a five-clinician practice: at the low end of independent estimates (about $2,500 per provider per year), five clinicians on Abridge would run roughly $12,500 per year, before implementation fees, and only if Abridge would sell to a five-clinician practice at all. BastionGPT Professional Plus runs $2,700 per year ($45 x 5 x 12), starting this week. ### A scribe that knows who is speaking BastionGPT's AI Scribe recognizes up to 10 speakers per session , live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role in the transcript: clinician, patient, or parent, for example. That matters for family sessions, group therapy, care conferences, and any visit with more people than a doctor and one patient. Abridge documents multi-party conversations, including in nursing settings, but publishes no speaker count and no session length limit. BastionGPT also produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes, and a Custom Note tab accepts any free-form prompt per recording : narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can upload example notes so BastionGPT matches your writing style. Abridge's note formats are specialty templates configured through the enterprise deployment, which means format changes route through your organization rather than through you. Accuracy evidence exists on both sides: BastionGPT achieved 99.3% sentence-level accuracy across 1,225 physician-reviewed sentences in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, and Abridge's technology was evaluated in a Kaiser Permanente study published in NEJM AI. The difference is access: BastionGPT's documented accuracy ships in the $20 self-serve product, while Abridge's ships inside enterprise deployments. ### A full AI assistant, not just a visit note Abridge's outputs are anchored to the clinical encounter: the note, coding specificity, orders, and a patient after-visit summary. Documentation work that starts outside the exam room sits outside the product. There is no document upload and analysis feature, no spreadsheet support, and no general assistant for drafting a referral letter or an insurance appeal from a records review. BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. The scribe and the assistant share one workspace, so a transcript can flow directly into any downstream document. ### Document analysis Abridge does not attempt On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Abridge generates its outputs from the recorded encounter and EHR context configured by the health system. Uploading outside records, insurance documents, or spreadsheets for analysis is not part of the product. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Abridge builds on OpenAI foundation models through its proprietary Contextual Reasoning Engine, and announced early access to GPT-5.5 in April 2026. The clinician has no visibility into or choice over which model reads a given encounter; those decisions live inside the platform and the enterprise configuration. ### Your data is never training data The two platforms take different positions here, in writing. BastionGPT never uses your prompts, documents, or audio to train AI models, and never shares them with OpenAI or any other AI provider for training. Chats and transcripts are wiped after 30 days by default, and you can delete sooner or retain longer. You keep full ownership and full control. Abridge's privacy policy states that it uses de-identified data to refine its algorithms and machine learning applications. Audio recordings and transcripts in health system deployments are automatically deleted on the organization's schedule, commonly around 30 days (Kaiser Permanente has stated 14 days for its deployment), and that schedule is set by the enterprise, not the clinician. Abridge's clinician terms also place responsibility for obtaining patient consent on the user. That allocation of responsibility has drawn attention: in December 2025 and April 2026, patients filed proposed class actions against Sharp HealthCare, Sutter Health, and MemorialCare alleging that visits were recorded with Abridge without informed consent. Abridge is not a defendant in those suits, and the claims are unproven, but they underline a practical truth: with an ambient scribe, consent workflows and data handling are your risk to manage. Choosing a platform that never trains on your data and puts retention in your hands removes an entire category of that risk. ### A signed BAA on every plan, without a procurement cycle BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , and supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. Customers in Canada and Australia keep their data hosted in country by default, a published option Abridge does not offer. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. Abridge executes BAAs with covered entities as part of its enterprise contracts, and its Trust Center lists SOC 2 Type 2, HIPAA, CCPA, TX-RAMP, and WCAG 2.2 AA. For a health system, that coverage is solid. For a solo practitioner or small practice, it is unreachable: there is no self-serve BAA because there is no self-serve product. For clinicians handling substance use disorder records (42 CFR Part 2) or school-based health data (FERPA), Abridge publishes no coverage for either framework; BastionGPT supports both. ### Published pricing all the way up, including the API Transparency extends past the plan cards. Developers and organizations building healthcare AI features get a published BastionGPT API : a $45 per month developer plan with a 30-day developer trial, published token and audio rates, a 99.9% uptime SLA, and no enterprise minimums. Abridge references custom integrations for enterprise customers, but publishes no API pricing, rate limits, or self-serve developer path. If your practice, platform, or billing company wants to budget an integration from the website, only one of the two makes that possible. ### Verified ratings from buyers like you BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. Abridge earned Best in KLAS for ambient AI in 2025 and 2026, a meaningful award based on surveys of enterprise health system leaders and clinicians. What Abridge does not have is a meaningful footprint on the self-serve review directories where individual buyers compare tools, because individual buyers cannot purchase it. Physician forum threads about Abridge consistently raise the same two frustrations: pricing that requires a sales call to learn, and a product that is unavailable outside a contracted health system. Ratings and recognition shift over time, so this page cites the source and date for each figure; the links in the sources section lead to the live numbers. ## When Abridge stands out Abridge is a serious platform for large health systems, and honesty about that helps you choose well. Its Epic integration runs deeper than any copy-paste workflow: notes generate in real time inside Epic, its Linked Evidence feature maps each note passage back to the transcript and audio behind it, and the platform extends into orders, coding, nursing documentation, clinical decision support with UpToDate, NEJM, and JAMA content, and real-time prior authorization through a January 2026 Availity partnership. Abridge supports 28 languages to BastionGPT's 15+, serves 55+ specialties across 300+ health systems including Mayo Clinic, Johns Hopkins, Duke Health, and Kaiser Permanente, and backs its accuracy with the largest peer-reviewed ambient AI study to date (1,306 Kaiser Permanente clinicians, published in NEJM AI). The company has raised more than $1 billion, including a Series E at a $5.3 billion valuation, to keep building. All of it comes through one door: an enterprise contract your organization signs. If you are a health system CIO running a system-wide ambient AI program on Epic with a multi-month deployment budget, Abridge belongs on your shortlist. If you are a physician, therapist, or practice owner who wants a HIPAA-compliant scribe and assistant working this week, Abridge does not sell to you, and BastionGPT delivers the scribe plus records analysis, appeal letters, and spreadsheet review for $20 to $75 per user per month. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Abridge cost in 2026? Abridge does not publish pricing. It is sold to health systems through negotiated enterprise contracts. Independent analyses peg enterprise licenses at roughly $2,500 per provider per year (about $208 per month), with other 2026 estimates ranging up to $800+ per provider per month, plus implementation costs. BastionGPT publishes its pricing: plans run $20 to $75 per user per month, with a 7-day free trial and a 45-day money-back guarantee. Can individual clinicians or small practices buy Abridge? No. Abridge offers no self-serve signup, no individual plan, and no free trial for independent clinicians; purchasing requires an organizational contract with a health system. BastionGPT offers self-serve signup for individuals and practices of any size, starting at $20 per user per month. Does Abridge sign a BAA? Abridge executes Business Associate Agreements with covered entities as part of its enterprise contracts. Because there is no individual product, there is no self-serve BAA path for solo practitioners. BastionGPT includes a signed BAA on every plan, including the free trial. Does Abridge use patient data to train its AI? Abridge's privacy policy states that it uses de-identified data to refine its algorithms and machine learning applications. BastionGPT never uses your prompts, documents, or audio to train AI models and never shares them with AI providers for training. How long does Abridge keep audio recordings? In published health system deployments, Abridge audio and transcripts are automatically deleted after roughly 14 to 30 days, on a schedule set by the organization. BastionGPT wipes chats and transcripts after 30 days by default, and each user can delete sooner or retain longer. What AI models power BastionGPT and Abridge? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Abridge runs a proprietary Contextual Reasoning Engine built on OpenAI foundation models, including GPT-5.5 as of April 2026, with no model visibility or choice for the clinician. Can Abridge analyze uploaded documents or spreadsheets? No. Abridge generates notes, orders, coding, and after-visit summaries from the recorded encounter; document upload and analysis is not part of the product, and spreadsheets are not supported. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context, including PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. How many speakers can BastionGPT's AI Scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role. Abridge supports multi-party conversations but does not publish a speaker count or session length limit. Is BastionGPT's AI scribe as accurate as Abridge's? Both publish peer-reviewed evidence. BastionGPT achieved 99.3% sentence-level accuracy across 1,225 physician-reviewed sentences in a study by Baylor College of Medicine and Texas Children's Hospital researchers — the clinical validation page covers the methodology — and Abridge's technology was evaluated in a Kaiser Permanente study published in NEJM AI. The practical difference is access: BastionGPT's documented accuracy is available self-serve from $20 per user per month, while Abridge ships through enterprise contracts. Both platforms expect clinicians to review and approve every note. Does Abridge integrate with Epic? Yes, deeply. Real-time note generation inside Epic workflows is Abridge's core strength, with its Linked Evidence feature mapping each note passage back to the transcript, and deployments extending to Oracle Health and athenahealth. That integration is available only through an enterprise contract and IT rollout. BastionGPT works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) through simple copy-paste and document workflows, with no IT project required. Does Abridge offer data hosting outside the United States? Abridge's support documentation states that data is stored and processed in US-based HIPAA-secure data centers; non-US hosting options are not published. BastionGPT hosts customer data in the United States, Canada, and Australia, with PHI kept in country by billing region, and supports PIPEDA, PHIPA, and Australian APP requirements. Which is better for solo practitioners and small practices, BastionGPT or Abridge? BastionGPT, as a structural matter: Abridge does not sell to solo practitioners or small practices. BastionGPT offers self-serve signup, a signed BAA on every plan, HIPAA, 42 CFR Part 2, and FERPA support, US-based human support by phone, chat, email, and video, and full-featured access at $45 per user per month. Is Abridge Best in KLAS? Yes. Abridge earned Best in KLAS recognition for ambient AI in 2025 and 2026, based on surveys of enterprise health systems. BastionGPT's ratings come from verified individual buyers: a 4.8 out of 5 average across Capterra, Software Advice, and GetApp. Both signals are real; they measure different buyers. Does BastionGPT have a free option like Abridge? Abridge has no free option for individuals. BastionGPT offers a 7-day free trial of its full Professional and Professional Plus plans, with a signed BAA active during the trial, no charge until the trial ends, and a 45-day money-back guarantee on paid plans starting at $20 per user per month. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/nabla # Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) Jump to Comparison Table Nabla and BastionGPT both turn patient conversations into structured, HIPAA-compliant clinical notes, and both are trusted by tens of thousands of clinicians. They are built for different buyers. Nabla is an enterprise ambient AI platform that health systems deploy inside Epic and other EHRs through a sales process, with pricing available by quote. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant that any clinician can start using today: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at a published $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, what happens beyond the visit note, how document handling and data retention compare, and what compliance coverage you get at each price. Nabla's product details were verified against nabla.com, help.nabla.com, docs.nabla.com, and trust.nabla.com on July 5, 2026. If you are considering Nabla alternatives, this comparison gives you the specifics: published pricing, what happens beyond the visit note, and compliance coverage. BastionGPT is a Nabla alternative that adds a full AI assistant, document analysis, and a BAA on every plan at $20 per month. For the wider field, our AI scribe comparison covers other options such as Freed, Heidi Health, and DeepScribe. The quick answer: Choose BastionGPT if you want published pricing from $20 per user per month, a signed BAA on every plan including the trial, an unlimited AI scribe with 10-speaker recognition, a full AI assistant for referral letters, appeals, and records review, and document analysis across 1,000+ pages. Choose Nabla if you are a health system deploying ambient AI natively inside Epic or another EHR at enterprise scale, with quote-based contracts, SMART on FHIR identity integration, and a dedicated implementation team. For solo clinicians and small US practices, BastionGPT delivers more capability at a price you can see today. Feature BastionGPT Nabla Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp No rated listings on the major B2B review directories; feedback comes from vendor testimonials, KLAS commentary, and app store reviews Published pricing $20/user/month (Professional); $45/user/month (Professional Plus); $75/user/month (Ultra) Not published; quotes through sales. Reported Pro pricing of about $119/user/month Free option 7-day free full trial, plus a 45-day money-back guarantee Free tier reported at 30 encounters/month; unlimited use requires a paid plan Signed BAA ✓ Every plan, including the free trial BAA incorporated into Terms of Service for clients; third-party reviews report the free tier excludes it HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP GDPR, SOC 2 Type 2, ISO 27001, Texas RAMP Level 2 AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Proprietary in-house model; no model choice Beyond the visit note Full AI assistant: referral letters, appeals, patient education, coding review, document and spreadsheet analysis Documentation plus ICD-10 coding suggestions; no clinical Q&A or general assistant Document analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra) No general document upload and analysis Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not applicable (encounter audio only) Audio sources Live recording or uploaded audio files Live capture: ambient recording or dictation Max recording length 240 minutes per session Ambient visits up to 3 hours; dictated notes 20 to 30 minutes Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published for its production notes Multi-speaker recognition Up to 10 speakers, with automatic AI labeling of each speaker's name and role Multi-speaker supported; speaker count not published Note retention 30 days by default; users can retain longer or delete sooner 14 days by default (organization-configurable) Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Template library across 55+ specialties; customization via instructions and dot phrases Saved prompts with team sharing ✓ All plans, with folder organization and slash commands Custom instructions and dot phrases for note templates Languages 15+ 35 EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) via copy-paste, no IT project Native integrations with 20+ EHRs (Epic, Oracle Health, athenahealth, NextGen, Greenway); enterprise implementation SSO and provisioning SSO on Ultra ($75/user/month, published price) SSO, SAML, SCIM, and SMART on FHIR login on enterprise contracts (quote-based) Enterprise deployment Option to run within your existing enterprise cloud (Ultra) Web app, mobile apps, Chrome extension, API; no published private-cloud option Data used for AI training Never Not by default; opt-in feedback audio is de-identified, stored indefinitely, and used to train Nabla Uptime guarantee 99.90% on every plan Not published; SLAs handled in enterprise contracts Support Email, chat, phone, and video with real humans, on every plan Self-serve support below enterprise; dedicated success teams for health system contracts ## Where BastionGPT and Nabla are similar Both platforms generate structured clinical notes from natural patient conversations in seconds, work across dozens of specialties, and run on desktop and mobile. Both state that patient data is not used to train AI models by default, both encrypt data in transit and at rest, and both let an individual clinician start with a self-serve trial instead of a demo call. If your entire requirement is turning a spoken visit into a standard note, both tools cover that core workflow well. The differences appear the moment you ask for a price, need documentation that goes beyond the visit, upload a record, or want to keep your notes longer than two weeks. ## Where BastionGPT goes further ### Pricing you can see, budget, and act on today Nabla does not publish pricing. Its website routes buyers to a demo request, and quotes come through a sales conversation. Reported pricing for its Pro tier has been approximately $119 per user per month (Axios and multiple 2025-2026 industry reviews), with enterprise contracts priced by custom quote, and 2026 roundups report Nabla now sells primarily through demo and contract. BastionGPT publishes its prices and has held them steady for over two years. Professional starts at $20 per user per month . Professional Plus, which adds the full document analysis suite, image and spreadsheet reading, and manual model selection, costs $45 . Teams that need SSO and deployment inside their own enterprise cloud also get a published number: Ultra at $75 per user per month . Annual billing includes a free month, and every plan carries a 45-day money-back guarantee. The math for a five-clinician practice: at Nabla's reported $119 rate, a year costs $7,140 ($119 x 5 x 12). BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12), and Professional runs $1,200. That is a difference of $4,440 to $5,940 per year , and with BastionGPT you know the number before you ever talk to anyone. ### A scribe built for real-world sessions BastionGPT's AI Scribe recognizes up to 10 speakers per session with speaker diarization, and its AI automatically labels each speaker's name and role in the transcript: clinician, patient, parent, interpreter. Sessions run up to 240 minutes, and BastionGPT transcribes both live audio and uploaded recordings , so a dictation captured in the car or a recorded team meeting produces the same structured output as a live visit. Nabla supports multi-speaker encounters but publishes no speaker count, and its workflow centers on live capture: ambient visits up to 3 hours, and dictated notes capped at 20 to 30 minutes per its help center. Its developer API accepts uploaded audio files, but that capability is for software builders, not a workflow inside the clinician app. Accuracy is documented, not asserted: BastionGPT received 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Nabla publishes no peer-reviewed accuracy study of its production notes. ### Your notes stay yours for more than two weeks Nabla retains encounter data for 14 days by default , a window its help center describes as time to review, edit, and export notes to the EHR. The default is configurable at the organization level, and audio is not stored at all. That minimal-retention posture is a genuine security strength for hospital deployments, and a real constraint for an individual clinician: if a note was not exported within the window, it is gone. BastionGPT wipes chats and transcripts after 30 days by default, and each user controls the window : retain records longer when a case is still open, or delete them sooner, including immediate deletion of specific items or all data. You get the same zero-training guarantee with retention that fits how a practice actually works. One more difference worth knowing: Nabla's help center states that optional user feedback, which may include de-identified audio with PHI replaced by a beep, is stored indefinitely and used both to train Nabla and for troubleshooting. BastionGPT never uses customer inputs, uploads, or audio to train AI models, with no exceptions to document. ### A full AI assistant that keeps working after the note is signed Nabla is a documentation platform. It transcribes the encounter, generates the note, and suggests ICD-10 codes. Per Glass Health's 2026 analysis, Nabla does not answer clinical questions, search literature, or generate content the clinician did not verbalize during the visit; the assessment and plan reflect what was said in the room. BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. A Saved Prompts library with folder organization and team sharing ships on every plan, so a practice can standardize documentation by sharing one vetted prompt set and inserting any prompt with a slash command. The distinction is simple: Nabla ends when the note is signed. BastionGPT keeps working on the prior authorization, the appeal, the records summary, and the parent-friendly explanation of the diagnosis. ### Document analysis Nabla does not offer On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. Nabla's product is built around the encounter: live ambient audio and at-cursor dictation. It does not offer a general document upload and analysis workflow, and it does not accept spreadsheets, records, or images for review. For practices that handle records requests, independent evaluations, legal referrals, or insurance documentation, that is an entire category of work Nabla leaves on your desk. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Nabla runs on a model that is not disclosed publicly, so you cannot benchmark it against alternatives, or switch to a stronger engine for a complex task. ### A signed BAA on every plan, plus the compliance depth US practices need BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , and supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. Nabla maintains a strong public-facing enterprise security posture: HIPAA and GDPR compliance, SOC 2 Type 2 and ISO 27001 certifications, Texas RAMP Level 2, and a BAA incorporated into its Terms of Service for clients. Third-party reviews report that the free tier does not include a BAA, which matters because using a scribe on patient encounters without a BAA is itself a HIPAA compliance gap. Nabla also does not list 42 CFR Part 2 or FERPA support, which substance-use treatment programs and school-based providers should note before piloting. ### Human support and a published uptime guarantee BastionGPT provides email, chat, phone, and video support with real humans on every plan, backed by a 99.90% uptime guarantee . Nabla's self-serve users rely on self-service support, with dedicated customer success reserved for enterprise contracts, and it publishes no uptime commitment outside those contracts. ## Where Nabla stands out Nabla is a serious platform, and for its intended buyer it is one of the strongest options on the market. Enterprise EHR integration is Nabla's core strength. It integrates natively with 20+ EHRs, including Epic, Oracle Health, athenahealth, NextGen, and Greenway, writing structured notes directly into EHR fields, with SMART on FHIR login, SSO/SAML, and SCIM provisioning for enterprise identity teams. Its Nabla Connect module lets EHR vendors embed its ambient AI in days. Health systems including LCMC Health, M Health Fairview, University of Iowa Health Care, Denver Health, and CVS Health have deployed it systemwide, and Nabla reports 85,000+ clinicians across 130+ healthcare organizations. If you are a CMIO rolling out ambient AI to 2,000 providers inside Epic, Nabla built its business for you. Language coverage is broader. Nabla supports encounters in 35 languages; BastionGPT supports 15+. Practices that regularly document in languages BastionGPT does not yet cover should weigh that directly. Dictation is a first-class product. Nabla's at-cursor dictation works anywhere on the desktop, supports Dragon PowerMic hardware, and positions Nabla as a replacement for legacy dictation tools, a genuine differentiator inside hospital workflows. The evidence base is deep. Nabla's outcomes have been studied in peer-reviewed settings, including NEJM Catalyst publications and a randomized trial at an academic health system, and it publishes a large named case-study library. Those advantages serve organizations with procurement teams, implementation timelines, and IT departments. For an individual clinician or a small practice, they come bundled with the things that make enterprise software frustrating: no published price, a sales process to get one, a 14-day window on your own notes, and a product that ends at the visit note. For US practices that want scribing plus records review, appeal letters, document analysis, and a BAA from day one at a price posted on a public page, BastionGPT delivers more capability with none of the procurement overhead. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Nabla cost in 2026? Nabla does not publish pricing; quotes come through a sales conversation. Reported pricing for its Pro tier has been approximately $119 per user per month, with custom enterprise contracts for health systems, and 2026 industry roundups report Nabla now sells primarily via demo and contract. BastionGPT publishes its pricing: plans run $20 to $75 per user per month, with Professional at $20, Professional Plus at $45, and Ultra at $75. Does Nabla have a free plan? Nabla has offered a free tier reported at up to 30 encounters per month, with unlimited use requiring a paid plan. Third-party reviews report the free tier does not include a Business Associate Agreement, which limits its use on real patient encounters. BastionGPT offers a 7-day free trial of its full paid plans, with a signed BAA in place from the first day of the trial. Is Nabla HIPAA compliant? Yes. Nabla is HIPAA compliant, holds SOC 2 Type 2 and ISO 27001 certifications, lists Texas RAMP Level 2, and incorporates a BAA into its Terms of Service for its clients. BastionGPT is also HIPAA compliant, includes a signed BAA on every plan including the free trial, and adds 42 CFR Part 2 and FERPA support for substance-use treatment programs and school-based providers. Does Nabla sign a BAA? Yes, for its clients: Nabla incorporates a Business Associate Agreement into its Terms of Service. Third-party reviews report that the free tier is not covered by a BAA. BastionGPT includes a signed BAA on every plan, including the free trial. Does Nabla use patient data to train its AI? Not by default. Nabla does not store encounter audio, and it states it does not train models on customer data. Its help center notes one exception path: clinicians can optionally share feedback that may include de-identified audio (PHI replaced with a beep), and that feedback audio is stored indefinitely and used both to train Nabla and for troubleshooting. BastionGPT never uses customer inputs, documents, or audio for AI training on any plan. How long does each platform keep my notes? Nabla retains encounter data for 14 days by default, configurable at the organization level, after which notes are gone unless exported to the EHR. BastionGPT retains chats and transcripts for 30 days by default, and each user can retain records longer or delete them sooner, including immediate deletion. Can Nabla analyze uploaded documents? No. Nabla's clinician product is built around live encounter capture (ambient recording and dictation) and does not offer general document upload and analysis. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, including PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. How long can recordings be, and how many speakers are supported? BastionGPT records sessions up to 240 minutes, live or from uploaded audio, recognizes up to 10 speakers, and automatically labels each speaker's name and role in the transcript. Nabla supports ambient visits up to 3 hours and dictated notes of 20 to 30 minutes per its help center; it supports multi-speaker encounters but does not publish a speaker count, and uploaded audio is available through its developer API rather than the clinician app. Is BastionGPT's AI scribe as accurate as Nabla's? BastionGPT received 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. Nabla publishes no peer-reviewed accuracy study of its production notes. Both platforms expect clinicians to review and approve every note. What AI models power BastionGPT and Nabla? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Nabla runs on an undisclosed model, with no model visibility or selection for users. Which EHRs does each platform work with? Nabla integrates natively with 20+ EHRs, including Epic, Oracle Health, athenahealth, NextGen, and Greenway, with structured write-back into EHR fields; full integrations are implemented through enterprise deployment. BastionGPT works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks, and more) through simple copy-paste with no IT project, so a solo clinician is running the same day they sign up. Does Nabla support more languages than BastionGPT? Yes. Nabla supports encounters in 35 languages, while BastionGPT supports 15+. If multilingual transcription is your deciding requirement, weigh it against the differences in pricing transparency, document analysis, note retention, assistant capabilities, and BAA coverage. Which is better for a solo practitioner or small practice? BastionGPT, for most US practices. It publishes its pricing ($20 to $75 per user per month), includes a signed BAA on every plan including the trial, retains notes for a user-controlled 30 days, and bundles an unlimited AI assistant for referral letters, appeals, records review, and document analysis. Nabla is engineered for health systems: native EHR write-back, systemwide Epic deployments, and quote-based enterprise contracts. Which is better for a large health system? Health systems that need native, structured EHR integration at scale, SMART on FHIR and SCIM identity integration, at-cursor dictation to replace legacy tools, and an enterprise implementation partner should evaluate Nabla; that is the buyer it was built for. Organizations that also want document analysis, administrative drafting, and a general clinical AI assistant across roles should evaluate BastionGPT alongside it, and BastionGPT's published per-user pricing makes piloting both straightforward. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/suki # Suki Alternative: BastionGPT vs Suki AI Scribe (2026) Jump to Comparison Table Suki and BastionGPT both offer HIPAA-compliant AI scribes that turn patient conversations into structured clinical notes. The overlap largely ends there. Suki is a voice-first ambient assistant built for health systems: it integrates deeply with four EHRs, stages orders by voice, and sells through a demo-to-contract enterprise process at an industry-reported $299 to $399 per user per month, with no pricing published on suki.ai. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at $20 per user per month with a signed BAA on every plan and a self-serve signup that takes minutes. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, how you buy it, whether your data trains anyone's AI models, what happens beyond the visit note, how document handling compares, what real users say on independent review sites, and what compliance coverage you get at each price. Suki's product details, plan structure, data policies, and reported pricing were verified against suki.ai, Suki's developer documentation and press releases, its Capterra and Apple App Store listings, and independent industry coverage on July 5, 2026. If you are looking at Suki alternatives, the sections below cover what changes at each price point: scribe capability, document analysis, and compliance depth. BastionGPT is a Suki alternative that pairs an unlimited scribe with a full AI assistant at $20 per month. For the wider field, our AI scribe comparison covers other options such as Freed, DeepScribe, and Nabla. The quick answer: Choose BastionGPT if you want published pricing from $20 per user per month, a signed BAA on every plan including the free trial, a strict no-training policy on customer data, mental health note formats (DAP, BIRP, intake assessments, treatment plans), 1,000+ pages of document analysis, and a scribe plus full AI assistant that works alongside any EHR the same day you sign up. Choose Suki if you run a large health system standardized on Epic, Oracle Health, athenahealth, or MEDITECH and want voice commands, ambient order staging, and ambient CPT/E/M coding written directly into the chart, with the budget for its industry-reported $299 to $399 per user per month enterprise contracts. Feature BastionGPT Suki Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 4.2/5 on the Apple App Store (63 reviews); 2.0/5 on Capterra from its single posted review Published pricing ✓ $20 to $75 per user per month, posted publicly Not published; industry reports place Suki at $299 to $399 per user per month Entry paid plan $20/user/month (Professional) ~$299/user/month reported (Suki Compose) Full-featured individual plan $45/user/month (Professional Plus) ~$399/user/month reported (Suki Assistant) How you buy Self-serve signup in minutes; no sales call required Demo-to-contract enterprise sales process Free option 7-day free full trial, plus a 45-day money-back guarantee None; Capterra lists a free trial as not available Contract terms Monthly or annual; annual billing includes a free month Annual commitments typical; independent reviews report setup fees of $500 to $2,000 per practice Signed BAA ✓ Every plan, including the free trial Provided to customers through enterprise contracting HIPAA compliant ✓ ✓ Customer data used to train AI models Never; prompts, documents, and outputs are not retained for model training De-identified and anonymized data is used for model training, per Suki's developer security FAQ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP Not published AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Foundation models not disclosed; Google Cloud Vertex AI powers patient summaries and Q&A Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles ✓ Identifies speakers; speaker count not published Max recording length 240 minutes per session No published limit Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published Document upload and analysis ✓ 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra) Not offered; works from EHR chart data on integrated EHRs Spreadsheet analysis ✓ Excel (.xls, .xlsx, .xlsm) and CSV Not offered Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not applicable (no document upload analysis) Full AI assistant ✓ Referral letters, insurance appeals, patient education, document review, lab and spreadsheet analysis Visit-anchored: notes, patient instructions, chart Q&A, order staging inside integrated EHRs Ambient coding automation Coding suggestions plus documentation review for suspected under- or over-coding ✓ ICD-10, HCC, CPT, and E/M codes generated from ambient listening and staged in the EHR Voice commands and EHR order staging Not offered; documentation-focused workflow ✓ "Suki, order metformin 500mg" style commands; ambient order staging Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, intake assessments, treatment plans, progress, custom free-form prompt SOAP, H&P, progress, custom templates; behavioral health formats such as DAP and BIRP not marketed Languages 15+ 80+ ambient session languages, with notes generated in English Saved prompts with team sharing ✓ All plans, with folder organization Custom templates; template management varies by deployment EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) via copy-paste, no IT project Deep bidirectional integration with 4 EHRs: Epic, Oracle Health, athenahealth, MEDITECH; independent reviews report Epic rollouts taking several weeks Uptime guarantee 99.90% No published SLA; availability marketed as 24/7/365 Support Email, chat, phone, and video with real humans, on every plan Email, chat, phone, and 24/7 live rep per its Capterra listing; rollout support tied to enterprise contracts Built for Individuals, small practices, and organizations of any size Health systems and large groups; Suki markets deployments from 100 to 100,000 clinicians ## Where BastionGPT and Suki are similar Both platforms generate structured, specialty-aware clinical notes from ambient listening, and both keep the clinician in control: AI output is a draft that the provider reviews and approves. Both are HIPAA compliant, encrypt data with AES-256 at rest and TLS in transit, and delete session audio and transcripts after 30 days by default. Retention differs after that point: Suki keeps the final clinical note for the duration of the service contract, while BastionGPT lets users retain content longer or delete it sooner, with deleted data held up to 30 days in a secure audit vault. If your entire requirement is turning a spoken visit into a standard note, both tools cover that core workflow. The differences appear the moment you ask for a price, ask what happens to your data after the visit, work outside a deeply integrated EHR, upload a record for review, or need documentation that goes beyond the visit. ## Where BastionGPT goes further ### Price you can see, at a fraction of the cost Suki does not publish pricing. Its Capterra listing reads "Contact vendor for pricing," and buying requires a demo and an enterprise contract. Industry reporting from 2025 and 2026 consistently places Suki Compose at roughly $299 per user per month and Suki Assistant at roughly $399, with enterprise contracts reported at $350 to $500+ depending on EHR integration depth. Independent reviews also report setup fees of $500 to $2,000 per practice and one-year minimum commitments. BastionGPT posts its prices publicly: Professional starts at $20 per user per month , and Professional Plus, which includes the full AI assistant, multi-document analysis, image and spreadsheet reading, and manual model selection, costs $45 . Annual billing includes a free month, and the organizational Ultra tier is published at $75 per user per month. The math for a five-clinician practice: Suki Assistant at its reported rate runs $23,940 per year ($399 x 5 x 12). BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12). That is a difference of $21,240 per year. Even against Suki Compose's reported $299 entry point, a practice on BastionGPT Professional ($1,200 per year for five users) keeps $16,740 that would otherwise go to documentation software. ### Signing up takes minutes, not a procurement cycle Suki is built and sold for enterprise deployment. Independent reviews describe implementation timelines that range from minutes on athenahealth to several weeks on Epic, a learning curve of one to two weeks to master voice commands, and a sales process involving demos, negotiation, and annual contracts. For a hospital IT department, that is normal. For a solo practitioner or a five-provider clinic, it is a barrier. BastionGPT deploys the day you sign up. There is no integration project, no implementation fee, and no sales call required: start a 7-day free trial , record your first session, and paste the note into the EHR you already use. Every paid plan carries a 45-day money-back guarantee . ### A scribe that knows who is speaking BastionGPT's AI Scribe is included on every plan: record live or upload audio, capture sessions up to 240 minutes , and let speaker diarization separate up to 10 speakers while the AI labels each speaker's name and role automatically. It also captures telehealth visits through Zoom, Teams, and Google Meet with no visible bot in the meeting. Suki identifies speakers but publishes no speaker count and no session length limit. Accuracy is documented, not asserted: BastionGPT earned 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Suki publishes no peer-reviewed accuracy study. ### Your data trains nothing This difference is easy to miss on a features page and matters more than most features. BastionGPT states that customer data is never sold or used to train AI models : prompts, uploaded documents, session audio, and outputs are not retained for model training on any plan. Suki's policy is different. Its developer security FAQ states that Suki uses de-identified and anonymized data for model training purposes , breaking audio into isolated chunks and removing PII from transcripts before training. De-identification is a recognized HIPAA practice, and Suki documents its method openly. It still means patient encounters contribute to model training under Suki's process. Some organizations accept that tradeoff; others want the simpler answer. With BastionGPT, the answer to "does our data train the AI" is one word: no. ### Document analysis Suki does not offer Suki reads patient data from the EHR chart on its four integrated systems. It does not offer document upload and analysis: there is no way to hand it a 300-page records dump, a psychological evaluation, a scanned fax, or an insurance denial letter and ask questions about it. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, records from multiple providers, and a lab spreadsheet, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. ### A full healthcare assistant, not only a visit workflow Suki's assistant capabilities are anchored to the encounter and the EHR: it drafts the note, generates patient instructions, answers questions about chart data, and stages orders. That is real value inside the visit, and it is where Suki has focused. BastionGPT includes an unlimited general AI assistant on every plan that covers the desk work around the visit too. Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Suki does not disclose which foundation models generate its notes. It has announced that patient summaries and clinical Q&A are built on Google Cloud's Vertex AI platform, and it describes its stack as medically tuned ASR, NLU, and LLMs, without naming the models behind them. ### Note flexibility for mental health and beyond BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, intake assessments, treatment plans, and progress notes, plus specialty outputs such as group therapy summaries, neuropsychological evaluation write-ups, and safety plans. A Custom Note tab accepts any free-form prompt per recording : narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can also upload example notes so BastionGPT matches your writing style and formatting. Suki generates SOAP, H&P, progress, and custom-template notes across 100+ specialties, with strong depth in physician workflows. Behavioral health formats such as DAP and BIRP are not part of its marketed template set, which matters for therapists, counselors, and group mental health practices that document in those structures every day. ### A signed BAA on every plan, including the trial BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial , with countersigned and custom BAAs available for larger organizations. The platform supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements, and data for US, Canada, and Australia customers defaults to storage in the customer's own country. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. Suki is HIPAA compliant and SOC 2 Type II certified, and it provides BAAs to customers through its standard enterprise contracting. Because there is no self-serve plan, there is no self-serve BAA: the agreement arrives with the contract, after the sales process. Coverage for 42 CFR Part 2 or FERPA is not published in Suki's public trust materials. ### User ratings on independent review sites BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp. It is used by 10,000+ healthcare organizations, processes 1.3 million clinical documents each month, and saves clinicians a reported 15,000 hours every day. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. Suki's public review footprint is thin for a company of its size. Its Capterra listing shows a 2.0 out of 5 overall rating from its single posted review, which describes glitchy dictation behavior; with one review, that number says less about the product than about how few individual buyers reach Suki through self-serve channels. On the Apple App Store, Suki holds a 4.2 out of 5 rating across 63 reviews, mixing praise for fast note generation with complaints about generic notes and missed medication names. Suki's strongest independent validation comes from KLAS, where it scores 93.8/100 among surveyed healthcare organizations. Ratings shift over time, so this page cites the source and date for each figure. The links in the sources section lead to the live numbers. ## When Suki stands out Suki's clearest advantage is EHR depth. Its bidirectional integrations with Epic, Oracle Health, athenahealth, and MEDITECH pull chart data into the note and write finished notes back without copy-paste. Voice commands ("Suki, order metformin 500mg") and ambient order staging let clinicians work the EHR hands-free, and in January 2026 Suki extended its ambient coding to generate CPT and E/M codes alongside ICD-10 and HCC, reporting a 48% reduction in amended encounters among its customers. Suki also supports ambient sessions in 80+ languages with notes generated in English, and its SDKs include an offline mode that keeps recording during network drops and uploads when the connection returns. Suki's enterprise proof is real. It reports 400+ health systems and partners, scores 93.8/100 with KLAS-surveyed organizations, and points to a 2026 KLAS ROI validation citing 21 to 27% documentation time savings and revenue impact of up to $2,629 per provider per month in customer case studies. Those results assume an enterprise context: an organization standardized on one of the four integrated EHRs, an IT team to run the rollout, and a budget that absorbs an industry-reported $299 to $399 per user per month plus setup fees on an annual contract. For individual clinicians, small and mid-sized practices, and any team that needs records review, appeal letters, document analysis, mental health note formats, or a flat no-training data policy, BastionGPT delivers a broader assistant, a signed BAA from day one, and published pricing at roughly one-ninth of Suki's reported full rate. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Suki cost in 2026? Suki does not publish pricing; its Capterra listing reads "Contact vendor for pricing." Industry reporting from 2025 and 2026 places Suki Compose at roughly $299 per user per month and Suki Assistant at roughly $399, with enterprise contracts reported at $350 to $500+ per user per month depending on EHR integration depth, plus reported setup fees of $500 to $2,000 per practice. BastionGPT publishes its plans at $20 to $75 per user per month. Does Suki offer a free trial or free tier? No. Suki's Capterra listing shows that a free trial is not available, and there is no free tier. Evaluation happens through a sales demo followed by an enterprise contract. BastionGPT offers a 7-day free trial of its full Professional and Professional Plus plans, and paid plans from $20 per user per month carry a 45-day money-back guarantee. Does Suki sign a BAA? Yes, Suki provides Business Associate Agreements to customers through its enterprise contracting process, alongside HIPAA compliance and SOC 2 Type II certification. Because Suki has no self-serve plan, the BAA arrives with the negotiated contract. BastionGPT includes a signed BAA on every plan, including the free trial, and also supports 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Does Suki use patient data to train its AI models? Suki's developer security FAQ states that Suki uses de-identified and anonymized data for model training purposes: audio is broken into isolated chunks and PII is removed from transcripts before training. BastionGPT never uses customer data to train AI models; prompts, documents, session audio, and outputs are not retained for model training on any plan. Is Suki worth it for a solo practitioner or small practice? Suki is built for health systems: its deepest value comes from bidirectional integration with Epic, Oracle Health, athenahealth, and MEDITECH, and its buying process involves demos, annual contracts, and reported setup fees. Independent reviews note that smaller clinics wanting posted pricing and a copy-to-EHR workflow find the enterprise contracting heavier than needed. BastionGPT is self-serve, starts at $20 per user per month, and works alongside 50+ EMR/EHR systems without an integration project. Can Suki analyze uploaded documents? No. Suki reads patient data from the EHR chart on its integrated systems and answers questions about that chart data, but it does not offer document upload and analysis. BastionGPT reads 10+ uploaded documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, including PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files with OCR for scans. Can either platform analyze Excel spreadsheets? BastionGPT reads Excel (.xls, .xlsx, .xlsm), CSV, PowerPoint, and HTML files on Professional Plus and above, alongside PDF, Word, text, and image files. Suki does not offer spreadsheet or document upload analysis; its data access is EHR chart data on integrated systems. What AI models power BastionGPT and Suki? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Suki does not disclose the foundation models behind its notes; it has announced that its patient summaries and clinical Q&A are built on Google Cloud's Vertex AI platform. How many speakers can BastionGPT's AI Scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role, for example clinician, patient, or parent. Suki identifies speakers in ambient sessions but does not publish a supported speaker count. Is BastionGPT's AI scribe as accurate as Suki's? BastionGPT earned 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. Suki publishes no peer-reviewed accuracy study. Both platforms expect clinicians to review and approve every note. How do BastionGPT and Suki user ratings compare? BastionGPT holds a 4.8 out of 5 average across the verified review directories Capterra, Software Advice, and GetApp. Suki holds a 4.2 out of 5 rating on the Apple App Store across 63 reviews and a 2.0 out of 5 on Capterra from its single posted review; its strongest independent result is a 93.8/100 KLAS score from surveyed healthcare organizations. Ratings change over time, so check each review site for current figures. Does Suki support more languages than BastionGPT? Yes. Suki supports ambient sessions in 80+ languages, generating the clinical note in English, and produces patient instructions in 80 languages. BastionGPT supports 15+ languages. If multilingual transcription is your deciding requirement, weigh it against the differences in document analysis, assistant capabilities, data-training policy, BAA access, and price. Does BastionGPT integrate with EHRs the way Suki does? The approaches differ. Suki offers deep bidirectional integration with four EHRs (Epic, Oracle Health, athenahealth, MEDITECH), writing notes and staged orders directly into the chart; independent reviews report Epic rollouts taking several weeks. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and eClinicalWorks, through a simple copy-paste workflow that requires no IT project and deploys the same day. Which is better for US-based small practices, BastionGPT or Suki? For most US small practices, BastionGPT is the stronger fit: published pricing from $20 per user per month, a signed BAA on every plan including the trial, no use of customer data for AI training, HIPAA, 42 CFR Part 2, and FERPA support, mental health note formats (DAP, BIRP), document and spreadsheet analysis, a 99.90% uptime guarantee, and human support by phone, chat, email, and video on every plan. Suki is the stronger fit for large health systems standardized on Epic, Oracle Health, athenahealth, or MEDITECH that want voice commands, ambient order staging, and ambient CPT/E/M coding inside the EHR, and that have the budget for its reported $299 to $399 per user per month enterprise contracts. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/dax-copilot # DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) Jump to Comparison Table Microsoft Dragon Copilot, the product formerly sold as Nuance DAX Copilot and Dragon Medical One, and BastionGPT both turn patient conversations into structured clinical notes on HIPAA-eligible infrastructure. The overlap largely ends there. Dragon Copilot is an enterprise ambient documentation platform sold through Microsoft sales representatives and Cloud Solution Provider partners: Microsoft publishes no subscription price, every subscription carries a 12-month term, and an authorized reseller lists small-practice licensing at $159.99 per user per month plus a $350 one-time implementation fee and a $175 annual account maintenance fee. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on named frontier models (GPT, Claude, and Gemini), starting at $20 per user per month with a signed BAA on every plan and no contract, no implementation fee, and no IT project. Three Microsoft product names lead buyers to this comparison, so here is the taxonomy in plain English. Dragon Medical One is Microsoft's medical dictation and speech-recognition product. DAX Copilot is its ambient note-generation product. Dragon Copilot is the newer unified product that combines Dragon Medical One's dictation with DAX Copilot's ambient listening and generative AI. Microsoft acquired Nuance in 2022 and merged the two products into Dragon Copilot in March 2025. Searches for Nuance Dragon, Nuance DAX, DAX Copilot, and Dragon Medical One all lead to the same product line today, and this page compares BastionGPT against all of it. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, how you buy and deploy each one, what happens beyond the visit note, how each vendor handles your data, and what independent research says. The stakes are real: the AMA reported that 45.2% of physicians showed at least one symptom of burnout in 2023, with documentation burden a leading driver. Dragon Copilot's licensing, pricing, and data practices were verified against Microsoft's licensing guidance, Microsoft Learn and Microsoft Support documentation, Microsoft Marketplace listings, and authorized reseller listings on July 5, 2026. If you are searching for a DAX Copilot alternative without an enterprise contract, this comparison gives you the specifics: pricing, usage limits, data handling, and what each tool covers beyond dictation. BastionGPT is a Dragon Copilot alternative with $20 published pricing and a signed BAA on every plan. For the wider field, our AI scribe comparison covers other options such as Abridge, Suki, and DeepScribe. The quick answer: Choose BastionGPT if you want published pricing from $20 per user per month with no contract, an unlimited AI scribe and full AI assistant on every plan, 10-speaker recognition with automatic name and role labeling, 1,000+ pages of document analysis, named AI models (GPT, Claude, and Gemini), and a signed BAA on every plan including the free trial. Choose Dragon Copilot if you are a large health system standardized on Epic, buying through Microsoft enterprise agreements, with IT staff to manage Entra ID, the Dragon admin center, and Azure billing. For independent practices, small groups, and mental health teams, BastionGPT delivers more capability at a fraction of the total cost. Feature BastionGPT Dragon Copilot (formerly DAX Copilot / Nuance Dragon) Published pricing ✓ $20, $45, and $75 per user per month, listed publicly Microsoft publishes no subscription price; quotes come through Microsoft sales or CSP partners Entry cost for a small practice $20/user/month (Professional), no contract $159.99/user/month at an authorized reseller, plus $350 implementation fee and $175/year account maintenance fee (July 2026) Contract term Month to month; annual billing optional (includes a free month) 12-month term required on all subscriptions Free option 7-day free full trial, plus a 45-day money-back guarantee No self-serve trial; trials arranged through resellers or Microsoft sales How you buy Self-serve signup in minutes Microsoft Customer Agreement, CSP partner, or Enterprise Agreement; Microsoft Entra ID required; Marketplace listings require pre-purchase coordination with DragonMarketplace Usage limits Unlimited AI Scribe and AI assistant on every plan Flex licenses include 10 AI-Assisted Sessions per month, then per-session Azure pay-as-you-go billing Signed BAA ✓ Every plan, including the free trial; countersigned copy available on request Covered under Microsoft's volume licensing Product Terms, accepted through enterprise purchase channels HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP PIPEDA, GDPR, GDPR-UK Customer data used for AI model improvement Never; inputs, documents, and audio are never used to train AI models Microsoft states data is deidentified or pseudonymized for AI model improvement Data retention Chats and transcripts wiped after 30 days; users can retain longer or delete sooner Audio, transcript, and flowsheet data retained up to 90 days AI models GPT, Claude, Gemini (named and selectable) "Fine-tuned generative AI"; model versions not named or user-selectable Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles Multi-party capture supported; no published speaker count. In multilingual mode, US speaker separation covers English and Spanish only Max recording length 240 minutes per session No published limit for Dragon Copilot; DAX Copilot mobile recordings documented at a 120-minute maximum Telehealth capture ✓ Zoom, Teams, and Google Meet, with no visible bot in the meeting Not addressed in the public product documentation reviewed Documents per upload 10+ at once (Professional Plus and Ultra) Document upload and analysis is not a documented capability Document capacity 1,000+ pages of combined context Not applicable; chat is anchored to encounter transcripts and clinical information search Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not applicable; spreadsheet analysis is not a documented capability Full AI assistant ✓ Referral letters, appeals, patient education, document review, coding review, spreadsheet analysis Referral letters, after-visit summaries, evidence summaries, and orders, tied to the encounter workflow Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt; matches your writing style from sample notes Specialty models plus section-based customizable templates (e.g., HPI, Allergies, Physical Exam, Assessment and Plan), configured per organization Mental health documentation ✓ Dedicated therapy workflows: DAP, BIRP, intake assessments, treatment plans Public materials center on physician encounter and clinical-summary workflows Note language English output; 15+ transcription languages Captures 58 spoken languages; notes are written in English EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks, SimplePractice, AdvancedMD) with no integration project Embedded Epic integration; athenahealth and MEDITECH integrations; Dragon Medical One dictation integrates with 200+ EHRs; enterprise IT setup required Deployment time Minutes, self-serve Provisioning through Dragon admin center; enterprise deployments commonly reported at 3 to 6 months Uptime guarantee 99.90% Not published for Dragon Copilot Support Email, chat, phone, and video with real humans, on every plan Microsoft Standard Support included; faster response tiers cost extra Independent evidence 4.8/5 average across Capterra, Software Advice, and GetApp; 99.3% sentence-level accuracy in a peer-reviewed Baylor College of Medicine and Texas Children's Hospital study UCLA randomized trial (NEJM AI, Nov 2025): DAX documentation time reduction was not statistically significant; burnout measures improved modestly ## Where BastionGPT and Dragon Copilot are similar Both platforms transcribe the patient encounter with ambient AI, generate structured draft notes for clinician review, and support recording from a phone or desktop. Both run on Microsoft Azure infrastructure, encrypt data with AES-256 at rest and TLS in transit, and operate under HIPAA obligations with a Business Associate Agreement in place. Both extend past the note itself: each can draft referral letters and after-visit or patient-facing summaries. If you practice inside a large health system that has already deployed Dragon Copilot through its Epic environment, the core scribe workflow is covered. The differences appear the moment you try to buy it yourself, price it for a small team, upload a records packet, document a therapy session, or ask for anything the encounter workflow does not include. ## Where BastionGPT goes further ### Price you can see, and a total cost that is a fraction of Dragon's Microsoft publishes no subscription price for Dragon Copilot. Its official licensing guidance directs buyers to a Microsoft representative or a Cloud Solution Provider partner for a quote, and the Microsoft Marketplace listings for DAX Copilot and Dragon Medical One require pre-purchase coordination with DragonMarketplace before an organization can even buy. The concrete numbers that are public come from authorized resellers: as of July 2026, DictationOne lists Dragon Copilot Physician Practice licensing at $159.99 per user per month with a required 12-month term, a $350 one-time implementation fee, and a $175 annual account maintenance fee. Historic pricing for the same product line ran higher: reseller listings and industry reporting placed DAX Copilot between $369 and $830+ per provider per month before Microsoft's May 2026 price restructuring, and older reviews still cite those figures. BastionGPT publishes every price. Professional starts at $20 per user per month . Professional Plus, the plan most comparable to a full Dragon Copilot license, costs $45 and includes the full AI assistant, multi-document analysis, image and spreadsheet reading, and manual model selection. There is no implementation fee, no maintenance fee, and no required contract. Prices have held steady for over two years, and annual billing includes a free month. The math for a five-clinician practice, using the reseller's published figures: Dragon Copilot runs $9,599 per year in subscriptions ($159.99 x 5 x 12), plus $1,750 in first-year implementation fees and $875 per year in account maintenance, for a first-year total near $12,224 . BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12). That is a difference of more than $9,500 in year one , and roughly $7,770 every year after. ### No meters, no Azure billing plan, no session math Dragon Copilot's lower-cost license, Physician Flex, covers front-end dictation only. Ambient recording and generative AI are metered: each Flex license includes 10 AI-Assisted Sessions per month , pooled across the organization, with unused sessions expiring at month end. Anything beyond the allotment bills per session through a linked Azure subscription, which the organization must set up with its own resource group and billing plan. Microsoft's licensing guidance notes that an AI-Assisted Session is consumed by ambient note generation, chat queries, referral letters, and other generative features. A clinician who sees 20 patients a day exhausts a month of included sessions before lunch on day one. BastionGPT includes unlimited AI Scribe sessions and unlimited AI assistant use on every plan , including the $20 Professional plan. There is no session meter, no consumption unit, and no cloud billing plan to configure. ### A full AI assistant that reads your documents Dragon Copilot's generative features are anchored to the encounter: it drafts referral letters, after-visit summaries, evidence summaries, and orders from the visit, and its chat can query shift transcripts and search clinical information. Uploading and analyzing outside documents is not a documented capability, and spreadsheet analysis does not appear anywhere in Microsoft's product documentation. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a specialist evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The assistant also drafts insurance appeal letters, prior authorizations, treatment plans, and patient education materials; reviews documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyzes lab summaries and spreadsheets. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. ### A scribe that knows who is speaking BastionGPT recognizes up to 10 speakers per session , live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role (clinician, patient, parent, interpreter) in the transcript. It captures telehealth visits through Zoom, Teams, and Google Meet with no visible bot in the meeting. That matters for family sessions, group therapy, care conferences, and any visit with more than two voices in the room. Dragon Copilot supports multi-party conversation capture but publishes no speaker count, and Microsoft's support documentation lists a 120-minute maximum recording time for the DAX Copilot mobile app, half of BastionGPT's 240-minute ceiling. Microsoft's own release notes state that when multilingual recording is enabled, speaker separation in the US is available only for English and Spanish, and in Canada the multilingual output arrives as a single paragraph with no speaker identification at all. ### Notes in your format, your specialty, and your voice BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, intake assessments, progress notes, and treatment plans as first-class outputs, and a Custom Note option accepts any free-form prompt per recording . A single recording generates six output tabs at once: transcription, summary, SOAP, Detailed SOAP, DAP, and a custom note. Upload sample notes and BastionGPT matches your writing style, so the draft reads like you wrote it. Dragon Copilot's customization is real but differently shaped. The clinician's primary specialty determines which AI model generates the note, and customizable templates are section-based: Microsoft documents template customization for Allergies, Physical Exam, and Assessment and Plan in the DAX Copilot for Dragon Medical One workflow, with the Epic workflow adding History of Present Illness, Results, and Procedure. Template configuration happens at the organization level, so an individual clinician adjusts within the sections IT has set up. ### A strong fit for mental health and multidisciplinary teams BastionGPT publishes dedicated therapy and behavioral health workflows: DAP, BIRP, and custom therapy notes, intake assessments, and treatment plans, with 10-speaker recognition that handles group therapy and family sessions cleanly. It works alongside the systems mental health practices actually run, including SimplePractice, AdvancedMD, and Kareo, as well as hospital EHRs. Microsoft's public DAX and Dragon materials center on physician encounter documentation and clinical summaries. For a practice with therapists, psychiatrists, counselors, and case managers on one roster, BastionGPT is one platform that covers the entire organization. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of GPT, Claude, and Gemini . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Microsoft describes Dragon Copilot as running on "fine-tuned generative AI with healthcare-adapted safeguards." It does not name the model versions in use, and clinicians cannot select a model. ### Your data is never training data This difference is worth reading twice. Microsoft's Dragon Copilot FAQ states that audio, transcript, and flowsheet data is retained for up to 90 days, and that data is deidentified or pseudonymized for AI model improvement. Your patient encounters, in deidentified form, help train Microsoft's models. BastionGPT never uses your prompts, documents, or audio to train any AI model, deidentified or otherwise. Chats and transcripts are wiped after 30 days by default, and you can retain them longer or delete them sooner. Customer data is segregated by tenant, encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial ; the BAA is built into the standard terms of service, and a countersigned copy is available on request. BastionGPT supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. ### Buy it today instead of procuring it this quarter Dragon Copilot is purchased through a Microsoft Customer Agreement, a CSP partner, or an Enterprise Agreement, and the Marketplace listings require pre-purchase eligibility verification through DragonMarketplace. Deployment requires Microsoft Entra ID for authentication, provisioning through the Dragon admin center, and, for Flex licensing, a linked Azure subscription with a billing plan. The legacy DAX Copilot for Dragon Medical One workflow adds its own dependencies: recording runs through PowerMic Mobile on iOS, and note transfer to the EHR happens through Dragon Medical One on desktop. Independent reviews of the DAX product line describe enterprise implementation timelines of 3 to 6 months, and the reseller channel bills a per-user implementation fee before the first note is generated. BastionGPT deploys in minutes. A solo practitioner or a 50-clinician group signs up online, signs the BAA electronically, and starts documenting the same day. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, eClinicalWorks, SimplePractice, and AdvancedMD, via simple copy-paste, with no integration project and no IT department required. A Saved Prompts library with folder organization and team sharing is included on every plan, so a practice can standardize documentation across all providers from day one. ### What the independent evidence shows BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. Dragon Copilot is sold enterprise-only, so the most rigorous public evidence comes from independent research rather than review directories. A randomized clinical trial at UCLA Health, published in NEJM AI in November 2025 and funded by the university rather than any vendor, assigned 238 physicians across 14 specialties to DAX Copilot, a competing scribe, or usual care across roughly 72,000 patient encounters. The competing scribe reduced time-in-note by a statistically significant 9.5% versus control. DAX Copilot's reduction was smaller and did not reach statistical significance. Both tools produced modest improvements in burnout measures, on the order of 7%. Physician communities echo the mixed picture: recurring complaints include verbose assessment-and-plan sections that need trimming and pricing widely described as out of reach for anyone outside a large health system. Figures shift over time, so this page cites the source and date for each number. The links in the sources section lead to the live figures. ## When Dragon Copilot stands out Dragon Copilot's clearest advantage is enterprise depth inside Epic. It generates notes directly in the Epic chart environment, supports order suggestions from ambient recordings, and is deployed at more than 600 healthcare organizations with over 100,000 clinicians using it daily, including large academic health systems. If your health system has already purchased Dragon Copilot, standardized on Epic, and staffed the IT resources to run it, that embedded workflow is real value. Its dictation heritage matters too. Dragon Medical One's front-end speech recognition, with its medical vocabulary, cursor-level dictation into Windows applications, voice commands, auto-texts, PowerMic Mobile support, and integrations with more than 200 EHRs, remains a strength for clinicians who built their workflow around dictation. Microsoft also brings substantive trust signals: the DAX and Dragon Medical One products are HITRUST CSF certified on Azure, and Microsoft states that DAX clinical summaries follow AHDI documentation standards and stay grounded in what was said in the recording. Dragon Copilot also captures a wider range of spoken languages: Microsoft documents 58 languages for conversation capture, against 15+ for BastionGPT. Weigh that against the output side: Dragon Copilot's notes are written in English regardless of the spoken language, the platform does not auto-detect language, and US speaker separation in multilingual mode covers English and Spanish only. For independent practices, small groups, mental health teams, and any clinician who wants full AI capability without enterprise procurement (records review, appeal letters, spreadsheet analysis, multi-document context), BastionGPT delivers more breadth at published prices that run a fraction of Dragon Copilot's total cost. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS — Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does DAX Copilot (Microsoft Dragon Copilot) cost in 2026? Microsoft does not publish Dragon Copilot subscription pricing; quotes come through Microsoft sales representatives or Cloud Solution Provider partners. As of July 2026, authorized reseller DictationOne lists Physician Practice licensing at $159.99 per user per month with a required 12-month term, plus a $350 one-time implementation fee and a $175 annual account maintenance fee. Before Microsoft's May 2026 restructuring, DAX Copilot was widely reported between $369 and $830+ per provider per month. BastionGPT publishes its prices: $20 to $75 per user per month with no contract or fees. What is the difference between DAX Copilot, Dragon Medical One, and Dragon Copilot? Dragon Medical One is Microsoft's medical dictation and speech-recognition product: dictation at the cursor, voice commands, and EHR transfer. DAX Copilot is the ambient documentation product: it records the patient conversation and generates a draft clinical summary. Dragon Copilot is the unified product Microsoft introduced in March 2025 that combines both. All descend from Nuance, which Microsoft acquired in 2022. Is Dragon Copilot replacing DAX Copilot and Dragon Medical One? Yes. Microsoft positions Dragon Copilot as the successor experience combining Dragon Medical One's dictation with DAX Copilot's ambient listening and generative AI, and its licensing guidance and sales channels now center on Dragon Copilot. Searches for Nuance Dragon, Nuance DAX, DAX Copilot, and Dragon Medical One all lead to the same product line today. Does DAX Copilot require Dragon Medical One or PowerMic Mobile? In the DAX Copilot for Dragon Medical One workflow, yes: Microsoft's support documentation says the DAX mobile experience is accessed through PowerMic Mobile on iOS, recordings max out at 120 minutes, summaries are edited on mobile or through Dragon Medical One on desktop, and transfer to the EHR happens through Dragon Medical One. Microsoft also offers a distinct Epic-embedded DAX workflow in Epic Haiku, Canto, and Hyperdrive. BastionGPT records live or from uploaded audio in a browser or mobile app with no companion product required. Can an individual clinician or small practice buy Dragon Copilot directly? Not through a self-serve checkout. Dragon Copilot is sold through a Microsoft Customer Agreement, CSP partners, or Enterprise Agreements, requires Microsoft Entra ID for authentication, and carries a 12-month term on all subscriptions; Microsoft Marketplace listings require pre-purchase coordination with DragonMarketplace. BastionGPT offers self-serve signup with a 7-day free trial, a signed BAA on every plan including the trial, and month-to-month billing from $20 per user. Does Dragon Copilot have usage limits? The full Physician Per User license includes unlimited use of ambient and AI features. The lower-cost Physician Flex license covers dictation only and includes 10 AI-Assisted Sessions per month per license, pooled across the organization with no rollover; additional sessions bill through a linked Azure subscription on a pay-as-you-go basis. BastionGPT includes unlimited AI Scribe sessions and unlimited assistant use on every plan. Does Microsoft use Dragon Copilot data to train AI models? Microsoft's Dragon Copilot FAQ states that data is deidentified or pseudonymized for AI model improvement, and that audio, transcript, and flowsheet data is retained for up to 90 days. BastionGPT never uses customer prompts, documents, or audio to train AI models, and wipes chats and transcripts after 30 days by default, with user control to retain longer or delete sooner. Does Dragon Copilot sign a BAA? Yes, through Microsoft's standard channels: HIPAA Business Associate terms are included in the Microsoft Product Terms that organizations accept when they license the product through enterprise purchase channels. BastionGPT enters a signed BAA directly with every customer on every plan, including the free trial; the BAA is built into BastionGPT's standard terms of service, and a countersigned copy is available on request. Does BastionGPT require an EHR integration project? No. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, eClinicalWorks, SimplePractice, and AdvancedMD, through copy-paste and document-upload workflows. A practice signs up, signs the BAA electronically, and starts documenting the same day, with no IT project. Can BastionGPT do more than medical notes? Yes. BastionGPT is both an AI scribe and a secure general AI assistant for regulated healthcare work: therapy notes, patient education, referral letters, prior authorizations, insurance appeals, revenue-cycle documentation, legal and compliance record review, spreadsheet and lab-data analysis, and everyday administrative drafting, all inside a HIPAA-compliant environment. How many speakers can BastionGPT's AI Scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role, for example clinician, patient, or parent. Dragon Copilot supports multi-party capture but publishes no speaker count, and in multilingual mode its US speaker separation covers only English and Spanish. Can Dragon Copilot analyze uploaded documents or spreadsheets? Document upload and analysis is not a documented Dragon Copilot capability. Its generative features are anchored to the encounter: notes, referral letters, after-visit summaries, evidence summaries, orders, and chat over shift transcripts. Spreadsheet analysis does not appear in Microsoft's product documentation. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context, including PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. What AI models power BastionGPT and Dragon Copilot? BastionGPT uses licensed versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Microsoft describes Dragon Copilot as fine-tuned generative AI with healthcare-adapted safeguards and does not name the model versions or offer model selection. What does independent research say about DAX Copilot? A university-funded randomized clinical trial at UCLA Health, published in NEJM AI in November 2025, studied 238 physicians across 14 specialties and roughly 72,000 encounters. DAX Copilot's reduction in documentation time did not reach statistical significance versus usual care, while a competing scribe's 9.5% reduction did. Both tools showed modest burnout improvements of about 7%. Clinicians in the trial also reported occasional inaccuracies requiring ongoing vigilance. Does Dragon Copilot support more languages than BastionGPT? For capture, yes: Microsoft documents 58 spoken languages for recording the conversation, while BastionGPT supports 15+. Dragon Copilot writes all notes in English regardless of the spoken language, does not auto-detect language, and limits US speaker separation in multilingual mode to English and Spanish. If multilingual capture is your deciding requirement, weigh it against the differences in cost, contracts, document analysis, and data handling. Which is better for a small practice, BastionGPT or Dragon Copilot? For independent practices and small groups, BastionGPT is the stronger fit: published pricing from $20 per user per month, no 12-month contract, no implementation or maintenance fees, self-serve setup in minutes, a signed BAA on every plan, unlimited scribe and assistant use, and document analysis Dragon Copilot does not offer. Dragon Copilot fits large health systems that run Epic, buy through Microsoft enterprise agreements, and have IT teams to manage Entra ID, Dragon admin center, and Azure billing. Which is better for therapists and mental health practices? BastionGPT. It publishes dedicated behavioral health workflows with DAP, BIRP, intake, and treatment plan formats, handles group and family sessions with 10-speaker recognition, matches your writing style from sample notes, and works alongside SimplePractice, AdvancedMD, and Kareo. Microsoft's public DAX and Dragon materials center on physician encounter documentation. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/deepscribe # DeepScribe Alternative: BastionGPT vs DeepScribe (2026) Jump to Comparison Table DeepScribe and BastionGPT both offer HIPAA-compliant AI scribes that turn patient conversations into structured clinical notes. The overlap largely ends there. DeepScribe is an enterprise ambient scribe built for specialty care, strongest in oncology, sold through a sales process with custom Order Form pricing that the company does not publish. Third-party pricing guides estimate $350 to $500 per provider per month. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform actually costs in 2026, how you buy each one, what happens beyond the visit note, how each vendor handles your data, and what real users say on independent review sites. DeepScribe's features, published claims, and third-party ratings were verified against deepscribe.ai, its Terms of Use, its trust center, G2, and Capterra on July 8, 2026. If you are evaluating DeepScribe alternatives, this comparison lays out the specifics: published pricing vs quotes, scribe capability, and data handling. BastionGPT is a DeepScribe alternative with $20/month published pricing, a 10-speaker scribe, and a signed BAA on every plan. For the wider field, our AI scribe comparison covers other options such as Freed, Heidi Health, and Abridge. Quick answer: DeepScribe fits mid-size and enterprise specialty groups — strongest in oncology — that want deep EHR integration and ambient coding and can work through a sales process with custom Order Form pricing. BastionGPT is the DeepScribe alternative you can start today: unlimited scribing on every plan with up to 10 labeled speakers and 240-minute sessions, peer-reviewed 99.3% documentation accuracy, published $20 to $75 per user per month pricing, a full clinical AI assistant DeepScribe does not offer, and a signed BAA on every plan, including the 7-day free trial. Feature BastionGPT DeepScribe Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 4.1/5 on G2 (26 reviews); 4.3/5 on Capterra (8 reviews) as of July 2026 Published pricing $20 to $75/user/month, listed on the website Not published; fees set in a custom Order Form. Third-party guides estimate $350 to $500/user/month Entry paid plan $20/user/month (Professional) Custom enterprise quote only Free option 7-day free full trial, plus a 45-day money-back guarantee Demo through sales; no self-serve trial How you buy Self-serve signup in minutes Enterprise sales process with contract and implementation; reported setup fees of $500 to $1,000 Signed BAA ✓ Every plan, including the free trial Standard BAA executed as part of customer contracts HIPAA compliant ✓ ✓ Additional compliance and attestations 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP; runs on HITRUST CSF certified and SOC 2 Type II attested Azure infrastructure SOC 2 Type II (report available via its trust center); support for 42 CFR Part 2 or FERPA not published Data residency US, Canada, or Australia, based on billing geography United States, per its privacy policy Use of your data for AI training Never; inputs, documents, and audio are not used to train models Has trained proprietary models, including HealAI, on 3+ million de-identified clinical conversations captured on its platform AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Proprietary in-house models; published research describes HEAL, a 13B LLaMA 2-based model. Current production models not named Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles in the transcript ✓ Identifies patient, clinician, and third parties; maximum speaker count not published Max recording length 240 minutes per session No published limit Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers Vendor research and internal benchmarks on its HEAL model; no independent peer-reviewed accuracy study of production notes Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra) No user document analysis; AI pre-charting pulls context from EHR data instead Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG User uploads are not part of the documentation workflow; spreadsheets not supported Full AI assistant ✓ Referral letters, appeals, patient education, document review, spreadsheet and lab analysis Scribe, coding, and pre-charting; no general clinical AI chat AI coding Coding suggestions through the AI assistant Automated ICD-10, E/M, and HCC coding built into the ambient workflow Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Specialty-tuned templates with a Customization Studio that learns from edits Languages 15+ English and Spanish commercially; roughly 50 more in a beta program EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks) Bidirectional integration with Epic (SmartData), athenahealth, eClinicalWorks, AdvancedMD, OncoEMR, and iKnowMed Developer API ✓ Public API from $45/month, deployable in US, Canada, and Australia regions No public developer API offering listed Enterprise deployment Ultra plan runs within your existing enterprise cloud AWS-hosted; a customer-cloud deployment option is not published Target buyer Individual clinicians, small practices, and organizations Mid-size groups and enterprise health systems, with a specialty focus on oncology Uptime guarantee 99.90% Not published Support Email, chat, phone, and video with real humans, on every plan Chat, email, and phone, Monday to Friday, 5am to 5pm Pacific; onboarding through enterprise contracts ## Where BastionGPT and DeepScribe are similar Both platforms capture the natural conversation between clinician and patient and produce a structured, EHR-ready note for the clinician to review and sign. Both work for in-person and telehealth visits, both encrypt data with AES-256, and both execute Business Associate Agreements with covered entities. If your entire requirement is turning a spoken visit into a clinical note, both tools cover that core workflow. The differences appear the moment you ask for a price, request a trial, need documentation beyond the visit, upload a record for analysis, or ask what happens to your data after the encounter. ## Where BastionGPT goes further ### Published pricing you can act on today DeepScribe does not publish pricing. Its Terms of Use state that fees are set in an Order Form negotiated with each customer, and buying requires a sales consultation. Third-party pricing guides estimate $350 to $500 per provider per month, with some estimates ranging from $200 to $750 depending on volume, integration depth, and features, plus reported setup fees of $500 to $1,000 and annual contracts. BastionGPT Professional starts at $20 per user per month , listed publicly. Professional Plus costs $45 and includes the full AI assistant, multi-document analysis, image and spreadsheet reading, and manual model selection. Prices have held steady for over two years, and annual billing includes a free month. The math for a five-clinician practice, using the low end of the third-party estimates for DeepScribe: $350 x 5 x 12 comes to $21,000 per year. BastionGPT Professional Plus runs $2,700 ($45 x 5 x 12). That is a difference of $18,300 per year , and it grows to $27,300 at the high end of the estimate range. A solo clinician pays $540 per year for BastionGPT Professional Plus versus an estimated $4,200 or more for DeepScribe. ### A trial you can start today DeepScribe is enterprise sales only. There is no self-serve trial: evaluation starts with a demo request, and deployment runs through contracting and implementation. That process fits a health system procurement cycle and rules out most solo and small-group buyers. BastionGPT offers a 7-day free trial with full access to the AI scribe and AI assistant, no sales call required, and every paid plan carries a 45-day money-back guarantee. You can record your first session the same day you sign up. ### A scribe that knows who is speaking BastionGPT's AI Scribe recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes . Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role, for example clinician, patient, or parent. That matters for family sessions, group therapy, care conferences, and interpreter-assisted visits. DeepScribe identifies the patient, clinician, and third parties in the room, but publishes no maximum speaker count and no session length limit, so buyers cannot verify capacity for group settings before signing a contract. Accuracy is documented, not asserted: BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, and reports that AI scribe users reclaim an average of 90 minutes per day. DeepScribe publishes research on its in-house HEAL model and reports internal benchmarks; an independent peer-reviewed accuracy study of its production notes is not published. ### A full AI assistant, not just the visit note DeepScribe's platform centers on the encounter: ambient scribing, AI pre-charting that pulls prior notes, labs, and medications from the EHR, and automated coding. It does not include a general clinical AI chat, and it does not analyze documents you upload. BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. ### Document and spreadsheet analysis On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a full patient history, a psychological evaluation, and records from multiple providers, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. DeepScribe approaches records differently: its AI pre-charting aggregates context from data already inside the EHR. That helps at the point of care, and it also means there is no path for uploading and analyzing outside records, insurance documents, legal correspondence, or spreadsheets. Excel and CSV analysis is not part of the product. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Naming the models lets you verify exactly what is reading your records. DeepScribe builds on proprietary in-house models. Its published research describes HEAL, a 13-billion-parameter model based on LLaMA 2 and trained on clinical conversations, and the company reports strong internal benchmarks. It does not name the models that power production notes today, so note quality rests on a stack you cannot inspect or compare. ### Your data stays yours, and stays out of training The two platforms take opposite approaches to training data. DeepScribe has publicly described processing and labeling more than 3 million de-identified clinical conversations captured on its platform to train proprietary models, including HealAI, and describes this dataset as the core advantage behind its AI. Customer encounters, once de-identified, feed model improvement. BastionGPT never uses your inputs, documents, or audio to train AI models , and enters a signed Business Associate Agreement with every customer on every plan, including the free trial. BastionGPT supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements, with PHI residency in the US, Canada, or Australia based on billing geography. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. Retention stays in your control: chats persist until you delete them, documents are removed on deletion or after 30 days of account inactivity, and transcript retention is user-configurable from 1 to 365 days. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. DeepScribe executes a standard BAA with its customers as part of contracting, encrypts data with AES-256, and maintains a public trust center with a SOC 2 Type II report. Healthcare-specific frameworks beyond HIPAA, such as 42 CFR Part 2 for substance use disorder records or FERPA for school-based services, are not published on its site. For a practice handling SUD or school-based records, that difference is worth a direct question during procurement. ### Note flexibility on every plan BastionGPT produces SOAP, Detailed SOAP, DAP, BIRP, H&P, and progress notes, and a Custom Note tab accepts any free-form prompt per recording : narrative formats, parent-friendly summaries, forensic documentation, and other specialty outputs. You can also upload example notes so BastionGPT matches your writing style. A Saved Prompts library with folder organization and team sharing is included on every plan. DeepScribe generates notes from specialty-tuned templates, and its Customization Studio learns from clinician edits over time. Customization is a real strength of the product, and it is delivered through enterprise configuration rather than a per-recording free-form prompt you control from day one. ### Deployment paths from solo practice to enterprise cloud BastionGPT deploys three ways. Individual clinicians and practices run it in the browser on any device with no IT project. Organizations that want to embed AI or build direct EHR connections use the public BastionGPT API, priced from $45 per month with deployment regions in the US, Canada, and Australia. Large organizations on the Ultra plan ($75 per user per month, starting at 100 users) can run BastionGPT within their existing enterprise cloud with SSO. DeepScribe delivers as an AWS-hosted service and is available through the AWS Marketplace. A public developer API product or a customer-cloud deployment option comparable to BastionGPT Ultra is not published. ### User ratings on independent review sites BastionGPT holds a 4.8 out of 5 average across the three major verified B2B review directories: Capterra, Software Advice, and GetApp. Reviewers most often cite time saved on documentation and the freedom to use AI on real cases because data protection is built in. DeepScribe holds a 4.1 out of 5 on G2 across 26 verified reviews and a 4.3 out of 5 on Capterra across 8 reviews as of July 2026. Positive reviews praise reliability, responsive support, and reclaimed evenings and weekends. Critical reviews most often report transcriptions lost during server outages, findings placed in the wrong note sections, occasional speaker attribution errors, and notes stuck in processing when connectivity drops. Ratings shift over time, so this page cites the source and date for each figure. The links in the sources section lead to the live numbers. ## When DeepScribe stands out DeepScribe's clearest advantage is EHR depth. It offers bidirectional integration with Epic (including SmartData elements and mobile capture through Haiku and Canto), athenahealth, eClinicalWorks, AdvancedMD, and the oncology EHRs OncoEMR and iKnowMed: the patient schedule syncs into the app, and finished notes write back into discrete EHR fields. BastionGPT works alongside 50+ EMR/EHR systems through simple copy-paste and document workflows without an IT project, which deploys faster but does not write into discrete fields. DeepScribe has also built real depth in specialty care. It reports capturing more than 5 million oncology visits per year, offers SmartPrep pre-visit intelligence for oncology, and automates ICD-10, E/M, and HCC coding inside the ambient workflow. Its about page reports 2,800+ clinics and health systems served and a 99.92% note satisfaction rating. In January 2025 it scored 98.8 out of 100 in a KLAS Research Emerging Company Spotlight, a strong result that KLAS notes was based on limited data, and KLAS's 2025 ambient speech report scored it 97.6 across seven organizations. A published Ochsner Health case study reports 78% clinician adoption and 96% patient satisfaction. Those capabilities are packaged for enterprise buyers: custom contracts, unpublished pricing that third parties estimate at $350 to $500 per provider per month, implementation projects, and no self-serve trial. For a large oncology group standardizing on Epic write-back and automated coding, DeepScribe is built for that procurement. For individual clinicians, small and mid-size practices, and any team that needs documentation beyond the visit note (records review, appeal letters, reports, spreadsheet analysis), BastionGPT delivers a broader toolset at a small fraction of the estimated price, with a trial you can start today. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does DeepScribe cost in 2026? DeepScribe does not publish pricing; its Terms of Use set fees in a custom Order Form negotiated with each customer. Third-party pricing guides estimate $350 to $500 per provider per month, with reported setup fees of $500 to $1,000 and annual contracts. BastionGPT publishes its plans at $20 to $75 per user per month. Does DeepScribe offer a free trial? No self-serve trial is available. DeepScribe is sold through an enterprise sales process that starts with a demo request and runs through contracting and implementation. BastionGPT offers a 7-day free trial with full access, plus a 45-day money-back guarantee on every paid plan. Does DeepScribe sign a BAA? Yes. DeepScribe's Terms of Use include a standard Business Associate Agreement executed with its customers as part of contracting. BastionGPT includes a signed BAA on every plan, including the free trial, through a self-serve process that takes minutes. Does DeepScribe use patient data to train its AI? DeepScribe has publicly described processing and labeling more than 3 million de-identified clinical conversations captured on its platform to train proprietary models, including HealAI, and describes this dataset as the core advantage behind its AI. BastionGPT never uses customer inputs, documents, or audio to train AI models, on any plan. What AI models power BastionGPT and DeepScribe? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. DeepScribe uses proprietary in-house models; its published research describes HEAL, a 13B LLaMA 2-based model, and it does not name current production models. Can DeepScribe analyze uploaded documents? No. DeepScribe's AI pre-charting pulls context from data already inside the EHR, and user document upload and analysis is not part of the product. Spreadsheets are not supported. BastionGPT reads 10+ documents and 1,000+ pages at once, including PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. How many pages and documents can BastionGPT analyze at once? On Professional Plus and Ultra, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context. The Professional plan covers a single PDF, Word, or text document up to 30 pages. How many speakers can BastionGPT's AI Scribe recognize? BastionGPT recognizes up to 10 speakers per session, live or from uploaded audio, with sessions up to 240 minutes. Speaker diarization separates the transcript by voice, and BastionGPT's AI automatically labels each speaker's name and role. DeepScribe identifies the patient, clinician, and third parties but does not publish a maximum speaker count. Is BastionGPT's AI scribe as accurate as DeepScribe's? BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. DeepScribe publishes research on its in-house HEAL model and reports internal benchmarks, but no independent peer-reviewed accuracy study of its production notes. Both platforms expect clinicians to review and approve every note. How do BastionGPT and DeepScribe user ratings compare? BastionGPT holds a 4.8 out of 5 average across the verified review directories Capterra, Software Advice, and GetApp. DeepScribe holds a 4.1 out of 5 on G2 (26 reviews) and a 4.3 out of 5 on Capterra (8 reviews) as of July 2026. Ratings change over time, so check each review site for current figures. Does DeepScribe work for solo practitioners or small practices? DeepScribe is built and sold for mid-size groups and enterprise health systems, with custom contracts, implementation projects, and third-party price estimates of $350 to $500 per provider per month. BastionGPT serves individual clinicians and small practices with self-serve signup from $20 per user per month and a signed BAA on every plan. Does DeepScribe integrate with Epic? Yes. DeepScribe offers bidirectional Epic integration, including SmartData elements, along with integrations for athenahealth, eClinicalWorks, AdvancedMD, OncoEMR, and iKnowMed. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, through simple copy-paste and document workflows with no IT project, and organizations can build direct EHR connections with the BastionGPT API. Can BastionGPT run inside our own cloud environment? Yes. The BastionGPT Ultra plan ($75 per user per month, starting at 100 users) runs within your existing enterprise cloud with SSO, and the public BastionGPT API (from $45 per month) supports deployment regions in the US, Canada, and Australia. DeepScribe is delivered as an AWS-hosted service; a customer-cloud deployment option is not published. Which is better for most practices, BastionGPT or DeepScribe? For a large specialty group, particularly in oncology, that wants Epic write-back and automated E/M, ICD-10, and HCC coding and can run an enterprise procurement, DeepScribe is built for that buyer. For individual clinicians, small and mid-size practices, and any team that needs a scribe plus a full AI assistant, document analysis, and a signed BAA from day one, BastionGPT delivers more capability at published prices starting at $20 per user per month. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/ambience-healthcare # Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 Jump to Comparison Table Ambience Healthcare and BastionGPT both turn patient conversations into structured, HIPAA-compliant clinical notes. They are built for different buyers. Ambience is an enterprise AI platform for documentation, coding, and clinical documentation integrity (CDI), sold to health systems through custom contracts: there is no published pricing, no self-serve signup, and no free trial. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant that any clinician or practice can start using today. It recognizes up to 10 speakers and labels each speaker's name and role automatically, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, and runs on licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: who can actually buy each platform, what each one costs in 2026, what happens beyond the visit note, how each vendor handles patient data, and how long deployment takes. Ambience Healthcare's public claims were verified against ambiencehealthcare.com, its Trust Center, its privacy policy, and independent industry coverage on July 8, 2026. Ambience does not publish pricing; where estimates appear below, the third-party source is named. If you are researching Ambience Healthcare alternatives that fit a smaller practice, this comparison covers the details: published pricing, setup time, and compliance coverage. BastionGPT is an Ambience alternative you can set up the same day, with $20 published pricing and a signed BAA on every plan. For the wider field, our AI scribe comparison covers other options such as Abridge, DeepScribe, and Suki. Quick answer: Ambience Healthcare fits large health systems buying an enterprise ambient platform — deep EHR integration, AutoCDI coding, 200+ specialty templates — through a sales process; third-party estimates put it at $233 to $417 per provider per month, and a small group typically cannot purchase it at all. BastionGPT is the Ambience alternative a practice can run today: unlimited scribing with up to 10 labeled speakers and 240-minute sessions on every plan, peer-reviewed 99.3% documentation accuracy, a full clinical AI assistant, published $20 to $75 per user per month pricing, and a signed BAA on every plan, including the free trial. Feature BastionGPT Ambience Healthcare Who can buy it Any clinician, practice, or organization; self-serve signup Health systems via enterprise sales; no self-serve signup, trial, or small-practice tier Published pricing $20 to $75/user/month, listed publicly None published; third-party estimates of $233 to $417/provider/month plus a one-time implementation fee (DeepCura, March 2026) Free option 7-day free full trial, plus a 45-day money-back guarantee No free trial Signed BAA ✓ Every plan, including the free trial HIPAA Business Associate; privacy policy states customer agreements may include BAAs as applicable and required HIPAA compliant ✓ ✓ Additional compliance 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP GDPR, SOC 2 Type I and Type II Patient data used outside the care relationship Never shared with AI providers or used for training; chats and transcripts wiped after 30 days Privacy policy permits disclosure of de-identified patient information to third-party vendors and academic researchers where customer contracts and HIPAA allow AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models (named and selectable) Built on OpenAI models via Microsoft Azure; no user model selection Full AI assistant ✓ Referral letters, appeals, patient education, document review, coding review EHR-anchored modules (AutoScribe, AutoCDI, AutoAVS, AutoRefer); no general-purpose assistant Document uploads 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra) Reads the patient chart through EHR integration; no published support for user-uploaded documents Supported upload types PDF, Word, Excel, CSV, PowerPoint, HTML, text, JPG, PNG Not applicable (chart data via EHR) Multi-speaker recognition ✓ Up to 10 speakers, with automatic AI labeling of speaker names and roles ✓ Speaker attribution supported; speaker count not published Max recording length 240 minutes per session Not published Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers KLAS satisfaction recognition (97.7, limited data); no peer-reviewed accuracy study published Note types SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt Specialty-tuned templates across 200+ specialties Point-of-care coding Coding review and suggestions through the AI assistant ✓ AutoCDI surfaces ICD-10, CPT, E/M, and HCC guidance in real time EHR compatibility Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, eClinicalWorks); HIPAA-compliant API for custom integrations Native integration with Epic, Oracle Health (Cerner), and athenahealth; enterprise implementation required Deployment time Minutes; no IT project 4-week average implementation published by Ambience; independent guides cite 2 to 6 weeks of EHR configuration, longer for full rollouts Languages 15+ Live translation and multilingual patient summaries publicized; total language count not published Third-party validation 4.8/5 average across the verified review directories Capterra, Software Advice, and GetApp KLAS recognition: 97.7 satisfaction score (limited data) and the 2026 KLAS/CHIME Trailblazer Award; limited consumer review-site coverage Uptime guarantee 99.90% No public SLA; terms set per enterprise contract Support Email, chat, phone, and video with real humans, on every plan Enterprise implementation and support teams ## Where BastionGPT and Ambience Healthcare are similar Both platforms generate structured clinical notes from ambient conversation, attribute speech to different people in the room, and support high-complexity specialties, including psychiatry and other mental health settings. Both encrypt patient data, operate as HIPAA Business Associates, and run on independently audited infrastructure. Both have meaningful third-party validation: BastionGPT holds a 4.8 out of 5 average across the major verified software review directories, and Ambience has earned strong marks from KLAS among enterprise health systems. If the only question is whether the underlying scribe technology works, both companies clear that bar. The differences appear the moment you ask who can buy it, what it costs, what happens beyond the visit note, and where your patients' data can travel. ## Where BastionGPT goes further ### You can buy it today Ambience sells exclusively to health systems through an enterprise sales process. Independent reviews confirm there is no self-serve signup, no free trial, and no small-practice tier, and position the platform as best suited to organizations with 50 or more providers (DeepCura, March 2026). Deployment is an IT project: Ambience's own informatics page lists an average implementation time of 4 weeks with white-glove onboarding, independent guides describe 2 to 6 weeks of EHR configuration and security sign-off (OmniMD, July 2026), and full health-system rollouts can stretch to months depending on system size. BastionGPT deploys in minutes. Sign up, start the 7-day free trial, and record your first session the same day , with a signed BAA in place from the moment you begin. There is no procurement cycle, no committee, and no IT dependency. For a clinician whose health system has not rolled out an ambient tool, or has rolled one out that they cannot use the way they want, that difference is the whole decision. ### Published pricing you can verify Ambience does not publish pricing anywhere on its website. Third-party analysis estimates base AutoScribe contracts at roughly $2,800 to $3,200 per provider per year ($233 to $267 per month), with the full suite reaching $4,000 to $5,000 per provider per year ($333 to $417 per month), plus a one-time implementation fee covering EHR integration and template configuration (DeepCura, March 2026). Other industry guides place entry pricing near $200 per provider per month before enterprise add-ons (AssemblyAI, June 2026). Because every contract is custom, the only way to learn your actual price is to enter the sales process. BastionGPT publishes its prices: Professional at $20 per user per month, Professional Plus at $45, and Ultra at $75 , with annual billing that includes a free month. Prices have held steady for over two years. The math for a five-clinician practice: at the third-party full-suite estimate, Ambience would run $20,000 to $25,000 per year before implementation fees. BastionGPT Professional Plus runs $2,700 per year ($45 x 5 x 12). That is a difference of roughly $17,000 to $22,000 per year , and in practice a five-clinician group typically cannot purchase Ambience at all. ### A scribe that knows who is speaking BastionGPT's AI Scribe is included on every plan: record live or upload audio, capture sessions up to 240 minutes , and let speaker diarization separate up to 10 speakers while the AI labels each speaker's name and role automatically. Family sessions, group visits, care conferences, and multi-provider discussions come out as a transcript that reads like a script. Ambience supports speaker attribution but publishes no speaker count and no session length limit. Accuracy is documented, not asserted: BastionGPT cites 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Ambience's KLAS recognition measures user satisfaction; a peer-reviewed accuracy study is not published. ### A full AI assistant, not only a scribe Every Ambience module anchors to the EHR encounter: AutoScribe documents the visit, AutoCDI validates the coding, AutoAVS writes the after-visit summary, AutoRefer drafts the referral, and chart-aware tools prepare the clinician beforehand. That is a deep pipeline for one workflow. It does not include a general-purpose assistant, and Ambience publishes no support for uploading your own documents, spreadsheets, or images for analysis. BastionGPT includes an unlimited general AI assistant on every plan . Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. ### Read a whole record, not just the chart in front of you On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once, up to 1,000+ pages of combined context . Supported formats include PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned PDFs, charts, and graphs. A clinician can upload a psychological evaluation, outside records from multiple providers, and an insurance denial letter, then query, summarize, and cross-reference all of it in one conversation. Ambience reads the patient's longitudinal chart through its EHR integration, which is valuable inside a deployed health system. It publishes no capability for records that live outside that EHR: faxed outside records, legal documents, spreadsheets, or anything a clinician receives directly. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models . Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your records. Ambience builds on OpenAI models delivered through Microsoft Azure, per its Microsoft Marketplace listing and its own product announcements. That is a single-vendor model stack, and clinicians cannot see or select which model handles a given task. ### A stricter boundary around patient data BastionGPT enters a signed Business Associate Agreement with every customer on every plan, including the free trial . Prompts, documents, and audio are never shared with OpenAI or any other AI provider for training, chats and transcripts are wiped after 30 days by default, and data is encrypted with AES-256 at rest and TLS in transit on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, with independent penetration testing at least annually. Organizations with stricter requirements can host regionally in the United States, Canada, or Australia, and the Ultra plan can deploy within your organization's own enterprise cloud. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. Ambience operates as a HIPAA Business Associate and maintains SOC 2 Type I and Type II certification. Its privacy policy also states that it discloses de-identified patient information to third-party service providers and academic researchers where its customer contracts and HIPAA permit, using the safe harbor or expert determination methods. That is legal under HIPAA, and it is also a clause worth reading closely during procurement if your organization has told patients their data stays inside the care relationship. ### Compliance coverage beyond HIPAA BastionGPT supports HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements, with DocuSign BAAs available on request. Substance-use treatment programs and school-based providers get named regulatory coverage on every plan. Ambience publishes HIPAA and GDPR adherence alongside its SOC 2 certifications. It does not publish 42 CFR Part 2 or FERPA support. ### Human support on every plan BastionGPT provides support by email, chat, phone, and video with real humans on every plan , backed by a 99.90% uptime guarantee. Ambience provides enterprise implementation and support teams to its health system customers, with SLA terms set per contract. If you are one clinician inside a large deployment, your support path runs through your organization's IT department. ## When Ambience Healthcare stands out Ambience's clearest advantage is native EHR integration at health-system scale. It lives directly inside Epic Toolbox (Hyperdrive and Haiku) using the Epic Ambient Module and FHIR APIs, integrates with Oracle Health's Cerner Millennium, and deploys through the athenahealth Marketplace, writing notes and structured data straight into the chart. If your requirement is an enterprise-wide, EHR-embedded rollout with no copy and paste, that is what Ambience was engineered for. Its second advantage is point-of-care revenue integrity and documented enterprise outcomes. AutoCDI surfaces ICD-10 and CPT codes, E/M level guidance, and HCC opportunities in real time. Ambience's own customer stories report a 38.8% decrease in documentation time, a 40.2% drop in after-hours documentation, 22.8% more patient face time, and a 25% reduction in severe clinician burnout at St. Luke's Health System, plus roughly $3 million in estimated savings tied to reduced provider turnover at John Muir Health. Its homepage cites 80% average clinician utilization and 45% less charting time across deployments. The enterprise market has responded: Cleveland Clinic selected Ambience for an exclusive five-year contract after a six-month, five-vendor head-to-head pilot, deployments run at Houston Methodist, UCSF Health, and Memorial Hermann, and the company raised a $243 million Series C at a $1.25 billion valuation in July 2025. KLAS reports a 97.7 satisfaction score (limited data) and named Ambience the 2026 KLAS/CHIME Trailblazer Award winner. All of that arrives through an enterprise contract, an unpublished price, and an IT-led implementation. For solo practitioners, small and mid-size practices, and clinicians whose organizations have not deployed an ambient platform, BastionGPT delivers the scribe plus a full AI assistant today, at a published price, with a signed BAA from the first minute of the free trial. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Ambience Healthcare cost in 2026? Ambience Healthcare does not publish pricing; every contract is negotiated through enterprise sales. Third-party estimates place base AutoScribe contracts at $2,800 to $3,200 per provider per year and full-suite contracts at $4,000 to $5,000 per provider per year, plus a one-time implementation fee (DeepCura, March 2026). BastionGPT publishes its plans at $20 to $75 per user per month. Can solo practitioners or small practices buy Ambience Healthcare? No. Ambience sells to health systems through an enterprise sales process, with no self-serve signup, no free trial, and no small-practice tier; independent reviews position it for organizations with 50 or more providers. BastionGPT offers self-serve signup for individual clinicians and practices of any size, with a 7-day free trial and a signed BAA on every plan. Does Ambience Healthcare offer a free trial? No. Ambience has no free trial and no self-serve signup; evaluation happens through vendor-led pilots negotiated by health systems. BastionGPT offers a 7-day free trial with full access to the AI scribe and AI assistant, no charge during the trial, and a 45-day money-back guarantee on paid plans. Does Ambience Healthcare sign a BAA? Ambience operates as a Business Associate under HIPAA, and its privacy policy states that its customer agreements may include BAAs as applicable and required. There is no self-serve BAA path because there is no self-serve product. BastionGPT includes a signed BAA on every plan, including the free trial, so a solo clinician is covered from day one. Does Ambience Healthcare share or train on patient data? Ambience's privacy policy states that it discloses de-identified patient information to third-party service providers and academic researchers when its customer contracts and HIPAA permit, using HIPAA's safe harbor or expert determination methods. BastionGPT never shares prompts, documents, or audio with OpenAI or other AI providers for training, and wipes chats and transcripts after 30 days by default. What AI models power BastionGPT and Ambience Healthcare? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection. Ambience builds on OpenAI models delivered through Microsoft Azure, per its Microsoft Marketplace listing and product announcements, and does not offer clinicians a model choice. Is BastionGPT's AI scribe as accurate as Ambience Healthcare's? BastionGPT cites 99.3% sentence-level accuracy on clinical documentation tasks from a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers; the clinical validation page covers the methodology. Ambience's KLAS scores reflect user satisfaction rather than a published accuracy rate. Both platforms expect clinicians to review and approve every note. How long does Ambience Healthcare take to implement? Ambience's own informatics page lists an average implementation time of 4 weeks with white-glove onboarding. Independent guides describe 2 to 6 weeks of IT-supported EHR configuration and security review, and full health-system rollouts can run weeks to months (OmniMD, July 2026; DeepCura, March 2026). BastionGPT deploys in minutes: clinicians sign up, accept the BAA, and record their first session the same day. Which EHRs does each platform work with? Ambience integrates natively with Epic (through Epic Toolbox and the Ambient Module), Oracle Health's Cerner Millennium, and athenahealth, writing directly into the chart; that integration requires enterprise implementation. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and eClinicalWorks, with no integration project required, and offers a HIPAA-compliant API for custom integrations. Does BastionGPT do medical coding like Ambience AutoCDI? Ambience's AutoCDI surfaces ICD-10, CPT, E/M, and HCC guidance in real time inside the EHR, and that point-of-care coding depth is its standout feature for health systems. BastionGPT's AI assistant reviews documentation for suspected under- or over-coding and drafts coding suggestions on request, alongside referral letters, appeals, and document analysis that Ambience does not offer. Can Ambience Healthcare analyze uploaded documents or spreadsheets? Ambience reads the patient's longitudinal chart through its EHR integration, but publishes no support for user-uploaded documents, spreadsheets, or images. BastionGPT reads 10+ uploaded documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, covering PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files. How do BastionGPT and Ambience Healthcare ratings compare? BastionGPT holds a 4.8 out of 5 average across the verified review directories Capterra, Software Advice, and GetApp. Ambience's strongest third-party validation comes from KLAS rather than consumer review sites: a 97.7 KLAS satisfaction score (limited data) and the 2026 KLAS/CHIME Trailblazer Award. Ratings change over time, so check each source for current figures. Which is better for a small or mid-size practice, BastionGPT or Ambience? For practices below enterprise scale, BastionGPT is the practical choice: Ambience does not sell to small practices, publishes no pricing, and requires an IT-led implementation. BastionGPT costs $20 to $75 per user per month, includes a signed BAA and unlimited AI scribe on every plan, adds a full AI assistant for documents and correspondence, and deploys the same day. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/mentalyc # Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc Jump to Comparison Table Mentalyc and BastionGPT both turn therapy sessions into HIPAA-compliant progress notes, and both sign a Business Associate Agreement with every account. They are built around different ideas of what a clinician needs. Mentalyc is a therapy-specialist note taker: it converts session audio or a typed recap into SOAP, DAP, or BIRP notes, adds treatment planning and progress tracking, and sells through tiers from $14.99 to $99.99 per month (billed annually) that meter documentation with monthly note credits (40 to 330), cap sessions at 75 to 130 minutes, and reserve couples, family, group, EMDR, and psychiatry features for higher tiers. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: unlimited notes and transcription on every plan, sessions up to 240 minutes with up to 10 auto-labeled speakers, document analysis across 1,000+ pages, named frontier models (GPT, Claude, and Gemini), and named support for 42 CFR Part 2 and FERPA records, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform costs at a real caseload, where note credits and session caps bite, what happens beyond the progress note, how each vendor handles your clients' data, and which records each platform is built to touch. Mentalyc's pricing, plan limits, features, and compliance claims were verified against mentalyc.com, its pricing page, and its FAQ on July 11, 2026. If you are weighing Mentalyc alternatives for your practice, the sections below compare what matters to therapists: note limits, session length, pricing, and compliance. BastionGPT is a Mentalyc alternative with unlimited therapy notes and a full AI assistant at $20 per month. For the wider field, our AI scribe comparison covers other options such as Freed, Twofold Health, and Heidi Health. Quick answer: Mentalyc fits solo therapists who want a therapy-only note tool with monthly note credits inside a web-based EHR. BastionGPT gives behavioral health clinicians and group practices unlimited notes and transcription from $20 per user per month, 240-minute sessions, every note format and client type on every plan, and a full clinical AI assistant that also reads records, drafts letters and appeals, and supports psychiatry, billing, and admin teams. Feature BastionGPT Mentalyc Verified user ratings 4.8/5 average across Capterra, Software Advice, and GetApp 4.5/5 on Capterra (60 reviews) and 4.4/5 on Trustpilot (65 reviews) as of July 2026 Published pricing $20 to $75/user/month, listed on the website $14.99 to $99.99/month billed annually ($19.99 to $119.99 billed monthly); Team seats $49.99/month billed annually ($59.99 monthly) Monthly note limits None; notes, chat, and transcription are unlimited on every plan 40 notes (Mini), 100 (Basic), 160 (Pro), or 330 (Super) per month; unlimited requires a Team seat Max session recording length 240 minutes per session, on every plan 75 minutes on Mini and Basic; 130 minutes on Pro, Super, and Team Free option 7-day free full trial, plus a 45-day money-back guarantee 14-day trial with Pro-level access, capped at 15 notes; no credit card required Signed BAA ✓ Every plan, including the free trial ✓ Every plan; downloadable from account settings HIPAA compliant ✓ ✓ Additional compliance and attestations 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP; runs on HITRUST CSF certified and SOC 2 Type II attested infrastructure SOC 2 Type II; PIPEDA, PHIPA, PIPA, PHIA (Canada); Australian APP 42 CFR Part 2 (substance use records) ✓ Named support on every plan Its FAQ states it provides HIPAA-aligned safeguards and that Part 2 consent and disclosure obligations remain the provider's responsibility Data residency US, Canada, or Australia, based on billing geography United States (its FAQ cites US-based data residency) Use of your data for AI training Never; inputs, documents, and audio are not used to train models States session data is never used to train AI models AI models GPT, Claude, Gemini (named and selectable) Not named; its FAQ describes machine learning algorithms, large language models, and statistical models Full AI assistant ✓ Unlimited clinical AI chat on every plan: referral letters, appeals, patient education, document review, spreadsheet and lab analysis Not offered; the platform centers on session documentation and therapy workflow tools Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra); single documents up to 30 pages on Professional Not part of the workflow; inputs are session audio, dictated recaps, or typed summaries Ways to capture a session Live recording or uploaded audio, with bot-free capture alongside Zoom, Teams, and Google Meet; typed summaries through the AI assistant Live listening, uploaded audio, dictated recap, or typed summary Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles Maximum speaker count not published; group sessions produce a collective note plus per-member notes on Super and Team Note formats at the entry price SOAP, Detailed SOAP, DAP, BIRP, and free-form custom formats on every plan, with writing-style matching from your sample notes SOAP, DAP, and intake notes on Mini and Basic; BIRP, PIRP, GIRP, PIE, SIRP, and 100+ templates require Pro Couples, family, and group sessions Included on every plan Individual adults only on Mini and Basic; couples, kids, and family on Pro; group therapy notes on Super and Team Specialty modalities (EMDR, play therapy, psychiatry) Any modality via custom prompts and writing-style matching, on every plan EMDR, play therapy, and psychiatry templates on Pro and above Treatment plans and progress tracking Treatment plans, discharge summaries, and progress summaries through the AI assistant ✓ Dedicated AI Treatment Planner (Basic and above), session-by-session progress tracking, and Alliance Genie insights CPT coding Coding suggestions and documentation review through the AI assistant Auto-computed CPT codes on Pro and above EHR workflow Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, SimplePractice), web-based or desktop, with no browser extension required Chrome browser extension autofills notes into web-based EHRs (SimplePractice, TherapyNotes, Jane, TheraNest, Osmind, Tebra, Kareo); supports EHRs that run in a browser Developer API ✓ Public API from $45/month, deployable in US, Canada, and Australia regions No public API documentation; the published integration path is its Chrome extension Enterprise deployment Ultra plan runs within your existing enterprise cloud, with SSO Multi-tenant SaaS hosted on AWS Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published; vendor-reported metrics include progress notes in about 3 minutes and savings of 3 to 5 hours per week Uptime guarantee 99.90% Not published Support Email, chat, phone, and video with real humans, on every plan In-app guides and live support; priority onboarding and support on Super and Team Target buyer Clinicians, practices, and organizations across healthcare, including behavioral health Solo therapists and behavioral health group practices ## Where BastionGPT and Mentalyc are similar Both platforms generate structured, insurance-ready therapy notes from a live session, an uploaded recording, or a typed summary, and both let clinicians review and edit before anything is final. Both sign a Business Associate Agreement with every account, state that your data is never used to train AI models, encrypt data in transit and at rest, and offer self-serve signup that takes minutes with no sales call. Both serve mental health clinicians at price points a solo practice can afford, which is exactly why this comparison gets asked. If your entire requirement is turning a standard 50-minute individual session into a compliant SOAP or DAP note, both tools cover that core workflow. The differences appear when your caseload outgrows a note credit tier, a session runs past a recording cap, you need couples, family, or group documentation at the entry price, a client's records fall under 42 CFR Part 2 or FERPA, you need to analyze documents rather than record sessions, or your practice includes people who are not therapists. ## Where BastionGPT goes further ### Unlimited notes, with no credit tiers Mentalyc meters documentation with monthly note credits: 40 on Mini ($14.99 per month billed annually), 100 on Basic ($29.99), 160 on Pro ($59.99), and 330 on Super ($99.99). Billed monthly, those plans run $19.99 to $119.99. Unlimited notes require a Team plan seat at $49.99 per month billed annually. A full-time caseload of 25 to 30 sessions per week produces roughly 100 to 120 notes per month. That volume exceeds Basic's 100-credit cap, so the working tier for a full-time solo therapist is Pro: $719.88 per year billed annually, or $839.88 billed monthly. BastionGPT does not meter notes. Professional costs $20 per user per month ($240 per year), and notes, transcription, and AI assistant chat are all unlimited. A full-time therapist saves roughly $480 to $600 per year compared with Mentalyc Pro, and there is no month where a wave of intakes or catch-up documentation runs the meter out. A five-clinician group pays $2,999.40 per year for Mentalyc Team seats ($49.99 x 5 x 12); the same group runs $1,200 on BastionGPT Professional, or $2,700 on Professional Plus with document analysis and manual model selection included. ### Sessions up to 240 minutes on every plan Mentalyc caps recording length by plan: 75 minutes on Mini and Basic, 130 minutes on Pro, Super, and Team. A standard 50-minute session fits comfortably. Extended intakes, EMDR intensives, family sessions that run long, and group therapy push against those caps, and the only way to raise a cap is to change plans. BastionGPT records up to 240 minutes per session on every plan, live or from uploaded audio. A three-hour group session, a long psychiatric intake, or a full IEP meeting fits inside one recording at the $20 entry price. ### Every note format, client type, and modality at the entry price On Mentalyc, plan choice determines who you can document and how. Mini and Basic cover individual adult therapy only, with SOAP, DAP, and intake notes. Couples, kids and teens, and family client types arrive at Pro, along with BIRP, PIRP, GIRP, PIE, and SIRP formats, EMDR, play therapy, and psychiatry modalities, and auto-computed CPT codes. Group therapy notes arrive at Super. The 14-day free trial runs at the Pro level, so the plan that matches what you tried is Pro, at $59.99 per month billed annually. BastionGPT includes every documentation capability on every plan. The AI scribe produces SOAP, Detailed SOAP, DAP, BIRP, and progress notes, a Custom Note tab accepts any free-form prompt per recording, and you can upload sample notes so BastionGPT matches your writing style, whether that style is EMDR documentation, play therapy observations, or psychiatric medication management. Couples, family, and group sessions are covered at $20 per user per month, and a Saved Prompts library with folder organization and team sharing is included on every plan. ### Ten speakers, labeled by name and role Couples work, family sessions, group therapy, and care-team meetings involve more voices than a solo session. BastionGPT recognizes up to 10 speakers per session, separates the transcript by voice, and automatically labels each speaker's name and role, for example clinician, client, or parent. It also captures telehealth sessions alongside Zoom, Microsoft Teams, and Google Meet without a visible recording bot. Mentalyc supports group documentation on its Super and Team plans, generating a collective group note plus individual notes per member, and it publishes no maximum speaker count and no per-voice transcript attribution. For a therapist whose week includes one couples session and one process group, speaker-labeled transcripts are the difference between reviewing what was said and reconstructing who said it. ### A full clinical AI assistant, not only session notes Mentalyc's platform centers on the session: audio in, note out, plus treatment plans and progress tracking built from session content. It does not include a general clinical AI assistant. BastionGPT includes an unlimited AI assistant on every plan. Therapists use it to draft insurance appeal letters when sessions are denied, write coordination-of-care and school accommodation letters, build treatment plans and discharge summaries, create patient education materials, and review their own documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, and reports that AI scribe users reclaim an average of 90 minutes per day. ### Reads the records your practice receives Behavioral health work involves documents that never came from a recorded session: psychological testing reports, prior treatment records, hospital discharge paperwork, school IEPs and 504 plans, and insurance denials. Mentalyc does not take documents; its inputs are session audio, dictated recaps, or typed summaries. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context, covering PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned records. A clinician can upload a psychological evaluation, three years of prior notes, and a denial letter, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models. Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your session content. Mentalyc does not name its models. Its FAQ describes note generation by machine learning algorithms, large language models, and statistical models, with no user visibility into which model handles a session and no way to choose. ### Named coverage for substance use and school-based records Behavioral health practices routinely handle records with protection beyond HIPAA. Substance use disorder treatment records fall under 42 CFR Part 2, and school-based services generate FERPA-covered education records. BastionGPT names support for HIPAA, 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements on every plan, with PHI residency in the United States, Canada, or Australia based on billing geography. Mentalyc's FAQ addresses Part 2 directly: it provides HIPAA-aligned safeguards under a BAA, states that Part 2-required consent and disclosure obligations remain the provider's responsibility, and does not provide a certificate confirming Part 2 compliance. FERPA support is not published. Mentalyc states compliance with Canadian standards (PIPEDA, PHIPA, PIPA, PHIA) and the Australian Privacy Principles, with data hosted in the United States. For a SUD program, a school-based therapist, or a practice outside the US, those are exactly the lines to compare during procurement. ### A data lifecycle you control Both vendors state that your data is never used to train AI models. The differences sit in where data lives, how long it stays, and who sets the rules. Mentalyc anonymizes transcripts, deletes session audio after processing (its FAQ notes audio is stored for up to 3 days in case a note needs to be fixed or retried), holds SOC 2 Type II certification, and hosts data in the United States on AWS. BastionGPT puts retention under your control: chats persist until you delete them, documents are removed on deletion or after 30 days of account inactivity, and transcript retention is user-configurable from 1 to 365 days, which matters if your board or malpractice carrier expects source material kept longer than a note cycle. Data is encrypted with AES-256 at rest and TLS in transit, runs on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, and is independently penetration tested at least annually. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. ### Works with any EHR, and grows with your organization Mentalyc's EHR integration is a Chrome browser extension that inserts finished notes into web-based EHRs: SimplePractice, TherapyNotes, Jane, TheraNest, Osmind, Tebra, Kareo, and others. Its FAQ is precise about the boundary: the extension works with EHRs that run in a browser. Practices on desktop or client-server EHR systems transfer notes manually, and organizations whose IT policies block third-party browser extensions cannot use the integration at all. BastionGPT works alongside 50+ EMR/EHR systems, web-based or desktop, through simple copy-paste and document workflows that require no extension and no IT approval. Organizations that want deeper integration use the public BastionGPT API, priced from $45 per month with deployment regions in the US, Canada, and Australia, and large organizations on the Ultra plan ($75 per user per month, starting at 100 users) run BastionGPT within their existing enterprise cloud with SSO. One platform covers the therapists, the psychiatrists and PMHNPs, and the billing and admin staff who never touch a session recording. ## When Mentalyc stands out Mentalyc's clearest advantage is reported by the vendor as therapy-native clinical intelligence built around the note itself. Alliance Genie analyzes each session across five clinical dimensions grounded in Bordin's working alliance model, flags ruptures and missed moments with transcript quotes, and tracks alliance trends over time (limited access on Basic, full access on Pro and above). Its AI Treatment Planner (Basic and above) auto-updates plans as clients progress, and session-by-session progress tracking fills the gaps between PHQ-9 or GAD-7 administrations. Supervision workflows include supervisee note access, supervisor sign-offs structured for licensing boards, and shared session review. BastionGPT drafts treatment plans and progress summaries on request through its assistant, and it does not offer an equivalent to Alliance Genie's automated session analysis. Mentalyc has also built a large sample library of therapy templates and web-EHR insertion. Pro and above include 100+ editable templates spanning BIRP, PIRP, GIRP, and PIE, and its Chrome extension writes each note section into the correct field of SimplePractice, TherapyNotes, Jane, and some other browser-based EHRs in one click. Mentalyc also offers a pre-licensed and intern discount. All of that arrives inside a metered package: note credits of 40 to 330 per month, session caps of 75 to 130 minutes, individual-adults-only documentation on the two entry tiers, group notes reserved for Super, a browser extension that reaches only web-based EHRs, and US-only hosting. For a solo therapist with a predictable caseload, an all-web practice stack, and no need for AI beyond the session note, Mentalyc is a credible specialist tool. For a practice where volume grows, sessions run long, couples and group work sit alongside individual clients, records fall under Part 2 or FERPA, or psychiatrists, billing, and admin staff need the same secure AI, BastionGPT covers the whole workload at $20 per user per month with nothing metered. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Mentalyc cost in 2026? Mentalyc's solo plans cost $14.99 (Mini), $29.99 (Basic), $59.99 (Pro), and $99.99 (Super) per month billed annually, or $19.99 to $119.99 billed monthly, with monthly note credits of 40, 100, 160, and 330. Team seats cost $49.99 per month billed annually with unlimited notes. BastionGPT costs $20 to $75 per user per month with unlimited notes, transcription, and AI assistant chat on every plan. Does Mentalyc limit how many notes you can create each month? Yes. Mentalyc meters usage with note credits: 40 per month on Mini, 100 on Basic, 160 on Pro, and 330 on Super; unlimited notes require a Team plan seat. A full-time caseload of 25 to 30 sessions per week produces roughly 100 to 120 notes per month, which forces the Pro tier. BastionGPT has no note limits on any plan, starting at $20 per user per month. Is Mentalyc HIPAA compliant, and does it sign a BAA? Yes. Mentalyc is HIPAA compliant, holds SOC 2 Type II certification, and includes a BAA you can download from account settings. BastionGPT is also HIPAA compliant with a signed BAA on every plan including the free trial, and adds named support for 42 CFR Part 2 and FERPA records, which Mentalyc does not publish. What is Mentalyc's maximum session recording length? 75 minutes on the Mini and Basic plans and 130 minutes on Pro, Super, and Team. BastionGPT records up to 240 minutes per session on every plan, which covers extended intakes, EMDR intensives, and multi-hour group sessions without a plan upgrade. Which is better for therapists, BastionGPT or Mentalyc? For a solo therapist who wants therapy-only note automation with treatment planning and alliance insights inside a web-based EHR, and whose caseload fits a monthly note credit, Mentalyc is built for exactly that. For therapists and group practices that want unlimited notes at $20 per user per month, sessions up to 240 minutes, every client type and note format on every plan, document analysis, named AI models, and coverage for Part 2 and FERPA records, BastionGPT delivers more capability for less money. Does Mentalyc offer a free trial? Yes. Mentalyc offers a 14-day trial with Pro-level access, capped at 15 notes, with no credit card required. BastionGPT offers a 7-day free trial with full, unlimited access to the AI scribe and AI assistant, plus a 45-day money-back guarantee on every paid plan. What AI models power BastionGPT and Mentalyc? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Mentalyc does not name its models; its FAQ describes machine learning algorithms, large language models, and statistical models, with no user model choice. Can Mentalyc analyze uploaded documents or records? No. Mentalyc's inputs are session audio (live or uploaded), dictated recaps, and typed summaries; document upload and analysis is not part of the workflow. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, covering PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files such as scanned records. Does Mentalyc use my data to train AI? Mentalyc states that session data is never used to train AI models, anonymizes transcripts, and deletes session audio after processing (its FAQ notes audio is stored for up to 3 days in case a note needs to be fixed). BastionGPT never uses inputs, documents, or audio to train models on any plan, keeps data in your chosen region (US, Canada, or Australia), and makes transcript retention user-configurable from 1 to 365 days. Does Mentalyc support couples, family, and group therapy notes? Yes, on its higher tiers: couples, kids, and family client types require the Pro plan ($59.99 per month billed annually), and group therapy notes require Super or Team. Mini and Basic support individual adult therapy only. BastionGPT documents couples, family, and group sessions on every plan, recognizes up to 10 speakers, and labels each speaker's name and role automatically. Is Mentalyc compliant with 42 CFR Part 2 for substance use records? Mentalyc's FAQ states that it provides HIPAA-aligned safeguards under a BAA, that Part 2-required consent and disclosure obligations remain the provider's responsibility, and that it does not provide a certificate confirming Part 2 compliance. BastionGPT names 42 CFR Part 2 support on every plan, alongside HIPAA, FERPA, PIPEDA, PHIPA, and Australian APP coverage. Does Mentalyc have an API? Mentalyc publishes no API documentation or developer portal; its published integration path is a Chrome browser extension for web-based EHRs. BastionGPT offers a public, HIPAA-compliant API from $45 per month with deployment regions in the US, Canada, and Australia, and its Ultra plan can run within your organization's own enterprise cloud. Does BastionGPT generate therapy notes like SOAP, DAP, BIRP, and EMDR? Yes. BastionGPT's AI scribe produces SOAP, Detailed SOAP, DAP, BIRP, and progress notes on every plan, and a Custom Note tab accepts any free-form format per recording. You can upload sample notes so BastionGPT matches your documentation style, including EMDR, play therapy, and psychiatric medication management formats, with no plan upgrade required. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/twofold-health # Twofold Alternative: BastionGPT vs Twofold Health (2026) Jump to Comparison Table Twofold Health and BastionGPT both turn patient conversations into structured, HIPAA-compliant clinical notes, and both sell self-serve at prices a solo clinician can afford. The products are shaped differently. Twofold is a focused AI scribe: four ways to capture a session (live conversation, dictated summary, typed key points, or an uploaded recording), unlimited notes with treatment plans and progress tracking on a single Personal plan at $69 per month ($49 per month billed annually as a $588 upfront payment), and mobile and desktop apps. BastionGPT pairs an unlimited AI scribe with a full healthcare AI assistant: it records sessions up to 240 minutes with up to 10 auto-labeled speakers, drafts referral letters and insurance appeals, reads 10+ documents at once with 1,000+ pages of context, runs on named frontier models (GPT, Claude, and Gemini), and supports 42 CFR Part 2 and FERPA records, starting at $20 per user per month with a signed BAA on every plan. This comparison covers the differences that matter when you choose: what each platform costs per year, what you get beyond the visit note, how each vendor documents its security posture, which records and regulations each platform is built for, and what each one publishes about its own capacity limits. Twofold Health's pricing, features, and security claims were verified against trytwofold.com, its Security & Trust page, and its help center on July 11, 2026. If you are exploring Twofold Health alternatives, this comparison covers the specifics: what a scribe-only tool includes, what a full assistant adds, and what each costs. BastionGPT is a Twofold alternative that pairs unlimited scribing with a complete AI assistant at $20 per month. For the wider field, our AI scribe comparison covers other options such as Freed, Mentalyc, and Heidi Health. Quick answer: Twofold Health fits solo clinicians who want a simple, fast scribe with unlimited notes and nothing else. BastionGPT delivers the scribe plus a full clinical AI assistant at less than half Twofold's monthly price: unlimited notes and chat from $20 per user per month, 240-minute sessions with 10 labeled speakers, 1,000+ page document analysis, named AI models, coverage for 42 CFR Part 2 and FERPA records, and an enterprise path from solo practice to your own cloud. Feature BastionGPT Twofold Health Published pricing $20 to $75/user/month, listed on the website Personal plan $69/month, or $49/month billed annually as a $588 upfront payment; group plans are custom quotes What the entry price includes Unlimited AI scribe, unlimited transcription, and a full AI assistant at $20/month Unlimited notes, treatment plans, custom templates, and progress tracking at $69/month ($49 billed annually) Free option 7-day free full trial, plus a 45-day money-back guarantee 7-day free trial; no credit card required Signed BAA ✓ Every plan, including the free trial ✓ Executed automatically at signup; organization-wide BAA on group plans HIPAA compliant ✓ ✓ Additional compliance and attestations 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP; runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure SOC 2 Type II in progress per its security page; 42 CFR Part 2 and FERPA support not published Data residency US, Canada, or Australia, based on billing geography United States only; PHI is not stored or processed outside the US Use of your data for AI training Never; inputs, documents, and audio are not used to train models Never; PHI is explicitly excluded from AI training AI models GPT, Claude, Gemini (named and selectable) underlying models not disclosed Full AI assistant ✓ Referral letters, appeals, patient education, document review, spreadsheet and lab analysis Not offered; the product is a documentation scribe Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra); single documents up to 30 pages on Professional Not offered; inputs are session audio, dictated summaries, or typed key points Ways to capture a session Live recording or uploaded audio, with bot-free capture alongside Zoom, Teams, and Google Meet Live conversation, dictated summary, text input, or uploaded recording; the app must stay open and active during capture Max recording length 240 minutes per session Not published on its website, help center, or app listings Multi-speaker recognition ✓ Up to 10 speakers, with speaker diarization and automatic AI labeling of speaker names and roles No published speaker maximum; its own group-therapy guide says diarization works reasonably with 2 to 3 voices and degrades as participant count grows Note formats SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt, with writing-style matching from your sample notes SOAP, DAP, BIRP, GIRP, SIRP, and other templates, plus custom templates; learns each clinician's writing style Treatment plans and progress tracking Treatment plans, discharge summaries, and progress summaries through the AI assistant ✓ Included on the Personal plan Transcript access Full transcripts with per-speaker labeling, in the browser on any device Transcript viewing is desktop-only, per its FAQ EHR workflow Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, SimplePractice), web-based or desktop, with no extension required Copy and export into any EHR, plus a Chrome extension for web-based EHRs Developer API ✓ Full healthcare AI API (chat, documents, and audio) from $45/month, deployable in US, Canada, and Australia regions Medical speech-to-text API returning transcripts, notes, and structured data; partner and white-label program Enterprise deployment Ultra plan runs within your existing enterprise cloud, with SSO Cloud SaaS on Azure and GCP; a customer-cloud deployment option is not published Peer-reviewed clinical evidence 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published, and no numeric accuracy rate stated; its Most Accurate AI Scribe page cites reviewer feedback rather than a measured figure Uptime guarantee 99.90% Public status page; no published SLA Support Email, chat, phone, and video with real humans on every plan, plus free 30-minute one-on-one training Help center with in-app chat and email; dedicated success manager on group plans Target buyer Individual clinicians, practices, and organizations across healthcare Mental health clinicians and small-to-medium group practices, up to 200 clinicians per its own positioning ## Where BastionGPT and Twofold Health are similar Both platforms capture a clinical encounter and produce a structured, EHR-ready note for the clinician to review and sign. Both offer live capture and uploaded audio, both include unlimited note generation on their standard plans, both sign a Business Associate Agreement automatically at signup, and both state plainly that your data is never used to train AI models. Both sell self-serve with a 7-day free trial and no sales call, and both work alongside your existing EHR through copy-paste rather than requiring an integration project. If your entire requirement is turning a routine visit into a clean note at a fair price, both tools clear that bar. The differences appear when you ask what happens beyond the note: analyzing records, drafting correspondence, documenting a three-hour group session, handling substance use or school-based records, verifying published capacity limits, or deploying across an organization. ## Where BastionGPT goes further ### A full AI assistant, not only a scribe Twofold is deliberately focused: audio or a summary goes in, a structured note comes out, with treatment plans and progress tracking attached to the documentation workflow. It does not include a general clinical AI assistant, and there is no chat interface for work that falls outside the note. BastionGPT includes an unlimited AI assistant on every plan, alongside the scribe. Clinicians use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for incorrect names, pronouns, vague language, and suspected under- or over-coding; and analyze lab summaries and spreadsheets. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, and reports that AI scribe users reclaim an average of 90 minutes per day. ### Less than half the price at the entry tier Twofold's Personal plan costs $69 per month billed monthly. The $49 rate requires annual billing as a single $588 upfront payment. A solo clinician pays $588 to $828 per year for a scribe. BastionGPT Professional costs $20 per user per month, $240 per year, with no upfront commitment, and annual billing adds a free month. That is a saving of roughly $350 to $590 per clinician per year, and the $20 plan includes the full AI assistant, not only the scribe. Even BastionGPT Professional Plus at $45, which adds 1,000+ page document analysis, image and spreadsheet reading, and manual model selection, costs less per year ($540) than Twofold's annual rate ($588). Twofold publishes no group pricing; a practice must request a custom quote, while BastionGPT's team pricing is the same published $20 to $75 per user at any size. ### Reads the records your practice receives Clinical work involves documents that never came from a recorded visit: prior treatment records, psychological and neuropsychological evaluations, hospital discharge paperwork, insurance denials, and faxed outside records. Twofold does not take documents; its inputs are session audio, dictated summaries, or typed key points. On Professional Plus and Ultra plans, BastionGPT reads 10+ documents at once with 1,000+ pages of combined context, covering PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and JPG or PNG images, including scanned records, charts, and graphs. A clinician can upload three years of records, an evaluation, and a denial letter, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single PDF, Word, or text document up to 30 pages. ### Published capacity: 240 minutes and 10 labeled speakers BastionGPT publishes its scribe specifications: recordings up to 240 minutes per session, live or uploaded, with up to 10 speakers separated by voice and automatically labeled with each speaker's name and role, for example clinician, patient, or parent. Extended intakes, family sessions, care conferences, and multi-hour group therapy fit inside one recording, and the transcript shows who said what. Twofold publishes no maximum recording length and no maximum speaker count anywhere on its website, help center, or app store listings, so buyers cannot verify capacity for long or crowded sessions before committing. Its own group-therapy guide is candid about the multi-speaker problem: speaker diarization "works reasonably with 2 to 3 voices and degrades as participant count grows," and it recommends documenting the clinician's facilitation narrative and key themes rather than attributing dialogue per member. That is honest guidance, and it describes exactly the gap BastionGPT's 10-speaker acoustic fingerprinting was built to close. Twofold's FAQ also notes that full transcripts are viewable in the desktop version only. ### AI models you can see and choose BastionGPT names its engines: licensed and HIPAA-compliant versions of the latest GPT, Claude, and Gemini models. Every plan selects the best model for each task automatically, and Professional Plus adds manual model selection plus early access to new models as they ship. Model quality drives note quality, and naming the models lets you verify exactly what is reading your patient records. Twofold describes in-house, clinician-trained models and does not name the underlying AI providers or versions, so note quality rests on a stack you cannot inspect or compare. ### Compliance coverage beyond HIPAA Both vendors handle HIPAA well. The differences start with everything else. BastionGPT names support for 42 CFR Part 2 substance use disorder records, FERPA-covered school-based records, and Canadian (PIPEDA, PHIPA) and Australian (APP) privacy requirements on every plan, with PHI residency in the United States, Canada, or Australia based on billing geography. Twofold's security page covers HIPAA and HITECH and states that PHI is stored and processed exclusively on United States infrastructure. Support for 42 CFR Part 2 or FERPA is not published anywhere on its site: Twofold maintains a dedicated solutions page for SUD and addiction treatment documentation, and even that page does not mention Part 2. There is also no non-US hosting option. For a SUD treatment program, a school-based clinician, or a Canadian or Australian practice, those are the lines to compare during procurement. ### A security posture you can inspect, and retention you control Twofold's security page is clear and well organized: audio is processed to generate the transcript and note, then immediately deleted without ever being written to disk; staff cannot access transcripts, notes, or patient names; data is encrypted in transit and at rest; and PHI is excluded from AI training. Its SOC 2 Type II audit is in progress, with the report to be shared on completion. BastionGPT runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure, encrypts data with AES-256 at rest and TLS in transit, and is independently penetration tested at least annually. Security teams can review the evidence today: a trust center available under NDA holds security documentation in addition to extensive public documentation such as the HIPAA BAA and the sub-processor list. Retention stays under your control rather than fixed by the vendor: chats persist until you delete them, documents are removed on deletion or after 30 days of account inactivity, and transcript retention is user-configurable from 1 to 365 days, which matters when a board, payer, or malpractice carrier expects source material kept on a schedule you set. In three years serving more than 10,000 healthcare organizations, BastionGPT has had zero breaches of protected health information. ### Telehealth capture without babysitting the app Twofold's telehealth workflow is manual: keep the Twofold window open on the computer running the visit, or set a phone running the app next to it. Its FAQ states that the app must remain open and active during capture, and that if the app becomes inactive or the screen goes dark, the visit will not be captured correctly. BastionGPT captures telehealth sessions alongside Zoom, Microsoft Teams, and Google Meet without a visible recording bot and without a foreground-app requirement, live or from uploaded audio, on sessions up to 240 minutes. ### An API for more than speech-to-text Twofold offers a developer API that returns transcripts, notes, and structured data from audio, plus a partner program spanning referral, reseller, co-brand, and white-label models. That is a real offering for platforms that want to embed scribing. The BastionGPT API covers the full platform, not only transcription: HIPAA-compliant chat, document analysis, and audio in one OpenAI-format API, priced from $45 per month with deployment regions in the US, Canada, and Australia and a BAA that extends to API traffic. ### Human support and training on every plan BastionGPT provides support by email, chat, phone, and video with real humans on every plan, plus a free 30-minute one-on-one training session and team lunch-and-learns, backed by a 99.90% uptime guarantee. Twofold provides a help center with in-app chat and email support on the Personal plan and reserves a dedicated success manager for group plans, and it publishes a status page but no uptime SLA. ## When Twofold Health stands out Twofold's clearest strength is focused simplicity. It does one job with very little friction: four capture modes (live conversation, dictated summary, typed key points, or uploaded audio), notes generated in under one minute per its Fastest AI Scribe page, a claimed documentation-time reduction of up to 80%, editable templates across SOAP, DAP, BIRP, GIRP, SIRP, and other formats that learn each clinician's writing style, and native mobile and desktop apps. Treatment plans, custom templates, and client progress tracking are included in the single Personal plan, so there is no feature ladder to climb, and the 7-day trial requires no credit card. Its privacy-by-design defaults are genuinely reassuring for clinicians who want a minimal data footprint: session audio is never written to disk, staff cannot see clinical content, and an example patient consent form ships with the product. Twofold also shows real market traction: it reports 20,000+ clinicians on the platform, displays a 4.8 rating across 3,079 reviews on its homepage, holds a 4.8 on the Apple App Store, publishes customer case studies, and offers a speech-to-text API with a white-label partner program for platforms that want to embed scribing. All of that comes packaged as a scribe and only a scribe, at $69 per month (or $49 per month paid as $588 upfront), with US-only hosting, unpublished recording and speaker limits, an in-progress SOC 2 audit, and no document analysis or clinical assistant. For a solo clinician who wants fast notes and nothing else, Twofold is a credible pick. For clinicians and practices that also need records review, correspondence drafting, Part 2 or FERPA coverage, regional data residency, published capacity specs, or a path to enterprise deployment, BastionGPT covers the scribe and everything after it, starting at $20 per user per month. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs How much does Twofold Health cost in 2026? Twofold's Personal plan costs $69 per month billed monthly, or $49 per month billed annually as a $588 upfront payment, with unlimited notes; group plans are custom quotes. BastionGPT publishes all of its plans at $20 to $75 per user per month, includes a full AI assistant alongside the unlimited scribe at $20, and needs no upfront annual commitment. What is the difference between BastionGPT and Twofold Health? Twofold is a focused AI scribe: it turns session audio, dictation, or typed key points into clinical notes with treatment plans and progress tracking. BastionGPT is a healthcare AI platform: an unlimited scribe plus an AI assistant that drafts letters and appeals, reviews documentation, and analyzes up to 1,000+ pages of uploaded records, running on named models (GPT, Claude, and Gemini). Is Twofold Health HIPAA compliant, and does it sign a BAA? Yes. Twofold is HIPAA and HITECH compliant, and its BAA is executed automatically at signup, with an organization-wide BAA on group plans. BastionGPT is also HIPAA compliant with a signed BAA on every plan including the free trial, and adds named support for 42 CFR Part 2, FERPA, PIPEDA, PHIPA, and Australian APP requirements. Does Twofold Health have SOC 2 certification? Not yet. Twofold's security page states its SOC 2 Type II audit is in progress, with the report to be shared on completion. BastionGPT runs exclusively on HITRUST CSF certified and SOC 2 Type II attested Microsoft Azure infrastructure, is independently penetration tested at least annually, and provides a trust center under NDA with its penetration test summary, HIPAA Security Rule mapping, and infrastructure attestations. Which is better, BastionGPT or Twofold Health? For a solo clinician who wants only a simple, fast note-taker with unlimited notes, Twofold does that job well at $49 to $69 per month. For clinicians and practices that want the scribe plus a full clinical AI assistant, document analysis, named AI models, 240-minute sessions with 10 labeled speakers, and coverage for Part 2 and FERPA records, BastionGPT delivers substantially more capability at $20 per user per month. Does Twofold Health offer a free trial? Yes. Twofold offers a 7-day free trial with no credit card required. BastionGPT also offers a 7-day free trial with full access to the AI scribe and AI assistant, and adds a 45-day money-back guarantee on every paid plan. What AI models power BastionGPT and Twofold Health? BastionGPT uses licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, selecting the best model for each task automatically; Professional Plus adds manual model selection and early access to new models. Twofold describes in-house, clinician-trained models and does not name its underlying AI providers or versions. Can Twofold Health analyze uploaded documents? No. Twofold's inputs are session audio, dictated summaries, or typed key points; document upload and analysis is not part of the product. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, covering PDF, Word, Excel, CSV, PowerPoint, HTML, text, and image files, including scanned records. What is the maximum recording length for BastionGPT and Twofold Health? BastionGPT publishes a 240-minute maximum per session, live or uploaded, on every plan. Twofold does not publish a maximum recording length or speaker count on its website, help center, or app listings; its group-therapy guide notes that speaker diarization works reasonably with 2 to 3 voices and degrades as participant count grows, and its FAQ notes the app must remain open and active during capture for the visit to record correctly. Does Twofold Health use my data to train AI? No. Twofold's security page states PHI is explicitly excluded from AI training and that session audio is deleted immediately after processing, never written to disk. BastionGPT also never uses inputs, documents, or audio to train models, and adds user-controlled retention: transcripts configurable from 1 to 365 days, and chats kept until you delete them. Does Twofold Health have an API? Yes. Twofold offers a medical speech-to-text API that returns transcripts, notes, and structured data, plus a partner and white-label program. The BastionGPT API covers the full platform (HIPAA-compliant chat, document analysis, and audio) in one OpenAI-format API from $45 per month, with deployment regions in the US, Canada, and Australia. Does Twofold Health support 42 CFR Part 2 or FERPA records? Twofold does not publish support for 42 CFR Part 2 substance use records or FERPA education records; its published compliance coverage is HIPAA and HITECH on US-only infrastructure. BastionGPT names 42 CFR Part 2 and FERPA support on every plan, alongside PIPEDA, PHIPA, and Australian APP coverage with regional data residency. Does BastionGPT work for therapy notes like Twofold Health? Yes. BastionGPT's AI scribe produces SOAP, Detailed SOAP, DAP, BIRP, and progress notes on every plan, a Custom Note tab accepts any free-form format per recording, and you can upload sample notes so BastionGPT matches your documentation style. Unlimited notes and transcription are included at $20 per user per month. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/commure # Commure Alternative: BastionGPT vs Commure (2026) Jump to Comparison Table Commure and BastionGPT both use AI to take clinical documentation off your plate, and they sit at opposite ends of the healthcare market. Commure is an enterprise healthcare operations platform valued at $7 billion: ambient AI documentation, autonomous coding, and revenue cycle automation sold to more than 130 health systems, including HCA Healthcare and Tenet, through a sales and implementation process. BastionGPT is a HIPAA-compliant AI assistant and unlimited scribe that healthcare professionals set up themselves in minutes: it records sessions up to 240 minutes with up to 10 auto-labeled speakers, drafts referral letters and insurance appeals, reads 1,000+ pages of records at once, and starts at $20 per user per month with a signed BAA on every plan. This comparison covers what each platform costs, what the entry price includes, how each vendor handles your data, what each publishes about its capacity and accuracy, and which buyers each one fits. Commure's pricing, product claims, and company facts were verified against commure.com, its self-serve Commure Scribe pricing page, and its own published announcements on July 17, 2026. Most searches for a Commure alternative come down to fit. Commure is engineered and priced for health systems: its deepest features, such as direct EHR sync and ROI analytics, sit in its enterprise tier, and a solo practice or small group rarely needs revenue cycle automation or a multi-week implementation. BastionGPT is the Commure alternative built for the practice side of that line: individual healthcare professionals, small and independent practices, and mental health teams that want the scribe, the assistant, and the BAA without a procurement cycle. For the wider field, our AI scribe comparison covers enterprise ambient options such as Abridge, DAX Copilot, and Ambience Healthcare. Quick answer: Commure is the stronger choice for large health systems that need deep bi-directional EHR integration, autonomous coding, and revenue cycle automation at enterprise scale. BastionGPT is the better Commure alternative for solo, small, independent, and mental health practices: $20 per user per month includes the unlimited AI scribe, a full healthcare AI assistant, and a signed BAA on every plan, including the free trial, with no contract, no implementation fee, and nothing for IT to install. Feature BastionGPT Commure Best for Individual healthcare professionals, small and independent practices, and mental health teams Enterprise health systems and hospital networks; 130+ health systems including HCA Healthcare and Tenet Published pricing $20 to $75/user/month, listed on the website Scribe Pro $89/month, or $59/month billed annually ($708/year); platform and enterprise pricing by sales quote What the entry price includes Unlimited AI scribe, unlimited transcription, and a full AI assistant at $20/month Free Scribe Core covers unlimited notes for the first 7 days; Scribe Pro adds unlimited transcription, automated coding, and custom templates Contract and implementation No contract, no implementation fee; annual billing adds a free month Enterprise deployments run through sales, contracting, and implementation with forward-deployed engineers Free option 7-day free full trial, plus a 45-day money-back guarantee Free Scribe Core tier: unlimited notes for the first 7 days, no credit card Signed BAA ✓ Every plan, including the free trial ✓ On paid plans HIPAA compliant ✓ ✓ Additional compliance and attestations 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP; runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure SOC 2 Type II referenced; onshore US data processing Data residency US, Canada, or Australia, based on billing geography United States Use of your data for AI training Never; inputs, documents, and audio are not used to train models Its scribe FAQ states its proprietary LLMs "continuously enhance their performance with each use"; request the data-use terms in writing Data retention User-controlled: chats persist until you delete them; transcript retention configurable from 1 to 365 days Not published on the self-serve scribe site; request current retention terms in writing AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, named and selectable Proprietary fine-tuned models; providers and versions not disclosed Full AI assistant ✓ Referral letters, appeals, patient education, document review, spreadsheet and lab analysis on every plan AI copilot for clinical decision support on Scribe Pro; broader automation is sold as enterprise modules Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra); single documents up to 30 pages on Professional Not part of the self-serve scribe; document automation lives in enterprise RCM modules Max recording length 240 minutes per session Not published for the self-serve scribe Multi-speaker recognition ✓ Up to 10 speakers, with diarization and automatic labeling of speaker names and roles ✓ Multi-speaker recognition across 60+ languages on the self-serve scribe; no published speaker maximum Note formats SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt, with writing-style matching from your sample notes Specialty-aligned templates with automated ICD-10/CPT coding suggestions; custom template builder on Pro EHR workflow Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, SimplePractice), web-based or desktop, with no extension required Deep bi-directional integration with 60+ EHRs (Epic Toolbox, Cerner, MEDITECH, athenahealth); direct note sync is gated to Scribe Enterprise Revenue cycle and billing automation AI assistant reviews documentation for coding gaps and drafts appeals Core strength: autonomous coding, denials automation, and $25B+ in annual claims processed Deployment Self-serve in the browser; producing notes in minutes Sales-led; enterprise ambient rollouts involve contracting, workflow configuration, and implementation Independent validation 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers 93.3 in the 2025 KLAS First Look report, with 100% of interviewed customers saying they would buy again Company profile Bootstrapped healthcare AI and security company, founded 2018 in Dallas, Texas; BastionGPT on the market since 2023 General Catalyst-incubated roll-up of Athelas, Augmedix, Memora Health, and PatientKeeper; roughly $750M raised, valued at $7B as of May 2026 Vendor stability signals Zero reportable PHI breaches in three years on market; no layoffs or litigation on record Business Insider reported staff cuts after the 2023 Athelas merger; a federal court paused new Strongline Pro sales in April 2025 (Commure is appealing) Support Email, chat, phone, and video with real humans on every plan, plus free 30-minute one-on-one training Enterprise implementation and customer success teams; KLAS grades its executive involvement A+ ## Where BastionGPT and Commure are similar Both platforms listen to a patient visit and return a structured, specialty-appropriate note for a healthcare professional to review and sign. Both handle in-person and telehealth encounters, both support multi-speaker sessions, both offer automated coding help (Commure through automated ICD-10 and CPT suggestions, BastionGPT through assistant-driven documentation and coding review), and both publish strong third-party validation: a 93.3 KLAS First Look score for Commure Ambient AI, and a peer-reviewed 99.3% sentence-level accuracy study for BastionGPT. Both companies are US-based, focused entirely on healthcare, and sign Business Associate Agreements. The overlap ends at the buyer. Commure's platform reaches full depth inside a health-system contract: ambient documentation wired into coding, billing, denials, and staff workflows. BastionGPT reaches full depth the day you sign up: unlimited scribe, unlimited assistant, and a BAA from the first minute of the free trial. The rest of this page walks through what that difference means in practice. ## Where BastionGPT goes further ### Published pricing without a procurement cycle Commure does not publish pricing for its platform; the enterprise products are quoted through sales. Its self-serve Commure Scribe lists Scribe Pro at $89 per month, or $59 per month billed annually as a $708 upfront payment, and gates capabilities by tier: automated coding and custom templates arrive at Pro, while direct EHR sync, custom AI workflows, and ROI analytics require the custom-quoted Scribe Enterprise plan. BastionGPT publishes every plan on its website: Professional at $20 per user per month, Professional Plus at $45, and Ultra at $75. The $20 plan includes the unlimited AI scribe, unlimited transcription, and the full AI assistant. There is no contract, no implementation fee, and no feature ladder for the core product, and annual billing adds a free month. A solo practice pays $240 a year instead of $708 to $1,068, and a group pays the same published rate at any size. ### A signed BAA on every plan, including the free trial BastionGPT executes a Business Associate Agreement automatically at signup on every plan, including the 7-day free trial, so PHI is covered from the first note. Commure signs BAAs on paid plans, and its free Scribe Core tier is where most self-serve evaluations start: if you trial Commure Scribe, confirm your BAA status before recording real patient encounters. BastionGPT also backs every paid plan with a 45-day money-back guarantee. ### A full healthcare AI assistant alongside the unlimited scribe Commure Scribe is documentation-focused: notes, coding suggestions, and an AI copilot for clinical decision support, with broader automation reserved for enterprise modules. BastionGPT includes an unlimited AI assistant on every plan alongside the scribe. Healthcare professionals use it to draft referral letters, insurance appeal letters, treatment plans, and patient education materials; review documentation for vague language and suspected under- or over-coding; and analyze lab summaries and spreadsheets. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, and reports that AI scribe users reclaim an average of 90 minutes per day. ### Depth in mental and behavioral health Commure's platform language is the language of health systems: wRVUs, denials, throughput, and revenue integrity. BastionGPT's center of gravity is the mental health professional. It drafts DAP, BIRP, and detailed SOAP notes; separates and labels up to 10 speakers for couples, family, and group sessions; supports 42 CFR Part 2 substance use records and FERPA-covered school records on every plan; and handles sensitive clinical topics such as self-harm and abuse that consumer AI tools often refuse. Therapists, psychologists, counselors, and psychiatric providers make up the largest share of BastionGPT's user base, and the product reflects it. ### Reads the records your practice receives Clinical work involves documents that never came from a recorded visit: prior treatment records, evaluations, discharge paperwork, insurance denials, and faxed outside records. Commure's document automation lives in its enterprise revenue cycle pipeline rather than in the self-serve scribe. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, covering PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and scanned images with OCR. Upload three years of records, an evaluation, and a denial letter, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single document up to 30 pages. ### Your data is never used to train AI models BastionGPT's policy is unambiguous: inputs, documents, and audio are never used to train AI models, never shared with model providers for training, and never sold. Retention stays under your control: chats persist until you delete them, and transcript retention is configurable from 1 to 365 days. Commure Scribe's own FAQ answers the training question differently, stating that its "proprietary large language models (LLMs) continuously enhance their performance with each use." That approach can be engineered in privacy-protective ways, so ask Commure exactly what is retained, for how long, and how model improvement works, in writing, before onboarding. With BastionGPT there is nothing to clarify. ### Self-serve today instead of an implementation project Commure's enterprise deployments are genuine projects: sales, contracting, workflow configuration, and rollout support from forward-deployed engineers. That model makes sense at hospital scale. BastionGPT runs in the browser and is producing notes within minutes: sign up, record or upload a session, and review the finished note before you place it in your chart. There is nothing for IT to install and no minimum seat count, and the free 30-minute one-on-one training is optional rather than required. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and SimplePractice. ### AI models you can see, from vendors you know BastionGPT names its engines: licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, with the platform selecting the best model for each task and Professional Plus adding manual model selection plus early access to new models as they ship. Commure runs proprietary fine-tuned models and does not disclose the providers or versions underneath. Model quality drives note quality, and naming the models lets you verify exactly what is reading your patient records. ### A focused vendor without roll-up turbulence BastionGPT is built by Bastion Intelligence, a bootstrapped healthcare AI and security company founded in Dallas, Texas in 2018 by a former academic medical center CISO and CTO. The product has been on the market since 2023, serves more than 10,000 healthcare organizations, and has recorded zero reportable PHI breaches. Commure has grown by merger: Athelas, Augmedix, Memora Health, and PatientKeeper in roughly two years, with about $750 million raised and a $7 billion valuation as of May 2026. Scale brings resources, and it has also brought turbulence: Business Insider reported staff cuts shortly after the 2023 Athelas merger, and in April 2025 a federal court barred new sales of Commure's Strongline Pro staff-safety product while a trade secrets case proceeds (Commure disputes the ruling and is appealing). None of that touches the quality of Commure's scribe directly; it is context worth weighing when you choose the vendor that will hold your documentation workflow for years. ### Human support on every plan BastionGPT provides support by email, chat, phone, and video with real humans on every plan, plus a free 30-minute one-on-one training session and team lunch-and-learns, backed by a 99.90% uptime guarantee. Commure provides enterprise implementation and customer success teams for platform customers, and KLAS interviewees grade its executive involvement A+. For a self-serve Commure Scribe user, support runs through the product's help resources. ## When Commure stands out Commure's strengths are real, and at health-system scale they are substantial. Its ambient platform supports 40 million appointments a year across more than 130 health systems, including HCA Healthcare and Tenet. It integrates deeply with 60+ EHRs, appears in Epic's Toolbox program, and its enterprise scribe tier syncs notes directly into the chart on desktop or web. Beyond documentation, Commure automates the revenue cycle itself: autonomous coding, denials work, and more than $25 billion in annual claims processed. The 2025 KLAS First Look scored Commure Ambient AI 93.3, with 100% of interviewed customers saying they would buy again and an A+ for executive involvement, and deployments come with implementation support from forward-deployed engineers. If you are a CIO or CMIO buying ambient documentation plus revenue cycle automation for hundreds of professionals, Commure belongs on your shortlist alongside the other enterprise platforms. If you run a practice, most of that machinery is capacity you would pay for and never use: the platform assumes a health-system contract behind it, direct EHR sync and ROI analytics sit in the enterprise tier, and the self-serve Commure Scribe is a narrower product than the platform its case studies describe. BastionGPT gives a practice the parts that matter at that scale, the unlimited scribe, the full AI assistant, and the BAA, at a published $20 per user per month. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs Is BastionGPT a good Commure alternative for a small practice? Yes, that is the exact fit. Commure is built and priced for enterprise health systems, with its deepest features in custom-quoted tiers. BastionGPT is self-serve: $20 per user per month includes the unlimited AI scribe, a full healthcare AI assistant, and a signed BAA on every plan, including the 7-day free trial, with no contract or implementation fee. How much does Commure cost compared to BastionGPT? Commure's enterprise platform is quoted through sales. Its self-serve Commure Scribe lists Scribe Pro at $89 per month, or $59 per month billed annually ($708 per year), with a free 7-day Core tier; direct EHR sync and ROI analytics require a custom-quoted Enterprise plan. BastionGPT publishes all plans at $20 to $75 per user per month, and the $20 plan includes the unlimited scribe and full AI assistant. What is the difference between Commure and BastionGPT? Commure is an enterprise healthcare operations platform: ambient AI documentation, autonomous coding, and revenue cycle management for 130+ health systems. BastionGPT is a HIPAA-compliant AI assistant plus unlimited scribe for individual healthcare professionals and practices: notes, referral letters, appeals, document analysis, and patient education from $20 per user per month, set up in minutes. Does BastionGPT sign a BAA like Commure? Yes. BastionGPT executes a Business Associate Agreement automatically at signup on every plan, including the free trial. Commure signs BAAs on paid plans; if you evaluate its free Scribe Core tier, confirm your BAA status before recording real patient encounters. Which is better for therapists and mental health professionals, Commure or BastionGPT? BastionGPT. It drafts DAP, BIRP, and detailed SOAP notes, labels up to 10 speakers for couples, family, and group sessions, supports 42 CFR Part 2 and FERPA records on every plan, and handles sensitive topics that consumer AI tools often refuse. Commure's platform centers on health-system operations such as coding, denials, and revenue integrity. Is Commure only for large health systems? Mostly. Commure's platform, pricing, and implementation model target health systems, and it reports more than 130 of them as customers. It offers a self-serve Commure Scribe with a free Core tier and an $89 per month Pro plan, but its strongest capabilities, such as direct EHR sync, custom AI workflows, and ROI analytics, require the custom-quoted Enterprise tier. Does BastionGPT integrate with EHRs like Commure does? The two take different approaches. Commure's enterprise platform integrates bi-directionally with 60+ EHRs and appears in Epic's Toolbox program, with direct note sync on its Enterprise scribe tier. BastionGPT works alongside 50+ EMR/EHR systems, including Epic, Oracle Health, Athenahealth, and SimplePractice: you review and edit each finished note, then place it in your chart, with nothing for IT to install. Does Commure have a free version? Commure Scribe offers a free Core tier with unlimited notes for the first 7 days, no credit card required. BastionGPT offers a 7-day free trial with full access to the scribe and assistant, a signed BAA during the trial, and a 45-day money-back guarantee on every paid plan. Does Commure use patient data to train its AI? Commure Scribe's FAQ states that its "proprietary large language models (LLMs) continuously enhance their performance with each use." Ask Commure for its data-use and retention terms in writing to understand exactly what that means for your PHI. BastionGPT never uses your inputs, documents, or audio to train AI models, and transcript retention is user-configurable from 1 to 365 days. What is the difference between Commure Scribe and the Commure platform? Commure Scribe is the self-serve documentation product: a free Core tier, an $89 per month Pro plan, and a custom Enterprise tier. The Commure platform is the enterprise suite health systems buy: Ambient AI at scale, autonomous coding, revenue cycle automation, patient engagement, and staff safety tools, sold through sales with implementation. The case-study results Commure advertises come from platform deployments. How accurate are BastionGPT and Commure? BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. Commure Ambient AI scored 93.3 in the 2025 KLAS First Look report, with 100% of interviewed customers saying they would buy again; KLAS measures customer experience rather than note-level accuracy, and Commure publishes no peer-reviewed accuracy study. Has Commure had layoffs or lawsuits? Two documented events. Business Insider reported that Commure cut an undisclosed number of staff in November 2023, shortly after its merger with Athelas. In April 2025, a federal court in California barred Commure from selling its Strongline Pro staff-safety product to new customers while a trade secrets case proceeds; Commure disputes the ruling and is appealing. Neither involves its scribe directly, but both are worth weighing as vendor-stability signals. When is Commure the better choice than BastionGPT? Choose Commure if you are a health system buying at enterprise scale: deep bi-directional EHR write-back, autonomous coding and revenue cycle automation, ambient documentation wired into billing, and implementation support from forward-deployed engineers. For a solo, small, independent, or mental health practice that needs powerful, compliant AI today, BastionGPT delivers more capability per dollar at $20 per user per month. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs DeepCura Alternative: BastionGPT vs DeepCura (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/bastiongpt-vs/deepcura # DeepCura Alternative: BastionGPT vs DeepCura (2026) Jump to Comparison Table DeepCura and BastionGPT both use AI to take work off a small practice's plate, and they package it in very different shapes. DeepCura is an "agent-operated" clinical platform at $129 per provider per month: an AI scribe plus a 24/7 AI receptionist, billing and coding suggestions, patient intake, and fax automation, all drawing on a shared pool of 1,000 monthly credits and wired into the nine EHRs it lists. BastionGPT is a HIPAA-compliant AI assistant and unlimited scribe that healthcare professionals set up themselves in minutes: it records sessions up to 240 minutes with up to 10 auto-labeled speakers, drafts referral letters and insurance appeals, reads 1,000+ pages of records at once, and starts at $20 per user per month with a signed BAA on every plan. This comparison covers what each platform costs, what the entry price actually includes, how the credit system works, how each vendor handles your data, what each publishes about security and accuracy, and which buyers each one fits. DeepCura's pricing, product claims, and security statements were verified against deepcura.com, its published pricing page, and its Apple App Store listing on July 17, 2026. Most searches for a DeepCura alternative come down to a single question: how much of the platform would you actually use? DeepCura bundles a receptionist, billing help, intake, and faxing into one credit-metered subscription, and that breadth is the point. But if what you need every day is documentation and a capable, compliant AI assistant, you are paying platform prices for capacity you will not touch, with a monthly cap on the parts you do use. BastionGPT is the DeepCura alternative built for that majority: individual healthcare professionals, small and independent practices, and mental health teams that want unlimited scribing and a full AI assistant with no credit math. For the wider field, our AI scribe comparison covers enterprise ambient options such as Abridge, DAX Copilot, and Ambience Healthcare. Quick answer: DeepCura is the stronger choice if you want one subscription to automate the front office too: phone answering, intake, billing suggestions, and faxes, with native write-back into its list of supported EHRs. BastionGPT is the better DeepCura alternative for solo, small, independent, and mental health practices that want documentation and a full healthcare AI assistant without usage caps: $20 per user per month includes the unlimited AI scribe, unlimited transcription, and a signed BAA on every plan, including the free trial, backed by a peer-reviewed 99.3% accuracy study. Feature BastionGPT DeepCura Best for Individual healthcare professionals, small and independent practices, and mental health teams Solo and small practices that want scribe, phone, billing, and intake automation in one credit-based platform Published pricing $20 to $75/user/month, listed on the website $129/provider/month or $999/year; volume discounts to $101/month at 11+ seats What the entry price includes Unlimited AI scribe, unlimited transcription, and a full AI assistant at $20/month All seven AI agents (scribe, receptionist, billing, intake, fax, tasks, messaging) with 1,000 monthly credits per seat Usage limits None: unlimited transcription and assistant use, no credit system Credit-capped: 1,000 credits/month per seat across all agents; a $59/month add-on doubles it to 2,000 Free option 7-day free full trial with a signed BAA Free trial with starter credits, no credit card Refunds 45-day money-back guarantee No refunds for partial periods; access continues to the end of the billing cycle Signed BAA ✓ Every plan, including the free trial ✓ "BAA Included" per its homepage HIPAA compliant ✓ ✓ Additional compliance and attestations 42 CFR Part 2, FERPA, PIPEDA, PHIPA, APP; runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure Google CASA Tier 2 security assessment (verified by TAC Security); completed SOC 2, HITRUST, or ISO 27001 certifications not published on its website Data residency US, Canada, or Australia, based on billing geography Not published on its public site Use of your data for AI training Never; inputs, documents, and audio are not used to train models States customer data is not used for training, while its homepage advertises AI "trained on real clinical workflows, not synthetic data"; request the terms in writing Data retention User-controlled: chats persist until you delete them; transcript retention configurable from 1 to 365 days Not published on its public site AI models Licensed, HIPAA-compliant versions of the latest GPT, Claude, and Gemini models, named and selectable ✓ Side-by-side choice of frontier models (GPT, Claude, Gemini, Grok) General AI assistant ✓ Referral letters, appeals, patient education, document review, spreadsheet and lab analysis on every plan ✓ Agentic chat chains workflows; parallel multi-model answers to clinical questions AI receptionist and phone Not offered ✓ 24/7 call answering, scheduling, triage, and payments Dedicated billing and coding agent AI assistant reviews documentation for coding gaps and drafts appeals ✓ E&M levels, ICD-10/CPT suggestions, under- and over-coding flags Fax, intake, and comms automation Not offered ✓ AI fax/email/SMS hub and nurse-intake agent EHR workflow Works alongside 50+ EMR/EHR systems (Epic, Oracle Health, Athenahealth, SimplePractice); you review and edit each note before it enters the chart ✓ Native write-back with the nine systems it lists (Epic, athenahealth, eClinicalWorks, AdvancedMD, Practice Fusion, OpenEMR, OptiMantra, CharmHealth, Veradigm) Document upload and analysis 10+ documents at once, 1,000+ pages of combined context (Professional Plus and Ultra); single documents up to 30 pages on Professional Record uploads and summaries up to 1,000 pages Max recording length 240 minutes per session Not published Multi-speaker recognition ✓ Up to 10 speakers, with diarization and automatic labeling of speaker names and roles ✓ Diarization; no published speaker maximum Note formats SOAP, Detailed SOAP, DAP, BIRP, H&P, progress, custom free-form prompt, with writing-style matching from your sample notes Specialty-tuned templates across 50+ specialties with customizable fields Independent validation 99.3% sentence-level accuracy in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers None published; reports 6,000+ clinicians and 1M+ encounters (vendor figures) Company profile Bootstrapped healthcare AI and security company, founded 2018 in Dallas, Texas; BastionGPT on the market since 2023, serving more than 10,000 healthcare organizations Founded 2024 in San Francisco; bootstrapped, self-described as run by a very small human team plus its own AI agents Support Email, chat, phone, and video with real humans on every plan, plus free 30-minute one-on-one training AI-first: "Emily" voice onboarding agent and in-app help from a small human team ## Where BastionGPT and DeepCura are similar Both platforms listen to a patient visit and return a structured, specialty-appropriate note for a healthcare professional to review and sign. Both are self-serve with free trials, both publish their pricing in full on their websites, both offer a choice of named frontier AI models, both handle multi-speaker and multilingual sessions, and both accept large record uploads for AI review. Both companies are bootstrapped and healthcare-focused, and both sign Business Associate Agreements. The overlap ends at the operating model. DeepCura spreads one pool of 1,000 monthly credits across seven AI agents that reach into scheduling, billing, and faxing. BastionGPT concentrates on documentation and clinical support and removes the meter entirely: unlimited scribe, unlimited assistant, and a BAA from the first minute of the free trial. The rest of this page walks through what that difference means in practice. ## Where BastionGPT goes further ### Unlimited use instead of a monthly credit meter DeepCura's $129 price covers all seven of its AI agents, but usage is metered: each seat gets 1,000 credits per month shared across every agent, and a $59 per month add-on doubles that to 2,000. Notes, receptionist calls, intake, and faxes all draw from the same pool, so a busy month can hit the cap, and public reviewers note the credit balance needs watching. BastionGPT has no credit system anywhere: transcription is unlimited on every plan, sessions run up to 240 minutes, and the AI assistant is unlimited too. A heavy documentation week costs the same as a quiet one, and nobody on your team has to ration AI use in the middle of a full clinic day. ### A sixth of the entry price, published in full DeepCura costs $129 per provider per month, or $999 per year, with volume discounts that bottom out at $101 per month for 11 or more seats. BastionGPT publishes every plan on its website: Professional at $20 per user per month, Professional Plus at $45, and Ultra at $75. The $20 plan includes the unlimited AI scribe, unlimited transcription, and the full AI assistant. A solo practice pays $240 a year instead of $999 to $1,548, and annual billing adds a free month. There is no contract and no implementation fee, and every paid plan carries a 45-day money-back guarantee, versus DeepCura's policy of no refunds for partial periods. ### A signed BAA on every plan, including the free trial BastionGPT executes a Business Associate Agreement automatically at signup on every plan, including the 7-day free trial, so PHI is covered from the first note, and the trial includes the full product rather than a starter allotment of credits. DeepCura's homepage says a BAA is included; if you evaluate its free trial, confirm your BAA is in place before recording real patient encounters. ### Formal compliance depth beyond HIPAA DeepCura advertises HIPAA compliance, 256-bit encryption, an included BAA, and a Google CASA Tier 2 security assessment verified by TAC Security. Those are real controls, but a completed SOC 2 Type II audit, HITRUST certification, or ISO 27001 certification is not published on its website. BastionGPT runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure and is pursuing its own SOC 2 Type II attestation and HITRUST CSF certification. Coverage extends beyond HIPAA to 42 CFR Part 2 substance use records, FERPA-covered school records, PIPEDA and PHIPA in Canada, and the Australian Privacy Principles, with data residency in the US, Canada, or Australia based on billing geography. ### Peer-reviewed accuracy evidence instead of vendor metrics DeepCura reports 6,000+ clinicians and 1M+ generated encounters; both figures are the vendor's own, and no independent accuracy study is published. BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers, published in Critical Care Explorations in 2026. If you need to justify an AI tool to a compliance officer, a group practice owner, or a hospital committee, a citable study carries different weight than a marketing counter. ### Depth in mental and behavioral health DeepCura markets across 50+ specialties, with primary care, psychiatry, orthopedics, and dermatology up front. BastionGPT's center of gravity is the mental health professional. It drafts DAP, BIRP, and detailed SOAP notes; separates and labels up to 10 speakers for couples, family, and group sessions; supports 42 CFR Part 2 substance use records and FERPA-covered school records on every plan; and handles sensitive clinical topics such as self-harm and abuse that consumer AI tools often refuse. Therapists, psychologists, counselors, and psychiatric providers make up the largest share of BastionGPT's user base, and the product reflects it. ### Reads the records your practice receives Clinical work involves documents that never came from a recorded visit: prior treatment records, evaluations, discharge paperwork, insurance denials, and faxed outside records. DeepCura accepts record uploads up to 1,000 pages within its credit allotment. BastionGPT reads 10+ documents at once with 1,000+ pages of combined context on Professional Plus and Ultra, covering PDF, Word, Excel, CSV, PowerPoint, HTML, plain text, and scanned images with OCR, with no per-use metering. Upload three years of records, an evaluation, and a denial letter, then query, summarize, and cross-reference all of it in one conversation. The Professional plan covers a single document up to 30 pages. ### Clear answers on training data and retention BastionGPT's policy is unambiguous: inputs, documents, and audio are never used to train AI models, never shared with model providers for training, and never sold. Retention stays under your control: chats persist until you delete them, and transcript retention is configurable from 1 to 365 days. DeepCura also states that customer data is not used for model training, yet its homepage separately advertises that its AI is "trained on real clinical workflows, not synthetic data," and its retention terms are not published on its public site. Those statements can be reconciled in privacy-protective ways, so ask DeepCura exactly what is retained, for how long, and what that training language means for your patients' records, in writing, before onboarding. With BastionGPT there is nothing to clarify. ### Human support instead of an AI-first help desk DeepCura is deliberately lean: the company has publicly described running with a very small human team amplified by its own AI agents, and an onboarding agent named Emily handles setup by voice. That model keeps costs down, but when PHI, billing, or a mid-session failure is involved, many practices want a person on the line. BastionGPT provides support by email, chat, phone, and video with real humans on every plan, plus a free 30-minute one-on-one training session and team lunch-and-learns, backed by a 99.90% uptime guarantee. ### A vendor track record measured in years BastionGPT is built by Bastion Intelligence, a bootstrapped healthcare AI and security company founded in Dallas, Texas in 2018 by a former academic medical center CISO and CTO. The product has been on the market since 2023, serves more than 10,000 healthcare organizations, and has recorded zero reportable PHI breaches. DeepCura was founded in 2024 in San Francisco and describes itself as bootstrapped and profitable; its iOS App Store rating was recently reset and currently shows too few ratings to display an average. None of that makes DeepCura a bad product, but it is a young one, and the vendor holding your documentation workflow is a multi-year choice. ## When DeepCura stands out DeepCura's breadth is real. For one credit-based subscription it bundles an ambient AI scribe, a 24/7 AI receptionist that answers calls, schedules, triages, and takes payments, a billing agent that suggests E&M levels and ICD-10/CPT codes and flags under- and over-coding, an AI nurse agent for patient intake, and a communications hub that handles faxes, email, and SMS. It writes notes back natively into the EHRs it lists, including Epic, athenahealth, eClinicalWorks, AdvancedMD, Practice Fusion, OpenEMR, OptiMantra, CharmHealth, and Veradigm, and it lets you run multiple frontier AI models side by side and pick the output you prefer. If your practice runs on one of those EHRs and your biggest pain is the front office, missed calls, intake paperwork, and coding cleanup, DeepCura is a genuine contender that may replace several tools at once. If your biggest pain is documentation itself, most of that machinery is capacity you would pay $129 a month for and rarely touch, with a credit meter on the parts you use daily. BastionGPT gives a practice the parts that carry the daily load, the unlimited scribe, the full AI assistant, and the BAA, at a published $20 per user per month. ## What BastionGPT users say "BastionGPT has a lot more capabilities [than ChatGPT] with data protection, so clinicians can more freely use AI." David Lopis, Director, Psychology Squared "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" Catherine Maxted, RN, Nurse Coordinator "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Try BastionGPT free for 7 days Full access to the AI scribe and AI assistant. You won't be charged today, you can cancel anytime, and every plan includes a 45-day money-back guarantee. Start your free trial See the AI Scribe features ## FAQs Is BastionGPT a good DeepCura alternative for a small practice? Yes, if documentation and a compliant AI assistant are what you need most. BastionGPT costs $20 per user per month with no credit caps: unlimited AI scribe, unlimited transcription, a full healthcare AI assistant, and a signed BAA on every plan, including the 7-day free trial. DeepCura charges $129 per provider per month with usage metered at 1,000 monthly credits. How much does DeepCura cost compared to BastionGPT? DeepCura costs $129 per provider per month or $999 per year, with volume discounts to $101 per month at 11+ seats, and meters usage at 1,000 credits per month per seat (a $59 add-on doubles it). BastionGPT publishes all plans at $20 to $75 per user per month; the $20 plan includes the unlimited scribe and full AI assistant with no credit system. What is the difference between DeepCura and BastionGPT? DeepCura is an all-in-one practice automation platform: an AI scribe plus AI receptionist, billing, intake, and fax agents with native EHR write-back, metered by monthly credits. BastionGPT is a HIPAA-compliant AI assistant plus unlimited scribe: notes, referral letters, appeals, document analysis, and patient education from $20 per user per month, with no usage caps. Is DeepCura HIPAA compliant? DeepCura advertises HIPAA compliance, an included BAA, 256-bit encryption, and a Google CASA Tier 2 security assessment. It does not publish a completed SOC 2 Type II audit, HITRUST, or ISO 27001 certification. BastionGPT is HIPAA compliant, signs a BAA on every plan including the free trial, and runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure. Which is better for therapists and mental health professionals, DeepCura or BastionGPT? BastionGPT. It drafts DAP, BIRP, and detailed SOAP notes, labels up to 10 speakers for couples, family, and group sessions, supports 42 CFR Part 2 and FERPA records on every plan, and handles sensitive topics that consumer AI tools often refuse. DeepCura markets across 50+ specialties, with primary care, psychiatry, and procedural fields at the front. What are DeepCura credits and how do they work? Every DeepCura seat includes 1,000 monthly credits shared across all of its AI agents: scribe notes, receptionist calls, billing suggestions, intake, and faxes all draw from the same pool. A $59 per month add-on doubles the allotment to 2,000, and no refunds are issued for partial periods. BastionGPT has no credit system: transcription and assistant use are unlimited on every plan. Does DeepCura integrate with Epic, athenahealth, and eClinicalWorks? Yes. DeepCura lists native integrations with Epic, athenahealth, eClinicalWorks, AdvancedMD, Practice Fusion, OpenEMR, OptiMantra, CharmHealth, and Veradigm, and can write notes back into those systems. BastionGPT takes a different approach: it works alongside 50+ EMR/EHR systems with nothing for IT to install, and you review and edit each finished note before it enters the chart. Does BastionGPT have an AI receptionist or billing agent like DeepCura? No. BastionGPT focuses on documentation and clinical support: the unlimited scribe, referral letters, insurance appeals, patient education, document analysis, and a coding review that flags vague language and suspected under- or over-coding. If a 24/7 phone agent and automated billing workflows are your core need, that breadth is DeepCura's genuine advantage. Does DeepCura have a free trial? Yes. DeepCura offers a free trial with no credit card, limited by trial credits. BastionGPT's 7-day free trial asks for a card up front, with no charge until the trial ends, and includes the full product: unlimited scribe, full assistant, and a signed BAA active during the trial, plus a 45-day money-back guarantee once you subscribe. Does DeepCura use patient data to train its AI? DeepCura states customer data is not used for model training, while its homepage separately advertises that its AI is "trained on real clinical workflows, not synthetic data." Those can be reconciled, but ask DeepCura what is retained and how model improvement works, in writing, before onboarding. BastionGPT never uses your inputs, documents, or audio to train AI models, and transcript retention is user-configurable from 1 to 365 days. How accurate are BastionGPT and DeepCura? BastionGPT achieved 99.3% sentence-level accuracy on clinical documentation in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. DeepCura publishes no independent accuracy study; its reviewers praise transcription quality, and its reported 6,000+ clinicians and 1M+ encounters are the vendor's own figures. Is DeepCura SOC 2 certified? DeepCura publishes a Google CASA Tier 2 security assessment verified by TAC Security, alongside HIPAA compliance, an included BAA, and 256-bit encryption. A completed SOC 2 Type II audit, HITRUST certification, or ISO 27001 certification is not published on its website. BastionGPT runs exclusively on HITRUST CSF certified and SOC 2 Type II attested infrastructure and is pursuing its own SOC 2 Type II attestation and HITRUST CSF certification. When is DeepCura the better choice than BastionGPT? Choose DeepCura if your practice runs on one of its supported EHRs and you want one credit-based subscription to automate the front office: a 24/7 AI receptionist, billing and coding suggestions, intake, and faxing, with notes written back natively into the chart. For solo, small, independent, and mental health practices whose core need is documentation and a compliant AI assistant without usage caps, BastionGPT delivers more capability per dollar at $20 per user per month. ## More comparisons See how BastionGPT stacks up against other AI tools for healthcare: BastionGPT vs Lyrebird Health Alternative: BastionGPT vs Lyrebird Health (2026) → BastionGPT vs Doximity Alternative: BastionGPT vs Doximity (2026) → BastionGPT vs CompliantChatGPT Alternative: BastionGPT vs CompliantChatGPT (2026) → BastionGPT vs Commure Alternative: BastionGPT vs Commure (2026) → BastionGPT vs Twofold Alternative: BastionGPT vs Twofold Health (2026) → BastionGPT vs Mentalyc Alternative for Therapists: BastionGPT vs Mentalyc → BastionGPT vs Ambience Alternative: BastionGPT vs Ambience Healthcare 2026 → BastionGPT vs DeepScribe Alternative: BastionGPT vs DeepScribe (2026) → BastionGPT vs DAX Copilot Alternative: BastionGPT vs Dragon Copilot (2026) → BastionGPT vs Suki Alternative: BastionGPT vs Suki AI Scribe (2026) → BastionGPT vs Abridge Alternative: BastionGPT vs Abridge AI Scribe (2026) → BastionGPT vs Nabla Alternative: BastionGPT vs Nabla AI Scribe (2026) → BastionGPT vs Claude Alternative for Healthcare: BastionGPT vs Claude Team → BastionGPT vs Microsoft Copilot Alternative: BastionGPT vs Copilot (2026) → BastionGPT vs Hathr AI Alternative: BastionGPT vs Hathr AI (2026) → BastionGPT vs Heidi Health Alternative: BastionGPT vs Heidi Health (2026) → BastionGPT vs Freed Alternative: BastionGPT vs Freed AI Scribe (2026) → --- > Source: https://bastiongpt.com/templates # Behavioral Health Note Templates & Examples Free, copy-ready templates for the notes and documents healthcare and behavioral health professionals actually write. Every template includes a plain-language definition, the structure section by section, a complete fictional sample, and the compliance details that matter, reviewed by the BastionGPT Clinical Advisory Board. The library covers 100+ document types and new templates publish routinely. Don't see the one you need? Contact our support team and we'll prioritize it. A complete sample Every page includes a full fictional example at realistic length, plus a blank structure you can copy into your own workflow. Compliance built in Psychotherapy-note boundaries, payer and audit expectations, and privacy rules, explained for each document type. US, Canada, and Australia Key templates include a jurisdiction table covering naming, governing bodies, and billing context in all three countries. ## Browse the library No templates match. Try a shorter term, or tell us what you need . Session & Progress Notes Intake & Clinical Assessments Treatment Planning Risk & Safety Evaluation Reports Test Write-Up Guides Clinical Scales & Scores Care Coordination Consent & Privacy Billing & Practice Ops Work, School & Disability ## Session & Progress Notes The recurring note formats clinicians write after every session: SOAP, DAP, BIRP, PIE, and the specialized individual, group, couples, crisis, and telehealth session notes. ### BIRP Note A BIRP note is a four-part progress note format that organizes a session into Behavior, Intervention, Response, and Plan. Its signature strength is forcing the intervention-and-response thread onto the page, which is exactly what payer reviewers read for. A typical BIRP note runs 150 to 300 words. View template → ### Canada Telepsychology & Virtual Care Note A Canadian telepsychology and virtual care note is an ordinary progress note with telepresence elements added: session modality, the client's physical location, consent to virtual care, an emergency plan naming local resources, and a technology-failure fallback. Canadian psychologists and psychotherapists write one after every video or telephone session. No college mandates a template; expectations come from provincial standards and the CPA telepsychology guidelines. Most run 200 to 400 words. View template → ### Crisis Psychotherapy Note A crisis psychotherapy note documents an encounter where psychotherapy addressed an acute crisis: the precipitating event, urgent assessment, mental status, risk findings, interventions, client response, and disposition. Therapists, psychologists, counselors, and clinical social workers write one after crisis encounters in office, telehealth, or community settings. In the US it supports the time-based crisis codes 90839 and 90840. A typical crisis note runs 250 to 700 words. View template → ### DAP Note A DAP note is a three-part progress note format that organizes a therapy session into Data, Assessment, and Plan. It merges SOAP's Subjective and Objective material into a single Data section, which suits the more qualitative flow of psychotherapy documentation. A typical therapy DAP note runs 150 to 300 words. View template → ### Family & Couples Therapy Note A family and couples therapy note is the clinical record of a conjoint psychotherapy session, one where partners or family members attend together. It documents who attended, the relational patterns addressed, the interventions, each person's response, and progress toward treatment goals. Most run 200 to 500 words. Unlike an individual note, it must serve several people's privacy rights inside one story. View template → ### Group Therapy Note A group therapy note is the progress note a clinician writes for each member of a group psychotherapy session. It pairs a brief group-level summary of the session's topic and interventions with individualized documentation of that member's participation, response, and progress. Group therapists write one note per member after every session. A typical note runs 100 to 250 words per member. View template → ### Individual Therapy Session Note An individual therapy session note is the clinical record of one one-on-one psychotherapy session: what was addressed, the interventions used, the client's response, current risk, and the plan for the next session. Therapists, psychologists, counselors, and clinical social workers write one after every session, in SOAP, DAP, BIRP, or narrative form. Most run 150 to 350 words. View template → ### Narrative Progress Note A narrative progress note is a free-text progress note that documents a therapy session in flowing prose instead of the labeled sections of SOAP or DAP. Therapists use it when a session's chronology and reasoning matter more than a template, and it is fully compliant as long as the required content is present. Most narrative notes run 150 to 500 words. View template → ### PIE Note A PIE note is a three-part progress note format that organizes a session into Problem, Intervention, and Evaluation. Each entry hangs off a numbered problem from the client's problem list; there is no separate assessment or plan section, so interpretation and next steps live inside the entry. Behavioral health and nursing-adjacent teams use it for problem-focused charting. A typical PIE note runs 100 to 250 words. View template → ### Psychotherapy Progress Note A psychotherapy progress note is the session-by-session record of psychotherapy in the standard clinical chart: how the client presented, the interventions used, the response, and the plan. Despite the similar name, it is legally distinct from HIPAA's separately kept psychotherapy notes. Most therapy progress notes run 100 to 350 words. View template → ### SOAP Note A SOAP note is a progress note format that organizes a clinical encounter into four sections: Subjective, Objective, Assessment, and Plan. Therapists, psychologists, and integrated care teams use SOAP notes to document sessions in a structure that other providers, payers, and auditors can read quickly. A typical therapy SOAP note runs 150 to 350 words. View template → ### Telehealth Psychotherapy Note A telehealth psychotherapy note is the progress note for a therapy session delivered by video or phone. It records everything a standard session note does, plus the telehealth layer: modality, client location, provider location, consent, technology issues, and a session-specific emergency plan. Therapists, psychologists, and counselors write one after every virtual session. A typical telehealth note runs 150 to 400 words. View template → ## Intake & Clinical Assessments First-contact and interview-based assessment documents, from the intake note and psychiatric diagnostic evaluation to the biopsychosocial, trauma, and culturally responsive assessments. ### Aboriginal & Torres Strait Islander Culturally Safe Assessment A culturally safe assessment is a clinical assessment conducted and documented so that Aboriginal and Torres Strait Islander clients experience care as culturally safe, a judgment that belongs to the client, not the clinician. Australian psychologists, GPs, and multidisciplinary teams apply the standard to intake, mental health, and cognitive assessment. No law prescribes a form; the record evidences a culturally safe process. View template → ### Biopsychosocial Assessment A biopsychosocial assessment is the comprehensive intake document that examines a client's presenting problem across biological, psychological, and social domains and ties them together in an integrated summary. Therapists, clinical social workers, counselors, and case managers complete one at admission to treatment, and it anchors the diagnosis, level of care, and treatment plan. A typical biopsychosocial assessment runs 600 to 1,500 words. View template → ### Child & Adolescent Developmental Assessment A child and adolescent developmental assessment documents a child's developmental history and milestone attainment across motor, language, cognitive, social-emotional, and adaptive domains, then ties the findings to eligibility or diagnosis. Child psychologists, developmental-behavioral teams, and early intervention programs complete one when a screen raises concern or a school, pediatric, or Part C referral asks what is going on. Most run 500 to 1,500 words. View template → ### Cultural Formulation Interview (CFI) Intake A cultural formulation interview (CFI) intake documents what the 16-question CFI from DSM-5-TR elicits: the client's own definition of the problem, explanatory model, stressors and supports, cultural identity, and preferences for care. Psychologists, therapists, and psychiatrists fold it into initial evaluations when cultural context shapes diagnosis or engagement. The interview takes about 20 minutes; the documented section usually runs 200 to 500 words. View template → ### Eating Disorder Assessment An eating disorder assessment is a structured intake evaluation that combines a validated screener such as the SCOFF or EDE-Q, DSM-5-TR criterion mapping, eating and weight history, and a medical-risk appraisal with a level-of-care rationale. Therapists, psychologists, and counselors complete it at intake, coordinating with medical and nutrition colleagues. Most run 400 to 1,000 words. View template → ### General Risk Screening Note A mental health risk screening note documents a brief, standardized screen for suicide, violence, abuse, substance use, and psychosis risk: the instrument used, the result for each domain, a positive or negative determination, and the follow-up action. Therapists and intake staff complete it at intake or when concerns arise, often inside the intake or progress note. Most run 50 to 150 words. View template → ### Indigenous & First Nations Wellness Assessment Addendum An Indigenous wellness assessment addendum is a culturally grounded section added to a standard mental health intake for First Nations, Inuit, Métis, American Indian/Alaska Native, or Aboriginal and Torres Strait Islander clients. It documents strengths, family and community connections, language and culture, spiritual factors, and community supports. No payer or regulator mandates the format; it reflects wellness frameworks such as the FNMWC. Most run 300 to 600 words. View template → ### Intake Note A therapy intake note documents a new client's first clinical session: presenting problem, relevant history, a risk screen, mental status observations, an initial clinical impression, and the starting plan. Therapists, counselors, and psychologists write one at the start of each new episode of care, and the rest of the chart builds on it. A typical intake note runs 300 to 900 words. View template → ### Mental Status Exam (MSE) Note A mental status exam (MSE) is the clinician's structured description of how a client presents at one point in time: appearance, behavior, speech, mood, affect, thought process, thought content, perception, cognition, insight, and judgment. Psychologists, psychiatrists, therapists, and medical teams record a complete MSE at intake or diagnostic evaluation and briefer focused versions afterward. Most written MSEs run 80 to 250 words. View template → ### Parent & Caregiver Collateral Note A parent and caregiver collateral note documents information a clinician gathers from someone other than the client, such as a parent, guardian, or caregiver, and what the clinician did with it. Child and adolescent therapists write one after any substantive caregiver contact. The note lives in the client's chart, attributes every statement to its source, and typically runs 100 to 300 words. View template → ### Psychiatric Diagnostic Evaluation A psychiatric diagnostic evaluation is the comprehensive initial assessment that establishes a mental health diagnosis and treatment plan: psychiatric, medical, family, and social history, a mental status exam, and a risk assessment. Psychiatrists, psychologists, therapists, and psychiatric nurse practitioners complete one at the start of care; US payers bill it as CPT 90791, or 90792 when medical services are included. A typical evaluation runs 500 to 1,200 words. View template → ### School & Teacher Collateral Note A school and teacher collateral contact note is a therapist's chart record of a conversation with a teacher, school counselor, or other school staff member about a minor client. Child and adolescent clinicians write one after each school contact to capture classroom observations and connect them to the treatment plan. Most run 100 to 300 words. View template → ### Substance Use Assessment A substance use assessment is the comprehensive clinical evaluation that establishes whether a substance use disorder is present, how severe it is, and which level of care fits. Behavioral health clinicians and alcohol and drug counselors complete one at entry to treatment, usually organized around the six ASAM Criteria dimensions. A typical substance use assessment runs 400 to 1,200 words. View template → ### Trauma Assessment A trauma assessment is a structured clinical evaluation of a client's trauma history, current trauma- and stressor-related symptoms, dissociation, and immediate safety. Psychologists, therapists, and counselors complete one at intake, or when trauma surfaces during treatment, usually pairing a clinical interview with a validated screen such as the PCL-5 or PC-PTSD-5. A typical trauma assessment runs 300 to 900 words. View template → ## Treatment Planning Plans, reviews, and formulations that set treatment goals and tie every session back to medical necessity. ### Better Access Mental Health Treatment Plan Referral A Mental Health Treatment Plan is the document an Australian GP or prescribed medical practitioner prepares under Better Access to assess a diagnosed mental disorder and support Medicare-rebated psychology sessions. A separate signed and dated referral, not the plan, authorises those sessions. GPs prepare both; the psychologist holds the referral and reports back after each course. Most plan-and-referral pairs run 600 to 1,200 words. View template → ### Case Formulation Note A case formulation note (also called a case conceptualization) is the working hypothesis of a case: how predisposing, precipitating, perpetuating, and protective factors fit together to produce the presenting problem, and what that means for treatment. It is written at intake and revised at major reviews rather than every session. Most run one to two pages, and no law or payer prescribes its format. View template → ### Outcome Measure Review Note An outcome measure review note is the documentation entry of measurement-based care: it records which standardized instrument was administered, the score, the change from baseline, the clinical interpretation, the discussion with the client, and the treatment decision that follows. Therapists, psychologists, and collaborative care teams write one each time a measure like the PHQ-9 or GAD-7 is given. Most run 75 to 300 words beside a score flowsheet. View template → ### Relapse Prevention Plan A relapse prevention plan is a written plan a clinician and client build together that names the client's triggers, early warning signs, coping strategies, support contacts, and the steps to take after a lapse. Substance use and mental health clinicians write one mid-treatment or before step-down, usually inside the treatment plan. Most run 300 to 800 words, built in a 20 to 45 minute session. View template → ### Treatment Plan A mental health treatment plan is the roadmap document for a course of therapy: diagnoses, presenting problems, measurable goals and objectives, the interventions that will address them, and how often care happens. Unlike a progress note, it is future-focused and spans many sessions. Most run one to three pages and are reviewed on a set schedule. View template → ### Treatment Plan Review & Update A treatment plan review is the periodic note that measures progress against each goal in the active mental health treatment plan, records barriers and new clinical information, and ends in a decision: continue the plan unchanged or update it. Clinicians complete one on their program's cadence and whenever the clinical picture changes. Most run 150 to 500 words. View template → ## Risk & Safety Documentation for suicide and violence risk assessment, self-harm, safety planning, and mandated reporting duties. ### Mandated Reporting & Duty-to-Protect Documentation Mandated reporting and duty-to-protect documentation is the chart record a clinician creates when reporting suspected child or elder abuse, or acting on a client's threat of serious violence against an identifiable person. It captures the facts observed, the legal basis, who was notified and when, and the clinical reasoning. State law mandates the report itself; the note is the clinician's proof the duty was met. View template → ### Safety Plan A safety plan is a brief document a clinician and client build together during or after a suicidal crisis: personal warning signs, coping strategies, people and places that help, professional and crisis contacts, and steps to make the environment safer. The client keeps a copy. Most follow the six-step Stanley-Brown format, often with an optional reasons-for-living step, and take 20 to 45 minutes to complete together. View template → ### Self-Harm (NSSI) Assessment A self-harm assessment, also called an NSSI assessment, is the clinical documentation of how a clinician evaluated non-suicidal self-injury: the behavior's history and severity, the suicide risk screen that separates it from suicidal self-injury, its function, and the risk formulation with the response. Clinicians document one when self-injury is disclosed or discovered. Most run 300 to 800 words, and no regulator prescribes a format. View template → ### Suicide Risk Assessment A suicide risk assessment is the clinical documentation of how a clinician evaluated a client's suicide risk: the inquiry into ideation, plan, intent, and access to means, the risk and protective factors weighed, a risk formulation, and the actions taken. Mental health professionals document one at intake, when risk indicators emerge, and after a crisis. Most run 200 to 800 words, and the formulation must support the disposition. View template → ### Violence & Homicide Risk Assessment A violence and homicide risk assessment is the clinical record of how a clinician evaluated a client's risk of harm to others: the threat and its context, target identifiability, intent and access to means, risk and protective factors, a formulation, and the protective actions taken or reasonably ruled out. Clinicians document one when a client threatens an identifiable person or homicidal ideation emerges. Most run 300 to 800 words. View template → ## Psychological Evaluation Reports Complete report structures psychologists write after testing: psychological, neuropsychological, psychoeducational, ADHD, autism, and forensic evaluations. ### ADHD Evaluation Report An ADHD evaluation report is the written product of a diagnostic assessment for attention-deficit/hyperactivity disorder. It integrates the referral question, history, observations, and multi-informant rating scales into a DSM-5-TR criteria mapping, a differential, a diagnosis, and recommendations. Psychologists, psychiatrists, pediatricians, and primary care prescribers write them for children and adults. Most run 3 to 10 pages and are shared with families, schools, and prescribers. View template → ### Autism Spectrum Evaluation Report An autism spectrum evaluation report is the written product of a diagnostic evaluation for autism spectrum disorder. It integrates developmental history, caregiver interview, direct observation, and standardized instruments into a DSM-5-TR criteria mapping, a diagnosis with severity and support levels, and functional recommendations. Psychologists, developmental pediatricians, and multidisciplinary teams write them for children and adults. Most run 4 to 15 pages and feed schools, payers, and disability programs. View template → ### Capacity & Competency Evaluation A capacity evaluation is a clinician's structured assessment of whether a person can make a specific decision at a specific time, such as consenting to treatment, managing finances, or making a will. Psychologists, psychiatrists, and physicians write them for treating teams, courts, and tribunals. Capacity is the clinical judgment; competency is the court's legal determination. Most reports run 2 to 8 pages. View template → ### Fitness-for-Duty Evaluation A fitness-for-duty evaluation (FFDE) is a formal psychological examination of an employee, ordered by the employer when objective evidence suggests a mental health condition may be impairing safe or effective job performance. Psychologists and psychiatrists retained by the employer conduct them; the employer, not the examinee, is the client. Most reports run 2 to 8 pages and answer a specific referral question. View template → ### Forensic Psychological Evaluation Report A forensic psychological evaluation report is the written product of an evaluation conducted to answer a legal question, such as competency to stand trial, criminal responsibility, or parenting capacity. A court or attorney is the client, the person examined is an evaluee rather than a patient, and confidentiality is limited. Psychologists and psychiatrists write them; length varies by matter, and no statute sets a maximum. View template → ### NDIS Functional Capacity Assessment Report An NDIS functional capacity assessment report is an allied health report documenting how a person's disability affects everyday functioning across six domains: communication, social interaction, learning, mobility, self-care and self-management. It provides evidence for NDIS access and planning decisions and recommends supports with frequency and duration. Occupational therapists, psychologists and other qualified professionals write them. Most run 15 to 30 pages after 1 to 10 hours of assessment. View template → ### NDIS Plan Reassessment & Progress Report An NDIS progress report, called a plan reassessment report in current NDIA language, is the document a treating allied health provider writes toward the end of a participant's plan period. It summarises the supports delivered, evidences progress against a functional baseline using named outcome measures, and justifies recommended supports for the next plan against the NDIS reasonable and necessary criteria. Most run 3 to 6 pages. View template → ### Neuropsychological Evaluation Report A neuropsychological evaluation report is the written product of neuropsychological testing: it integrates the referral question, history, behavioral observations, and standardized cognitive test scores into a domain-by-domain interpretation, a diagnosis, and functional recommendations. Neuropsychologists write one after evaluating conditions such as brain injury, dementia, ADHD, or epilepsy, addressed to the referring physician and increasingly read by the patient. Most clinical reports run 5 to 20 pages. View template → ### Psychoeducational Evaluation Report A psychoeducational evaluation report is the written product of an assessment that integrates cognitive ability, academic achievement, and behavioral data to answer a question about learning. School psychologists write them to inform special-education eligibility and instruction planning; private evaluators write them as independent educational evaluations. Most run 4 to 15 pages, and parents receive a copy of a school evaluation report at no cost. View template → ### Psychological Evaluation Report A psychological evaluation report is the written product of psychological testing: it integrates the referral question, history, behavioral observations, and standardized test results into an interpretation, a diagnosis, and recommendations. Psychologists write one after an assessment battery, addressed to the referrer and increasingly read by the patient. Most clinical reports run 3 to 12 pages. View template → ### Psychological Testing Administration Note A psychological testing administration note documents a test administration and scoring encounter: which standardized instruments were given, who administered each one (psychologist or technician), and the total face-to-face minutes supporting each billed unit of CPT 96136 to 96139. Psychologists, neuropsychologists, and testing technicians write one for every administration session. Most notes run 150 to 350 words plus the time log. View template → ### Psychological Testing Scoring Note A psychological testing scoring note records the professional evaluation side of psychological or neuropsychological testing: verifying and summarizing scores, integrating them with clinical data, and logging the dated minutes that support CPT 96130 to 96133. Psychologists and neuropsychologists write one per assessment episode, alongside the administration record and the evaluation report. Most run 200 to 400 words plus the score summary and time log. View template → ## Test Write-Up Guides Instrument-specific write-up guides with scored fictional samples: WISC-V, WAIS, BASC-3, Conners 4, MMPI-3, ADOS-2, and the other measures psychologists report on. ### ABAS-3 Report Write-Up The ABAS-3 (Adaptive Behavior Assessment System, Third Edition) is a norm-referenced measure of adaptive behavior, the everyday conceptual, social, and practical skills a person uses to function independently, rated entirely by forms: parent, teacher, and adult versions covering birth through age 89. Psychologists use it in intellectual-disability, autism, and disability evaluations. This page covers how to write up ABAS-3 results, with a fictional two-rater sample and a template. View template → ### ADI-R Report Write-Up The ADI-R (Autism Diagnostic Interview-Revised) is a standardized, semi-structured caregiver interview that gathers developmental history and current behavior when autism is part of the question. Psychologists and other trained clinicians use it as one input to an autism evaluation, and it yields a categorical classification, not a diagnosis. This page covers how to write up ADI-R results, with a fictional sample and a results-section template. View template → ### ADOS-2 Report Write-Up The ADOS-2 (Autism Diagnostic Observation Schedule, Second Edition) is a semi-structured observation used to assess social communication, play, and restricted or repetitive behaviors when autism is a question. Psychologists and other trained clinicians use it as one input to an autism evaluation, and it yields a classification, not a diagnosis. This page covers how to write up ADOS-2 results, with a fictional sample and a results-section template. View template → ### ASQ-3 Score Documentation The ASQ-3 (Ages & Stages Questionnaires, Third Edition) is a parent-completed developmental screen for ages 1 to 66 months, scored in five domains against interval-specific cutoffs. Pediatric, family, and early-childhood programs use it at well-child visits and program intakes. Each domain lands above cutoff, in the monitoring zone, or below cutoff, and each zone obligates something different. This page covers how to document ASQ-3 results, with a fictional sample note. View template → ### ASRS v1.1 Score Documentation The ASRS v1.1 (Adult ADHD Self-Report Scale) is a six-question adult ADHD screener developed through the WHO World Mental Health initiative and published by Kessler and colleagues in 2005. Primary care, psychiatry, and telehealth teams use it at intake and to track treatment. It screens; it never diagnoses. This page covers how to document ASRS results defensibly, with a fictional sample note. View template → ### AUDIT-C Score Documentation The AUDIT-C is a three-item alcohol consumption screen taken from the World Health Organization's ten-item AUDIT, validated as a standalone test by Bush and colleagues in 1998 and scored 0 to 12. Primary care, VA, and behavioral health clinicians use it to find unhealthy drinking and decide who needs a fuller assessment; it never diagnoses. This page covers how to document and interpret AUDIT-C results, with a fictional sample note. View template → ### BASC-3 Report Write-Up The BASC-3 (Behavior Assessment System for Children, Third Edition) is a multi-informant rating system measuring behavioral, emotional, and adaptive functioning from age 2 through college age, with teacher, parent, and self-report forms. Psychologists and school psychologists use it in ADHD, emotional, and autism-related evaluations. This page covers how to write up BASC-3 results, with a fictional sample and a results-section template. View template → ### Battelle Developmental Inventory (BDI-3) Report Write-Up The BDI-3 (Battelle Developmental Inventory, Third Edition) is a norm-referenced battery for birth through 7 years 11 months, scoring five IDEA-aligned domains through structured tasks, observation, and interview. Early-intervention and preschool teams use it for eligibility and federal outcome reporting. It is unrelated to the Beck Depression Inventory (BDI-II). This page covers how to write up BDI-3 results, with a fictional sample. View template → ### Bayley-4 Report Write-Up The Bayley-4 (Bayley Scales of Infant and Toddler Development, Fourth Edition) is an individually administered developmental assessment for children 16 days to 42 months old, published by Pearson in 2019. Psychologists, therapists, and early-intervention teams use it to measure cognitive, language, and motor development and to inform eligibility decisions. This page covers how to write up Bayley-4 results, with a fictional sample and a results-section template. View template → ### BDI-II Score Documentation The BDI-II (Beck Depression Inventory, second edition) is a 21-item, licensed depression severity measure published by Beck, Steer, and Brown in 1996 and scored 0 to 63. Behavioral health, medical, and evaluation practices use it for baselines, outcome monitoring, and assessment batteries. This page covers how to document BDI-II results defensibly, with a fictional sample note. View template → ### Beery VMI Report Write-Up The Beery VMI (Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition) is a norm-referenced battery for ages 2 to 100: a core copying test plus optional Visual Perception and Motor Coordination supplemental tests whose contrast localizes a weakness. OTs, school psychologists, and evaluators use it in handwriting and developmental referrals. Reporting only one score discards the design. This page covers how to write up Beery VMI results, with a fictional sample. View template → ### Boston Naming Test Score Documentation The Boston Naming Test (BNT-2) is a 60-item confrontation naming test: the examinee names line drawings that run from common to rare words. Neuropsychologists and speech-language pathologists use it in aphasia, dementia, and epilepsy surgery evaluations. A total is uninterpretable without the form, edition, cue accounting, and norm source. This page covers how to document BNT results, with a fictional sample. View template → ### BRIEF-2 Report Write-Up The BRIEF-2 (Behavior Rating Inventory of Executive Function, Second Edition) is a rating scale of everyday executive function for ages 5 to 18, completed by parents, teachers, and adolescents themselves, scored as T scores across three indexes and a global composite. Psychologists use it in ADHD, autism, TBI, and special-education evaluations. This page covers how to write up BRIEF-2 results, with a fictional sample and a template. View template → ### CAARS-2 Report Write-Up The CAARS-2 (Conners Adult ADHD Rating Scales, 2nd Edition) is a 2023 self-report and observer rating scale measuring ADHD symptoms, impairment, and response style in adults 18 and older. Psychologists and psychiatrists use it to quantify symptom severity inside adult ADHD evaluations. This page covers how to write up CAARS-2 results, with a fictional sample, reviewer-flagged errors, and free templates. View template → ### CAGE Questionnaire Documentation The CAGE questionnaire is a four-item, yes-or-no alcohol screen published by John A. Ewing in JAMA in 1984, worded for lifetime experience and built to detect dependence rather than current risky drinking. Primary care, emergency, and behavioral health clinicians use it, and the CAGE-AID variant for drugs, to decide who needs a fuller assessment. This page covers how to document and interpret CAGE results, with a fictional sample note. View template → ### CANS Assessment Documentation The CANS (Child and Adolescent Needs and Strengths) is a communimetric assessment, copyrighted by the Praed Foundation and free to use with annual certification, that rates a child's needs and strengths on four action levels rather than producing a score. Care coordinators, caseworkers, and behavioral health clinicians complete it at intake and reassessment in Medicaid and child welfare systems. This page covers how to document CANS ratings, with a fictional sample. View template → ### CAPS-5 Documentation The CAPS-5 (Clinician-Administered PTSD Scale for DSM-5) is the structured-interview reference standard for diagnosing PTSD: a trained clinician rates 20 symptoms against an index trauma, applies the DSM-5 algorithm, and sums a 0-80 severity total. VA, research, and trauma clinics rely on it. The write-up must name the form, show the algorithm, and keep PCL-5 numbers separate. This page covers CAPS-5 documentation, with a fictional sample. View template → ### CARS-2 Report Write-Up The CARS-2 (Childhood Autism Rating Scale, Second Edition) is a clinician-completed rating scale that quantifies autism symptom severity from direct observation, record review, and collateral information. Psychologists and other trained clinicians use it as one severity input within a broader autism evaluation, never as a standalone diagnosis. This page covers how to write up CARS-2 results, with a fictional sample and a results-section template. View template → ### CELF-5 Report Write-Up The CELF-5 (Clinical Evaluation of Language Fundamentals, Fifth Edition) is a norm-referenced language battery for ages 5:0 through 21:11 that yields a Core Language Score and index scores from age-selected tests. Speech-language pathologists use it in school eligibility and clinical language evaluations. The write-up must name the tests behind each composite and the eligibility rule applied. This page covers how to write up CELF-5 results, with a fictional sample. View template → ### CIWA-Ar Protocol Documentation The CIWA-Ar (Clinical Institute Withdrawal Assessment for Alcohol, revised) is a 10-item alcohol-withdrawal severity scale published by Sullivan and colleagues in 1989 and scored 0 to 67. Hospital, emergency, and detoxification teams use it for serial monitoring under symptom-triggered and fixed-schedule protocols. This page covers how to document CIWA-Ar assessments defensibly, with a fictional sample note. View template → ### Clock Drawing Test Score Documentation The clock drawing test is a brief constructional screen in which the patient draws a clock face and sets a requested time. It has no single version: at least a dozen named scoring systems use different ranges, conditions, and anchors. Primary care, geriatrics, and neuropsychology teams use it in cognitive screening. A score means nothing without its system. This page covers how to document clock drawing results, with a fictional sample note. View template → ### Conners 4 Report Write-Up The Conners 4 (Conners 4th Edition) is a multi-informant rating scale measuring ADHD symptoms, co-occurring problems, and functional impairment for ages 6 to 18, with parent, teacher, and self-report forms. Psychologists, school psychologists, and pediatric clinicians use it in ADHD evaluations, school assessments, and treatment monitoring. This page covers how to write up Conners 4 results, with a fictional sample and a results-section template. View template → ### COWS Protocol Documentation The COWS (Clinical Opiate Withdrawal Scale) is an 11-item, clinician-administered opioid-withdrawal measure published by Wesson and Ling in 2003 and scored 0 to 48. Emergency, hospital, addiction-medicine, and opioid treatment program teams use it around buprenorphine initiation and withdrawal monitoring. This page covers how to document COWS assessments defensibly, with a fictional sample note. View template → ### C-SSRS Documentation The C-SSRS (Columbia-Suicide Severity Rating Scale) is a widely adopted instrument that classifies suicidal ideation and behavior into defined categories. Hospitals, clinics, schools, and research programs use it to standardize screening and the response that follows. This page covers how to document C-SSRS administration and results in the chart, with a fictional, structure-only sample note. View template → ### DASS-21 Score Documentation The DASS-21 (Depression Anxiety Stress Scales, 21-item short form) is a public-domain self-report measure with three seven-item subscales, published by Lovibond and Lovibond in 1995 and scored over the past week. Psychologists, GPs, and counselors use it at intake and review, especially in Australia. It measures symptom severity; it never diagnoses. This page covers how to document and interpret DASS-21 results, with a fictional sample note. View template → ### DAST-10 Score Documentation The DAST-10 (Drug Abuse Screening Test) is a ten-item, yes-or-no self-report screen for problems related to drugs other than alcohol over the past 12 months, derived from Harvey Skinner's 1982 DAST and published by CAMH. Primary care, emergency, and behavioral health clinicians use it to decide who needs a substance use assessment; it never diagnoses. This page covers how to document and interpret DAST-10 results, with a fictional sample note. View template → ### DAYC-2 Report Write-Up The DAYC-2 (Developmental Assessment of Young Children, Second Edition) is a norm-referenced battery for birth through 5 years 11 months, scoring five IDEA-aligned domains through observation, caregiver interview, and direct assessment. Early-intervention teams use it for Part C eligibility evaluations. The write-up must name each domain's method and the state's exact eligibility rule. This page covers how to write up DAYC-2 results, with a fictional sample. View template → ### D-KEFS Report Write-Up The D-KEFS (Delis-Kaplan Executive Function System, 2001) is a battery of nine co-normed stand-alone tests of executive function for ages 8 to 89, with no overall composite by design. Neuropsychologists, clinical psychologists, and school psychologists select tests from it in ADHD, TBI, epilepsy, and dementia evaluations. This page covers how to write up D-KEFS results, with a fictional sample and free templates. View template → ### EPDS Score Documentation The EPDS (Edinburgh Postnatal Depression Scale) is a 10-item perinatal depression screen published by Cox, Holden, and Sagovsky in 1987 and scored 0 to 30 over the past seven days. Obstetric, midwifery, pediatric, and primary care teams use it through pregnancy and the postpartum year. It screens; it never diagnoses. This page covers how to document EPDS results defensibly, with a fictional sample note. View template → ### Epworth Sleepiness Scale Documentation The Epworth Sleepiness Scale (ESS) is an eight-item self-report measure of usual daytime dozing propensity, published by Murray Johns in 1991 and scored 0 to 24. Sleep, respiratory, primary care, and occupational clinicians use it at evaluation and to track treatment. It measures sleepiness; it never diagnoses sleep apnea. This page covers how to document and interpret ESS results in the chart, with a fictional sample note. View template → ### GAD-7 Score Documentation The GAD-7 is a seven-item anxiety severity measure scored 0 to 21, developed by Spitzer, Kroenke, Williams, and Löwe in 2006 and free to reproduce under the official PHQ Screeners terms. Primary care and behavioral health teams use it for anxiety screening and symptom monitoring. This page covers how to document and interpret GAD-7 results in the chart, with a fictional sample note. View template → ### GAF Score Documentation The GAF (Global Assessment of Functioning) is a clinician-rated 1 to 100 score of psychiatric symptom severity and functioning, published as Axis V of DSM-IV-TR and dropped from DSM-5 in 2013. Psychiatrists, psychologists, and community mental health clinicians still enter one when a payer form, legacy EHR field, disability file, or court asks for it. This page covers how to document a defensible legacy GAF, with a fictional sample note. View template → ### GARS-3 Report Write-Up The GARS-3 (Gilliam Autism Rating Scale, Third Edition) is an informant rating scale for ages 3 through 22: a parent, teacher, or clinician rates items across six subscales, producing an Autism Index and probability language. Schools and autism teams use it as supporting evidence, never as a diagnosis. The write-up must name the respondent and index used. This page covers how to write up GARS-3 results, with a fictional sample. View template → ### Geriatric Depression Scale (GDS-15) Score Documentation The Geriatric Depression Scale (GDS) is a public-domain, yes-or-no self-report depression screen for older adults, with a 30-item original (Yesavage and colleagues, 1982) and the 15-item GDS-15 short form (Sheikh and Yesavage, 1986). Primary care, geriatric, and behavioral health clinicians use it to decide who needs a full mood assessment; it screens, never diagnoses. This page covers how to document and interpret GDS-15 results, with a fictional sample note. View template → ### GFTA-3 Report Write-Up The GFTA-3 (Goldman-Fristoe Test of Articulation, Third Edition) is a norm-referenced single-word articulation test for ages 2:0 through 21:11 that scores consonant production against same-age, same-sex norms. Speech-language pathologists use it in school eligibility and clinical speech sound evaluations. The write-up must pair the score with pattern, stimulability, and connected-speech evidence. This page covers how to write up GFTA-3 results, with a fictional sample. View template → ### Insomnia Severity Index Documentation The Insomnia Severity Index (ISI) is a seven-item self-report measure of perceived insomnia severity and impact, developed by Charles Morin and validated by Bastien and colleagues in 2001, scored 0 to 28. CBT-I therapists, primary care, and sleep clinics use it at baseline, mid-course, and completion. It measures perceived severity; it never diagnoses insomnia disorder. This page covers how to document and interpret ISI results, with a fictional sample note. View template → ### KTEA-3 Report Write-Up The KTEA-3 (Kaufman Test of Educational Achievement, Third Edition) is an individually administered academic achievement battery for ages 4:0 through 25:11, measuring reading, math, written language, and oral language with standard scores set to a mean of 100. Psychologists and school psychologists use it in specific learning disability evaluations for its norm-referenced error analysis and parallel forms. This page covers how to write up KTEA-3 results, with a fictional sample. View template → ### MADRS Score Documentation The MADRS (Montgomery-Åsberg Depression Rating Scale) is a clinician-rated, ten-item depression severity scale, published in 1979 and designed to be sensitive to change, with a total of 0 to 60. Psychiatrists, nurse practitioners, and research raters use it to track treatment-resistant depression, esketamine, TMS, and trial outcomes. It measures severity and change; it never diagnoses. This page covers how to document and interpret MADRS results, with a fictional sample note. View template → ### M-CHAT-R/F Score Documentation The M-CHAT-R/F (Modified Checklist for Autism in Toddlers, Revised, with Follow-Up) is a two-stage autism screen for toddlers: a 20-item parent questionnaire, then a structured Follow-Up interview that resolves medium scores. Pediatric and family practices use it at 18- and 24-month visits. A medium score without the Follow-Up is an incomplete screen, not a result. This page covers how to document M-CHAT-R/F results, with a fictional sample. View template → ### MCMI-IV Report Write-Up The MCMI-IV (Millon Clinical Multiaxial Inventory-IV, 2015) is a 195-item true/false personality inventory for adults 18 and older who are already in mental health assessment or treatment, published by NCS Pearson. Psychologists use it to clarify personality-pattern questions and plan treatment, reading results as prevalence-anchored Base Rate scores rather than T scores. This page covers how to write up MCMI-IV results, with a fictional sample and a results-section template. View template → ### MIGDAS-2 Report Write-Up The MIGDAS-2 (Monteiro Interview Guidelines for Diagnosing the Autism Spectrum, Second Edition) is a sensory-based diagnostic interview that yields a qualitative three-domain profile instead of scores. Psychologists and evaluation teams use it within autism evaluations, almost always paired with a scored instrument. This page covers how to write up MIGDAS-2 findings, with a fictional sample and a results-section template. View template → ### Mini-Cog Score Documentation The Mini-Cog is a three-minute cognitive screen scored 0 to 5, combining three-item recall with a pass-or-fail clock drawing, published by Soo Borson and colleagues in 2000. Primary care, geriatrics, and hospital teams use it at wellness visits and as a first-pass check when memory concerns arise. It screens; it never diagnoses. This page covers how to document and interpret Mini-Cog results in the chart, with a fictional sample note. View template → ### MMPI-3 Report Write-Up The MMPI-3 (Minnesota Multiphasic Personality Inventory-3, 2020) is a 335-item true/false broadband personality inventory for adults 18 and older, published by the University of Minnesota Press and distributed by Pearson. Psychologists use it in clinical, forensic, public-safety, and presurgical evaluations to assess emotional, thought, and behavioral dysfunction. This page covers how to write up MMPI-3 results, with a fictional sample and a results-section template. View template → ### MMSE Score Documentation The MMSE (Mini-Mental State Examination) is a 30-point cognitive screen published by Folstein, Folstein, and McHugh in 1975, now a licensed instrument sold per form by its publisher. Hospitals, clinics, and drug-coverage programs still key decisions to it. This page covers how to document and interpret MMSE results in the chart, with a fictional sample note. View template → ### MoCA Score Documentation The MoCA (Montreal Cognitive Assessment) is a 30-point cognitive screening instrument developed by Ziad Nasreddine and validated in 2005, built to detect mild cognitive impairment. Primary care, neurology, and geriatrics teams use it at wellness visits and in memory workups. This page covers how to document and interpret MoCA results in the chart, with a fictional sample note. View template → ### NEPSY-II Report Write-Up The NEPSY-II (2007) is a developmental neuropsychological battery for ages 3 to 16 that yields subtest-level scores across six domains, with no overall composite by design. Psychologists and school psychologists use it to profile attention, language, memory, sensorimotor, social perception, and visuospatial skills in pediatric evaluations. This page covers how to write up NEPSY-II results, with a fictional sample and free templates. View template → ### PAI Report Write-Up The PAI (Personality Assessment Inventory, 1991) is a 344-item self-report inventory of adult personality and psychopathology for ages 18 to 89, authored by Leslie C. Morey and published by PAR. Psychologists use it in clinical, forensic, correctional, and post-offer public safety evaluations, valued for its treatment planning scales and fourth-grade reading level. This page covers how to write up PAI results, with a fictional sample and a results-section template. View template → ### PCL-5 Score Documentation The PCL-5 (PTSD Checklist for DSM-5) is a free, public-domain, 20-item self-report measure of PTSD symptoms scored 0 to 80, stewarded by the VA's National Center for PTSD. Trauma therapists, VA and DoD clinicians, and primary care teams use it for screening and treatment monitoring. This page covers how to document and interpret PCL-5 results in the chart, with a fictional sample note. View template → ### PCL-R Report Write-Up The PCL-R (Hare Psychopathy Checklist-Revised, Second Edition, 2003) is a 20-item clinician rating of psychopathic traits scored from a semi-structured interview plus collateral records. Forensic psychologists and psychiatrists use it in sentencing, civil commitment, parole, and dangerous-offender evaluations. A defensible write-up reports the total as a measurement band against a named comparison group and never as a diagnosis. This page covers how to document PCL-R results, with a fictional sample. View template → ### PHQ-9 Score Documentation The PHQ-9 is a nine-item, public-domain depression severity measure scored 0 to 27, validated by Kroenke, Spitzer, and Williams in 2001. Primary care and behavioral health teams use it for depression screening and measurement-based care. This page covers how to document and interpret PHQ-9 results in the chart, including item 9 follow-up, with a fictional sample note. View template → ### PLS-5 Report Write-Up The PLS-5 (Preschool Language Scales, Fifth Edition) is a norm-referenced language test for birth through 7 years 11 months that yields Auditory Comprehension, Expressive Communication, and Total Language standard scores. Speech-language pathologists and early-intervention evaluators use it in Part C and preschool eligibility evaluations. The write-up must label caregiver-reported evidence and apply the state's exact rule. This page covers how to write up PLS-5 results, with a fictional sample. View template → ### PSC-17 Score Documentation The PSC-17 (Pediatric Symptom Checklist-17) is a free 17-item caregiver questionnaire screening children's psychosocial functioning, scored as one total plus internalizing, attention, and externalizing subscales, each with its own cutoff. Pediatric, family, and school-based clinicians use it at well visits and EPSDT checkups. A child can screen negative overall and positive on a subscale. This page covers how to document PSC-17 results, with a fictional sample. View template → ### RBANS Report Write-Up The RBANS (Repeatable Battery for the Assessment of Neuropsychological Status) is a roughly 30-minute neuropsychological screening battery: 12 subtests yield five index scores and a Total Scale score, with four parallel forms for retesting. Neuropsychologists, psychologists, and medical teams use it to screen and track cognition in dementia, stroke, and psychiatric care. This page covers how to write up RBANS results, with a fictional sample. View template → ### SASSI Documentation The SASSI (Substance Abuse Subtle Screening Inventory) is a proprietary self-report screen published by The SASSI Institute in adult SASSI-4 and adolescent SASSI-A3 editions, whose publisher decision rules return a high or low probability of a substance use disorder. Addiction, DUI, probation, child-welfare, and EAP evaluators use it to decide who needs full assessment; it never diagnoses. This page covers how to document a SASSI result, with a fictional sample. View template → ### Sensory Profile 2 Report Write-Up The Sensory Profile 2 (SP2) is a family of five caregiver and teacher questionnaires (birth through 14 years 11 months) measuring sensory processing patterns through Dunn's four quadrants. OTs and evaluation teams use it in autism, feeding, and school evaluations. The write-up must name the form and respondent, and report patterns without diagnosing. This page covers how to write up SP2 results, with a fictional sample. View template → ### SLUMS Score Documentation The SLUMS (Saint Louis University Mental Status examination) is a 30-point cognitive screen developed by Tariq and colleagues at Saint Louis University with the St. Louis VA and published in 2006, interpreted through education-adjusted bands. Primary care, geriatrics, and VA teams use it as a no-fee alternative to the licensed MMSE. It screens; it never diagnoses. This page covers how to document SLUMS results, with a fictional sample note. View template → ### SRS-2 Report Write-Up The SRS-2 (Social Responsiveness Scale, Second Edition) is a 65-item rating scale quantifying autism-related social impairment from age 2 and a half through adulthood, completed by parents, teachers, and adults about themselves. Psychologists use it as a corroborating measure in autism evaluations and a progress measure in ABA programs. This page covers how to write up SRS-2 results, with a fictional sample and a results-section template. View template → ### STOP-Bang Score Documentation The STOP-Bang is an eight-item obstructive sleep apnea risk screen published by Frances Chung and colleagues at Toronto's University Health Network in 2008, scored 0 to 8 with one point per positive domain. Anesthesia, primary care, sleep, and occupational teams use it to stratify OSA risk before surgery and referral. It screens; it never diagnoses. This page covers how to document and interpret STOP-Bang results, with a fictional sample note. View template → ### Stroop Test Score Documentation The Stroop test is a family of timed color-word interference measures, not one instrument: Golden, Victoria, Comalli and Kaplan, Trenerry SNST, and the D-KEFS Color-Word Interference Test each use different scores and norms. Neuropsychologists and psychologists use them to assess performance under response conflict. A score is uninterpretable without its version and norm source. This page covers how to document Stroop results, with a fictional sample. View template → ### T.O.V.A. Documentation The T.O.V.A. (Test of Variables of Attention) is an FDA-cleared computerized continuous performance test measuring sustained attention and response inhibition across a 21.6-minute go/no-go task. ADHD evaluators use it as an aid to assessment, never as a standalone diagnostic. Documentation must name the version, conditions, and validity before scores. This page covers how to document T.O.V.A. results, with a fictional sample. View template → ### Trail Making Test Score Documentation The Trail Making Test is a timed paper-and-pencil measure of visual search, sequencing, and set shifting: Part A connects numbered circles, Part B alternates numbers and letters. Neuropsychologists, psychologists, and occupational therapists use it in cognitive batteries and driving screens. Raw seconds mean nothing without a named norm system. This page covers how to document and interpret Trail Making results, with a fictional sample note. View template → ### Vanderbilt ADHD Scale Documentation The NICHQ Vanderbilt Assessment Scales are parent- and teacher-completed ADHD rating scales published by NICHQ with the American Academy of Pediatrics in 2002, with initial and follow-up forms and a third edition in 2019. Pediatric and primary care clinicians use them to gather cross-setting symptom, impairment, and follow-up data. They support a diagnosis; they never make one. This page covers how to document Vanderbilt results, with a fictional sample note. View template → ### Vineland-3 Report Write-Up The Vineland-3 (Vineland Adaptive Behavior Scales, Third Edition) is a norm-referenced measure of adaptive behavior: the communication, daily living, and social skills a person uses to function independently, from birth through adulthood, gathered by interview or by caregiver and teacher rating forms. Psychologists and developmental teams use it in autism, intellectual-disability, and disability evaluations. This page covers how to write up Vineland-3 results, with a fictional sample and a template. View template → ### WAIS-5 / WAIS-IV Report Write-Up The Wechsler Adult Intelligence Scale (WAIS) is the standard adult intelligence battery for ages 16 to 90: the WAIS-5 (2024) is the current edition, and the WAIS-IV (2008) remains in use during the transition. Psychologists and neuropsychologists use it in diagnostic, disability, accommodation, and forensic evaluations. This page covers how to write up WAIS results, with a fictional sample and the errors reviewers flag. View template → ### WHODAS 2.0 Score Documentation The WHODAS 2.0 (World Health Organization Disability Assessment Schedule 2.0) is WHO's ICF-based measure of functioning and disability over the past 30 days, in 36-item and 12-item versions, self-, proxy-, or interviewer-administered. Psychiatrists, psychologists, and disability evaluators use it as the DSM-5 Section III functioning measure and as evidence in SSA, VA, 1915(i), and NDIS files. This page covers how to document WHODAS 2.0 results, with a fictional sample note. View template → ### WIAT-4 Report Write-Up The WIAT-4 (Wechsler Individual Achievement Test, Fourth Edition) is an individually administered academic achievement battery for ages 4 through 50, measuring reading, mathematics, written expression, and oral language with standard scores set to a mean of 100. Psychologists, school psychologists, and specific-learning-disability evaluation teams use it to quantify academic skills against age or grade norms. This page covers how to write up WIAT-4 results, with a fictional sample and a results-section template. View template → ### WISC-V Report Write-Up The WISC-V (Wechsler Intelligence Scale for Children, Fifth Edition) is an individually administered intelligence test for ages 6 to 16, yielding a Full Scale IQ and five index scores covering verbal comprehension, visual spatial, fluid reasoning, working memory, and processing speed. Psychologists use it in special-education, learning-disability, and disability-benefit evaluations. This page covers how to write up WISC-V results, with a fictional sample and a results-section template. View template → ### Woodcock-Johnson (WJ IV / WJ V) Report Write-Up The Woodcock-Johnson is an individually administered, norm-referenced battery of cognitive and achievement tests built on Cattell-Horn-Carroll theory, currently sold in two editions: the WJ IV (2014) and the digital WJ V (2025). Psychologists, school psychologists, and educational diagnosticians use it in special-education, learning-disability, gifted, and accommodation evaluations. This page covers how to write up Woodcock-Johnson results, with a fictional sample and a results-section template. View template → ### WPPSI-IV Report Write-Up The WPPSI-IV (Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition) is an individually administered intelligence test for ages 2 years 6 months to 7 years 7 months, published in 2012. Psychologists use it for preschool special-education eligibility, kindergarten and gifted decisions, and the cognitive core of early-childhood evaluations. This page covers how to write up WPPSI-IV results, with a fictional sample and a results-section template. View template → ### Y-BOCS Score Documentation The Y-BOCS (Yale-Brown Obsessive Compulsive Scale) is a clinician-rated measure of OCD severity published by Goodman and colleagues in 1989, scored 0 to 40 across obsession and compulsion subtotals, with a 0 to 50 second edition and child versions. OCD specialists, therapists, and prescribers use it for baselines and treatment tracking. It rates severity; it never diagnoses. This page covers how to document Y-BOCS results, with a fictional sample note. View template → ## Clinical Scales & Scores Bedside scales and scores that nursing, critical care, obstetric, wound care, and home health teams document every shift: Glasgow Coma Scale, Braden, Apgar, RASS, Morse, NEWS2, CAM-ICU, pressure injury staging, Wong-Baker FACES, and OASIS. ### Apgar Score Documentation The Apgar score is a five-component description of a newborn's condition at 1 and 5 minutes after birth (heart rate, respiratory effort, muscle tone, reflex irritability, and color), devised by Virginia Apgar in 1953. Labor and delivery nurses, midwives, obstetricians, and neonatal teams use it to record the newborn's transition and response to resuscitation. This page covers how to document Apgar scores with the support in progress, with a fictional sample. View template → ### Bishop Score Documentation The Bishop score is a five-component summary of a digital cervical examination (dilation, effacement, station, consistency, and position) proposed by Edward Bishop in 1964 to judge how ready the cervix is for induction. Obstetricians, midwives, and labor and delivery nurses record it before and during cervical ripening and induction of labor. This page covers how to document a Bishop score with its version, examiner, indication, and plan, with a fictional sample. View template → ### Braden Scale Documentation The Braden Scale for Predicting Pressure Sore Risk is a copyrighted nursing instrument (Bergstrom, Braden, Laguzza, and Holman, 1987) that rates six subscales for a total of 6 to 23, lower meaning higher pressure injury risk. Bedside, long-term-care, and home health nurses use it to decide who needs prevention measures and which ones. This page covers how to document the score and the interventions it triggers, with a fictional sample. View template → ### CAM-ICU Documentation The CAM-ICU (Confusion Assessment Method for the Intensive Care Unit) is a bedside delirium screen, adapted by Ely and colleagues in 2001 from Inouye's Confusion Assessment Method, that assesses four features once a RASS shows the patient is arousable. ICU and step-down nurses use it every shift to flag delirium for the provider who diagnoses it. This page covers how to document CAM-ICU results, with a fictional sample. View template → ### Glasgow Coma Scale (GCS) Documentation The Glasgow Coma Scale (GCS) is a bedside scale of impaired consciousness published by Teasdale and Jennett in 1974 that rates eye opening, verbal response, and motor response as three separate components summing to 3 to 15. Emergency, trauma, critical care, and neurosurgical nurses, paramedics, and physicians use it to describe a patient's level of response and to track change over time. This page covers how to document GCS findings, with a fictional sample. View template → ### MEWS & NEWS2 Early Warning Score Documentation MEWS (Modified Early Warning Score, Subbe and colleagues, 2001) and NEWS2 (National Early Warning Score 2, Royal College of Physicians, 2017) are aggregate track-and-trigger scores that turn a set of vital signs into a total and an escalation prompt. Ward and step-down nurses, rapid response teams, and hospitalists use them to recognize deterioration and summon help. This page covers how to document the score and the escalation it triggered, with a fictional sample. View template → ### Morse Fall Scale Documentation The Morse Fall Scale is a nurse-rated screen for a patient's risk of falling, published by Janice Morse and colleagues in 1989, that scores six items for a total of 0 to 125, higher meaning higher risk. Hospital, rehabilitation, long-term care, and home health nurses use it to decide who needs which fall prevention measures. This page covers how to document the score and the interventions it triggers, with a fictional sample. View template → ### OASIS Assessment Documentation The OASIS (Outcome and Assessment Information Set) is the assessment item set CMS requires of Medicare home health agencies, in force as OASIS-E2 since April 1, 2026 and collected at start of care, resumption, recertification, transfer, and discharge. Home health nurses and therapists complete it, and its responses drive PDGM payment, star ratings, HHVBP, and survey findings. This page covers how to document a defensible OASIS, with a fictional sample. View template → ### Pressure Injury Staging Documentation Pressure injury staging classifies a pressure injury by the deepest tissue exposed or damaged, using the NPIAP system of four numbered stages plus unstageable and deep tissue pressure injury, last revised in 2016. Wound care and bedside nurses stage the wound, the provider documents the diagnosis, and coders, surveyors, and payers read both entries against each other. This page covers how to document the stage and its present-on-admission status, with a fictional sample. View template → ### RASS Score Documentation The Richmond Agitation-Sedation Scale (RASS) is a ten-level observational scale, published by Sessler and colleagues in 2002, that places a patient on one continuum from deep sedation through calm alertness to severe agitation. ICU and step-down nurses, intensivists, respiratory therapists, and pharmacists use it to set and check sedation targets and to gate delirium screening. This page covers how to document RASS assessments, with a fictional sample. View template → ### Wong-Baker FACES Pain Scale Documentation The Wong-Baker FACES Pain Rating Scale is a self-report pain intensity tool on which a patient aged three or older chooses the face that matches their pain, recorded on a 0 to 10 metric. Pediatric, emergency, perioperative, and medical-surgical nurses use it when a patient cannot use a numeric scale; it is never rated by an observer. This page covers how to document a FACES rating, with a fictional sample. View template → ## Care Coordination & Transitions Referral letters, discharge and transfer summaries, and the notes that document coordinated care across providers and teams. ### Australia Telehealth & MBS Case Conference Note An MBS case conference note is the clinical record entry an Australian practitioner makes to substantiate a multidisciplinary or mental health case conference item under the Medicare Benefits Schedule. The regulations legislate recorded facts, not a form: the patient's agreement, the day with start and end times, participant names, and the matters discussed. Telehealth attendance notes add a documented eligibility pathway. Most run 150 to 400 words. View template → ### Care Coordination Note A care coordination note documents the work of organizing a client's care across providers: contacts made, referrals and their status, consent to share information, handoffs, and outcomes. Therapists, case managers, and integrated care teams write one whenever they arrange, share, or follow up on care with someone outside the session. Most run 75 to 250 words. View template → ### Case Conference Note A case conference note documents a pre-arranged meeting held by two or more providers to discuss and direct one identified client's care. It records the participants, times, decisions, and allocated tasks, plus whatever elements the payer requires when the conference is billed. Conferences are billable in Australia and Ontario and in limited US contexts. Most notes run 150 to 400 words. View template → ### Case Management Note A case management note documents one case management contact: work that helps a client gain access to medical, social, educational, and other services. Case managers, social workers, and community agency staff write one for every contact, including phone calls and collateral contacts, because the note is the evidence behind billed time. Most case management notes run 100 to 300 words. View template → ### Chronic Care Management (CCM) Note A chronic care management (CCM) note is the monthly documentation behind Medicare's CCM billing: a consent record, a comprehensive electronic care plan, and a time log showing at least 20 minutes of clinical staff care coordination for a patient with two or more chronic conditions. Care managers in primary care and integrated behavioral health teams keep it every enrolled month. Most months the record runs 150 to 400 words. View template → ### Collaborative Care (BHI/CoCM) Note A collaborative care note is the registry-backed record a behavioral health care manager keeps for each patient in the psychiatric Collaborative Care Model (CoCM) or general behavioral health integration (BHI). It logs contacts, validated symptom scores, psychiatric consultant recommendations, and clinical minutes across the calendar month. The billing practitioner's monthly claim under 99492, 99493, 99494, or 99484 stands on that time-and-registry trail. View template → ### Consultation Response & Report A consultation response, or consultation report, is the written reply a consultant sends to the provider who requested the consultation: findings, diagnostic impression, recommendations, and who does what next. It closes the referral loop the referral letter opened. Psychiatrists, psychologists, and therapists send one after an assessment or a completed course of treatment. Most run 150 to 700 words. View template → ### Discharge Summary A mental health discharge summary is the closing document of an episode of care: why treatment started, the course and response, final diagnoses, the client's condition and risk status at discharge, and the aftercare plan. Hospitals and community mental health centers must produce one under federal rules; in outpatient therapy it is professional convention, with some state exceptions. A typical summary runs 300 to 1,200 words. View template → ### Hospital Consult & Liaison Note A consultation-liaison (C-L) note records a psychiatric assessment of a hospital patient admitted under a non-psychiatric team, performed at that team's request. It answers the consult question, documents mental status, capacity, and risk, and returns numbered recommendations the requesting team can act on the same day. Most run 200 to 600 words. View template → ### Medication Management Follow-Up Note A medication management follow-up note is a prescriber's record of an established-patient psychiatric visit centered on medications: response, adherence, side effects, monitoring, and plan changes. Psychiatrists, psychiatric nurse practitioners, and physician assistants write one at every med check, billed in the US as an E/M visit (99212 to 99215). Most run 150 to 500 words. View template → ### Multidisciplinary Team Meeting Note A multidisciplinary team meeting note records an interdisciplinary discussion of one client's care: who attended, what each discipline reported, what the team decided and why, and the action items with owners and due dates. Treatment teams in inpatient, rehabilitation, hospice, and community behavioral health programs write one after each team meeting or staffing. Most run 100 to 400 words per client discussed. View template → ### Psychiatric Consultation Note A psychiatric consultation note records a psychiatrist's or psychiatric nurse practitioner's opinion on another provider's patient: the consult question, a focused history, mental status findings, an impression that answers the question, and numbered recommendations back to the referrer. Consultants write one after each referred evaluation in outpatient, primary care, and specialty settings. Most run 250 to 800 words. View template → ### Referral Letter A mental health referral letter is a signed, dated request from one clinician to another to assess or treat a named client. It states the referral question, relevant history, current medications, symptoms and measures, risk status, and what the referrer keeps doing while care continues. GPs, therapists, psychologists, and psychiatrists exchange them whenever care crosses providers. Most run 150 to 500 words. View template → ### Return-to-Primary-Care Summary A return-to-primary-care summary is the closing letter that transfers a stabilized behavioral health patient from specialty or integrated care back to their primary care clinician. It states diagnoses, treatment delivered, current outcome measures, medications and who prescribes them, a relapse prevention plan, and re-referral triggers. Psychologists, psychiatrists, and collaborative care teams send it at step-down; most run 300 to 600 words. View template → ### Transfer-of-Care Summary A transfer-of-care summary is the document a clinician writes when ongoing care moves to a new provider, giving the receiving clinician the client's current status, active risks, medications and supports, and pending items. Therapists, psychologists, counselors, and care coordinators write one whenever another clinician assumes the client. Most run 150 to 600 words, shorter than a discharge summary. View template → ### Transitional Care Management (TCM) Note A transitional care management (TCM) note is the medical-record documentation behind Medicare's 30-day post-discharge service, billed under CPT 99495 or 99496. It evidences the three billable components: interactive contact within two business days of discharge, a face-to-face visit within 14 days (99495) or 7 days (99496), and moderate or high complexity medical decision making. Practices keep one record per qualifying discharge; most run 250 to 600 words. View template → ## Consent & Privacy Forms Consent, authorization, and records-request documents that keep client information disclosures compliant. ### 42 CFR Part 2 Consent A 42 CFR Part 2 consent is the written patient authorization, required by federal regulation, that lets a substance use disorder treatment program share a patient's SUD records. Clinicians and records staff in federally assisted SUD programs complete it before any consent-based disclosure. Since February 16, 2026, one consent can cover all treatment, payment, and operations uses until the patient revokes it in writing. View template → ### Consent for Treatment of a Minor A consent for treatment of a minor is the record showing who authorized a child's mental health care and on what legal basis: parent, guardian, custody order, or the minor under a state statute. Therapists complete it before the first session with any client under the age of majority. No federal law prescribes the form; state law governs who may consent. View template → ### HIPAA Notice of Privacy Practices Acknowledgment A Notice of Privacy Practices acknowledgment is the intake form that documents a patient received a provider's HIPAA privacy notice. Front desk staff collect it at or before first service delivery. Federal law requires a good-faith effort to obtain it, not a patient signature, and refusal to sign never blocks treatment. Most forms run 100 to 250 words. View template → ### Informed Consent for Treatment An informed consent for treatment form is the intake document where a therapist explains the nature of therapy, its risks, benefits, and alternatives, fees, and the limits of confidentiality, and the client agrees to treatment, usually by signature. Therapists, counselors, psychologists, and social workers review it before care begins. Most forms run 800 to 2,000 words in plain language. View template → ### Psychotherapy Notes Authorization A psychotherapy notes authorization is the standalone HIPAA authorization required before a practice may use or disclose psychotherapy notes, the private process notes a clinician keeps separate from the medical record. Under 45 CFR 164.508(a)(2) it cannot be combined with any other release. Records staff and clinicians use it whenever a third party requests these notes. Most forms run one page. View template → ### Record Amendment & Correction Response A record amendment and correction response is the set of documents a practice creates when a client asks to change something in their health record: the decision letter, the appended amendment or the denial with its required elements, and the request log. Under HIPAA, US practices must act within 60 days of receipt (45 CFR 164.526). A typical decision letter runs 150 to 400 words. View template → ### Records Request & Chart Copy Packet A records request and chart copy packet is the set of documents a mental health practice produces when a client asks to inspect or copy their record: a cover letter, an itemized fee notice, a denial letter when one applies, and an internal log. US practices must act within 30 calendar days under 45 CFR 164.524. A typical packet runs 300 to 700 words. View template → ### Release of Information (ROI) Authorization A release of information (ROI) is the written authorization a client signs before a mental health practice may disclose their records to a third party. HIPAA does not publish an official form; it mandates the content: six core elements and three required statements (45 CFR 164.508). Records staff use one whenever a disclosure is not already permitted by law. Most forms run one to two pages. View template → ### Telehealth Consent A telehealth consent form documents that a client agreed to receive care by video or phone after learning the modality's risks, privacy limits, emergency procedures, and technology-failure backup plan. Clinicians obtain and document it before or at the first virtual session. Most US states require some form of telehealth consent by statute, regulation, or Medicaid policy, and documented verbal consent usually satisfies them. View template → ## Billing, Insurance & Practice Operations Claim-support notes, prior authorizations, appeals, and the financial and operational documents that keep a practice running. ### Advance Beneficiary Notice (ABN) An Advance Beneficiary Notice (ABN), Form CMS-R-131, is the standardized written notice that tells an Original Medicare patient, before a specific service, that Medicare is expected to deny payment, so the patient can choose whether to proceed and accept the cost. Practice and billing staff issue it when a case-specific denial is expected, such as psychotherapy beyond covered frequency. The completed notice is one page. View template → ### Appointment No-Show & Outreach Note An appointment no-show note is a brief clinical-record entry documenting that a client missed a scheduled session or cancelled late, along with any outreach attempts and a risk-informed next step. Front desk staff record the missed appointment; the clinician documents outreach and any risk decision. Most run 75 to 200 words. View template → ### Claim-Support Billing Note A claim-support billing note is the documentation that ties a billed CPT or HCPCS code to what happened in session: the service, the time, the diagnosis it addressed, medical necessity, and an authenticated author. United States payers require these elements in the clinical record rather than a separate document. Clinicians use the template as a pre-claim check, a records-request response, or a compliant addendum. Most run 100 to 300 words. View template → ### Communication Log & Portal Message Note A communication log or portal message note documents a client contact that happens outside a scheduled session: a patient portal message, phone call, text, or secure email. Therapists, psychologists, and practice teams use it to record what was communicated, their clinical read of it, and the follow-up. Most entries run 50 to 200 words; a billed e-visit entry runs longer. View template → ### Diagnosis & Problem List Update (ICD) A diagnosis and problem list update is the dated, authored entry a clinician makes when adding, refining, correcting, or resolving a diagnosis in the medical record, including the ICD-10-CM code that flows onto claims. Clinicians, coders, and CDI teams use it whenever the active diagnosis list changes. Most entries run 50 to 200 words: a diagnostic statement plus the clinical support for it. View template → ### Financial Policy & Fee Agreement A financial policy and fee agreement is the signed contract between a therapy practice and a client that sets fees, billing and insurance terms, cancellation and no-show charges, and collection rules before treatment starts. Practices present it at intake alongside informed consent. No US statute requires one; ethics codes and payer contracts define what it must say. Most run one to three pages. View template → ### Good Faith Estimate (No Surprises Act) A good faith estimate (GFE) is the written notice of expected charges that the No Surprises Act requires US providers to give uninsured and self-pay clients, including therapy clients who pay out of pocket. It itemizes expected services, codes, and charges for up to 12 months of care. Required since January 1, 2022 under 45 CFR 149.610; most therapy GFEs run one to two pages. View template → ### Incident & Adverse Event Report An incident or adverse event report is an internal risk-management record of an unexpected event affecting a client, visitor, or staff member: a fall, medication error, elopement, assault, or suicide attempt. Staff file it with the practice or facility, separate from the clinical record. Facility and program rules often mandate it; for solo private practice it is convention. Most run 300 to 600 words. View template → ### Insurance Appeal Letter An insurance appeal letter is a written request asking a health plan to reverse an adverse benefit determination: a denied, reduced, or terminated mental health claim. Clinicians, billing staff, patients, or authorized representatives file it within the plan's appeal deadline, pairing the denial code with a clinical rebuttal. No law prescribes its format; most run 1 to 3 pages. View template → ### Medical Necessity Letter A medical necessity letter (also called a letter of medical necessity or LMN) is a clinician-authored narrative explaining why a specific service, level of care, or item is medically necessary for a particular patient. Therapists, psychologists, and psychiatrists send one to payers, FSA and HSA administrators, schools, and benefit programs to support coverage, reimbursement, or accommodations. Most run 1 to 3 pages. View template → ### Patient Instructions & After-Visit Summary An after-visit summary (AVS), also called a clinical summary or patient instructions, is a plain-language handout a client takes home after an appointment. It covers what happened in the visit, medications, the between-session plan, follow-up, and who to call. No US, Canadian, or Australian law requires one for outpatient therapy. Most run 150 to 400 words, about half a page. View template → ### Peer-to-Peer Review Note A peer-to-peer review note is the treating clinician's record of a call with a payer's medical or behavioral health reviewer about the medical necessity of requested services. Clinicians write it right after a payer P2P call, most often around a prior authorization or concurrent review denial. Most run 150 to 300 words and are filed in the general chart, not psychotherapy notes. View template → ### Prior Authorization Request A prior authorization request is the clinical justification packet a provider submits to a health plan to get a mental health service approved for coverage before it starts or continues. Practice staff or the clinician assemble it from the record: diagnosis, symptom severity, functional impairment, treatment history, risk, and a plan with measurable goals. No universal form exists; each payer or carve-out vendor sets its own. View template → ### Prior Authorization Support Letter A prior authorization support letter is a narrative letter a treating clinician sends to a health plan to justify why a mental health service should be authorized before it is delivered, or continued. It states the diagnosis, symptoms, functional impairment, treatment history, and the exact service requested, mapped to the plan's medical-necessity criteria. Many plans do not require it for routine outpatient therapy. View template → ### Start-Stop Time Attestation A start-stop time attestation is the notation in a psychotherapy progress note that records when the session began and ended, or the total face-to-face minutes, to support time-based codes such as 90832, 90834, and 90837. Clinicians add it to every timed session; payers and auditors use it to verify the billed code. Most versions run one to two lines inside the note. View template → ### Utilization Review Summary A utilization review summary is the clinical document a treating clinician or facility submits to a payer to justify continuing an authorized course of treatment or level of care. Payers call the process concurrent or continued-stay review. It compiles current symptoms, objective measures, risk status, treatment response, and the barriers to step-down. Most run 300 to 600 words or fill a payer portal form. View template → ## Work, School, Disability & Legal Letters and reports clinicians write for employers, schools, disability programs, and courts. ### Court Report & Treatment Progress Update A court report or treatment progress update is a limited report a treating clinician sends to a court, probation officer, or attorney with a valid legal basis, summarizing attendance, engagement, progress toward treatment goals, and recommendations for continuing care. It stays within the treatment role and is not a forensic evaluation. Most run 1 to 3 pages, about 400 to 900 words. View template → ### Disability Evaluation & Medical Source Statement A medical source statement is a treating clinician's written opinion describing what a patient can still do despite mental impairment, rated across four work-related domains: understanding and memory, concentration and pace, social interaction, and adaptation. Clinicians complete one when a disability claim, most often Social Security, needs functional evidence from the treating source. Most combine a structured rating form with narrative comments, 2 to 6 pages. View template → ### FMLA & Leave Certification Support Note An FMLA certification support note is the documentation a mental health clinician prepares when an employer requests medical certification of a serious health condition: the completed WH-380-E entries plus the chart note that backs them. Clinicians treating employees of covered US employers write it within the employee's 15-day return window. Form entries plus the supporting note typically run 150 to 400 words. View template → ### IEP Psychoeducational Input IEP psychoeducational input is a treating psychologist's or other outside clinician's written report to a school team: findings, educationally relevant functional effects, and recommendations offered for the IEP process. Clinicians send it, with a signed authorization, before evaluations, eligibility determinations, and IEP reviews. A short input letter runs 300 to 900 words; a full private evaluation runs far longer. View template → ### Return-to-School Letter A return-to-school letter is a short note from a treating mental health clinician telling a school when and how a student can resume attendance after a mental health absence or hospitalization. It states the return date, attendance plan, functional limitations, recommended supports, and follow-up, with the minimum detail needed. Families, schools, and clinicians use it during re-entry. Most run under one page. View template → ### Return-to-Work & Accommodation Letter A return-to-work letter is a short, signed statement from a treating clinician telling an employer when an employee can resume work and under what functional restrictions or accommodations. Therapists, psychologists, and social workers write them after a mental health leave or to support a workplace accommodation request. Most run under one page, and the strongest ones describe function, not diagnosis. View template → ### Section 504 Clinician Statement A Section 504 clinician statement is a letter from a treating clinician that describes a student's health condition, its functional impact on learning, and accommodation suggestions for the school's Section 504 team. The team considers it as one source of evidence when deciding eligibility and accommodations under the Rehabilitation Act. Most run 300 to 600 words, about one page. View template → ### Subpoena Response & Records Affidavit A subpoena response and records affidavit is the packet a therapy practice assembles when legal process demands client records: an intake log, a written objection or privilege assertion where grounds exist, a custodian affidavit authenticating any copies produced, and a production letter. Records custodians and clinicians use it, with counsel, whenever a subpoena arrives without a court order or client authorization. A typical packet runs 400 to 900 words. View template → ### Workers' Compensation Initial Report A workers' compensation initial report is the first claims-facing report a treating clinician sends after evaluating a claimed work-related mental health condition. It identifies the condition, records the work event, addresses work-relatedness, states functional impact and work status, and proposes treatment for the claims administrator deciding the claim. Every state and scheme names its own form. Most run 600 to 1,500 words. View template → ### Workers' Compensation Progress Report A workers' compensation progress report is a periodic report from the treating clinician to the claims administrator, insurer, or state board during an active work-injury claim. It documents the accepted condition, interval symptoms and objective findings, response to treatment, current work restrictions, and return-to-work outlook. California standardizes it as Form PR-2; most run 300 to 700 words. View template → ## Frequently asked questions What note formats does BastionGPT support? BastionGPT is trained, tuned, and clinically tested across 100+ behavioral health document types: SOAP, DAP, BIRP, PIE, and narrative progress notes, intake and diagnostic evaluations, biopsychosocial assessments, treatment plans, risk assessments and safety plans, psychological testing reports, discharge summaries, referral and medical-necessity letters, prior authorizations and appeals, and consent and privacy forms. A custom note option also accepts any free-form format you describe. Are these templates free to use? Yes. Copy and adapt any template or sample for your own practice, no signup required. The templates are educational resources; follow your organization, board, payer, and jurisdiction requirements. Are the sample notes real patient records? No. Every sample in the library is fictional and written for education. No real patient information appears anywhere in these templates. Can BastionGPT convert a note from one format to another? Yes. BastionGPT converts between formats, such as DAP to SOAP or a narrative note to a payer-facing progress note, without losing clinically important content. BastionGPT drafts, you review and sign. Do these templates work in Canada and Australia? Yes. The formats themselves are conventions used in all three countries, and key templates include a US, Canada, and Australia table covering naming and billing differences. BastionGPT supports HIPAA, PIPEDA, and Australian APP requirements. Stop writing notes from scratch BastionGPT drafts complete, compliant notes from bullets, dictation, or a session transcript. Many users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. Start your free trial How we clinically test Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/clinical-validation Clinical validation # How BastionGPT Is Clinically Tested Every note and document type BastionGPT supports goes through a structured validation program before we describe it as trained, tuned, and clinically tested. This page explains what that program covers, how each document type is tested, and where human judgment stays in charge. 99.3% Sentence-level accuracy on clinical documentation tasks in a peer-reviewed study by Baylor College of Medicine and Texas Children's Hospital researchers. ## What we validate BastionGPT's validation catalog covers 100+ general and behavioral health document types across the workflows clinicians actually run: session and progress notes (SOAP, DAP, BIRP, and more), intake and diagnostic assessments, risk and safety documentation, treatment plans, psychological testing reports, speech-language evaluation reports, the bedside clinical scales and scores that nursing, critical care, obstetric, wound care, and home health teams document (Glasgow Coma Scale, Braden, Apgar, RASS, Morse, NEWS2, CAM-ICU, pressure injury staging, Wong-Baker FACES, and OASIS), care coordination and discharge records, consent and privacy forms, and the billing, school, disability, and legal documents practices are asked to produce. Coverage spans United States, Canadian, and Australian documentation expectations. You can explore these formats in our free note and document template library , which pairs each template with a section-by-section structure guide and a complete sample note. ## How we test each document type Each type is exercised across five task classes, matched to how that document is used in practice: Drafting Generate a complete note or report from bullet points, a dictation, or a session transcript, at the length and structure the document type calls for. Transformation Convert between formats, such as DAP to SOAP or a narrative note to a payer-facing progress note, without losing clinically important content. Extraction Pull structured facts such as dates, diagnoses, scores, risk status, and follow-ups into fields exactly. A wrong name or date in a record is never a small error. Privacy and segmentation Redact protected information correctly, keep psychotherapy notes separate from the standard record, and keep each person's content clean in group and family documentation. Administrative fit Support billing documentation accurately, rewrite for patient, school, insurer, or legal audiences, and flag tasks that need a human decision instead of guessing. ## What we measure Each document type is scored against the failure modes that matter for that type rather than a single generic quality score: Accuracy. Names, dates, scores, diagnoses, and service facts must match the source exactly. Completeness. Every required section appears, with no missing risk or follow-up fields. Clinical safety. Missed or invented risk indicators are treated as automatic failures. Privacy. Redaction correctness, psychotherapy-note segregation, and release-of-information scope. Billing support. Service, time, diagnosis, and medical-necessity linkage without unsupported codes. Legal defensibility. Neutral tone, clear chronology, and appropriate expression of uncertainty. Interoperability. Output keeps structure that maps cleanly into EHR templates and exchange formats. Cultural responsiveness. The client's own explanatory model and context are preserved, in audience-appropriate language. ## Stricter gates for high-risk documents Suicide and violence risk assessments, safety plans, mandated reports, and legal documents are held to our strictest acceptance gates. Fabricated or omitted risk content is an automatic failure and legal documents are tested for scope discipline: neutral language, accurate chronology, and clear boundaries around what the record can support. ## Human review by design BastionGPT is built as documentation support for licensed professionals. It drafts, converts, extracts, and quality-checks; you review and sign. That design mirrors what record-keeping guidance and payer rules actually reward: accuracy, completeness, privacy discipline, and supportable reasoning, with clinical judgment as the final step. Validation results and the clinical content we publish are reviewed by the BastionGPT Clinical Advisory Board . Put it to work on your own notes Try BastionGPT free for 7 days. HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. Start your free trial Read the security overview Updated July 2026 --- > Source: https://bastiongpt.com/template/birp-note All note templates # BIRP Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A BIRP note is a four-part progress note format that organizes a session into Behavior, Intervention, Response, and Plan. Its signature strength is forcing the intervention-and-response thread onto the page, which is exactly what payer reviewers read for. A typical BIRP note runs 150 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, case managers, community and public behavioral health teams Audience Treating clinician, care team, supervisors, payers, auditors Typical length 150 to 300 words · 10 to 20 minutes by hand (clinical team estimate) Format family Structured progress note (compare: SOAP, DAP; GIRP and SIRP are sibling variants) When it's used After each individual, group, or case-management contact Standards context A documentation convention, not a mandated form; California's Medi-Cal policy is now explicitly format-neutral On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a BIRP note? BIRP stands for Behavior, Intervention, Response, Plan. It is a four-part progress note format built around a simple thread: what you observed, what you did, how the client responded, and what happens next. Unlike SOAP, no primary source we reviewed identifies a creator or standards body for BIRP; it grew up as a workflow convention inside community and public behavioral health systems, alongside sibling variants like GIRP (Goal first) and SIRP (Situation first). Australian services often spell it out as Behaviour, Intervention, Response, and Plan. Notably, systems that once mandated these formats are moving away from doing so: California's Medi-Cal behavioral health policy is now explicitly format-neutral , and county guidance under CalAIM states the old BIRP/GIRP/SIRP/SOAP format choices are no longer required. The boundary that actually matters is legal, not structural. A working BIRP note contains progress, plan content, and often session timing, so under HIPAA it is a progress note in the standard clinical record, not a separately protected " psychotherapy note ": that narrow category excludes diagnosis, treatment plan, symptoms, prognosis, progress, and counseling start and stop times, and calling a chart note a psychotherapy note does not change its status (see the psychotherapy notes authorization page). ## Who uses BIRP notes and when BIRP's home turf is community mental health, public behavioral health systems, substance-use programs, and case management, settings where many hands touch a chart and reviewers need the intervention-and-response loop visible at a glance. It also serves solo clinicians who want their notes to read the way auditors read: Medicare's psychotherapy coverage policy asks for the interventions used, the patient's participation and reaction, and progress toward goal-oriented outcomes, which is BIRP's exact skeleton. Prefer SOAP in medically integrated settings, DAP when a single merged Data section fits a qualitative flow, and GIRP when your program organizes notes around the goal rather than the behavior. No payer or regulator we reviewed requires any of these by name. ## How to write a BIRP note in four steps Behavior: what you observed and what the client reported: presentation, mental-status highlights, measurable scores, and changes since last session. Intervention: the specific techniques you used this session, tied to a numbered treatment-plan goal. Response: how the client participated, reacted to the interventions, and what changed or did not change. Plan: concrete next steps: interventions for coming sessions, homework, referrals, next appointment, and any plan changes with the rationale. Each section in more detail, with the pitfall that most often undermines it: B: Behavior. Observables first: appearance, affect, engagement, quotes, screening scores (PHQ-9, GAD-7, PCL-5), plus the client's report. Pitfall: opinions written as observations; "manipulative" is a judgment, "asked three times to extend the session" is a behavior. I: Intervention. Name what you actually did: cognitive restructuring, exposure work, motivational interviewing, skills training, and tie it to the goal it serves. Pitfall: "supportive psychotherapy given"; Medicare's psychotherapy coverage policy says that phrase alone is not adequate documentation. R: Response. The section reviewers read first: participation, reaction to each intervention, and symptom or behavior change, or the lack of it. Pitfall: skipping response entirely, which turns the note into a record that something happened rather than evidence it did anything. P: Plan. Specific and dated: next interventions, homework, coordination, next appointment, and the rationale for any change to the treatment plan. Pitfall: a bare "continue treatment" with no thread back to the goal. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: [type/length] Start/stop or total time: B (Behavior): I (Intervention): R (Response): P (Plan): Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample BIRP note Scenario: adult client, PTSD, cognitive processing therapy, session 7 of a planned 12, community mental health clinic. All details are fictional. Client: J.M., 37 · Date: 07/09/2026 · Session: 7 · Service: Individual psychotherapy, 53 min, in office · Start/stop: 15:03 to 15:56 B: On time, casually dressed, cooperative. Affect constricted early in session, more expressive by the close. Reported nightmares twice this week, down from four, and one avoided grocery trip completed with moderate anxiety. PCL-5 today: 38 (down from 52 at intake). Speech normal rate; no psychomotor agitation. Denied suicidal ideation and self-harm urges. I: Cognitive processing therapy, session 7 content: reviewed the completed challenging-questions worksheet on the stuck point "I should have stopped it"; introduced patterns-of-problematic-thinking worksheet; Socratic questioning targeting self-blame; assigned one written account review. All interventions serve Goal 1 (reduce trauma-related self-blame and avoidance). R: Client completed the worksheet fully for the first time and generated two alternative statements with minimal prompting, stating "maybe there wasn't a right move that night." Became tearful during the account discussion but stayed engaged and used grounding without cueing. Avoidance decreasing (grocery trip completed); nightmare frequency halved. No change yet in sleep-onset difficulty. P: Continue weekly CPT; next session: patterns-of-problematic-thinking review and second written account. Homework: daily worksheet practice, one additional avoided-situation task. Re-administer PCL-5 at session 9. Coordinate with prescriber regarding ongoing sleep-onset difficulty. Next appointment 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The note stands alone: a reviewer can follow assessed need, goal, intervention, response, and next step without opening the rest of the chart, the "golden thread" state auditors teach. Behavior carries observables and a score (PCL-5 trend), not judgments. Interventions are named at protocol level (CPT session 7 content, specific worksheets) and tied to a numbered treatment-plan goal. Response records participation, reaction, and change and no-change honestly, which is what Medicare's coverage policy asks progress documentation to show. Start and stop times support the billed time-based code, and risk is addressed even though the client is low risk. Plan items are specific and dated, including cross-provider coordination the next note can evaluate. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A BIRP note is part of the designated record set: clients can request it, payers can audit it, and other providers may rely on it. Keep private process reflections in separately maintained psychotherapy notes with their own authorization rules . And do not rely on the BIRP label for payer acceptance one way or the other: the format is convention. California's Medi-Cal behavioral health policy now says it will not enforce format requirements beyond its stated content rules, a useful marker of where behavioral-health documentation policy is heading. What reviewers do enforce is content and authentication. Medicare's psychotherapy coverage policy wants the interventions described, the patient's participation and reaction, progress toward goal-oriented outcomes, and the rationale for plan changes, and it says a bare "supportive psychotherapy given" is not enough. For psychiatrists billing same-day E/M plus psychotherapy, CMS billing guidance requires the two services to be significant and separately identifiable, and E/M time cannot be counted as psychotherapy time. On signatures, Medicare's signature guidance is current enough to address AI directly: if AI or a scribe transcribes the entry, the clinician must still sign to authenticate it, and OIG audits have rejected pasted images of signatures. CMS's records rule adds a 7-year maintenance-and-access expectation for covered ordering and referral documentation. On retention, the rules differ even inside one country: HIPAA sets no patient-chart period ( state law governs ), Ontario's college expects at least 10 years while Québec's psychologist regulation sets at least 5 years from the last professional service, and Australia's Psychology Board code sets 7 years, or until age 25 for clients who were minors. Check your own jurisdiction before adopting any single number. ↑ Back to the template and downloads ## Common BIRP note errors auditors flag The audit record is blunt about how psychotherapy documentation fails. In a provider-level OIG audit , 111 of 120 sampled claims were noncompliant, with therapeutic maneuvers not specified, notes that did not support the services billed, no documented expectation of improvement, and, on 109 claims, notes "signed" with pasted images of signatures. Medicare's psychotherapy coverage policy singles out vague entries like "supportive psychotherapy given" as inadequate, and state behavioral-health training teaches that a progress note must stand alone and keep the golden thread from assessed need to next step. The BastionGPT Clinical Advisory Board sees the same patterns most often in BIRP note reviews: Intervention not specified. A modality name or "supportive therapy" with no actual technique a reviewer can identify. Response missing. No participation, reaction, or change/no-change statement, which severs the exact thread BIRP exists to show. The note does not stand alone. No link to the assessed need or a numbered treatment-plan goal, forcing a reviewer to reconstruct medical necessity from elsewhere in the chart. Authentication gaps. Unsigned or undated entries, pasted signature images, or AI-drafted notes never signed by the clinician. Time problems. Missing time support for time-based codes, or psychotherapy time not separated from a same-day E/M service. ### BIRP notes in the US, Canada, and Australia Aspect United States Canada Australia Status Convention; California's Medi-Cal policy is explicitly format-neutral, and Medicare regulates content, signatures, and access rather than headings Convention; provincial colleges make record content enforceable (Ontario CRPO; Québec regulation for psychologists) Convention; Psychology Board Code of Conduct (from 1 Dec 2025) and Ahpra expectations govern record quality Terminology BIRP note, BIRP progress note; GIRP and SIRP are sibling formats Clinical record, session note BIRP with the local spelling Behaviour, Intervention, Response, and Plan; case note What changes Intervention-response detail, same-day E/M separation, signature authentication (including AI-transcribed notes) Ontario: dated, attributable entries and a 10-year floor; Québec: prescribed record contents with a 5-year floor Records made at the time of events; continuity-of-care detail; Better Access reporting sits at the episode level Retention No HIPAA chart rule; state law governs; Medicare expects 7 years for covered ordering and referral documentation Ontario: at least 10 years from last interaction, or from the client's 18th birthday; Québec: at least 5 years from the last professional service 7 years since last entry; until age 25 for clients under 18 (Psychology Board Code) The intervention-response thread reads well to reviewers in all three countries. What changes is the regulatory layer around it, so anchor retention and signature practices to your own jurisdiction, not to the format. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on BIRP notes. Draft a complete BIRP note from a few bullets, a dictation, or a pasted session transcript, with the intervention-response thread intact. Convert between BIRP, SOAP, DAP, and narrative structures without losing clinical content. Check a finished note for unspecified interventions, a missing Response section, and treatment-plan linkage before you sign. BastionGPT drafts, you review and sign, which is exactly what Medicare's signature guidance requires for AI-assisted notes. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What is the difference between BIRP, SOAP, and DAP notes? All three are progress-note conventions. SOAP splits the client's report from your observations; DAP merges them into one Data section; BIRP organizes the note around the intervention-and-response loop. Choose BIRP when showing what you did and how the client responded matters most, not because any payer requires it. What belongs in the Response section? How the client participated, how they reacted to each intervention, and what changed or did not change in symptoms, behavior, or goal progress. Medicare's psychotherapy coverage policy asks for exactly that: participation, reaction to the session, and progress toward goal-oriented outcomes. An honest "no change yet" is better documentation than silence. Do insurers require or prefer BIRP notes? No source we reviewed shows a payer requiring BIRP by name, and California's Medi-Cal behavioral health policy is now explicitly format-neutral, with CalAIM county guidance stating the old BIRP/GIRP/SIRP/SOAP choices are no longer required. What payers enforce is content: interventions described, response documented, plan linkage, time support, and authenticated signatures. What are GIRP and SIRP notes? Sibling formats that swap the first section: GIRP leads with the Goal being worked, SIRP with the Situation, and both keep Intervention, Response, Plan. Public behavioral-health systems have historically treated BIRP, GIRP, SIRP, and SOAP as interchangeable structures for the same underlying record. Is a BIRP note a psychotherapy note under HIPAA? Usually not. HIPAA's psychotherapy-notes category excludes diagnosis, treatment plan, symptoms, prognosis, progress, and session timing, the working content of a BIRP note. A BIRP note kept in the chart is a progress note in the standard record, and labeling it otherwise does not change its access and disclosure status. If AI or a scribe drafts my BIRP note, who signs it? You do. Medicare's signature guidance says that when AI or a human scribe transcribes an entry, the responsible clinician must still sign the note to authenticate it, and OIG auditors have rejected pasted images of signatures on psychotherapy claims. That is why BastionGPT is built as draft-and-review: it drafts, you review and sign. How do I document psychotherapy billed with a same-day E/M visit? CMS billing guidance requires the two services to be significant and separately identifiable, and time spent on E/M activities cannot be counted as psychotherapy time. In BIRP terms: keep the medication or medical work visibly separate from the psychotherapy content, and record psychotherapy time on its own. How long do I need to keep BIRP notes? There is no universal rule. HIPAA sets no patient-chart retention period, so US state law governs, with Medicare expecting 7 years for covered ordering and referral documentation. Ontario's college expects at least 10 years, Québec's psychologist regulation at least 5 years from the last professional service, and Australia's Psychology Board code 7 years, or until age 25 for clients who were minors. Can BastionGPT write a BIRP note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces a structured BIRP draft for your review, with the intervention-response thread intact. It can also convert notes between BIRP, SOAP, DAP, and narrative structures. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and its exclusions. HHS, HIPAA FAQ 580 : the Privacy Rule sets no medical-record retention period; state law governs. CMS Medicare Coverage Database, Psychotherapy coverage policy L33252 : required documentation content, including why "supportive psychotherapy given" is inadequate. CMS Medicare Coverage Database, Billing and Coding Article A57480 : same-day psychotherapy and E/M must be significant, separately identifiable, and separately timed. CMS, Complying with Medicare signature requirements (MLN905364) : authentication rules, including the requirement that clinicians sign AI- or scribe-transcribed entries. CMS, Medical record maintenance and access requirements (MLN4840534) : the 7-year maintenance-and-access expectation for covered documentation. California DHCS, BHIN 23-068 behavioral health documentation requirements : explicitly format-neutral documentation policy. Washington State DSHS, Documenting medically necessary behavioral health services: progress notes : stand-alone notes and the golden thread from need to next step. HHS Office of Inspector General, On-Site Psychological Services audit (2020) : 111 of 120 claims noncompliant; unspecified maneuvers and pasted signature images. College of Registered Psychotherapists of Ontario, Clinical Records practice standard : record content and the 10-year retention expectation. Québec, Regulation respecting the records of psychologists (C-26, r. 221) : prescribed record contents and the 5-year retention floor. Psychology Board of Australia, Code of Conduct (effective 1 Dec 2025): record quality and the 7-year / age-25 retention floor. ## Related templates SOAP Note Psychotherapy Progress Note DAP Note Treatment Plan Case Formulation Note PIE Note Telehealth Psychotherapy Note Individual Therapy Session Note Narrative Progress Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/crisis-note All note templates # Crisis Psychotherapy Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A crisis psychotherapy note documents an encounter where psychotherapy addressed an acute crisis: the precipitating event, urgent assessment, mental status, risk findings, interventions, client response, and disposition. Therapists, psychologists, counselors, and clinical social workers write one after crisis encounters in office, telehealth, or community settings. In the US it supports the time-based crisis codes 90839 and 90840. A typical crisis note runs 250 to 700 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, crisis clinicians Audience Treating clinician, crisis and emergency teams, payers, auditors, legal reviewers Typical length 250 to 700 words · 15 to 25 minutes by hand (clinical team estimate) Format family Crisis encounter documentation (compare: progress note, suicide risk assessment, safety plan) When it's used During or immediately after a crisis encounter, in office, by telehealth, or in the community Standards context No mandated format; Medicare defines the service content and time rules, not the note layout On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a crisis psychotherapy note? A crisis psychotherapy note is the clinical record entry for an encounter where psychotherapy addressed an acute crisis: substantial distress from a problem serious and complex enough to need attention that day. The name comes from the billing side, not from a documentation standard. The US crisis service codes took effect January 1, 2013, and were developed at the request of the National Association of Social Workers, according to the American Psychiatric Association's contemporaneous coding materials . Clinicians and payers also call it a crisis note, a crisis intervention note, or simply a 90839 note. Two clarifications matter. First, no law, regulation, or payer policy prescribes a crisis note format. CMS describes what the service itself must involve: an urgent assessment and history of the crisis state, a mental status exam, psychotherapy, mobilization of resources to defuse the crisis and restore safety, and a disposition. Any note that clearly shows that work qualifies; the layout is a convention. Second, a crisis psychotherapy note is not a psychotherapy note in the HIPAA sense. It carries session times, diagnosis, risk findings, and the disposition, exactly the content HIPAA excludes from that specially protected category, so it belongs in the standard clinical record (see the psychotherapy notes authorization page for the difference). ## Who uses crisis psychotherapy notes and when Outpatient therapists, psychologists, counselors, and clinical social workers write one whenever a session became crisis work: an established client arrives in acute distress, a new client presents in crisis, or a routine appointment converts partway through. Medicare recognizes the service from physicians and a broad practitioner list that includes clinical psychologists, clinical social workers, nurse practitioners, physician assistants, marriage and family therapists, and mental health counselors, in facility and non-facility settings including homes, and by telehealth when requirements are met. An encounter qualifies by what happened in it, not by how it got on the calendar: a scheduled session that turned into genuine crisis stabilization can qualify, and an unscheduled visit is not automatically a crisis service. When the session stayed planned treatment, write a standard progress note or SOAP note instead. Mobile crisis and community stabilization teams usually document under separate state program codes with their own rules, so this page covers the individual psychotherapy-for-crisis encounter. ## Crisis psychotherapy note structure: what goes in each section The sections mirror what CMS says the service includes: an urgent assessment and history of the crisis state, a mental status exam, psychotherapy, mobilization of resources, and a disposition. Time is the sixth element: record start and stop times or total time, including each segment when the work was split across the day. Precipitating event and history of the crisis state. What happened, when it escalated, and why it needed attention today: the acute stressor, recent changes, relevant history. Pitfall: recording distress without the why-now. A reviewer should see a problem requiring immediate attention, not a difficult week. Mental status and presentation. Appearance, speech, affect, thought process, orientation, and judgment, focused on what is relevant to the crisis, with changes across the encounter noted. Pitfall: skipping the mental status exam in the urgency of the moment. It is part of the service Medicare describes, not an optional extra. Risk assessment. Ideation, intent, plan, access to means, and protective factors as relevant to this crisis, ending in a risk formulation. Pitfall: a bare "denies SI." The formulation is what supports the disposition, so a checkbox leaves the most consequential decision in the note unexplained. Interventions and resources mobilized. The crisis-focused therapeutic work delivered (de-escalation, grounding, structured problem solving, safety planning) and the supports activated, with consent noted for collateral contacts. Pitfall: a note that reads as monitoring or coordination only. The service is psychotherapy for crisis, so name the therapeutic work. Client response. How the client's state changed in response to the work, measurable where possible. Pitfall: no recorded response. Contractor reviews repeatedly flag notes with no documented outcome, and a disposition with no recorded response above it looks arbitrary. Disposition and follow-up. Where the client went and why that was safe: level of care, consultation, a dated next contact, and the escalation instructions given. Pitfall: "will follow up" with no timeframe. The disposition is the part of a crisis note that gets defended later, so give it reasoning and a date. ### Blank template (copy and adapt) Client: [initials] Date: Setting: [office/telehealth/home or community] Start/stop time(s), may be non-continuous: Total crisis time (min): Precipitating event & history of crisis state: Mental status & presentation: Risk assessment (ideation, intent, plan, means access, protective factors): Interventions & resources mobilized (consent noted for contacts): Client response: Disposition & follow-up (level of care, next contact, escalation plan): Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample crisis psychotherapy note Scenario: established adult client in outpatient CBT contacts the practice in acute distress after a sudden job loss and is seen the same day for a 65-minute crisis session. All details are fictional. Client: R.L., 42 · Date: 07/14/2026 · Service: Crisis psychotherapy, in office · Start/stop: 2:10 pm to 3:15 pm (65 minutes, continuous) Precipitating event and history: Client called the practice at 1:40 pm in acute distress after being terminated from a 12-year position this morning. Reports escalating panic since the meeting, one episode of hyperventilation on the drive over, and repetitive thoughts that the family's finances are ruined. No sleep concerns before today; symptoms of this intensity are new. Spouse drove client to the office and is in the waiting room. Mental status: Alert and oriented. Grooming appropriate but visibly distressed on arrival: tearful, wringing hands, speech rapid and pressured initially, slowing by mid-session. Affect anxious and constricted, brightening slightly by session end. Thought process organized. No hallucinations or delusions elicited. Judgment and insight intact. Risk assessment: Client reported passing thoughts this morning that the family "would be better off" without them, which resolved by early afternoon. Denied current suicidal ideation, intent, or plan, and denied any history of self-harm or prior attempts. Access to means reviewed with client and spouse; no concerns identified. Protective factors: engaged spouse, two school-age children, active treatment relationship, previously effective coping skills. Formulation: acute situational crisis with brief passive ideation, now resolved; risk currently low with strong protective factors; outpatient management with supports is appropriate. Interventions and resources mobilized: Paced breathing practiced in session until physical symptoms settled. Structured review of the precipitating event and immediate financial concerns; cognitive restructuring of catastrophic appraisals using established CBT skills. Safety plan reviewed and updated with client. With client's consent, spouse joined the final 15 minutes to review warning signs, the updated plan, and next steps. Crisis line (988) saved to client's phone. Client response: Visible de-escalation across the session. Breathing normalized, speech slowed, and client engaged actively in planning. Self-rated distress decreased from 8/10 at arrival to 4/10 at close. Client generated two concrete steps for the next 48 hours and committed to both. Disposition and follow-up: Client returns home with spouse. Phone check-in scheduled tomorrow at 10:00 am; next appointment 07/17/2026. Client verbalized the safety plan steps and agreed to call or text 988 or go to the nearest emergency department if distress escalates. Risk formulation supports outpatient disposition as documented above. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The precipitant and same-day escalation show why the encounter met crisis-level need, not just that the client was distressed. The risk section covers ideation, intent, plan, means access, and protective factors, and its formulation connects directly to the disposition. Start and stop times support the time-based crisis code, and the note states the time was continuous. Interventions are specific therapeutic work, and the response is measured (8/10 to 4/10), so stabilization is shown rather than asserted. The spouse's involvement is documented with consent, and follow-up is dated with a clear escalation path. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A crisis note is read by more people, under more scrutiny, than any other note you write: the colleague taking the next-day check-in, a care coordination handoff, payers, and sometimes attorneys or your licensing board. The disposition and its reasoning are what get examined afterward, so document the risk formulation that led to it, any consultation, and collateral contacts with the consent basis noted. Duty-to-warn and duty-to-protect rules are state law, and they vary: California and Virginia define triggers and protective actions without prescribing note content, while New York expressly requires the reasons for a danger-related disclosure to be "fully documented in the clinical record" ( Mental Hygiene Law 33.13 ). When a threat situation touches your note, consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). For Medicare, the crisis codes are time based and the time rules are specific. 90839 covers the first 60 minutes and 90840 each additional 30, and contractor guidance counts the total face-to-face crisis time on the date of service even when it is not continuous, with the patient present for all or some of it and the clinician's full attention on the crisis while time is counted. Since January 1, 2024, crisis psychotherapy furnished at an applicable site of service (anywhere the non-facility rate applies other than an office, such as a client's home) bills Medicare's G0017 and G0018 at 150 percent of the non-facility rate under the Consolidated Appropriations Act, 2023 . Do not bill the crisis codes on the same date as a psychiatric diagnostic evaluation or standard psychotherapy, and remember that a diagnosis, panic, or an unscheduled appointment by itself does not establish the service ( CMS, Psychotherapy for Crisis ). The format is a convention; the content and the time are the requirement. Crisis work also usually changes the treatment plan , so close the loop there once the client is stable. ↑ Back to the template and downloads ## Common crisis psychotherapy note errors auditors flag Crisis-specific audit data is thin: no national dataset publishes a 90839-only denial rate. The broader psychotherapy findings are blunt. Medicare's 2024 reporting-period data for outpatient psychiatric services put improper payments at 16.1 percent ($254.5 million), with insufficient documentation driving 78.3 percent of them, and a national OIG audit of psychotherapy claims estimated $580 million in improper payments, with undocumented time a leading cause. The BastionGPT Clinical Advisory Board sees the same errors most often in crisis note reviews: Time that does not support the code. No total time or start and stop times, under 30 countable minutes billed as 90839, or 90840 reported alone. In one OIG provider audit , 82 of 100 sampled psychotherapy services lacked required time documentation. Crisis criteria asserted, not shown. The note records distress but not why it required immediate attention. A diagnosis, panic, ideation on a screener, or an unscheduled slot does not by itself establish the service. No psychotherapy visible. The note reads as assessment, monitoring, or coordination only. The service is psychotherapy for crisis, so the therapeutic work has to be named. Wrong code for the site since 2024. Home or community crisis work billed as 90839 where G0017 applies, or the reverse, which misses the 150 percent differential in one direction and miscodes the claim in the other. Authentication gaps. Missing signatures or credentials, and late entries not identified as late. Medicare's signature rules accept compliant electronic signatures but never backdating. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on crisis psychotherapy notes. Draft a complete crisis note from a few bullets or a dictation, with the precipitant, risk assessment, interventions, response, and disposition in the right places. Check a finished note for missing time documentation, risk language, consent notes, or follow-up details before you sign. Flag when the documented work reads as assessment or coordination rather than crisis psychotherapy, and whether the recorded time supports the codes billed. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a crisis psychotherapy note be? Most run 250 to 700 words, noticeably longer than a routine progress note, because the note carries an urgent assessment, a risk formulation, interventions, the client's response, and a reasoned disposition. Length is not the goal: a reviewer should be able to reconstruct what the crisis was, what you did, and why the disposition was safe. Is there a required format for crisis psychotherapy notes? No. No regulation or payer policy prescribes headings, a named template, or a SOAP-style layout for crisis work. Medicare defines the service content instead: an urgent assessment and history of the crisis state, a mental status exam, psychotherapy, resource mobilization, and a disposition. Any note that clearly shows that work qualifies; the format is a convention. What time documentation do 90839 and 90840 need? They are time-based codes, so the record must make the crisis time countable. Medicare contractor guidance counts the total face-to-face time on the date of service even when it is not continuous, and the patient may be present for all or some of it, but the clinician must give the crisis full attention while time is counted. A clear total time can be sufficient; start and stop times are the safest habit. 90839 covers the first 60 minutes and 90840 each additional 30. Can a scheduled session qualify as crisis psychotherapy? Yes. Nothing in Medicare policy makes a scheduled appointment ineligible, and an unscheduled visit is not automatically a crisis. What controls is the work: whether the presentation met the immediate-attention criteria and the session became crisis psychotherapy with assessment and disposition work. When a routine session converts partway through, count only the crisis time. Can I bill 90839 with a regular psychotherapy code on the same day? No. CMS instructs practitioners not to report the crisis codes on the same date as a psychiatric diagnostic evaluation (90791, 90792) or standard psychotherapy (90832 through 90838). 90840 also never stands alone; it is an add-on to 90839. When a session becomes a crisis, bill the crisis service the time supports rather than stacking codes. What are G0017 and G0018? Medicare codes for crisis psychotherapy furnished at an applicable site of service: anywhere the non-facility rate applies other than an office, such as a client's home. Congress created the category in the Consolidated Appropriations Act, 2023 , and since January 1, 2024 these codes pay 150 percent of the non-facility rate. In an office or facility you keep using 90839 and 90840, so picking the code that matches the site matters. Is a crisis note a psychotherapy note under HIPAA? No. HIPAA's specially protected psychotherapy-notes category excludes exactly what a crisis note must contain: session times, diagnosis, symptoms, treatment plan, and progress. A crisis psychotherapy note belongs in the standard clinical record, where payers and other treating providers can rely on it. Keep private process reflections in separately maintained psychotherapy notes with their own authorization rules . Do I need a separate safety plan document with every crisis note? Not as a universal rule: no national 90839 requirement makes a standalone safety plan mandatory, so document the safety planning you actually did in the note itself. Since January 1, 2025, Medicare also recognizes structured safety-planning interventions as a separate service, G0560 , billed in 20-minute increments and distinct from the crisis psychotherapy itself. Can BastionGPT draft a crisis psychotherapy note? Yes. Give it bullets or a dictation from the encounter, and it drafts a structured crisis note with the precipitant, mental status, risk assessment, interventions, response, and disposition for your review. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, Psychotherapy for Crisis : service content, eligible practitioners, settings, same-day exclusions, and the G0017/G0018 site rule. CMS Medicare Coverage Database, Billing and Coding Article A57520 : crisis codes as time-based, non-continuous same-date time, full-attention rule, and patient presence for all or some of the service. Consolidated Appropriations Act, 2023, section 4123 : 150 percent payment for crisis psychotherapy at applicable sites of service, effective January 1, 2024. American Psychiatric Association, Spring 2013 CPT coding overview : the 2013 effective date and the National Association of Social Workers' role in the crisis codes. HHS Office of Inspector General, national psychotherapy audit (2023) : an estimated $580 million in improper payments, with undocumented time among the leading causes. CMS Medicare Learning Network, outpatient psychiatric care compliance tips : 2024 reporting-period improper-payment data, including the 78.3 percent insufficient-documentation share. HHS Office of Inspector General, Grand Desert Psychiatric Services audit (2020) : 82 of 100 sampled services lacking time documentation. CMS, Complying with Medicare Signature Requirements (MLN905364) : authentication and attestation rules. HHS, HIPAA FAQ 2088 : the psychotherapy-notes definition and its exclusions. New York State Senate, Mental Hygiene Law 33.13 : the requirement to document reasons for a danger-related disclosure. CMS, CY 2025 Physician Fee Schedule final rule summary (MM13887) : the G0560 safety-planning service. ## Related templates Psychotherapy Progress Note SOAP Note Trauma Assessment Treatment Plan Care Coordination Note Telehealth Psychotherapy Note Suicide Risk Assessment Safety Plan Mandated Reporting & Duty-to-Protect Documentation Violence & Homicide Risk Assessment Self-Harm (NSSI) Assessment Communication Log & Portal Message Note Incident & Adverse Event Report Consent for Treatment of a Minor C-SSRS Documentation COWS Protocol Documentation CIWA-Ar Protocol Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/dap-note All note templates # DAP Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A DAP note is a three-part progress note format that organizes a therapy session into Data, Assessment, and Plan. It merges SOAP's Subjective and Objective material into a single Data section, which suits the more qualitative flow of psychotherapy documentation. A typical therapy DAP note runs 150 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, psychologists, community and solo behavioral health clinicians Audience Treating clinician, care team, supervisors, payers, auditors Typical length 150 to 300 words · 10 to 20 minutes by hand (clinical team estimate) Format family Structured progress note (compare: SOAP, BIRP, PIE; DARP adds a Response section) When it's used After each individual, couples, group, or telehealth session Standards context A documentation convention, not a mandated form; law and payers regulate the record's content, not its headings On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a DAP note? DAP stands for Data, Assessment, Plan. It is a three-part progress note format that merges SOAP's Subjective and Objective material into a single Data section, then keeps Assessment and Plan. That merge is the point: psychotherapy sessions rarely split cleanly into "what the client said" and "what you measured," and DAP lets the session's material flow into one place. DAP appears to be a behavioral-health adaptation of the SOAP note and the problem-oriented medical record tradition behind it; no standards body or original publication established DAP as a formally standardized type, and no law or payer manual we reviewed requires the Data, Assessment, Plan headings by name. A common variant is the DARP note, which adds a Response section for the client's reaction to interventions. The legal footing matters more than the headings. A DAP note that contains diagnosis, treatment-plan content, symptoms, progress, or session timing is a progress note in the standard clinical record, not a HIPAA " psychotherapy note ": HHS's definition expressly excludes exactly those items. Keep private process reflections in separately maintained psychotherapy notes with their own authorization rules . ## Who uses DAP notes and when DAP is the default in much of behavioral health: solo therapists, counselors, community mental health teams, and group practices whose documentation is more qualitative than medical. It works well when your notes are read mostly inside your own practice. When records routinely travel to integrated or multidisciplinary settings, SOAP is often preferred because payers and medical teams already know how to read the subjective/objective split. Choose DARP when you want the client's response to interventions separated from the plan, and BIRP when documenting intervention and response as distinct fields matters operationally. No payer or regulator we reviewed requires one of these formats over another; what they regulate is whether the record supports the service. ## How to write a DAP note in three steps Data: everything the session produced: the client's report, your observations, mental-status highlights, scores, the interventions you delivered, and therapy time when billing requires it. Assessment: your clinical judgment: what the data means, progress toward a numbered treatment-plan goal, and current risk status. Plan: what happens next: interventions for coming sessions, homework, referrals, next appointment, and any treatment-plan changes. Each section in more detail, with the pitfall that most often undermines it: D: Data. The merged record of what was reported, observed, measured, and done. Include at least one observable or score (PHQ-9, GAD-7, SUDS) and name the interventions you actually delivered. Pitfall: Data that is all narrative with nothing observable or measurable behind it, or therapy time missing when a time-based code is billed. A: Assessment. Interpretation, not repetition: what the data means for diagnosis, functioning, and progress toward a numbered treatment-plan goal, plus risk status. Pitfall: restating the Data section instead of interpreting it, which breaks the medical-necessity chain payers look for. P: Plan. Specific and dated next steps: interventions, homework, referrals, coordination, next appointment. Pitfall: a bare "continue treatment" with nothing the next note can evaluate. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: [type/length] Start/stop or total time: D (Data): A (Assessment): P (Plan): Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample DAP note Scenario: adult client, panic disorder, CBT with interoceptive exposure, session 5 of a planned 12, delivered by telehealth. All details are fictional. Client: R.S., 29 · Date: 07/10/2026 · Session: 5 · Service: Individual psychotherapy, 45 min, telehealth video; client at home, in state; consent on file · Start/stop: 09:01 to 09:46 D: Client reported two panic episodes this week, down from four, both managed without leaving the situation. Panic log completed 6 of 7 days. On video: engaged, speech normal rate, affect mildly anxious when discussing an upcoming flight. Reviewed log entries together. Conducted interoceptive exposure (straw breathing, three trials; peak SUDS 65 falling to 40 by trial three) and cognitive reframing of "a racing heart means something is wrong with me," with evidence review. Denied suicidal ideation and self-harm urges. A: Panic frequency and duration improving; client is applying skills between sessions and tolerated exposure trials with decreasing distress, consistent with Goal 1 of the treatment plan (reduce panic frequency and situational avoidance). Willingness to plan the flight suggests reduced avoidance. Symptoms remain consistent with panic disorder, improving. Risk: denies ideation; no risk indicators observed; continue routine monitoring. P: Continue weekly CBT with interoceptive exposure. Homework: one straw-breathing trial daily, continue panic log. Begin flight-preparation exposure hierarchy next session. Re-rate symptom measures at session 8. Next appointment 07/17/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Start and stop times support the billed 45-minute code, and the telehealth line carries modality, client location, and consent, the details payers ask about first for virtual care. Data includes measurable anchors (panic count, log completion, SUDS trajectory), not just narrative. Interventions are named inside Data (interoceptive exposure, cognitive reframing), specific enough for a reviewer to tell what the service was. Assessment interprets rather than restates, and ties progress to a numbered treatment-plan goal, the medical-necessity link. Risk is addressed even though the client is low risk, so the record shows monitoring rather than silence. Plan items are specific and dated, giving the next note something concrete to evaluate. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A DAP note is part of the designated record set: clients can request it, payers can audit it, and other providers may rely on it. Keep private process reflections in separately maintained psychotherapy notes with their own authorization rules . For couples, family, and group work, two rules travel well: each participant needs documentation that can stand alone, and access is not all-or-nothing. Ontario's psychotherapy college, for example, releases a joint record in full only when all participants consent; a single requester is entitled to their own information plus communal content that is not attributable to another participant (see family and couples and group notes). On billing, the sharpest open question is time documentation, and the official guidance is genuinely mixed: Medicare's current billing article for psychotherapy services accepts start/stop times or total time when time is relevant to the code, while an older contractor fact sheet says exact times must be documented. In that environment, recording start and stop times remains the most audit-defensible choice. Authentication matters just as much: in a provider-level OIG audit , all 100 sampled beneficiary days failed Medicare requirements, with missing psychotherapy time, absent treatment notes, and unsigned treatment plans among the findings. On retention, HIPAA does not set a patient-chart period: HHS says state law generally governs. Ontario colleges use a 10-year floor (longer for minors). In Australia, Better Access adds two program rules generic DAP pages miss: a written report to the referrer at the completion of each course of treatment, and a requirement to keep referrals for two years from the first MBS service under that referral. ↑ Back to the template and downloads ## Common DAP note errors auditors flag Audit findings are about content, not the choice of DAP over SOAP. In Medicare's 2024 reporting-period data for outpatient psychiatric services, 16.1% of payments were improper, a projected $254.5 million, with insufficient documentation driving 78.3%. A national OIG audit estimated $580 million in improper psychotherapy payments, and in one provider-level audit , 100 of 100 sampled beneficiary days were noncompliant. The BastionGPT Clinical Advisory Board sees the same patterns most often in DAP note reviews: Therapy time not documented. A time-based code billed with no start/stop or total time anywhere in the note. Data with no measurable anchor. Pure narrative with no observation, score, or log the next reviewer can verify. An Assessment that restates the Data. No interpretation, no progress judgment, no link to a numbered treatment-plan goal. Treatment plans absent or incomplete. Missing frequency or duration, or plans never signed, both named findings in the provider-level OIG audit. Cloned notes. Copy-forward text repeating across sessions, which reads as a record that does not support a distinct billed service. ### DAP notes in the US, Canada, and Australia Aspect United States Canada Australia Status Convention: no law or payer manual requires the DAP headings; HIPAA governs the record's privacy status and payers govern documentation support Convention; Ontario college standards (CRPO, CPBAO) make record content and access enforceable Convention; Psychology Board Code of Conduct (from 1 Dec 2025) and Ahpra expectations govern record quality Terminology DAP note, DAP progress note, DARP (adds Response) Clinical record, session note Case note, session note What changes Time support for billed codes, signatures, treatment-plan linkage Ontario: progress notes may be kept in the therapy language (key info in English or French); joint-record access is severable, not all-or-nothing Better Access: written report to the referrer after each course of treatment; keep referrals 2 years from the first MBS service Retention No HIPAA chart rule; state law and payer contracts govern Ontario: at least 10 years from last interaction, or from the client's 18th birthday 7 years since last entry; until age 25 for clients under 18 (Psychology Board Code) The three-section structure travels well everywhere. What changes is the surrounding layer: payer time-and-signature rules in the US, college record and access standards in Canada, and program-cycle reporting in Australia. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on DAP notes. Draft a complete DAP note from a few bullets, a dictation, or a pasted session transcript. Convert between DAP, SOAP, BIRP, and narrative structures without losing clinical content. Check a finished note for missing time support, an Assessment that merely restates the Data, and treatment-plan linkage before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What is the difference between a DAP note and a SOAP note? DAP combines SOAP's Subjective and Objective sections into a single Data section, then keeps Assessment and Plan. DAP suits qualitative psychotherapy documentation and solo or behavioral-health settings; SOAP reads better in integrated and multidisciplinary settings where the subjective/objective split aids communication. Neither is legally required; the choice is workflow, not compliance. What goes in the Data section? Everything the session produced: the client's report, your observations and mental-status highlights, measurable scores or logs, the interventions you delivered, and therapy time when a time-based code is billed. Include at least one observable or measurable anchor; pure narrative is the section's most common weakness. What is a DARP note? A DAP variant that adds a Response section: Data, Assessment, Response, Plan. Teams use it when they want the client's reaction to interventions recorded separately instead of folded into Data or Assessment. If response-tracking matters operationally, BIRP is the other common choice. Is a DAP note a psychotherapy note under HIPAA? Usually not. HIPAA's psychotherapy-notes category excludes diagnosis, treatment plan, symptoms, prognosis, progress, and session timing, exactly what a working DAP note contains. A DAP note used for care and billing is part of the standard record; true psychotherapy notes must be kept separately and have their own authorization rules . Do insurance companies accept DAP notes? Yes. No payer in the sources we reviewed mandates a note format by name. What reviewers require is content: documentation that supports the billed service and time, links the session to an individualized treatment plan, and is signed and dated. A DAP note that carries those elements is as payer-ready as a SOAP note. Do I need start and stop times in a DAP note? The official guidance is mixed: Medicare's current billing article accepts start/stop times or total time for time-based psychotherapy codes, while an older contractor fact sheet says exact times must be documented. Because both float around audit desks, recording start and stop times remains the most defensible practice. How long should a DAP note be? Most run 150 to 300 words and take 10 to 20 minutes by hand (our clinical team's estimate; no published benchmark exists, and vendor time-saving statistics for DAP are marketing claims rather than research). Long enough to carry a measurable anchor, named interventions, an interpretation, and a dated plan. In couples or family therapy, can one client get the whole DAP record? Not automatically. Ontario's psychotherapy college, for example, releases a joint record in full only when all participants consent or request together; a single requester gets their own information plus communal content not attributable to another participant. Wherever you practice, write joint-therapy notes so each person's content can be severed cleanly. Can BastionGPT write a DAP note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces a structured DAP draft for your review. It can also convert notes between DAP, SOAP, BIRP, and narrative structures. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and its exclusions. HHS, HIPAA FAQ 2088 : psychotherapy notes exclude diagnosis, treatment plan, progress, and session timing; authorization rules for the rest. HHS, HIPAA FAQ 580 : the Privacy Rule sets no medical-record retention period; state law governs. CMS Medicare Coverage Database, Billing and Coding Article A57520 : start/stop times or total time for time-based psychotherapy codes. CMS contractor, Outpatient psychiatry and psychology services fact sheet : the older exact-times documentation wording. CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 improper-payment data (16.1%, $254.5M, 78.3% insufficient documentation). HHS Office of Inspector General, national audit of Medicare psychotherapy services (2023) : estimated $580 million in improper payments. HHS Office of Inspector General, New York provider psychotherapy audit (2022) : 100 of 100 sampled beneficiary days noncompliant; missing time, notes, and plan signatures. College of Registered Psychotherapists of Ontario, Clinical Records practice standard : required record content and the 10-year retention expectation. CRPO, Language of records and access to joint therapy records : notes may be kept in the therapy language; joint-record access is severable. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : authorship, co-signature, separate group records, and retention rules. MBS Online, Arrangements for eligible allied health professionals : written report to the referrer at completion of a course of treatment. MBS Online, Record-keeping changes under the Health Insurance Act : keep referrals for two years from the first MBS service under the referral. ## Related templates SOAP Note Psychotherapy Progress Note BIRP Note Treatment Plan Case Formulation Note Psychotherapy Notes Authorization Trauma Assessment Care Coordination Note PIE Note Telehealth Psychotherapy Note Group Therapy Note Individual Therapy Session Note Narrative Progress Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/couples-family-therapy-note All note templates # Family & Couples Therapy Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A family and couples therapy note is the clinical record of a conjoint psychotherapy session, one where partners or family members attend together. It documents who attended, the relational patterns addressed, the interventions, each person's response, and progress toward treatment goals. Most run 200 to 500 words. Unlike an individual note, it must serve several people's privacy rights inside one story. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Marriage and family therapists, psychologists, counselors, clinical social workers Audience Treating clinician, payers, auditors, legal reviewers Typical length 200 to 500 words · 10 to 15 minutes by hand (clinical team estimate) Format family Conjoint session note: shared session layer plus per-person response (compare: SOAP, group note) When it's used After each couples or family therapy session, conjoint or collateral Standards context No law mandates a form; payers bill one identified patient and privacy rights attach per person On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a family & couples therapy note? A family and couples therapy note is the clinical record of a conjoint psychotherapy session, one where partners or family members attend together as the treatment unit. Like the group therapy note , it is not a named format with a single origin: no professional body has standardized a conjoint note form, and the work it documents predates every modern charting framework (marriage counseling was institutionalized in the United States by the 1940s, and family therapy emerged as its own field in the 1950s, decades before the problem-oriented record gave medicine SOAP). What defines the document is its subject. An individual note documents one person's session; a conjoint note documents a relational system, which is exactly how Medicare's coverage policy describes the service: "treating the identified patients' mental illness by intervening in a family system" ( LCD L33252 ), with or without the patient present. So the note records who attended, the interaction patterns addressed, the interventions applied to the cycle rather than to one person, and each attendee's response. Clinicians and payers also call it a conjoint therapy note, family psychotherapy note, joint-session note, or couples counseling note. The disambiguation that matters most is where it lives. In insurance-billed US care there is no "couple's chart": the note is filed in the record of the identified patient, the person whose diagnosis makes the service medically necessary, and Medicare's national coverage rule states that family counseling is covered only where "the primary purpose of such counseling is the treatment of the patient's condition" ( NCD 70.1 ). So one document must describe several people while formally belonging to one of them, which is why access, consent, and release questions dominate this note type. A family and couples therapy note is a progress note, part of the standard record; it is not a psychotherapy note , HIPAA's separately protected category of private process notes. ## Who uses family & couples therapy notes and when Marriage and family therapists write them constantly, and since January 1, 2024 MFTs and mental health counselors can enroll and bill Medicare independently, per the current CMS Medicare and Mental Health Coverage booklet . Psychologists, professional counselors, and clinical social workers write the same note whenever treatment is conjoint: couples work in private practice, family sessions in children's and adolescent services, family involvement in SUD and eating disorder treatment, and family meetings inside an individual treatment episode. Use it for any session where the people in the room are a treatment unit; for a session with one client alone, use an individual progress note , and for therapy with unrelated members, use a group therapy note , which carries different confidentiality mechanics. A conjoint session conducted by video follows the same structure plus the location and consent elements of a telehealth therapy note . ## Family & couples therapy note structure: what goes in each section A conjoint note covers one session attended by several people, filed (in insurance-billed care) in one identified patient's chart. That double duty shapes every section: the note must document real relational work, support the code billed, and stay safe to release when one person, or one person's attorney, requests the record. Participants & session frame. Everyone present by first name and relationship, who was expected but absent, whether the identified patient attended, and start and stop times. Pitfall: a note that never states whether the patient was present; 90847 and 90846 differ on exactly that line, and a reviewer who cannot find it can reclassify or deny the claim. Session focus. The issue worked this session, stated so it connects to the identified patient's condition and treatment plan . Pitfall: framing the session around "the relationship" alone; Medicare's coverage rule pays for family psychotherapy only as treatment of the patient's condition, and relationship-problem Z codes do not appear in Medicare contractors' medical-necessity lists. Interaction patterns. The relational cycle in process language: who pursues, who withdraws, what triggers escalation, what interrupted it. Pitfall: verbatim arguments and blow-by-blow detail; conjoint records get requested in divorce and custody disputes, and a transcript reads as ammunition rather than treatment. Interventions. The specific clinical work: enactments, cycle de-escalation, communication or problem-solving training, reframing, restructuring tasks. Pitfall: "provided couples therapy"; an intervention a reviewer cannot picture is an intervention they cannot credit. Response, mental status & risk. Each attendee's response to the session's work, relevant observations, and a safety statement, screening individually when conflict, history, or presentation indicates. Pitfall: risk silence, or screening only the identified patient; every person in the room is a clinical responsibility, and conjoint sessions are where safety concerns often first surface. Progress toward plan goals. Progress tied to the numbered goal on the treatment plan, the same linkage an individual progress note carries. Pitfall: substituting "the couple communicated better" for the patient's own goal; medical necessity is judged against the identified patient's plan. Plan & next steps. Homework, the next session, who attends it, and any indicated individual referrals or collateral contacts. Pitfall: a bare "continue couples work"; say what continuing is meant to change and who needs to be in the room for it. ### Blank template (copy and adapt) Client / identified patient: [initials] Date: Session #: Type: couples / family Patient present: Y / N Participants present (name & relationship): Absent: Approach (e.g., EFT, Gottman, structural): Start/stop time: SESSION Focus this session (tied to the patient's condition and plan): Interaction patterns observed: Interventions: ATTENDEES Response by participant: Mental status & risk (per attendee, as indicated): PROGRESS & PLAN Progress toward treatment plan goal(s): Plan / homework / who attends next: Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample couples therapy note Scenario: married couple in emotionally focused therapy, session 6 of a planned 12. The identified patient carries an adjustment disorder diagnosis; the presenting pattern is a criticize-withdraw cycle around finances. All details are fictional. Client: J.M., 38 (identified patient) · Date: 07/14/2026 · Session: 6 of 12, couples · Patient present: Yes · Also present: T.M., spouse · Approach: EFT · Clinician: R. Alvarez, LMFT · Start/stop: 4:03 pm to 4:55 pm Session focus: Continued work on the criticize-withdraw cycle that maintains J.M.'s adjustment disorder symptoms, this session triggered by a disagreement over a large unplanned expense. Focus consistent with Goal 1 of J.M.'s treatment plan (reduce frequency and intensity of conflict episodes contributing to depressed mood and insomnia). Interaction patterns: Familiar cycle appeared within the first ten minutes: T.M. raised the expense with rising urgency, J.M. responded with brief factual answers and turned away. When slowed, each could name their position in the cycle, a change from session 3 when the cycle ran to completion twice. Interventions: Slowed and tracked the cycle as it ran; used an enactment to have T.M. voice the worry under the criticism ("I am afraid we are not a team on money") directly to J.M.; coached J.M. to stay engaged and respond rather than withdraw; reframed the cycle, not either partner, as the problem the couple faces together. Response & risk: J.M. stayed present through the enactment, made sustained eye contact, and reflected T.M.'s worry accurately; reported the exchange as "the first money conversation that didn't end in silence." T.M. was tearful during the enactment, settled, and expressed relief. Mental status unremarkable for both. Both deny suicidal ideation; no aggression history or indicators this session; no safety concerns identified. Progress: Goal 1: couple interrupted the cycle once without clinician prompting, the first spontaneous interruption of treatment. J.M. reports two conflict episodes this week, down from four at intake, and improved sleep on non-conflict nights. Plan: Continue weekly conjoint sessions; both partners to attend session 7. Homework: one 20-minute structured money conversation using the slow-down steps practiced today, before Friday. Re-administer PHQ-9 for J.M. at session 8. No individual sessions indicated at this time. Signed: R. Alvarez, LMFT, 07/14/2026. This sample is fictional and for educational purposes. It does not describe a real patient. Note how it documents relational process without transcribing the argument, and ties conjoint work to the identified patient's own diagnosis and plan goal. ↑ Back to the template and downloads ### Why this sample works Attendance and patient-present status are explicit in the header, so the code choice (90847) is supported on the face of the note. The session focus ties relational work to the identified patient's own diagnosis and numbered plan goal, keeping the medical-necessity chain intact. Conflict is documented as process ("criticize-withdraw cycle around finances"), not transcript, so the note can be released without handing either partner ammunition. Response and safety are stated for each attendee, not just the identified patient, so the record shows monitoring of everyone in the room. Start and stop times support a 50-minute service, and the plan says who attends next and what homework is due, so continuity does not depend on memory. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write every conjoint note knowing that each adult in the room may one day request the record it sits in. Privacy rights attach to individuals, not to couples: under 45 CFR 164.524 a patient's access is not automatically blocked because another person appears in the note, and denial on that ground requires a licensed professional's substantial-harm judgment. The marriage and family profession's own ethics code, revised effective January 1, 2026, sets the working convention: in couple or family treatment, records are not disclosed, and access is not provided, without "written authorization from each individual competent to provide written authorization," and one participant's confidences are not revealed to others in the client unit without that person's prior written permission ( AAMFT Code 2.2, 2.3 ). So agree in writing at intake how the record will be kept and who must authorize release, document any no-secrets policy, describe conflict as process rather than transcript, and keep individual disclosures out of the shared narrative. When a divorce or custody dispute arrives, a subpoena alone is not a release: 45 CFR 164.512(e) distinguishes court orders, which permit disclosing only what the order authorizes, from bare subpoenas, which require notice or a protective order. Preserve the chart as written, and consult your attorney or board before releasing anything, because state rules vary. Private process reflections belong in segregated psychotherapy notes , which Medicare reviewers are instructed not to request. For payers, the identified patient is the requirement and the format is a convention: Medicare's manuals say progress notes "may be in any form or format." The coverage rule ( NCD 70.1 ) pays for family counseling only when its primary purpose is treatment of the patient's condition, naming two qualifying uses: observing the patient's interaction with family members, and assessing and assisting the family in managing the patient. Counseling aimed at the other person's own distress is not reimbursable under the patient's benefit, and no Medicare contractor billing article for these services lists relationship-problem Z codes as supporting medical necessity. The contractor rules add the mechanics: whether the patient attended decides the code (90847 with, 90846 without), both codes "may not be reported for services less than 26 minutes," and neither covers consultation with paid institutional staff, per Article A56937 . Coverage guidance also expects periodic documentation of the patient's capacity to benefit, target symptoms, and measurable goals ( LCD L33252 ), so tie every conjoint session to the identified patient's own treatment plan goal. ↑ Back to the template and downloads ## Common family & couples therapy note errors auditors flag Conjoint documentation fails for identification problems: who was in the room, whose diagnosis anchors the service, and whose story the record tells. The audit stakes are documented. A national OIG audit of Medicare psychotherapy estimated $580 million in improper payments with 128 of 216 sampled enrollee-days failing at least one requirement, and an OIG audit of a single psychology practice found 111 of 120 sampled claims noncompliant , with treatment-plan defects on 111 claims and unspecified therapeutic maneuvers on 9, driving an estimated $3.3 million recovery. The BastionGPT Clinical Advisory Board sees the same errors most often in family and couples note reviews: No patient-present statement. 90847 (patient present) and 90846 (patient absent) differ on one documented fact. A note that never says who attended invites reclassification or denial, and a claim under the wrong code fails even when the session was real. The relationship as the diagnosis. Notes that document couple distress without linking the work to the identified patient's condition break the coverage chain: Medicare's NCD 70.1 covers family counseling only as treatment of the patient's condition, and contractor medical-necessity lists carry no relationship-problem Z codes. Transcript-style conflict detail. Verbatim accusations and blow-by-blow arguments create a record that damages both parties when released, and conjoint records are exactly the ones requested in divorce and custody disputes. One partner's secrets in the shared narrative. Individual disclosures from calls or one-on-one contacts, folded into the conjoint note, become part of a record other participants can seek. The 2026 AAMFT code bars revealing one participant's confidences to the client unit without prior written permission; the note has to keep the same promise. Mechanics that fail review. Sessions billed below the 26-minute floor in A56937 , missing session times, interventions never named (the OIG's "therapeutic maneuvers not specified" finding), and missing or image-stamped signatures. ### Family & couples therapy notes in the US, Canada, and Australia Aspect United States Canada Australia Status No law mandates a conjoint note form; Medicare covers family psychotherapy only as treatment of the identified patient (NCD 70.1), and MFTs and MHCs have billed Medicare independently since January 2024 College standards bind by profession; Ontario registered psychotherapists may keep one joint file while the same combination attends, and the psychologists' college defines a couple or family itself as a client Better Access serves a referred patient with a clinically diagnosed mental disorder; family and carer participation items bill against the patient, with a written record of the patient's consent Terminology Family psychotherapy, conjoint therapy note, couples counseling note (90846, 90847, 90849) Couple and family therapy, joint record, conjoint session note Family and carer participation services (MBS note MN.7.5); relationship counselling itself is not an MBS service What changes Whose chart it lives in: the identified patient's record carries the note, the diagnosis must be the patient's own, and contractor policy bars billing the family codes under 26 minutes File architecture: a joint file only while the same combination attends (Ontario); an access ruling treats the joint record as severable, one requester receiving their own plus communal information (PHIPA Decision 158); Quebec guidance accepts two architectures Two family or carer services per patient per calendar year, counted against the patient's allocation, patient not in attendance; the items expressly cannot be used for family therapy, group therapy, or couples/relationship counselling Retention No universal HIPAA rule for charts; Medicare expects claim-supporting records for 7 years (42 CFR 424.516(f)); state boards vary Ontario: 10 years from last contact, or 10 years after the client turns 18, whichever is later; BC's harmonized standard runs 16 years Psychology Board code: 7 years from last entry for adults; until the 25th birthday for clients under 18 The conjoint structure travels well across all three countries. What changes is who the system says the note is about: a US payer bills one identified patient, an Ontario college regulates the file architecture, and the Australian schedule counts family sessions against the referred patient's allocation. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on family and couples therapy notes. Draft a complete conjoint note from a few bullets about the session, from dictation, or from a transcript, with the relational work described as process rather than argument detail. Check a finished note before you sign: is patient-present status stated, is each attendee's response and safety documented, is the work tied to the identified patient's plan goal. Convert between formats when a payer or agency wants the same session as a SOAP or DAP note, keeping the conjoint elements intact. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Whose chart does a couples or family therapy note go in? In insurance-billed US care, the identified patient's: the person whose diagnosis makes the service medically necessary. Medicare's NCD 70.1 covers family counseling only as treatment of that patient's condition, so the note lives in their record even when three people attended, and copying one session into both partners' charts does not create two billable services. In private-pay work no law picks the architecture; a single relational chart or per-person records are both defensible if you tell clients at intake how the record is kept and who must authorize release. What is the difference between 90846 and 90847? One documented fact: whether the identified patient was in the room. 90847 is family psychotherapy with the patient present; 90846 is the same work with family members while the patient is absent. Both are 50-minute codes billed under the identified patient, and Medicare contractor guidance ( Article A56937 ) states both "may not be reported for services less than 26 minutes" and neither covers consultation with paid institutional staff. State patient-present status in the header so the code choice is supported on its face. Does insurance cover couples therapy? Not as relationship counseling. US coverage runs through family psychotherapy for a diagnosed identified patient: Medicare requires the primary purpose to be treatment of the patient's condition, and the contractor medical-necessity lists contain diagnosis codes, not relationship-problem Z codes. Australia is explicit: Better Access requires a clinically diagnosed mental disorder, and the official factsheet states its family and carer items "cannot be used in whole, or in part" for family therapy, group therapy, or couples therapy and relationship counselling. Commercial plans vary, and cash-pay couples work is common precisely because of this gap. How do I document a no-secrets policy? No law requires a no-secrets policy; it is an informed-consent choice, and the documentation is what makes it real: state the policy in your intake agreement, have every adult participant sign it, and note in the record when it is invoked. Write it carefully, because the 2026 AAMFT Code bars revealing one participant's confidences to others in the client unit without that person's prior written permission, so a policy promising automatic disclosure needs to be reconciled with your ethics code and local law. Whatever you promise, the record has to match it. Can one partner authorize release of the couple's record? Generally only for their own information. Privacy rights attach to individuals: under 45 CFR 164.524 each person can reach their own record, and the AAMFT code's convention is written authorization from each competent participant before the couple or family record is disclosed or accessed. Ontario has decided the question as law: in PHIPA Decision 158 , a family counselling record was not "dedicated primarily" to any one participant, so a lone requester received only their own reasonably severable information plus the family's communal information. What happens to couples therapy records in a divorce or custody case? They get requested, which is why the note you wrote in calmer times matters. A subpoena is not automatically a release: under 45 CFR 164.512(e) , a court order permits disclosing only what the order authorizes, and a bare subpoena requires satisfactory assurance of notice or a protective order; for SUD records under 42 CFR Part 2, a court order is required beyond the subpoena. Preserve the chart as written: do not rewrite, split, or selectively delete history after a dispute begins. Process-language documentation protects everyone; transcript-style notes hurt someone. Consult your attorney or board before releasing anything, because state rules vary. What do I document when only one partner or family member attends? Say who attended, who did not, and why the session ran that way. In US billing, family work without the patient is 90846, still billed under the identified patient and still aimed at their treatment; a parents-only session for a child patient follows the same logic when it addresses the child's condition rather than general parenting advice. Australia has a dedicated mechanism: up to 2 family and carer participation services per calendar year, billed against the patient, valid only when the patient is not in attendance, with a written record of the patient's consent required by MBS note MN.7.5 . In Ontario, an individual session gets its own file, separate from the couple's joint file. How is a family therapy note different from a group therapy note? A group therapy note documents unrelated people who each get their own note in their own chart, written so no member's identity leaks into another's record. A family note documents people who share their lives: they know everything that happened in the room, so the privacy work moves from hiding identities to managing release, consent, and whose chart the story sits in. Billing differs the same way: 90853 is one untimed code per member, while 90846 and 90847 are 50-minute codes billed once, under the identified patient. Can BastionGPT write my family and couples therapy notes? Yes. Give it a few bullets after session, who attended, the cycle you worked, interventions, each person's response, and it drafts the conjoint note for your review, with conflict described as process rather than transcript. It can also check a finished note for a missing patient-present statement, missing per-attendee risk, or individual disclosures that should not sit in the shared narrative. Medicare's record guidance accepts computer-generated notes; you still sign and date what you submit. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, NCD 70.1, Consultations With a Beneficiary's Family and Associates : family counseling is covered only where "the primary purpose of such counseling is the treatment of the patient's condition," with observing family interaction and assisting the family in managing the patient as the named qualifying uses. CMS, Medicare & Mental Health Coverage, MLN1986542 (March 2026) : family psychotherapy with or without the patient present, with patient treatment as the primary purpose; MFT and MHC independent billing since January 1, 2024. CMS Medicare Coverage Database, Billing and Coding Article A56937 : 90846 is used when the patient is not present, 90847 when present; both "may not be reported for services less than 26 minutes"; neither pertains to consultation with paid institutional staff. Novitas Solutions, Billing and Coding Article A57480 : the family codes are for psychotherapy treating a mental disorder, not family history taking or E/M counseling; its medical-necessity list carries no relationship-problem Z codes. First Coast Service Options, LCD L33252, Section IV : family psychotherapy treats the identified patient's mental illness "by intervening in a family system"; a family member is anyone who spends significant time with and psychologically supports the patient, including a significant other; periodic documentation of capacity to benefit, target symptoms, and measurable goals. CGS, Billing and Coding Article A57065 : required documentation elements including type of service (individual, group, family), the time element where duration determines the code, and per-encounter clinical note contents. CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes "may be in any form or format"; reviewers do not request HIPAA psychotherapy notes. CMS, Medical Record Maintenance & Access Requirements, MLN4840534 (January 2026) : records supporting Medicare services kept 7 years from the date of service under 42 CFR 424.516(f); computer-generated notes accepted, signed and dated. 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and the designated record set. 45 CFR § 164.512(e) : judicial-proceeding disclosures; court orders versus subpoenas, notice, and protective orders. 45 CFR § 164.524 : the individual access right and the substantial-harm standard for records that reference another person. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : an estimated $580 million in improper payments; 128 of 216 sampled enrollee-days failed at least one requirement. HHS Office of Inspector General, On-Site Psychological Services audit (2020) : 111 of 120 sampled psychotherapy claims noncompliant; treatment-plan defects and unspecified therapeutic maneuvers; estimated $3.3 million overpayment. AAMFT, Code of Ethics, effective January 1, 2026 : in couple, family, or group treatment, no disclosure or record access without written authorization from each competent individual, and no revealing one participant's confidences without prior written permission (Standards 2.2, 2.3). Information and Privacy Commissioner of Ontario, PHIPA Decision 158 (2021) : a family counselling record is not "dedicated primarily" to one participant; a lone requester receives reasonably severable personal information plus the family's communal information. College of Registered Psychotherapists of Ontario, Professional Practice Standards (revised December 2025) : joint records for couples and families attending in the same combination; separate files or sub-files when combinations vary; retention 10 years. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct (2024) : a client may be "a person, couple, family, or other group of individuals"; consent documentation in the record; retention 10 years after age 18 or last contact. MBS Online, Explanatory Note MN.7.5, Family and Carer Participation : claims raised against the patient, patient not in attendance, a written record of the patient's consent, and the items not for treating the family member or carer. Australian Government, Family and Carer Participation factsheet : the items "cannot be used in whole, or in part" for family therapy, group therapy, or couples therapy and relationship counselling. Psychology Board of Australia, Code of Conduct (effective December 2025) : record-keeping expectations and retention of 7 years, or beyond the 25th birthday for clients under 18. ## Related templates Psychotherapy Progress Note Group Therapy Note SOAP Note Treatment Plan Telehealth Psychotherapy Note Consent for Treatment of a Minor Parent & Caregiver Collateral Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/group-therapy-note All note templates # Group Therapy Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A group therapy note is the progress note a clinician writes for each member of a group psychotherapy session. It pairs a brief group-level summary of the session's topic and interventions with individualized documentation of that member's participation, response, and progress. Group therapists write one note per member after every session. A typical note runs 100 to 250 words per member. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Group therapists, psychologists, counselors, clinical social workers Audience Treating clinician, group facilitators, supervisors, payers, auditors Typical length 100 to 250 words per member · 5 to 10 minutes per member by hand (clinical team estimate) Format family Two-layer progress note: group summary plus individual sections (compare: SOAP, DAP) When it's used After each group psychotherapy session, one note per member Standards context No law mandates a group note form; multi-client confidentiality and payer individualization rules shape it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a group therapy note? A group therapy note is the progress note a clinician writes for each member of a group psychotherapy session. Unlike SOAP or DAP, it is not a named format with a single origin: no professional body has standardized a group note form, and group treatment itself is older than every charting framework used to document it (Joseph Pratt was running structured tuberculosis classes in Boston by 1905, six decades before the problem-oriented record gave medicine SOAP). What defines the document is a two-layer structure. One layer summarizes the session the whole group attended: topic, curriculum, facilitator interventions. The other documents one member's participation, response, mental status, risk, and progress. Clinicians and payers also call it a group psychotherapy progress note, group session note, or group counseling note. Two clarifications save a lot of confusion. First, for clinical and payer purposes there is one note per member, filed in that member's record; a narrative about the whole group with no per-member content is an attendance or program record, not clinical documentation of anyone's treatment. Medicare contractor guidance describes exactly this split: a common portion for the session and a second portion for each patient's record . Second, each member's note must stand alone without exposing the other members. Every member can request their own record, so peers appear by role ("another member"), never by name. A group therapy note is a progress note, part of the standard record; it is not a psychotherapy note , the separately protected category of private process notes under HIPAA, a definition that explicitly covers group counseling sessions. ## Who uses group therapy notes and when Anyone who runs a therapy group writes them, and that is most of the field: in SAMHSA's national facilities survey , 86 percent of US mental health treatment facilities offered group therapy. Group therapists in outpatient practice, facilitators of CBT and DBT skills groups, process group leaders, and clinicians in IOP and PHP programs, SUD treatment, inpatient units, and community mental health all write one note per member after every session. For a session with one client, use an individual progress note ; for partners or families treated together, a family or couple note carries different consent and access rules. Psychoeducational and support groups still deserve documentation, but whether they count as billable group psychotherapy depends on the payer and on whether the session was active treatment. ## Group therapy note structure: what goes in each section A group therapy note has two layers. The group summary describes the session everyone attended and can read the same across member notes. The individual sections describe one member's experience of it and must be different in every note. Keeping the layers visibly separate is what makes each note a distinct record of a distinct service. Group header & attendance. Group name and type, date, session number, facilitator and credentials, the count of members present, and start and stop times. Pitfall: listing other members by name in this member's note; the roster belongs in an attendance log, not in every chart. Group summary: topic & interventions. The shared layer: the session topic or curriculum item, what you did as facilitator, and group process in general terms. Pitfall: letting the shared summary be the whole note; the individualized sections below are what auditors read for. Individual participation & response. This member's engagement, the work they did in session, and how they responded to interventions. Describe peer interaction by role ("another member", "a peer"), never by name. Pitfall: quoting or describing another member's disclosure in this member's record. Mental status & risk. Member-specific observations and a risk statement each session, the same monitoring you would document in individual work . Pitfall: silence on risk; a group setting does not suspend individual monitoring. Progress toward treatment plan goals. Tie the session to the member's own numbered goal on their treatment plan . Pitfall: substituting the group's purpose for the member's goal; medical necessity is judged member by member. Plan & next steps. Homework, the next session, and any indicated individual follow-up. Pitfall: a bare "continue group"; say what continuing group is meant to achieve for this member. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Group name/type: Facilitator(s)/credentials: Members present (count): Start/stop time: GROUP SUMMARY (may be shared across member notes) Topic / curriculum: Facilitator interventions: INDIVIDUAL (this member only) Participation & response: Mental status & risk: Progress toward plan goal(s): Plan / next steps: Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample group therapy note Scenario: adult member of an eight-person CBT skills group for anxiety, session 5 of a planned 10. All details are fictional. Client: J.R., 41 · Date: 07/14/2026 · Group: Managing Anxiety (CBT skills), session 5 of 10 · Members present: 8 · Facilitator: L. Ortega, LPC · Start/stop: 5:02 pm to 6:00 pm Group summary: Session topic: cognitive restructuring, identifying and testing anxious predictions. Facilitator reviewed thought-record homework, taught the evidence-for and evidence-against technique with a worked example, and led a paired practice exercise. Group engagement was active; all members participated in at least one exercise. Participation & response: J.R. arrived on time and volunteered a work-presentation example, identifying "I will freeze and everyone will notice" as the hot thought. With coaching, generated three pieces of disconfirming evidence and re-rated belief in the prediction from 90% to 55%. Offered supportive feedback to another member during paired practice. Affect anxious early in session, settling with task focus; speech and behavior otherwise unremarkable. Mental status & risk: Alert, oriented, engaged; no abnormal movements; thought process organized. Denies suicidal ideation and self-harm urges; no risk indicators observed; routine monitoring continues. Progress: Working Goal 2 (reduce avoidance of feared work situations). Reports volunteering to answer one question in a meeting this week, the first time since intake. Plan: Continue weekly group. Homework: one thought record on a live work situation. Re-administer GAD-7 at session 6. No individual session indicated. Signed: L. Ortega, LPC, 07/14/2026. This sample is fictional and for educational purposes. It does not describe a real patient. The group summary paragraph may repeat across member notes for the same session; everything after it is individualized. ↑ Back to the template and downloads ### Why this sample works The group summary and the individual sections are visibly separate, so a reviewer can see the distinct service this member received, which is the defense against a cloned-note finding. Other members appear only by role ("another member"), never by name, so the note can be released on a records request without exposing anyone else. Progress ties to the member's own numbered treatment plan goal, which keeps the medical-necessity chain intact. Risk is stated for this member even though low, so the record shows monitoring rather than silence. The attendance count and start and stop times support the billed session without putting the group roster in this member's chart. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Every member's note is part of that member's designated record set: the member can request it, payers can audit it, and on a records request the presence of other people in the story does not automatically block access. Under 45 CFR 164.524 , denying access because a note references another person requires a licensed professional's judgment that access is reasonably likely to cause that person substantial harm. Write each note assuming it will be read, and keep the group roster in a separate attendance record rather than in every chart. Keep private process material out entirely; it belongs in segregated psychotherapy notes , which Medicare reviewers are instructed not to request. For substance use disorder groups, 42 CFR Part 2 adds a distinctive subpoena rule: a subpoena alone does not permit release, and a Part 2 court order is also required. For payers, individualization is the requirement and the format is a convention. One Medicare contractor's billing article, CGS A57065 , describes an optional two-portion group note: a common portion with the date, session length, and key issues ("Names of the patients in the group should not appear in this group note"), and a per-member portion on that patient's participation and any significant changes. Arizona Medicaid's June 2026 billing guidance is blunter: "Documentation cloning is strictly prohibited," and group progress notes must show each member's response and how members and visits differ. Tie every note to the member's own treatment plan goal, and document session duration even though 90853 is not a timed code, because local policies and other payers still expect the time to be visible. ↑ Back to the template and downloads ## Common group therapy note errors auditors flag Group documentation fails audits for individualization problems more than format problems. A national OIG audit of Medicare psychotherapy claims estimated $580 million in improper payments, with 128 of 216 sampled enrollee-days failing at least one requirement, and state payers now name group notes directly: Arizona Medicaid's June 2026 guidance requires exact duration, diagnosis, each member's response to services, and "documentation showing the differences and the needs of the patient for each visit." The BastionGPT Clinical Advisory Board sees the same errors most often in group therapy note reviews: Identical notes across members. The shared summary is the whole note, so every member's record reads the same. Reviewers treat cloned documentation as failing to show a distinct service for a distinct patient. Another member named in the chart. Peer names, and details that identify a peer without naming them, put one client's protected information in another client's releasable record. No individualized response. "Participated in group" describes attendance, not treatment. The note needs this member's engagement, work, and response to the interventions. No link to the member's own plan. The group's purpose is not the member's goal; medical necessity is judged member by member against their own treatment plan. Attendance, time, and signature gaps. Missing session length, missing attendance support, and missing or late signatures were among the most common findings in the OIG psychotherapy audits. ### Group therapy notes in the US, Canada, and Australia Aspect United States Canada Australia Status No law mandates a group note form; Medicare contractor billing articles describe an optional common-portion-plus-individual-portion format College standards bind by profession; Ontario psychologists must keep a separate record per group participant, while registered psychotherapists may keep joint records with controlled access Psychology Board code of conduct binds registered psychologists; MBS rules govern Better Access group services Terminology Group therapy note, group psychotherapy note (90853) Group or joint record, session note Group psychological therapy service note (MBS group items) What changes 90853 is untimed nationally and has no national size rule; local policies differ (one contractor caps groups at 12) No national billing code; provincial programs and college file-architecture rules lead Better Access groups run 4 to 10 members with set duration tiers, and the provider must record the date, time, and people attending Retention No universal HIPAA rule for patient charts; state law and payer contracts govern Ontario colleges expect 10 years from last contact, or 10 years after the client turns 18, whichever is later Psychology Board: 7 years from last entry for adults; until the 25th birthday for clients under 18 The two-layer structure travels well across all three countries. What changes is the payer layer and the file architecture: who may share a record, how attendance is proved, and how long everything is kept. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on group therapy notes. Draft a complete group therapy note from a few bullets about the session and this member, from dictation, or from a transcript. Write the group summary once, then generate an individualized note for each member documenting their own participation and response. Check a finished note for cloned text, missing individualized content, or another member's name before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I need a separate note for every group member? Yes. Each billed member needs their own note, in their own record, documenting their participation and response. A single narrative about the whole group is an attendance or program record, not clinical documentation for anyone. One Medicare contractor's billing article describes the standard architecture: a common portion plus a per-member portion. In Ontario the file structure depends on your college: psychologists must keep a separate record for each group participant, while registered psychotherapists may keep joint records with controlled access. Can group therapy notes be identical for every member? No. A shared factual paragraph about the session topic and facilitator interventions is acceptable, but everything after it must be individualized. Arizona Medicaid's June 2026 billing guidance states "Documentation cloning is strictly prohibited" and requires notes to show the differences among members and visits. Cloned notes read to a reviewer as one service copied to many charts. Can I mention other group members in a client's note? No federal rule bans it, but payer policy can. CGS, the Medicare contractor for Kentucky and Ohio, instructs that "Names of the patients in the group should not appear in this group note," meaning the common portion shared across records. The portable practice: describe peers by role ("another member"), never by name, and keep the roster in a separate attendance record rather than in each member's chart. Do I need start and stop times for group therapy sessions? 90853 is not a time-based code, and Medicare contractor guidance applies its start-and-stop-times rule to the individual psychotherapy codes 90832, 90834, and 90837. Record the session length anyway: the CGS group-note format documents "length of time for the session," Australian MBS group items are duration-tiered, and auditors read missing time as missing support. What is the minimum or maximum group size for 90853? No national CMS rule sets either number. The caps you see quoted are local or foreign rules: First Coast's LCD says groups should stay at "maximum 12 people" in its jurisdiction, and Australia's Better Access group items run 4 to 10 members. Check your own contractor and payer before treating any number as the rule. How is a group therapy note different from an individual progress note? An individual progress note documents a session that belongs to one client. A group therapy note adds a shared layer describing the session everyone attended, and a constraint individual notes never face: it must document this member's experience without exposing anyone else's. Billing differs too. Individual psychotherapy codes are time-based; 90853 is untimed. Does a psychoeducational group count as group psychotherapy? It depends on the payer and on what actually happened. For Medicare, group psychotherapy must be active treatment ordered as part of the member's treatment plan; teaching, recreation, and social activity do not become psychotherapy because a therapist leads them. A diagnosis-linked, clinically active skills group can qualify. Australia's Better Access group services explicitly pair psychoeducation with therapy such as CBT, so the same session can sit differently in different systems. What happens if a member requests their record or my group notes are subpoenaed? The member generally gets their own note. A reference to another person does not automatically block access: under 45 CFR 164.524 , denial on that ground requires a licensed professional's judgment that access would likely cause the other person substantial harm. For substance use disorder groups under 42 CFR Part 2, a subpoena alone is not enough; a Part 2 court order is also required. Notes written by role, with the roster kept separately, release cleanly. Can BastionGPT write the notes for my whole group? Yes. Give it the group summary once plus a line or two per member, and it drafts an individualized note for each member for your review. It can also check finished notes for cloned text, missing individualized content, or another member's name before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CGS, Billing and Coding Article A57065 : the optional two-portion group note format, and "Names of the patients in the group should not appear in this group note." CMS Medicare Coverage Database, Billing and Coding Article A57520 : start and stop time documentation applies to psychotherapy codes 90832, 90834, and 90837; 90853 is covered with ICD-10 medical-necessity linkage. First Coast Service Options, LCD L33252 : the local "maximum 12 people" group size guidance and the active-treatment-plan requirement. 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition, which covers notes from group counseling sessions kept separately from the record. 45 CFR § 164.524 : the access right and the substantial-harm standard for records that reference another person. 42 CFR § 2.61 : for Part 2 substance use disorder records, a subpoena alone does not permit disclosure without an authorizing court order. CMS, Medicare Program Integrity Manual, Chapter 3 : reviewers do not request HIPAA psychotherapy notes; providers may submit an extraction of non-protected information. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : an estimated $580 million in improper payments, with time and signature failures among the causes. AHCCCS, Claims Clues, June 2026 : "Documentation cloning is strictly prohibited," with required group note elements including each member's response. SAMHSA, National Mental Health Services Survey 2019 : group therapy offered by 86 percent of US mental health treatment facilities. MBS Online, Explanatory Note MN.6.2 : Better Access group sizes, the four-scheduled-one-absent exception, and the contemporaneous record of date, time, and people attending. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : a separate record for each group participant, and Ontario retention periods. College of Registered Psychotherapists of Ontario, Clinical Records standard : joint records and what a single participant may access without the others' consent. Psychology Board of Australia, Code of Conduct : Australian record-keeping and retention expectations for psychologists. ## Related templates Psychotherapy Progress Note SOAP Note DAP Note Treatment Plan Telehealth Psychotherapy Note Family & Couples Therapy Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/therapy-session-note All note templates # Individual Therapy Session Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An individual therapy session note is the clinical record of one one-on-one psychotherapy session: what was addressed, the interventions used, the client's response, current risk, and the plan for the next session. Therapists, psychologists, counselors, and clinical social workers write one after every session, in SOAP, DAP, BIRP, or narrative form. Most run 150 to 350 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers Audience Treating clinician, supervisors, care team, payers, auditors Typical length 150 to 350 words · 10 to 20 minutes by hand (clinical team estimate) Format family Progress note, any structure (SOAP, DAP, BIRP, or narrative) When it's used After every individual psychotherapy session Standards context Content is set by payer policy and board standards; no law mandates a format or deadline On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an individual therapy session note? An individual therapy session note is the working record of one one-on-one psychotherapy session, the note behind time-based CPT codes 90832, 90834, and 90837. It belongs to the progress-note family that grew out of Lawrence Weed's problem-oriented medical record, the same lineage as the SOAP note . Clinicians call it a session note, therapy note, encounter note, or simply the note; payer manuals call it the clinical record of each encounter. Whatever the label, it is the same document. Two rules do most of the compliance work. First, no law or Medicare manual mandates a format: Medicare's Program Integrity Manual says progress notes "may be in any form or format," so SOAP , DAP , BIRP , and narrative structures are all acceptable containers for the same content. Second, every billed session needs its own note: Carelon's documentation standards state that "All encounters must have a progress note to support the service billed." A session note is part of the standard chart, not a HIPAA psychotherapy note; that separate private-notes category and its authorization rules are covered on the psychotherapy progress note and psychotherapy notes authorization pages. ## Who uses individual therapy session notes and when Every licensed clinician delivering individual psychotherapy writes one per session: in private practice, group practice, community mental health, and platform-based work. The format is a fit-for-audience choice. When the note will be read outside your practice, in integrated care or routine payer review, a structured format such as SOAP reads fastest; solo practices often prefer DAP or a disciplined narrative. Sessions with more than one client in the room follow different confidentiality rules; see the couples and family and group therapy note pages. For virtual sessions, the telehealth therapy note adds modality and location documentation. ## Individual therapy session note structure: what goes in each section Whatever format you choose, auditors and payer documentation standards read the same anatomy. Each block below carries the pitfall that most often undermines it. Header and identifiers. Client identifier, date of service, service and CPT code, session number, and setting. Pitfall: a date of service that does not match the claim; payer record reviews check the two against each other. Time. Start and stop times or total time for time-based codes: 16 to 37 minutes supports 90832, 38 to 52 supports 90834, and 53 or more supports 90837. Pitfall: missing time was the most common deficiency in OIG's national psychotherapy audit, 60 of 216 sampled days. Presentation and symptoms. The client's report: current symptoms, stressors, functioning, and change since last session. Pitfall: mood adjectives with no anchor; pair the word with a report or behavior a reviewer can picture. Mental status observations. A brief, session-relevant mental status entry: appearance, engagement, speech, affect. Pitfall: a full boilerplate exam pasted into every note; identical text across sessions reads as cloning. Interventions. The named technique you used and the treatment plan goal it served. Pitfall: "provided supportive therapy" with no named technique; unspecified interventions are a recurring OIG finding. Response and progress. How the client responded in session and where they stand against the goal, with scores when you use them. Pitfall: notes that never say whether anything changed, which breaks the medical-necessity chain. Risk status. A statement either way, even when risk is low. Pitfall: silence; the record should show monitoring, not absence. Plan and signature. Next steps, homework, coordination, next appointment, then your name, credentials, and date signed. Pitfall: a bare "continue treatment" and a signature added long after the session; 31 of OIG's 216 sampled days were missing signatures. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: Individual psychotherapy [CPT] Setting: [office/telehealth] Start/stop time: Total time: Presentation and symptoms: Mental status observations: Interventions (technique + goal #): Response and progress: Risk status: Plan / next steps: Next appointment: Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample individual therapy session note Scenario: adult client, social anxiety disorder, CBT with graduated exposure, session 8 of a planned 16. All details are fictional. Client: M.R., 29 · Date: 07/09/2026 · Session: 8 · Service: Individual psychotherapy, 90834, office · Start/stop: 11:03 to 11:55 (52 min) Presentation and symptoms: Client reported completing both planned exposure tasks this week (ordering at a counter, asking one question in a team meeting), with anticipatory anxiety 6/10 that fell to 3/10 during each task. Reported steady sleep and appetite. Denied suicidal ideation, self-harm urges, and substance use. Mental status observations: On time, casually dressed, cooperative. Speech normal in rate and volume. Mildly anxious at session start, settling by mid-session; affect congruent and full range. Engaged in agenda setting without prompting. Interventions: Reviewed exposure log against Goal 2 of the treatment plan (complete graduated social-exposure hierarchy). Cognitive restructuring of the prediction "everyone will notice I'm nervous," comparing predicted versus actual outcomes. Planned the next hierarchy step collaboratively. Response and progress: Client identified the prediction-outcome gap unprompted and rated confidence in the next step 7/10. GAD-7 today: 8, down from 14 at intake. Progress is consistent with the plan's 16-session timeline. Risk status: Denied suicidal ideation and self-harm; no risk indicators observed. Routine monitoring continues. Plan: Continue weekly CBT. Homework: next hierarchy step (offering one comment in the staff meeting) with a thought record. Re-administer GAD-7 at session 10. Next appointment 07/16/2026. Signed: J.L., LPC · 07/09/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Time sits in the header: start and stop times supporting a 38-to-52-minute code, with no hunt through the chart. The intervention names a technique and cites a numbered treatment-plan goal, the linkage reviewers read as medical necessity. Progress is measurable twice over: an exposure log and a GAD-7 trend, so the next note has something concrete to evaluate. Risk is documented either way, so the record shows monitoring rather than silence. Signed with credentials the same day, inside every payer and platform deadline described on this page. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Start with the deadline question, because it is the one this note type generates most. No statute and no CMS manual sets a 24-, 48-, or 72-hour completion rule. Medicare's Program Integrity Manual expects services to be documented "at the time they are rendered," treats anything after the fact as an amendment or delayed entry with its own identification rules, and lists excessive late entries among the signs reviewers read as possible falsification. The binding hour-count comes from contracts and platforms: Optum expects entries on the date of service, with a late-entry notation when an entry lands more than 24 hours after it; Carelon requires documentation complete before the claim is submitted; Headway sets 72 hours, tightening to 48 for Medicare and Medicaid clients; and Rula requires signed notes within 48 hours. So the honest answer to "what is my deadline" is: whatever your payer contract or platform says, on top of a federal expectation that notes are written promptly and late additions are labeled, never backdated. Two more numbers belong in your planning. Optum publishes its treatment-record audit math: an 85% passing score, a written Corrective Action Plan below 85%, a CAP plus re-audit within six months below 80%, and record retention of at least seven years. And the alignment idea vendors call the golden thread is real payer policy wearing informal clothes: the term appears in no statute or Medicare manual we reviewed, but reviewers do expect each session's interventions to trace to the treatment plan and diagnosis. Even there, specifics vary: OIG found that only three of seven Medicare contractors had coverage policies requiring a treatment plan at all, so check your own contractor's rules rather than assuming a national standard. The wider payer mechanics, records requests, addenda, and time-rule detail live on the billing documentation page. ↑ Back to the template and downloads ## Common individual therapy session note errors auditors flag The failure data here is unusually specific. In OIG's national psychotherapy audit , 128 of 216 sampled enrollee days failed Medicare requirements, and the deficiency list was led by undocumented time (60 days), incomplete or missing treatment plans (43), and missing signatures (31), supporting an estimated $580 million in improper payments. In Medicare's 2024 reporting-period data for outpatient psychiatric services, insufficient documentation drove 78.3% of improper payments. Commercial audits run on the same rails with published thresholds, such as Optum's 85% passing score. The BastionGPT Clinical Advisory Board sees the same errors most often in individual session note reviews: No time support. Start and stop times or total time missing for 90832, 90834, or 90837, the single most common deficiency in the OIG sample. The note trails the claim. Documentation finished after the claim went out, or a late entry with no late-entry notation, date, and author. Interventions without a name or a goal. Generic phrases with no technique and no link to a treatment-plan goal, the second-largest OIG deficiency group. Authentication gaps. Missing signatures or credentials, or signing sessions in batches long after the dates of service. Cloned notes. Copy-forward text repeating across your own sessions, which reads to a reviewer as a record that does not support a distinct billed service. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on individual therapy session notes. Draft a complete session note from a few bullets, a dictation, or a pasted transcript, in SOAP, DAP, BIRP, or narrative form. Summarize a long session into a payer-ready note with time, interventions, response, risk, and plan fields complete. Check a finished note for missing time support, risk language, goal linkage, or signature details before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long do I have to write a therapy session note? No US law or CMS manual sets a 24-, 48-, or 72-hour deadline. Medicare expects documentation "at the time they are rendered" and treats later additions as labeled delayed entries. The binding numbers come from contracts and platforms: Optum expects a late-entry notation past 24 hours, Headway sets 72 hours (48 for Medicare and Medicaid), Rula sets 48 hours, and Carelon requires the note complete before the claim goes out. The billing documentation page covers amendments and addenda. Is a therapy session note the same as a progress note? Yes. The session note you write after an individual session is a progress note, part of the standard clinical record; payer manuals just say clinical record of each encounter. The psychotherapy progress note page covers the record class in depth, including why it is legally distinct from HIPAA's separately kept psychotherapy notes and their authorization rules . Which format should I use: SOAP, DAP, BIRP, or narrative? Whichever fits your setting; no law or payer mandates one. Reviewers check content, not headings: time, symptoms, named interventions tied to plan goals, response, risk, plan, signature. Structured formats such as SOAP travel best in integrated care; DAP and BIRP are leaner for outpatient therapy; a disciplined narrative works when nothing else fits. What has to be in the note for insurance? Payer documentation standards converge on the same list: a date of service matching the claim, session time (start and stop or total), current symptoms, the named intervention and the treatment-plan goal it addressed, the client's response and progress, next steps, and your signature with credentials and date. Carelon's standards also expect the units billed to be supported by the documented time, and Optum's expect documentation of missed appointments. Does every session really need its own note? Yes. Carelon's documentation standards state that "All encounters must have a progress note to support the service billed," and a billed session without its own note is recoupable on audit. Document missed appointments too: Optum's treatment-record requirements include them, and the pattern matters clinically. Is the golden thread a legal requirement? The term is vendor vocabulary; it appears in no statute or Medicare manual we reviewed. The expectation it describes is real payer policy: each session's interventions should trace to the diagnosis and treatment plan . The specifics vary more than most pages admit; OIG found only three of seven Medicare contractors required a treatment plan in their coverage policies. How long should an individual therapy session note be? Most run 150 to 350 words and take 10 to 20 minutes by hand (clinical team estimate); there is no peer-reviewed study of individual session note length or writing time, so treat any precise figure as an estimate. The documentation burden around it is measured, though: in a large physician survey, psychiatrists spent the highest share of their week on administration, 20.3%, and ambulatory physicians in a time-motion study spent 49.2% of the office day on EHR and desk work. Who can write and bill for an individual session note? Licensure and payer enrollment decide. In the US, marriage and family therapists and mental health counselors can bill Medicare independently since January 1, 2024; pre-licensed clinicians document under state board supervision rules, with co-signature where required. In Ontario, CRPO registrants keep clinical records at least 10 years past the last interaction (longer for minors), and in Australia the Psychology Board's code requires accurate, up-to-date records, with Better Access adding a written report to the referrer at the end of each course of treatment. Can BastionGPT write my session note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces a session note in your preferred format for your review. It can also convert notes between SOAP, DAP, BIRP, and narrative, and check a draft against the checklist on this page. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, Medicare Program Integrity Manual, Chapter 3 : documentation expected "at the time they are rendered," amendment and delayed-entry rules, and signature requirements. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : 128 of 216 sampled enrollee days deficient; time, treatment-plan, and signature counts; $580 million estimate. CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 reporting-period improper-payment data, including the 78.3% insufficient-documentation share. CMS Medicare Coverage Database, Billing and Coding Article A57520 : time documentation and code-time ranges for 90832, 90834, and 90837. 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and the progress-note content it excludes. Optum, Treatment Record Documentation Requirements : per-note elements, the 24-hour late-entry notation, the 85% and 80% audit thresholds, and seven-year retention. Carelon Behavioral Health, Provider documentation standards : one note per billable encounter, completion before claim submission, and progress-note elements. Headway, Compliance deadlines for clinical documentation : the 72-hour standard and 48-hour Medicare and Medicaid deadline. Rula, Timely documentation policy : notes signed and complete within 48 hours. CMS, Marriage and family therapists and mental health counselors : independent Medicare billing effective January 1, 2024. College of Registered Psychotherapists of Ontario, Clinical records standard : per-session documentation and 10-year retention. Psychology Board of Australia, Code of conduct : Australian record-keeping expectations. Services Australia, Mental health treatment referrals : Better Access course-of-treatment reporting to the referrer. Sinsky et al., Annals of Internal Medicine (2016) and Woolhandler and Himmelstein, Int J Health Services (2014) : physician documentation and administration time, cited as general-medicine figures. ## Related templates Psychotherapy Progress Note SOAP Note DAP Note BIRP Note Treatment Plan Start-Stop Time Attestation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pie-note All note templates # PIE Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A PIE note is a three-part progress note format that organizes a session into Problem, Intervention, and Evaluation. Each entry hangs off a numbered problem from the client's problem list; there is no separate assessment or plan section, so interpretation and next steps live inside the entry. Behavioral health and nursing-adjacent teams use it for problem-focused charting. A typical PIE note runs 100 to 250 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, psychiatric nurses, community behavioral health teams Audience Treating clinician, care team, supervisors, payers, auditors Typical length 100 to 250 words · 10 to 15 minutes by hand (clinical team estimate) Format family Structured progress note (compare: SOAP, DAP, BIRP; APIE and PIRP are sibling variants) When it's used After each session, one P-I-E entry per problem addressed Standards context A documentation convention, not a mandated form; payers regulate the record's content, not its headings On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a PIE note? PIE stands for Problem, Intervention, Evaluation. The format began in nursing: staff on a 35-bed medical unit at Craven County Hospital in New Bern, North Carolina built it to unify the care plan and progress notes into one problem-focused record, publishing it in the June 1985 Quality Review Bulletin (Siegrist, Dettor, and Stocks). Behavioral health teams later adopted the same three-section pattern for session charting, and close variants add sections: APIE opens with an assessment step, while PIRP and GIRP bring back an explicit plan. Some clinicians also call it PIE charting or a problem-focused note. Two clarifications save confusion. First, a PIE note is not the PIE system social workers learn for assessment: the Person-in-Environment classification (Karls and Wandrei, NASW Press, 1994) is a taxonomy for describing social functioning problems, not a progress note format. Second, no law or payer mandates PIE or any other heading scheme. Medicare's Program Integrity Manual states that progress notes "may be in any form or format," and what regulators and payers police is the content: identifiers, interventions, response, time, and signature. A PIE note is a progress note, part of the standard clinical record, not a psychotherapy note under HIPAA. ## Who uses PIE notes and when Because every entry hangs off a numbered problem, a reviewer can follow one problem across weeks of notes without rereading whole sessions. That suits problem-list settings: community mental health programs, inpatient psychiatric and other nursing-adjacent units where PIE charting began, case management, and clinics where several providers work the same treatment plan. A session that touches several problems produces several short entries rather than one long narrative. If your chart is shared with medical teams that expect separate observations and an explicit plan, SOAP reads more completely; solo outpatient work that tracks goals forward often prefers DAP , and payer-facing programs that want the intervention-and-response thread explicit choose BIRP . ## How to write a PIE note in three steps Problem: name the specific problem this entry addresses, numbered to match the client's problem list or treatment plan. Intervention: record what you did about that problem in this session, named specifically. Evaluation: record how the client responded and what happens next. Each section in more detail, with the pitfall that most often undermines it: P: Problem. The problem being worked this session, stated concretely and keyed to the problem list ("Problem 2: depressed mood with social withdrawal"). Include current status: the client's report, your observations, and any scores. Client quotes and newly reported symptoms belong here too, because PIE has no separate Subjective section. Pitfall: a vague problem ("client struggling") or a new, unnumbered problem each session; an entry that tracks to nothing on the treatment plan breaks the medical-necessity chain. I: Intervention. The specific therapeutic work you delivered against that problem: the technique used, the skill taught, the material reviewed. Pitfall: naming only the modality ("provided CBT"); auditors read this section for the actual intervention, not the brand of therapy. E: Evaluation. The client's response to the intervention and progress on the problem, measurable where possible. PIE has no plan section, so next steps, homework, and any treatment-plan change live here, with dates. Pitfall: grading attendance ("client engaged well") instead of response, and leaving next steps nowhere in the record. A session that works several problems gets several short entries: P1-I1-E1, then P2-I2-E2. Combining two problems under one P is acceptable when the interventions genuinely overlapped; clarity is the test. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: [individual/group/telehealth] Start/stop time: Problem # (from treatment plan): P (Problem): I (Intervention): E (Evaluation): Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample PIE note Scenario: adult client, major depressive disorder, community mental health clinic, behavioral activation, session 8 of a planned 16. All details are fictional. Client: R.L., 41 · Date: 07/09/2026 · Session: 8 · Service: Individual psychotherapy, 45 min, community clinic · Start/stop: 14:02 to 14:47 P: Problem 2: depressed mood with social withdrawal (treatment plan of 05/14/2026). Client described the week as "flat, but steadier than June," rating mood 4/10 most days. Attended one social outing since last session and left after 20 minutes. PHQ-9 today: 12, down from 16 at intake. Denied suicidal ideation, self-harm urges, and substance use. I: Reviewed the behavioral activation log line by line. Used graded task assignment to split the community college enrollment goal into three steps and rehearsed a coping statement for urges to leave social situations early. Scheduled two pleasant activities with the client for the coming week. E: Client completed 4 of 7 planned activities this week, up from 2 the week prior, and connected leaving the outing early to the avoidance pattern identified in session 6. Progress on Problem 2 is measurable and consistent with the PHQ-9 trend. Next steps: client will request her transcript (step one of enrollment) before next session; continue weekly sessions; re-administer PHQ-9 at session 10. Risk: denies ideation; continue routine monitoring. Next appointment 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The problem is numbered and tied to a dated treatment plan, so the entry tracks across sessions and supports medical necessity. Interventions are specific maneuvers (graded task assignment, rehearsal, activity scheduling), not a modality label. Evaluation records a measurable response: activity completion counts and a PHQ-9 trend a reviewer can follow. Next steps are dated and concrete even though PIE has no plan section, so nothing falls out of the record. Risk is addressed even in a low-risk session, and start and stop times support time-based billing without a hunt through the chart. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A PIE note is part of the designated record set: clients can request it, payers can audit it, and other providers may rely on it. Write each entry as if a colleague, a reviewer, and the client will read it, and keep private reflections and hypothesis-level process material out; that content belongs in segregated psychotherapy notes with their own authorization rules . For payer work, the missing plan section is the thing to manage. No payer mandates a heading scheme; what reviewers expect is content that ties the session to a current treatment plan and supports the code billed. Reference the plan's problem numbers in P, and treat plan currency as a billing requirement: Arizona's Medicaid program, for example, denies claims when the treatment plan on file is more than a year old at the date of service (edit MD466). For time-based psychotherapy codes (90832, 90834, 90837), Medicare contractor guidance expects start and stop times or total time in the record, and timed codes billed in 15-minute units, such as H0004, need the note's documented duration to support the units claimed. Telehealth sessions add modality, consent, and client-location details on top of whichever format you chart in. ↑ Back to the template and downloads ## Common PIE note errors auditors flag Audit findings target the note's content, never the choice of PIE over SOAP. In Medicare's 2024 reporting-period data for outpatient psychiatric services, insufficient documentation drove 78.3% of improper payments, and a national OIG audit of psychotherapy claims estimated $580 million in improper payments, with missing time documentation and missing signatures among the most common problems. State Medicaid programs publish the same lesson with denial codes attached: Arizona Medicaid's March 2025 documentation workshop shows claims denied when the note lacks patient identifiers (edit MD405), when the code or units billed do not match what the progress note documents (edit MD418), or when no valid treatment plan covers the date of service (edit MD466). The BastionGPT Clinical Advisory Board sees the same errors most often in PIE note reviews: Problem drift. The problem worked in session matches nothing on the treatment plan, or the numbering changed since intake, so no problem can be tracked across entries and the medical-necessity chain breaks. Modality-only interventions. "Provided CBT" or "supportive therapy" without the specific maneuver. Auditors read I for the distinct service behind the billed code; a brand of therapy is not an intervention. Evaluation that grades attendance. "Client engaged well" says nothing about response to the intervention. Record what changed, with a measure where one exists. Next steps nowhere. PIE has no plan heading, so homework, appointments, and plan changes silently drop out of the record unless Evaluation carries them, dated. Units the note cannot support. For timed codes billed in 15-minute units, the duration documented in the note must support the units on the claim, and identifiers and prompt, credentialed signatures round out what reviewers check first. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on PIE notes. Draft a complete PIE note from a few bullets, a dictation, or a pasted session transcript. Convert an existing SOAP, DAP, or BIRP note into PIE format without losing clinical content. Extract the problems, interventions, and responses buried in a long narrative into clean numbered P-I-E entries. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a PIE note be? Most PIE notes run 100 to 250 words per problem entry and take 10 to 15 minutes to write by hand. The format is deliberately brief: one problem, what you did about it, and how the client responded. A session that covers several problems produces several short entries rather than one long note. What is the difference between a PIE note and a SOAP or DAP note? PIE drops the sections the other formats keep separate. SOAP splits the client's report from your observations and ends with a plan; DAP merges those into a Data section but keeps Assessment and Plan. PIE folds everything into the problem statement, the intervention, and the evaluation of response. See the SOAP note and DAP note templates for the side-by-side structures. Is a PIE note the same as the Person-in-Environment system in social work? No. The Person-in-Environment system (Karls and Wandrei, NASW Press, 1994) is a classification taxonomy social workers use to describe problems in social functioning during assessment. The PIE progress note is a charting format that began in nursing. A social worker can use both in the same chart: one classifies problems, the other documents sessions. Where does the plan go in a PIE note? In the Evaluation section, with dates. PIE has no plan heading, so write next steps, homework, and the next appointment at the end of E, and keep the client's treatment plan current: payers can deny claims when no valid plan covers the date of service, whatever note format you use. If your agency wants a visible plan heading, PIRP and GIRP are the sibling formats that add one. Do insurance companies accept PIE notes? Yes. No payer mandates a note format: Medicare's Program Integrity Manual states progress notes "may be in any form or format." Requirements can vary by payer, organization, and jurisdiction, so check your contracts. What reviewers consistently look for is content: the diagnosis, the specific intervention delivered, the client's response, time for time-based codes, and a prompt signature with credentials. Do I write one PIE entry per problem or one per session? Convention is one P-I-E block per problem addressed, numbered to the problem list, so each problem can be followed across sessions. Combining two related problems under one P is acceptable when the interventions genuinely overlapped; clarity is the test, and no payer rule limits how many entries one session produces. Did the PIE note come from nursing? Yes. Nurses on a medical unit at Craven County Hospital in New Bern, North Carolina developed it to unify the care plan and progress notes, and published it in the June 1985 Quality Review Bulletin (Siegrist, Dettor, and Stocks). Behavioral health teams adopted it later, and nursing programs still teach it, sometimes as APIE with an assessment step first. How long do I need to keep PIE notes? There is no universal HIPAA retention rule for patient charts; HIPAA's six-year rule covers required documentation such as policies, not patient records. Retention comes from state, provincial, and payer rules: Ontario's CRPO expects at least 10 years from the last interaction, and New South Wales requires 7 years for adults and until age 25 for minors. Check your jurisdiction and payer contracts. Can BastionGPT write a PIE note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces numbered P-I-E entries for your review. It can also convert SOAP, DAP, or BIRP notes you already have into PIE format. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: Siegrist LM, Dettor RE, Stocks B, "The PIE system: complete planning and documentation of nursing care" , QRB Quality Review Bulletin, June 1985: the format's original publication. NASW Press, Person-in-Environment System Manual (Karls and Wandrei): the social work PIE classification this note format is often confused with. CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes "may be in any form or format." CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 reporting-period improper-payment data, including the 78.3% insufficient-documentation share. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : estimated $580 million in improper payments, with missing time documentation and signatures among the causes. Arizona Health Care Cost Containment System, Behavioral Health Claim and Documentation Technical Errors Workshop (March 2025) : claim edits MD405, MD418, and MD466 for identifier, units/code, and treatment-plan documentation failures. CMS Medicare Coverage Database, Billing and Coding Article A57520 : time documentation expectations for psychotherapy codes 90832, 90834, and 90837. ## Related templates SOAP Note DAP Note BIRP Note Psychotherapy Progress Note Treatment Plan Narrative Progress Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychotherapy-progress-note All note templates # Psychotherapy Progress Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychotherapy progress note is the session-by-session record of psychotherapy in the standard clinical chart: how the client presented, the interventions used, the response, and the plan. Despite the similar name, it is legally distinct from HIPAA's separately kept psychotherapy notes. Most therapy progress notes run 100 to 350 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, psychiatric prescribers Audience Treating clinician, care team, supervisors, payers, auditors Typical length 100 to 350 words · 10 to 20 minutes by hand (clinical team estimate) Format family Umbrella progress-note record (SOAP, DAP, BIRP, and narrative are common structures) When it's used After every psychotherapy session, in the standard clinical record Standards context Content is regulated (privacy status, retention, payment support); the structure itself is convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychotherapy progress note? A psychotherapy progress note documents one psychotherapy encounter: what was addressed, how the client presented, which interventions you used, how the client responded, and what happens next. Unlike SOAP or DAP, it is not a named format with a single inventor. "Progress note" is the umbrella record type, and SOAP , DAP, BIRP, and narrative notes are all common ways to structure one. What regulators and payers govern is the note's content, privacy status, retention, and payment support, not its headings. The naming trap matters more here than on any other note type. Under HIPAA, " psychotherapy notes " are a narrow legal category: the therapist's private process notes, kept separate from the rest of the chart. That definition expressly excludes medication information, session start and stop times, modality and frequency, test results, and summaries of diagnosis, functional status, the treatment plan, symptoms, prognosis, and progress to date. Those excluded items are exactly what belongs in a progress note. So the note you write after each session, the one this page covers, is part of the standard record no matter what your EHR labels it, and labeling it a "psychotherapy note" does not give it HIPAA's special protection (see the psychotherapy notes authorization page). ## Who writes psychotherapy progress notes and when Every licensed clinician delivering psychotherapy keeps one per encounter: therapists, psychologists, counselors, clinical social workers, and psychiatric prescribers documenting therapy time. When insurance is involved, the progress note is where medical necessity lives session to session. Medicare's outpatient psychiatric criteria expect documentation that connects each service to an individualized written treatment plan with expected goals. The structure is your choice: SOAP , DAP , BIRP , or a clean narrative all satisfy the same content expectations when the required elements are present. ## How to write a psychotherapy progress note in five steps Frame the encounter: client, date, service type and length, start/stop or total time, and setting. Presentation and status: symptoms, mental-status highlights, measurable scores, and changes since last session. Interventions: name what you actually did, specifically enough that a reviewer can tell what the service was. Response and progress: the client's response in session, movement toward a numbered treatment-plan goal, and current risk status. Plan, sign, date: next steps, homework, next appointment, then sign with credentials promptly. Each element in more detail, with the pitfall that shows up in federal audit findings: 1: Frame the encounter. Identify the client, date, service, and time support for the billed code. Pitfall: psychotherapy time not documented; missing time is a named cause of improper payment in OIG's national psychotherapy audit. 2: Presentation and status. What you saw and measured: presentation, mental status highlights, scores (PHQ-9, GAD-7), stressors, and changes since last session. Pitfall: vague global statements ("doing better") with nothing observable or measurable behind them. 3: Interventions. Name the actual techniques: cognitive restructuring, exposure with response prevention, behavioral activation, EMDR resourcing. Pitfall: "provided therapy" or "supportive counseling" alone; OIG flagged notes where therapeutic maneuvers were not specified. 4: Response and progress. Tie the session to a numbered treatment-plan goal and record the client's response and current risk status. Pitfall: no link to an individualized plan, which breaks the medical-necessity chain payers look for. 5: Plan, sign, date. Specific next steps, then authenticate: signed, dated, with credentials. Pitfall: unsigned or undated notes, or pasted images of signatures instead of controlled electronic signatures; both appear in federal audit findings. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: [type/length] Start/stop or total time: Presentation & status: Interventions provided: Response & progress toward goal: Risk status: Plan / next steps: Clinician signature & credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with the audit-tested element checklist; the DOCX is the blank template, ready to adapt. ## Sample psychotherapy progress note Scenario: adult client, major depressive disorder, behavioral activation with cognitive work, session 8 of a planned 16. Narrative structure. All details are fictional. Client: F.L., 42 · Date: 07/09/2026 · Session: 8 · Service: Individual psychotherapy, 45 min, in office · Start/stop: 13:02 to 13:47 Presentation & status: On time, adequately groomed, cooperative. Mood reported as "flat but functional," rated 5/10. Affect mildly constricted, brighter when describing weekend hike. Sleep 7 hours, appetite stable. PHQ-9 today: 11 (down from 18 at intake, 13 at session 5). Denied suicidal ideation and self-harm urges. Interventions provided: Reviewed activity log from prior week. Behavioral activation: graded task assignment, adding one social activity and one mastery activity for the coming week. Cognitive restructuring targeting "I'm a burden when I ask for help," using evidence-for/against and a behavioral test (asking brother for a ride). Reinforced paced planning over mood-dependent scheduling. Response & progress toward goal: Completed 4 of 5 scheduled activities this week; reported the hike "actually felt good for once." Engaged actively in restructuring work and generated alternative thoughts with minimal prompting. Progress consistent with Goal 2 of the treatment plan (increase engagement in valued activities to 5 per week). Symptoms consistent with MDD, improving. Risk: denies ideation; no risk indicators observed; continue routine monitoring. Plan / next steps: Continue weekly behavioral activation with cognitive work. Homework: activity schedule with one social and one mastery activity, complete the behavioral test with brother. Re-administer PHQ-9 at session 10; review treatment plan goals at session 12. Next appointment 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Start and stop times support the billed 45-minute psychotherapy code without a separate hunt through the chart. Interventions are named specifically (behavioral activation, graded task assignment, cognitive restructuring), the exact detail OIG auditors found missing in noncompliant psychotherapy notes. Response is recorded separately from intervention, so the note shows the service did something, not just that it happened. Progress is tied to a numbered treatment-plan goal, which is the medical-necessity link payers expect. A measurable score (PHQ-9) makes the trajectory auditable across sessions. Risk is addressed even though the client is low risk, so the record shows monitoring rather than silence. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A psychotherapy progress note is part of the designated record set: clients can request it, payers can audit it, and other providers may rely on it. Keep private reflections and hypothesis-level process material out of it; that content belongs in separately maintained psychotherapy notes with their own authorization rules . In substance-use settings, note that the same progress note can also become a 42 CFR Part 2 record when created by a Part 2 program, which changes disclosure rules without changing what good clinical content looks like. For insurance work, three things carry the audit weight. First, treatment-plan linkage: Medicare's outpatient psychiatric criteria expect an individualized written treatment plan stating the type, amount, frequency, and duration of services, with diagnoses and expected goals. Second, authentication: CMS documentation guidance lists unsigned, undated, or insufficiently detailed progress notes among the failures that trigger denials. Third, time support: the note must document psychotherapy time sufficiently to support the billed code; missing time was a named error in OIG's national psychotherapy audit . On retention, separate the myth from the rules: HIPAA does not set a retention period for patient charts. HHS says state law generally governs, while Medicare separately requires seven years for covered ordering, certifying, and referral documentation. Ontario and Australia set their own floors; see the table below. ↑ Back to the template and downloads ## Common progress note errors auditors flag The federal record is unusually specific about how psychotherapy documentation fails. In Medicare's 2024 reporting-period data for outpatient psychiatric services, 16.1% of payments were improper, a projected $254.5 million, and insufficient documentation drove 78.3% of it. A national OIG audit estimated $580 million in improper psychotherapy payments, and a provider-level OIG audit found 111 of 120 sampled psychotherapy claims noncompliant. The BastionGPT Clinical Advisory Board sees the same patterns most often in progress note reviews: Interventions not specified. "Provided supportive therapy" with no named technique; OIG cited treatment notes where therapeutic maneuvers were not specified. The note does not support the billed service. Psychotherapy time missing, or content too thin to justify the code and length billed. Authentication gaps. Unsigned or undated notes, or pasted signature images instead of controlled electronic signatures. No treatment-plan linkage. Sessions that never reference an individualized plan, its goals, or progress against them. Cloned notes. Copy-forward text repeating across sessions, which reads to a reviewer as a record that does not support a distinct billed service. ### Psychotherapy progress notes in the US, Canada, and Australia Aspect United States Canada Australia Status Content regulated, structure convention: HIPAA sets the privacy split (progress note vs psychotherapy notes); payers set documentation-support rules College standards make record-keeping enforceable (Ontario: CRPO for psychotherapists, CPBAO for psychologists); failure to keep proper records is professional misconduct Psychology Board Code of Conduct (from 1 Dec 2025) plus Ahpra record expectations govern record quality Terminology Progress note, therapy note, session note Clinical record, session note Case note, session note What changes Treatment-plan linkage, signatures, and time support for billed codes Ontario: documents with conclusions or diagnoses authored by a supervisee need both signatures Better Access: no standard referral form, but a written report to the referrer after each course of treatment Retention No HIPAA chart rule; state law governs; Medicare requires 7 years for covered ordering and referral documentation Ontario: at least 10 years from last interaction, or from the client's 18th birthday 7 years since last entry; until age 25 for clients under 18 (Psychology Board Code) The clinical core travels across all three countries. What changes is the regulatory layer: privacy categories and payer linkage in the US, college-enforced record standards in Canada, and program-cycle reporting in Australia. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychotherapy progress notes. Draft a complete progress note from bullets, a dictation, or a session transcript, in SOAP, DAP, BIRP, or narrative structure. Convert an existing note between structures without losing clinically important content. Check a finished note for unspecified interventions, missing time or signature elements, and treatment-plan linkage before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a psychotherapy progress note the same as a HIPAA psychotherapy note? No. The progress note is the shared treatment record. HIPAA's "psychotherapy notes" are a narrow category of private process notes kept separate from the chart, and the legal definition excludes session times, modality and frequency, diagnosis, functional status, treatment plan, symptoms, prognosis, and progress, exactly the content of a progress note. Can I make my session note a psychotherapy note just by labeling it that way? No. The label does not control the legal status; the content and separate maintenance do. If the note contains diagnosis summaries, treatment-plan content, session timing, or progress to date, it belongs to the ordinary record and is treated that way for access and disclosure, no matter what it is called. Do insurers require a specific format like SOAP? No payer in the sources we reviewed mandates a note structure. What Medicare's outpatient psychiatric criteria expect is content: linkage to an individualized treatment plan with goals, documentation sufficient to support the billed service and time, and signed, dated entries. SOAP, DAP, BIRP, and narrative notes can all carry that content; see the SOAP note template for the most payer-familiar structure. How long should a psychotherapy progress note be? Most run 100 to 350 words and take 10 to 20 minutes by hand (our clinical team's estimate; no published national benchmark exists). Long enough to name the interventions, the response, the risk status, and the plan; short enough to finish in the minutes after session. Do I need exact start and stop times for 90837? Document time well enough to support the billed code. Missing psychotherapy time is a named cause of improper payment in OIG's national audit, and some Medicare contractors expect start/stop or total time explicitly. We did not find one universal national clock-time wording across all settings, which is why the question persists on coding forums; recording start and stop times is the simplest way to be safe under any contractor. How long do I need to keep progress notes? HIPAA does not set a retention period for patient charts; state law generally governs, and Medicare separately requires 7 years for covered ordering and referral documentation. Ontario's CRPO expects at least 10 years from the last interaction (or from the client's 18th birthday), and Australia's Psychology Board code sets 7 years since last entry, or until age 25 for minors. Who has to sign a progress note? The clinician responsible for the service, signed and dated, with credentials. CMS lists unsigned and undated notes among documentation failures, and pasted images of signatures have been flagged in OIG audits. In Ontario, psychologist-board rules add a co-sign duty: documents containing conclusions or diagnoses authored by a supervisee must be signed by both the supervisee and the supervising registrant. What does Australia's Better Access program require beyond session notes? Two things clinicians often miss: there is no standard referral form (a signed, dated letter with the specified content is acceptable), and after each course of treatment the allied mental health professional must send the referrer a written report covering assessments, treatments provided, and recommendations for future management. Your session notes are what that report gets built from. Can BastionGPT write a psychotherapy progress note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces a structured draft in SOAP, DAP, BIRP, or narrative form for your review. It can also convert notes you already have between structures. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and the items it expressly excludes. HHS, HIPAA FAQ 580 : the Privacy Rule sets no medical-record retention period; state law governs. CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 improper-payment data (16.1%, $254.5M projected, 78.3% insufficient documentation) and treatment-plan expectations. HHS Office of Inspector General, national audit of Medicare psychotherapy services (2023) : estimated $580 million improper payments; missing time documentation among the causes. HHS Office of Inspector General, On-Site Psychological Services audit (2020) : 111 of 120 sampled claims noncompliant; unspecified therapeutic maneuvers and signature-image findings. CMS, Complying with Medicare medical record documentation requirements (MLN909160) : unsigned, undated, and insufficient progress notes as denial causes. 42 CFR § 424.516 : Medicare's 7-year documentation-maintenance requirement for covered ordering and referral records. College of Registered Psychotherapists of Ontario, Clinical Records practice standard : required record content and the 10-year retention expectation. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : record content and supervisee co-signature rules. Psychology Board of Australia, Code of Conduct (effective 1 Dec 2025): record quality and the 7-year / age-25 retention floor. Ahpra, Managing health records : Australian record-keeping expectations across registered professions. Services Australia, Allied health referrals for mental health treatment services : no standard referral form; written report to the referrer after each course of treatment. ## Related templates SOAP Note DAP Note BIRP Note Treatment Plan Case Formulation Note Psychotherapy Notes Authorization Trauma Assessment Care Coordination Note PIE Note Crisis Psychotherapy Note Telehealth Psychotherapy Note Group Therapy Note Family & Couples Therapy Note Intake Note Psychiatric Diagnostic Evaluation Safety Plan Discharge Summary Prior Authorization Support Letter Release of Information (ROI) Authorization Claim-Support Billing Note Treatment Plan Review & Update Individual Therapy Session Note Mental Status Exam (MSE) Note Outcome Measure Review Note Better Access Mental Health Treatment Plan Referral Medication Management Follow-Up Note Patient Instructions & After-Visit Summary Appointment No-Show & Outreach Note Incident & Adverse Event Report Record Amendment & Correction Response Start-Stop Time Attestation Narrative Progress Note PCL-5 Score Documentation Y-BOCS Score Documentation Insomnia Severity Index Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/soap-note All note templates # Mental Health SOAP Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A SOAP note is a progress note format that organizes a clinical encounter into four sections: Subjective, Objective, Assessment, and Plan. Therapists, psychologists, and integrated care teams use SOAP notes to document sessions in a structure that other providers, payers, and auditors can read quickly. A typical therapy SOAP note runs 150 to 350 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, integrated behavioral health teams Audience Treating clinician, care team, supervisors, payers, auditors Typical length 150 to 350 words · 10 to 20 minutes by hand (clinical team estimate) Format family Structured progress note (compare: DAP, BIRP, PIE) When it's used After each individual, group, or telehealth session Standards context A documentation convention, not a mandated form; sits on top of HIPAA and payer requirements On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a SOAP note? SOAP stands for Subjective, Objective, Assessment, Plan. The format grew out of Lawrence Weed's problem-oriented medical record framework, introduced in 1968, and spread from medical charting into behavioral health because primary care teams and payers already know how to read it. Some clinicians also call it a therapy SOAP note or counseling SOAP note. Two clarifications save a lot of confusion. First, no law requires the SOAP format: HIPAA, state boards, and payers require that records be accurate, complete, and appropriate to the service, and SOAP is simply one accepted way to get there. Medicare's Program Integrity Manual states that progress notes "may be in any form or format." Second, a SOAP note is a progress note, which means it is part of the standard clinical record. It is not a psychotherapy note , the separate category of private process notes that receives special protection under HIPAA. Keep payer-facing content in the progress note and keep psychotherapy notes segregated. ## Who uses SOAP notes and when SOAP is a strong default when your notes will be read outside your own practice: integrated primary care, hospital-affiliated clinics, multidisciplinary teams, and any setting where payer review is routine. The Objective section gives SOAP an edge over looser formats when you track measurable data such as screening scores, missed appointments, or medication changes. Solo therapists who never share records sometimes prefer DAP , which merges the subjective and objective material into one Data section. ## How to write a mental health SOAP note in four steps Subjective: record what the client reports: symptoms, stressors, progress since last session, and one relevant quote. Objective: add observable behavior, mental-status findings, and measurable scores such as the PHQ-9 or GAD-7. Assessment: interpret the information, relate it to treatment-plan goals, and document current risk. Plan: record specific interventions, homework, care coordination, and the next appointment. Each section in more detail, with the pitfall that most often undermines it: S: Subjective. The client's report in their own frame: presenting concerns, symptoms, stressors, progress since last session. One or two short quotes can anchor the note. Pitfall: filling this section with your interpretation instead of the client's report; interpretation belongs in Assessment. O: Objective. What you observed and measured: appearance, affect, engagement, mental status highlights, screening scores (PHQ-9, GAD-7), attendance. Pitfall: writing "client seemed anxious" as objective; describe the observable behavior instead, such as fidgeting or rapid speech. A: Assessment. Your clinical judgment: progress toward treatment plan goals, response to interventions, diagnostic impressions, risk status. This is where the note earns its medical necessity. Pitfall: repeating S and O instead of interpreting them. P: Plan. What happens next: interventions for coming sessions, homework, referrals, medication coordination, next appointment, any safety planning. Pitfall: a bare "continue treatment"; tie the plan to a goal. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Service: [individual/group/telehealth] Start/stop time: S (Subjective): O (Objective): A (Assessment): P (Plan): Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample SOAP note Scenario: adult client, generalized anxiety disorder, CBT, session 6 of a planned 12. All details are fictional. Client: M.T., 34 · Date: 07/08/2026 · Session: 6 · Service: Individual psychotherapy, 53 min, in office · Start/stop: 10:04 to 10:57 S: Client reported "a better week than most," rating average anxiety 4/10, down from 6/10 last session. Completed thought records on 5 of 7 days. Reported one episode of work-related rumination Tuesday lasting about an hour, managed with paced breathing. Sleep improved to 6 to 7 hours nightly. Denied suicidal ideation, self-harm urges, and substance use. O: On time, casually dressed, good hygiene. Speech normal rate and tone. Affect euthymic, broader range than prior sessions. Engaged actively in agenda setting. GAD-7 today: 9 (down from 13 at intake). No observable psychomotor agitation. A: Steady progress toward Goal 1 (reduce daily worry intensity and duration). Client is applying cognitive restructuring with less coaching and attributed the Tuesday episode recovery to skills practice. Symptoms remain consistent with GAD, moderating. Risk: denies ideation; no risk indicators observed; continue routine monitoring. P: Continue weekly CBT. Introduce worry-postponement experiment next session. Homework: continue thought records, add one scheduled 15-minute worry period daily. Re-administer GAD-7 at session 8. Next appointment 07/15/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Risk is addressed in both S and A, even though the client is low risk, so the record shows monitoring rather than silence. The Objective section carries a score (GAD-7) that makes progress auditable across sessions. Assessment links symptoms and session content to a numbered treatment plan goal, which supports medical necessity. Plan items are specific and dated, giving the next note something concrete to evaluate. Start and stop times support time-based billing without a separate hunt through the chart. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A SOAP note is part of the designated record set: clients can request it, payers can audit it, and other providers may rely on it. Write every section as if a colleague, a reviewer, and the client will read it. Keep private reflections and hypothesis-level process material out of it; that content belongs in segregated psychotherapy notes with their own authorization rules . For insurance work, the Assessment and Plan sections do the heavy lifting: they connect the session to the diagnosis and the treatment plan, which is the core of medical necessity. The format is a convention; the content is the requirement. For time-based psychotherapy codes (90832, 90834, 90837), Medicare contractor guidance expects start and stop times or total time in the record, documented separately from any E/M service furnished in the same visit. For telehealth sessions, add modality, client location, and consent details (see the telehealth note ). For group or couples work, each client's note must stand alone without exposing the other participants' clinical content (see group and family and couples notes). ↑ Back to the template and downloads ## Common SOAP note errors auditors flag Audit findings are about documentation, not format choice. In Medicare's 2024 reporting-period data for outpatient psychiatric services, insufficient documentation drove 78.3% of improper payments, and a national OIG audit of psychotherapy claims estimated $580 million in improper payments, with missing time documentation and missing signatures among the most common problems. The BastionGPT Clinical Advisory Board sees the same errors most often in SOAP note reviews: No time support for time-based codes. Start and stop times or total time are missing, or psychotherapy time is not separated from E/M time in a combined visit. Cloned notes. Copy-forward text repeats across sessions, which reads to a reviewer as a record that does not support a distinct billed service. An Assessment that restates instead of interprets. S and O content is repeated without judgment about progress toward a treatment plan goal, which breaks the medical-necessity chain. Authentication gaps. Notes missing a signature and credentials, or signed long after the session date. Silence on risk. No statement about risk status either way, so the record shows absence rather than monitoring. ### SOAP notes in the US, Canada, and Australia Aspect United States Canada Australia Status Convention, not law: Medicare says progress notes "may be in any form or format"; HIPAA and payer rules govern content Convention; record content governed by provincial privacy law and college standards (e.g. Ontario CRPO) Convention; AHPRA expects accurate, factual, contemporaneous, attributable records Terminology Progress note, SOAP note Clinical record, session note Case note, session note What changes Payer and audit fields (codes, time, medical necessity) Fewer payer-driven fields; college documentation standards lead MBS and Better Access contexts add referral and review fields Retention No universal HIPAA rule for patient charts; state law and payer contracts govern Provincial; Ontario CRPO expects 10+ years from last interaction, or from the client's 18th birthday State rules; NSW requires 7 years for adults and until age 25 for minors The structure itself travels well across all three countries. What changes is the administrative layer around it, so adapt billing language, privacy references, and retention practices to your jurisdiction. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on SOAP notes. Draft a complete SOAP note from a few bullets, a dictation, or a pasted session transcript. Convert an existing DAP, BIRP, or narrative note into SOAP format without losing clinical content. Check a finished note for missing risk language, time support, plan items, or signature details before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a SOAP note be? Most therapy SOAP notes run 150 to 350 words and take 10 to 20 minutes to write by hand. Long enough to show medical necessity and risk monitoring, short enough to write in the minutes after session. Length beyond that rarely adds defensibility. Do insurance companies require SOAP notes? Medicare does not mandate the SOAP format itself: its Program Integrity Manual says progress notes "may be in any form or format." Requirements can vary by payer, organization, and jurisdiction, so check your contracts. What reviewers consistently look for is content: documentation that supports medical necessity, links the session to the treatment plan, and records time for time-based codes. SOAP persists because it organizes that content well. What is the difference between a SOAP note and a DAP note? DAP combines SOAP's Subjective and Objective sections into a single Data section. SOAP separates the client's report from your observations, which reads better in medical and integrated care settings. See the DAP note template for a side-by-side. Are SOAP notes psychotherapy notes under HIPAA? No. HIPAA's psychotherapy-notes definition excludes session times, modality and frequency, diagnosis, functional status, treatment plan, symptoms, prognosis, and progress. A SOAP note kept in the chart is a progress note, part of the standard record. True psychotherapy notes must be kept apart and have their own authorization rules . Do I need exact start and stop times in a SOAP note? For time-based psychotherapy codes (90832, 90834, 90837), Medicare contractor guidance expects start and stop times or total time in the record, and psychotherapy time documented separately from any E/M service in the same visit. Recording start and stop times in the note header is the simplest way to stay audit-ready. What goes in the Objective section for psychotherapy? Observable and measurable material: appearance, behavior, mental status observations, engagement, and screening scores such as the PHQ-9 or GAD-7. The client's self-report belongs in Subjective; relabeling it as objective is one of the most common SOAP note mistakes. Can I use SOAP notes for couples or group sessions? Yes, but each client needs their own note, and no client's note should expose another participant's clinical content. The group therapy note and family and couples note pages cover the multi-client confidentiality rules. How long do I need to keep SOAP notes? There is no universal HIPAA retention rule for patient charts; HIPAA's six-year rule covers required documentation such as policies, not patient records. Retention comes from state, provincial, and payer rules: Ontario's CRPO expects at least 10 years from the last interaction (or from the client's 18th birthday), and New South Wales requires 7 years for adults and until age 25 for minors. Check your jurisdiction and payer contracts. Can BastionGPT write a SOAP note from a recording or transcript? Yes. Paste a transcript, dictate, or give it bullets, and it produces a structured SOAP draft for your review. It can also convert notes you already have into SOAP format. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes "may be in any form or format." CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 reporting-period improper-payment data, including the 78.3% insufficient-documentation share. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : estimated $580 million in improper payments, with missing time documentation and signatures among the causes. CMS Medicare Coverage Database, Billing and Coding Article A57520 : time documentation expectations for psychotherapy codes 90832, 90834, and 90837. 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and its exclusions. HHS, HIPAA FAQ 580 : the Privacy Rule does not require covered entities to keep medical records for any period; state laws govern retention. 45 CFR § 164.316 : HIPAA's six-year rule for required documentation such as policies, which is distinct from patient-chart retention. College of Registered Psychotherapists of Ontario, Professional Practice Standards : Ontario record-keeping and retention expectations. Ahpra, Managing health records : Australian record-keeping expectations. New South Wales, Health Records and Information Privacy Act 2002, s 25 : retention for 7 years, or until age 25 for records of minors. ## Related templates Psychotherapy Progress Note DAP Note BIRP Note Treatment Plan Case Formulation Note Psychotherapy Notes Authorization Trauma Assessment Care Coordination Note PIE Note Crisis Psychotherapy Note Telehealth Psychotherapy Note Group Therapy Note Family & Couples Therapy Note Intake Note Prior Authorization Support Letter Claim-Support Billing Note Individual Therapy Session Note Mental Status Exam (MSE) Note Records Request & Chart Copy Packet Record Amendment & Correction Response Narrative Progress Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/telehealth-therapy-note All note templates # Telehealth Psychotherapy Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A telehealth psychotherapy note is the progress note for a therapy session delivered by video or phone. It records everything a standard session note does, plus the telehealth layer: modality, client location, provider location, consent, technology issues, and a session-specific emergency plan. Therapists, psychologists, and counselors write one after every virtual session. A typical telehealth note runs 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers delivering virtual care Audience Treating clinician, care team, payers, auditors Typical length 150 to 400 words · 10 to 20 minutes by hand (clinical team estimate) Format family Structured progress note with a telehealth layer (compare: SOAP, DAP, BIRP) When it's used After each psychotherapy session delivered by video or phone Standards context No mandated form; state licensure, payer telehealth rules, and consent laws govern the content On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a telehealth psychotherapy note? A telehealth psychotherapy note is the clinical record of a psychotherapy session delivered by live video or phone. It is not a new document type: it is an ordinary progress note , in whatever format your practice already uses, plus the telehealth facts that matter for licensure, safety, consent, and billing. No standards body ever created a separate telehealth note format. The clinical core descends from ordinary progress-note practice, and the telehealth layer comes from licensure rules, consent statutes, telepsychology standards (the American Psychological Association first adopted telepsychology guidelines in 2013, and the Canadian Psychological Association issued its own in 2023), and payer policy. Clinicians and payers also call it a teletherapy note, virtual therapy note, telepsychology note, telebehavioral health note, or video-visit note. Two clarifications matter. First, no law or payer mandates a telehealth note format. Medicare's Program Integrity Manual states that progress notes "may be in any form or format." What makes the telehealth layer worth writing is that modality, location, and consent are materially decisive: they determine who may lawfully treat the client, how emergency services would reach them, and how the claim is coded, even though no national rule lists them as required note fields. Second, a telehealth psychotherapy note is not a psychotherapy note in the HIPAA sense. It carries session times, diagnosis, progress, and risk findings, exactly the content HIPAA excludes from that specially protected category, so it belongs in the standard clinical record (see the psychotherapy notes authorization page for the difference). ## Who uses telehealth psychotherapy notes and when Any clinician who delivers therapy by video or phone writes one after every virtual session: therapists, psychologists, counselors, and clinical social workers in fully virtual practices, hybrid practices, and integrated care teams. Virtual delivery is now a routine share of care: a 2026 JAMA Psychiatry analysis found that 27.8 percent of US adults in outpatient mental health treatment received all of their care by telehealth, and another 21.5 percent used a mix of virtual and in-person visits. The telehealth layer is format-agnostic: it sits on top of whatever structure you already use, whether SOAP , DAP , BIRP , or a narrative note. When a virtual session becomes an acute crisis encounter, document it as a crisis psychotherapy note instead, with the remote-session facts included. ## Telehealth psychotherapy note structure: what goes in each section The note is a standard session note with a telehealth header. Put the telehealth layer where a reviewer looks first: modality (video or audio-only, with the reason if audio-only), the client's location and setting, your location, consent status with its date, and start and stop times. Then document the session with the same clinical substance as an office visit. Technology and session setup. The platform used, connection quality, any interruptions with times and how they were handled, and who else is present at the client's location. Pitfall: silence about a mid-session drop. If video failed and the session finished by phone, the modality changed, and under Medicare the claim belongs on the audio-only modifier; an unexplained gap also reads as time the record does not support. Emergency plan for this session. The client's specific location today, confirmed precisely enough for emergency dispatch, a local contact or number, and the reconnect plan if the session drops during a risk discussion. Pitfall: a generic "safety plan on file" line. Ontario's psychology standards expressly require emergency planning tied to services where the client is located, and the operational point everywhere is where the client actually is right now. Session content and interventions. Presenting concerns, themes, and the therapeutic work delivered, with the same substance as an in-person note. Pitfall: telehealth logistics crowding out the psychotherapy. Medicare's national psychotherapy audit flagged notes with no visible therapeutic work; the billed service is therapy, not a video call. Mental status and response. What you actually observed on camera or heard by phone: appearance, affect, speech, engagement, and anything the modality limited. Pitfall: charting findings the modality cannot support, such as full appearance and motor detail from a head-and-shoulders frame, or visual observations in an audio-only session. Assessment, risk and progress. Progress toward treatment plan goals, risk status, and whether telehealth remains clinically appropriate for this client. Pitfall: never revisiting suitability. Ontario requires determining whether virtual care is clinically indicated, and Ahpra tells practitioners to assess and re-assess it; a one-line suitability statement shows that judgment. Plan and next steps. Interventions for coming sessions, homework, the next appointment, and the planned modality. Pitfall: a bare "continue treatment." Tie the plan to a goal and say whether the next session is video, phone, or in person. ### Blank template (copy and adapt) Client: [initials] Date: Session #: Modality: [video / audio-only, reason] Telehealth consent: [on file, date] Client location & setting this session: Provider location: Start/stop times: Technology & session setup (platform, interruptions, others present): Emergency plan for this session (location confirmed, local contact): Session content & interventions: Mental status & response (as observed by video/phone): Assessment, risk & progress (incl. continued suitability for telehealth): Plan & next steps (incl. planned modality): Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample telehealth psychotherapy note Scenario: established adult client, major depressive disorder, CBT delivered by video, session 8 of a planned 16. All details are fictional. Client: D.K., 29 · Date: 07/09/2026 · Session: 8 · Service: Individual psychotherapy by video · Start/stop: 3:02 to 3:55 pm · Psychotherapy time: 51 minutes Telehealth setup: Video session on the practice's HIPAA-compliant platform. Client at home in Fort Collins, Colorado; address on file confirmed at the start of session. Clinician at the practice office in Denver, Colorado. Telehealth consent on file, dated 03/12/2026. Client alone in a closed room, using headphones. Technology: Connection stable except one video drop from 3:27 to 3:29 pm. Session resumed by video, the interrupted material was repeated, and the two minutes are excluded from the psychotherapy time above. Emergency plan: Location confirmed as above. Spouse home and reachable; local emergency contact information on file verified as current. Plan reviewed: if the connection fails during any risk discussion, clinician calls the client back immediately and the client answers by phone. Session content and interventions: Reviewed behavioral activation homework: activity scheduling completed 5 of 7 days, with morning walks the most consistent entry. Client skipped both planned social activities, citing "not wanting to explain the job search." Cognitive restructuring targeted the guilt-driven avoidance thoughts; client generated an alternative appraisal and rated belief in it 60 percent. PHQ-9 today: 11, down from 14 at session 6. Mental status and response: Groomed, alert, engaged on camera. Speech normal in rate and tone. Affect mildly dysphoric, brightening when describing the walks. No psychomotor changes visible in frame; full appearance and motor observations limited by the head-and-shoulders video frame. Denied suicidal ideation and self-harm; no risk indicators observed or reported. Assessment, risk and progress: Steady progress toward Goal 1 (increase engagement in valued activities): activity adherence up and PHQ-9 down three points in two weeks. Symptoms remain consistent with major depressive disorder, moderate, improving. Risk low; routine monitoring continues. Client engages well by video and presentation is stable; telehealth remains clinically appropriate. Plan and next steps: Continue weekly CBT by video. Homework: maintain the activity schedule and complete one of the two planned social activities before next session. Re-administer PHQ-9 at session 10. Next appointment 07/16/2026, video. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The telehealth layer sits in the header and first section where a reviewer looks first: modality, both locations, consent with its date, and times in one place. The emergency plan is session-specific: the client's confirmed location, who else is in the home, and what happens if the connection drops during risk content. The connection loss is documented with times and the psychotherapy time excludes it, so the billed time stays defensible. Mental status records what video could actually show and names what it could not, so the note never overclaims. Assessment ties the session to a numbered goal and states that telehealth remains appropriate, which supports medical necessity for the modality itself. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A telehealth psychotherapy note is part of the designated record set: clients can request it, payers can audit it, and other providers rely on it. Keep private process reflections out of it; those belong in separately maintained psychotherapy notes with their own authorization rules . Medicare review contractors may not request those HIPAA-protected psychotherapy notes, but the Program Integrity Manual makes the flip side explicit: you must supply enough ordinary progress-note documentation to establish the service and its medical necessity, and a claim supported only by content locked inside protected notes can be denied. The telehealth layer itself is mostly not note-field law. No national rule in the US, Canada, or Australia requires a street address, an identity-verification sentence, a "no technology problems" statement, or a repeated consent paragraph in every session note; those facts earn their place because they are materially decisive for licensure, safety, and billing, and because narrower state, provincial, payer, or board rules can make any of them mandatory. Record what is material rather than pasting boilerplate. For US Medicare, the binding layer is statute and payer policy, and the dates matter right now. The Consolidated Appropriations Act, 2026 extended the major telehealth flexibilities through December 31, 2027 and postponed the behavioral-health in-person visit requirement to January 1, 2028, with clients already in home tele-mental-health care by the end of 2027 grandfathered past the six-month prior-visit rule ( CMS telehealth FAQ ). Audio-only psychotherapy stays covered for a client at home when you are video-capable but the client "is not capable of, or does not consent to, the use of video technology" ( 42 CFR 410.78 ). On claims, the client's site drives the rate: POS 10 (client at home) pays the non-facility rate since January 1, 2024, POS 02 (any other client location) pays the facility rate, and modifier 95 marks video while modifier 93 marks audio-only ( CMS Change Request 13582 ). Medicare's originating site is the client's location and the distant site is yours; templates sometimes reverse them. One caution when checking sources yourself: the displayed eCFR text of 42 CFR 410.78 still shows superseded 2024 dates, and the later 2026 statute controls. The format is a convention; the content, the time, and the claim fields are the requirement. Location and consent are where state and provincial law bind. Under PSYPACT , the psychologist must be physically in their compact Home State, the client's state is the Receiving State, and the Receiving State's rules govern the work, so both locations are legally consequential even though no rule prescribes a note field for them. On consent, California is a clean example: providers must inform the client and obtain verbal or written consent before delivering care by telehealth, and the statute says the consent "shall be documented" ( Business and Professions Code 2290.5 ). Rules differ by state and board, so check yours. Canada and Australia are summarized in the table below. A telehealth session that becomes an acute crisis gets a crisis note , and a lasting modality change usually also touches the treatment plan . ↑ Back to the template and downloads ## Common telehealth psychotherapy note errors auditors flag The national audit record here is unusually direct, because Medicare's biggest psychotherapy audit landed on the telehealth era. During the audit period, 57 percent of Medicare psychotherapy payments went to telehealth services, and the 2023 OIG national audit estimated $580 million in improper payments, $348 million of it for telehealth sessions. Reviewers did not fail notes for missing telehealth boilerplate: the failures were missing time, missing or inadequate treatment plans, missing signatures, and notes with no visible psychotherapy. Medicare's 2024 compliance data for outpatient psychiatric services tells the same story, with insufficient documentation driving 78.3 percent of improper payments. The BastionGPT Clinical Advisory Board sees the same errors most often in telehealth note reviews: Time that does not support the code. Missing start and stop times or total time appeared on 24 of the 111 telehealth service days in the OIG sample. Time-based psychotherapy codes need time support regardless of modality, and interruptions should be visibly excluded. Modality that does not match the claim. A session billed as video (modifier 95) after it actually finished by phone, or a home session coded POS 02 instead of POS 10, which also changes the payment rate. When the billed code is not covered as telehealth at all, the claim returns remark code N776. No link to a treatment plan. Incomplete or missing treatment plans were the most common telehealth deficiency in the OIG sample, on 29 of 111 days. Whether a written plan is expressly required varies by contractor (only three of the seven MACs reviewed had one in a coverage determination), but the medical-necessity chain always needs the note to connect to goals. Authentication gaps and late entries. Missing signatures appeared on 20 of the 111 telehealth days. Delayed entries must be identified as delayed; in one OIG example, records created more than eight months after the service were rejected as support. Boilerplate instead of substance. A cloned telehealth paragraph repeated across sessions ("consent confirmed, identity verified, no technical issues") adds nothing a reviewer credits. OIG's point is that documentation should describe the therapeutic interventions and the client's response. ### Telehealth psychotherapy notes in the US, Canada, and Australia Aspect United States Canada Australia Status No mandated note format; Medicare accepts progress notes "in any form or format"; the telehealth layer comes from statute, payer policy, and state law No national regime; CPA's 2023 telepsychology guideline is guidance, while provincial college standards bind (Ontario technology and emergency rules, BC's 2026 virtual-care standards) No separate format; Ahpra holds telehealth to the same standard as face-to-face care, and the Psychology Board's code of conduct governs records Terminology Telehealth, telemental health, teletherapy note Virtual care, telepsychology Telehealth, video and phone attendances What changes Claim fields: POS 02 or 10 and modifier 95 or 93, plus the December 2027 and January 2028 Medicare dates Authorization to practise across provinces, Ontario's emergency-plan and identity rules, consent expectations Separate MBS item numbers for video and phone under Better Access, plus referral and reporting rules Retention No universal HIPAA or Medicare rule for charts; state law and payer contracts govern Ontario: 10 years after age 18 or the last contact, whichever is later; BC: 16 years from the last encounter or from the age of majority Psychologists: 7 years after the last entry, or until the 25th birthday for clients who were minors The clinical core travels unchanged across all three countries. What changes is the jurisdictional layer around it: who may treat a client in another state or province, how the modality reaches the claim, and how long the record must survive. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on telehealth psychotherapy notes. Draft a complete telehealth session note from a few bullets, a dictation, or a pasted transcript, with the modality, locations, consent, and emergency plan in the right places. Check a finished note for missing telehealth elements such as client location, consent status, technology events, or time support before you sign. Flag when the documented modality does not match the billing route, such as an audio-only session drafted as a video visit. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a telehealth psychotherapy note be? Most run 150 to 400 words, about the same as an in-person progress note plus a few header lines for the telehealth layer. Length is not the compliance driver: the failures that cost providers money in Medicare's national psychotherapy audit were missing time, missing plans, and missing signatures, not short notes. Write enough to show the service, the response, and the telehealth facts that matter. Is a special telehealth note format required by law? No. No national rule in the US, Canada, or Australia prescribes a telehealth note format. Medicare accepts progress notes "in any form or format," and Canadian and Australian regulators require accurate, complete, contemporaneous records rather than a layout. What is binding sits around the note: licensure tied to location, consent statutes in some states and provinces, and payer claim rules. That is why this page labels each requirement as law, payer policy, or convention. Do I have to document the client's location every session? No universal federal rule says so, but location is the most consequential fact in the telehealth layer. It determines whose licensing rules apply (under PSYPACT, the client's state is the Receiving State), which emergency services can reach the client, and whether the claim is POS 10 (home) or POS 02 (anywhere else). Ontario expressly requires an emergency plan tied to services where the client is located. The defensible habit: record at least the city and state or province every session, and a precise location whenever risk is elevated or the client is somewhere new. What are POS 02, POS 10, and modifiers 95 and 93? Medicare claim fields, not note fields. POS 10 means the client was at home and has paid the non-facility rate since January 1, 2024; POS 02 means any other client location and pays the facility rate. Modifier 95 marks an audio-video session and modifier 93 audio-only ( CMS Change Request 13582 ). No rule requires repeating them inside the narrative, but the note's modality and location lines are what prove them right in an audit. One orientation check: Medicare's originating site is the client's location and the distant site is yours; templates sometimes reverse the two. Is audio-only (phone) psychotherapy still covered by Medicare? Yes. The broad audio-only flexibility runs through December 31, 2027. After that, a permanent behavioral-health exception still covers phone sessions for a client at home when you are capable of video but the client "is not capable of, or does not consent to, the use of video technology" ( 42 CFR 410.78 ). Document the reason for audio-only: that line is what supports modifier 93 and the exception. What changes on January 1, 2028? Unless Congress acts again, Medicare's in-person requirement for home tele-mental-health takes effect: an in-person visit within six months before the first home telehealth service and roughly annually afterward, with exceptions. Clients who began home tele-mental-health care by December 31, 2027 are treated as established and skip the six-month prior visit ( CMS telehealth FAQ ). The dates come from the Consolidated Appropriations Act, 2026; the displayed eCFR text still shows older dates, and the statute controls. The video failed mid-session. What do I document and bill? Document the interruption with times, what was missed, and how the session resumed, and make the psychotherapy time add up without the gap. If the session finished by phone, the modality changed: under Medicare that is modifier 93 territory, not 95. Australia's MBS handles the same situation with separate video and phone item numbers, so the item must match what actually happened. Commercial payers can differ, so do not generalize Medicare's coding rule to every insurer. Do I need telehealth consent at every session? No jurisdiction reviewed for this page requires re-documenting consent in every note. Consent is state and province law: California, for example, requires informed consent before care is delivered by telehealth and requires that the consent be documented. Federal guidance accepts written, electronic, or verbal consent . Obtain it properly once, record where it lives and its date in the note header, and renew it when something material changes: a new platform, recording or AI tools, new participants, or cross-border care. Can BastionGPT draft telehealth psychotherapy notes? Yes. Give it a few bullets, a dictation, or a transcript, and it drafts the full note with the telehealth layer in the right places: modality, locations, consent status, technology events, and the emergency plan, plus the clinical content. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes "may be in any form or format," the psychotherapy-notes request limits, and delayed-entry rules. CMS, Telehealth FAQ, updated February 26, 2026 : flexibility end dates, the 2028 in-person requirement, and the grandfathering rule. Public Law 119-75, section 6209 : the Consolidated Appropriations Act, 2026 telehealth extensions. 42 CFR 410.78 : originating and distant site definitions and the audio-only home exception; note the displayed text lags the 2026 statute. CMS, Claims Processing Manual, Chapter 12 : telehealth billing and licensure rules. CMS, Change Request 13582 : POS 02 and 10 payment rates and modifiers 93 and 95, effective January 1, 2024. HHS Office of Inspector General, national psychotherapy audit (2023) : the $580 million improper-payment estimate, including $348 million for telehealth. PSYPACT Rules, revised November 18, 2025 : Home State and Receiving State definitions. California Business and Professions Code 2290.5 : telehealth consent obtained and documented. HHS, informed consent for telebehavioral health : consent may be written, electronic, or verbal. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : technology, emergency-plan, identity, and retention rules. College of Health and Care Professionals of BC, Practice Standard: Records : the 16-year retention floor, effective April 1, 2026. Psychology Board of Australia, Code of conduct (effective December 1, 2025): record content and retention. Ahpra, information for practitioners who provide virtual care : identity, consent, and suitability expectations. Australian Government, MBS allied health telehealth arrangements : separate video and phone item numbers. ## Related templates Psychotherapy Progress Note SOAP Note DAP Note BIRP Note Crisis Psychotherapy Note Group Therapy Note Family & Couples Therapy Note Claim-Support Billing Note Australia Telehealth & MBS Case Conference Note Communication Log & Portal Message Note Telehealth Consent Canada Telepsychology & Virtual Care Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/biopsychosocial-assessment All note templates # Biopsychosocial Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A biopsychosocial assessment is the comprehensive intake document that examines a client's presenting problem across biological, psychological, and social domains and ties them together in an integrated summary. Therapists, clinical social workers, counselors, and case managers complete one at admission to treatment, and it anchors the diagnosis, level of care, and treatment plan. A typical biopsychosocial assessment runs 600 to 1,500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, clinical social workers, counselors, case managers Audience Treating team, supervisors, payers, auditors, utilization reviewers Typical length 600 to 1,500 words · 45 to 75 minutes by hand (clinical team estimate) Format family Intake and assessment documentation (compare: intake note, psychiatric diagnostic evaluation) When it's used At admission to treatment, and wherever a program requires the full three-domain picture Standards context No law prescribes one biopsychosocial form; requirements attach to programs such as opioid treatment programs, CMHCs, CCBHCs, and state Medicaid On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a biopsychosocial assessment? A biopsychosocial assessment is the comprehensive admission document that examines a client's presenting problem across biological, psychological, and social domains and integrates the findings into one clinical picture. Clinicians also call it a BPS assessment, biopsychosocial evaluation, or bio-psycho-social assessment; US opioid treatment regulations use the name psychosocial assessment for the same artifact, and Ontario's social work regulator defines the social work diagnosis as "also known as a biopsychosocial assessment" . The name traces to George Engel, whose 1977 Science paper proposed the biopsychosocial model as a challenge to purely biomedical thinking. The document came later, built from the older case-history tradition, with Mary Richmond's 1917 Social Diagnosis among its ancestors, as psychiatry, social work, and addiction treatment put the model to work. Engel proposed a model, not a form: no US, Canadian, or Australian rule prescribes one universal biopsychosocial template. Two boundaries do the most work. First, the biopsychosocial assessment is a document, while the psychiatric diagnostic evaluation (90791 or 90792) is a billable service: one Medicare contractor policy defines that service as "an integrated biopsychosocial assessment" , and Novitas policy says it plainly: "The diagnostic evaluation is a biopsychosocial assessment." Completing the form does not by itself establish that the billed service was performed or medically necessary. Second, depth: an intake note records the first session, while the biopsychosocial assessment works each domain in its own right and ends in an integrated summary; nothing requires them to be separate documents, and in many agencies they are one. However personal its contents, the assessment is part of the standard clinical record and the designated record set, not a HIPAA psychotherapy note . ## Who uses biopsychosocial assessments and when The biopsychosocial assessment is the standard admission workup wherever behavioral health care is organized around programs: community mental health centers, substance use treatment programs including opioid treatment programs, certified community behavioral health clinics, hospitals and residential programs, case management, and integrated care teams. Clinical social workers carry the strongest tradition with it, and therapists, counselors, and case managers write them daily. In a leaner private practice, the same ground is often covered by a shorter intake note ; the biopsychosocial assessment earns its length where programs, payers, or acuity call for the full three-domain picture. Specialized concerns still get their own deeper workups, such as a trauma assessment , and once treatment is underway, the thinking layer of the chart moves to the case formulation and treatment plan . ## Biopsychosocial assessment structure: what goes in each section Identifying information and referral. Client identifiers, date, clinician, referral source, service type, and start and stop times, together in one header block. Pitfall: scattering these details through the narrative; reviewers and records staff need them findable in seconds. Presenting problem. Why the client is here now, in their own words, plus onset, duration, severity, and what prompted the contact. Pitfall: recording only your clinical rephrasing; a short quote preserves the client's frame and anchors medical necessity. Biological domain. Medical conditions, current medications, allergies, relevant developmental history, family medical and psychiatric history, sleep, appetite, and pain. Pitfall: treating this as a checkbox pass; an unassessed medical contributor, a thyroid condition, a sedating medication, chronic pain, quietly undermines the whole formulation. Psychological domain. Psychiatric history, prior treatment and how the client responded, a trauma overview, coping style, and the course of current symptoms. Pitfall: restating the presenting problem instead of documenting history and pattern, or recording trauma detail past clinical need. Social domain. Family of origin and current relationships, housing, education, work and finances, legal involvement, cultural and spiritual factors, and community supports. Pitfall: recording bare circumstances with no clinical meaning attached; the point is what each fact does to the presentation. Substance use. Substances, pattern, frequency, most recent use, and past problem use or remission. Pitfall: burying substance findings inside the social history; reviewers expect an explicit screen they can find. Risk screen. Current and historical suicidal ideation, self-harm, and thoughts of harming others, protective factors, and any action taken. Pitfall: leaving risk blank when findings are negative; write the negative finding so the record shows you asked. Mental status observations. Appearance, behavior, speech, mood, affect, thought process, insight, judgment. Pitfall: template MSE strings that contradict the narrative, such as "euthymic mood" above a tearful presentation. Strengths and supports. What is working: coping that has succeeded before, relationships, motivation, community, plus baseline measure scores. Pitfall: a token line ("client is resilient") that never reaches the recommendations. Integrated summary. The paragraph that makes this an assessment: how the biological, psychological, and social findings interact to produce and maintain the presentation, and why this person is presenting now. Pitfall: the classic failure, data collection without synthesis; an assessment that ends at the last checkbox is a questionnaire. Diagnostic impression and recommendations. Provisional diagnosis with rule-outs, recommended level of care, referrals across domains, the initial plan, and consent documentation. Pitfall: recommendations that float free of the findings; a level of care no documented finding supports breaks medical necessity at the start of the episode. ### Blank template (copy and adapt) Client: [initials] DOB/Age: Date: Clinician: Referral source: Service: biopsychosocial assessment Start/stop time: PRESENTING PROBLEM In the client's words: Onset / duration / severity / precipitant: BIOLOGICAL DOMAIN Medical conditions / medications / allergies: Sleep / appetite / pain: Developmental + family medical and psychiatric history: PSYCHOLOGICAL DOMAIN Psychiatric history + prior treatment and response: Trauma overview (to clinical need, no more): Coping style / current course: SOCIAL DOMAIN Family + current relationships: Housing / education / work / finances / legal: Cultural, spiritual + community supports: SUBSTANCE USE (substances, pattern, last use, remission): RISK SCREEN SI / self-harm / harm to others (current + history): Protective factors / action taken: MENTAL STATUS (appearance, speech, mood, affect, thought process, insight): STRENGTHS & SUPPORTS: MEASURES (e.g., PHQ-9, GAD-7, AUDIT-C + scores): INTEGRATED SUMMARY (how the domains interact; why this person, why now): DIAGNOSTIC IMPRESSION (provisional dx + rule-outs): RECOMMENDATIONS (level of care, referrals, initial plan, next appt): Consent for treatment reviewed and signed: [Y/N] Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample biopsychosocial assessment Scenario: adult client referred by his primary care provider to a community behavioral health agency for low mood after a job loss. All details are fictional. Client: M.T., 41 · Date: 07/10/2026 · Clinician: J. Alvarez, LCSW · Service: Biopsychosocial assessment, 74 min, in office · Referral: Primary care provider · Start/stop: 10:04 to 11:18 Presenting problem: Client sought help for "not being able to get through a job application without shutting down." Reports about three months of low mood, loss of interest, early-morning waking, reduced appetite with an estimated 8-pound weight loss, guilt about the layoff, and poor concentration, beginning within weeks of losing his warehouse supervisor position in April 2026. Contacted his PCP in June; referred for behavioral health assessment. Biological: Hypertension managed with lisinopril 10 mg daily; no other medications; no known allergies. Sleep 4 to 5 hours with early waking; appetite reduced; no significant pain. Last physical February 2026, unremarkable per client report. Family history: mother treated for depression in her 50s; older brother with alcohol use disorder. No head injuries, seizures, or developmental concerns. Psychological: One prior depressive episode at age 33 after his divorce, treated with about ten sessions of outpatient CBT with good response; no psychiatric hospitalizations; no prior psychotropic medications. Trauma overview: grew up around his father's drinking and related conflict; prefers not to discuss further, and screens negative for intrusion, avoidance, and hyperarousal symptoms. A deeper trauma assessment is not indicated at this time. Usual coping has been work routine, exercise, and his recovery community; all three have lapsed since the layoff. Social: Divorced since 2019; shares custody of a 12-year-old son with a cooperative co-parenting arrangement. Lives alone in a one-bedroom apartment; rent is current but savings cover about two more months, which he names as his most pressing worry. Some college. Worked nine years as a warehouse supervisor until the April 2026 layoff; job searching with reduced follow-through in recent weeks. Sister lives nearby, weekly contact. Attends church and its men's recovery group, less regularly in the past two months. No legal involvement. Bilingual English and Spanish; prefers services in English. Substance use: Alcohol use disorder in sustained remission: drank up to six beers nightly from 2019 to late 2023, stopped with the support of his church recovery group, sober since November 2023; calls sobriety "the thing I'm proudest of." Denies tobacco, cannabis, and other drug use. Caffeine: two coffees daily. Risk screen: Reports passing thoughts of "what's the point" on the worst mornings, without active suicidal ideation, intent, plan, or preparatory behavior. Denies self-harm history and prior attempts; denies thoughts of harming others. Protective factors: parenting role, sister's support, recovery community, sobriety identity, and future orientation about work. Reviewed warning signs and crisis contacts; client agreed to call the agency or 988 if thoughts become active. Risk judged low, with monitoring at each visit. Mental status: On time, adequately groomed, cooperative. Speech soft and slightly slowed. Mood "worn down"; affect constricted, congruent. Thought process linear and goal-directed. No perceptual disturbance reported or observed. Insight and judgment good. Measures: PHQ-9 = 16 (moderately severe; item 9 = 1). GAD-7 = 7 (mild). AUDIT-C = 0. Strengths and supports: Two and a half years of sobriety with an intact recovery community, a motivating parenting role, prior full response to CBT, family support, and concrete goals about returning to work. Integrated summary: This is a second episode of recurrent major depression in a 41-year-old man, precipitated by a layoff that removed the three structures his stability was built on: daily routine, income security, and the provider role. Biological loading (family history of depression, disrupted sleep) deepens the episode; a self-critical reading of the layoff sustains guilt and withdrawal; and financial pressure with reduced recovery-group attendance strips out the reinforcement and support that maintained his previous recovery. Sobriety remains intact and is both a leading strength and a monitoring priority, given his own history and his brother's course. Prior full response to CBT and active help-seeking predict good engagement. Risk is currently low, with passive ideation to monitor. Diagnostic impression: Major depressive disorder, recurrent episode, moderate (F33.1). Alcohol use disorder, in sustained remission. Rule out persistent depressive disorder pending collateral history. Recommendations and initial plan: Weekly individual therapy, 50 minutes, CBT with a behavioral activation focus; collaborative treatment plan to be completed next session. Referral to agency case management for employment support and rent-assistance screening. With the client's consent, a summary letter to his PCP; medication evaluation to be revisited at session 6 if the PHQ-9 has not improved. Client will re-engage the recovery group this week and begin a sleep log. Re-administer PHQ-9 and GAD-7 at session 4. Limits of confidentiality, fees, and policies reviewed; informed consent signed. Next appointment 07/17/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The integrated summary synthesizes the three domains instead of repeating them, answering why this person is presenting now. The risk screen is written out with protective factors and an action taken, not left blank because the findings were reassuring. Each domain records clinical meaning, not bare circumstances: the layoff is documented as the collapse of the client's coping structure, not just an event. Strengths are documented and then used: sobriety supports, family, and prior treatment response all reappear in the recommendations. The diagnosis and each recommendation trace to findings documented above them, including referrals that answer the social-domain needs. Times, consent, signature, and credentials are all in the document. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The biopsychosocial assessment concentrates the most sensitive material in the chart: trauma exposure, substance history, legal involvement, finances, family conflict. All of it sits in the designated record set, where the client can request it and payers can audit it, so collect to clinical need and no further. What reviewers actually weigh is the synthesis. Ontario's social work regulator defines this document by its judgments rather than its fields: a social work diagnosis, also known as a biopsychosocial assessment, is "a series of judgments based on social work knowledge and skills," conducted through a trauma-informed, anti-oppressive, and culturally informed lens, that identifies underlying causes and supports the intervention plan ( OCSWSSW Standards, 3rd edition ). An assessment whose integrated summary could not support the level of care it recommends fails that test no matter how complete its checkboxes. In the US, no general law requires a document with this name; the mandates are program rules, and the deadlines belong to the program, not the document. LAW: opioid treatment programs must complete the full initial psychosocial assessment, care plan included, within 14 calendar days of admission ( 42 CFR 8.12 ). LAW: Medicare-certified community mental health centers owe an initial evaluation within 24 hours and a comprehensive assessment within 4 working days, updated at least every 30 days for partial hospitalization and intensive outpatient clients ( 42 CFR 485.914 ). PAYER POLICY: CCBHC certification criteria give clinics 60 calendar days for the comprehensive evaluation and expressly allow treatment to begin during that window. PAYER POLICY: California's Medi-Cal documentation rules ( BHIN 23-068 ) set no fixed general deadline at all: assessments proceed "as expeditiously as possible" per clinical need, medically necessary services are reimbursable during assessment, and the assessment "may be in any format" so long as the seven required domains are present. That last clause is the general shape everywhere: the format is a convention; the content is the requirement. When the admission interview is billed, the assessment is the record that supports the claim, and Medicare contractors diverge on how often that service is payable: one policy allows repeat evaluation after an extended treatment hiatus of about 6 months, an inpatient admission, or a significant change in mental status, while Novitas caps coverage at three evaluations per year, per beneficiary, by the same provider or group. There is no national once-per-year rule; check your own MAC and plan. In Australia, keep three artifacts distinct: the referring practitioner's mental health treatment plan (a living document that does not expire), the referral (which the treating allied health professional must retain for 2 years, a rule about the referral rather than the clinical record), and your own assessment under the Psychology Board's record standards . The assessment then hands off to the treatment plan , and the baselines recorded here are what later progress notes move against. ↑ Back to the template and downloads ## Common biopsychosocial assessment errors auditors flag Behavioral health audit findings rarely turn on which template was used; they turn on what the record proves. In Medicare's 2024 reporting-period data for outpatient psychiatric services, the improper-payment rate was 16.1%, a projected $254.5 million, and insufficient documentation drove 78.3% of it. A national OIG psychotherapy audit estimated $580 million improperly paid, citing missing session time and missing signatures among the failures. Across the public audit materials reviewed for this page, failures route through documentation, medical necessity, frequency, eligibility, time, and signatures; none names a denial code for a missing biopsychosocial assessment. The BastionGPT Clinical Advisory Board sees the same errors most often in biopsychosocial assessment reviews: Data without synthesis. Every domain is filled in and nothing connects them. The integrated summary is what separates an assessment from a questionnaire, and it is the section supervisors and reviewers read first. Domains asserted rather than assessed. "Noncontributory" or "denies all" with nothing showing the question was asked, most often in the biological and trauma sections. To an auditor, and to the next clinician, it reads as not assessed. The client's paperwork standing in for the assessment. Self-report intake forms copied forward without verification, synthesis, or the clinician's own conclusions, leaving a long document with no professional judgment in it. Overcollected sensitive detail. Trauma narrative, family conflict, and legal history recorded past clinical need, inside the most-requested document in the chart. Recommendations that do not follow. A level of care or referral list with no documented findings behind it. The medical-necessity chain starts here, and a disconnect at the assessment breaks every later note that builds on it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on biopsychosocial assessments. Draft a complete biopsychosocial assessment from interview bullets, a dictation, or a pasted transcript. Merge referral packets, prior records, and screening scores into the right domains so the history arrives organized. Check a finished assessment before you sign: an empty domain, a missing risk screen, or a summary that repeats instead of synthesizing. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a biopsychosocial assessment be? Most biopsychosocial assessments run 600 to 1,500 words and take 45 to 75 minutes by hand, the longest routine document in most behavioral health charts. The time range is a clinical team estimate: no published study benchmarks completion time for this specific document. No regulation in the US, Canada, or Australia sets a word count or page length either; where rules exist, they govern completion deadlines and content domains for specific programs. Write to what the document must support: the diagnosis, the level of care, and the treatment plan. What is the difference between a biopsychosocial assessment and an intake note? They cover the same event, the start of care, at different depths. An intake note is the leaner first-session record a private practice typically writes; the biopsychosocial assessment is the fuller workup, with each domain assessed in its own right and an integrated summary tying them together. They are not required to be separate documents: in many agencies, the biopsychosocial assessment is the intake documentation. Is a biopsychosocial assessment required by law? Not by any general law. In the US, the mandates are program rules: federally certified opioid treatment programs must complete a full initial psychosocial assessment within 14 calendar days of admission ( 42 CFR 8.12 ), Medicare-certified community mental health centers owe a comprehensive assessment within 4 working days (42 CFR 485.914), CCBHC certification adds a 60-day comprehensive evaluation, and state Medicaid programs such as California's specify required assessment domains. In Canada and Australia, regulator record standards require sufficient, secure records without prescribing a biopsychosocial form. For everyone else the document is professional convention, which is not the same as optional: it is how the chart proves the diagnosis and level of care were grounded. Is a biopsychosocial assessment the same as a psychosocial assessment? The names are often used interchangeably, and agencies define their own forms. US opioid treatment regulations call the required admission document a psychosocial assessment, and it covers medical, psychiatric, economic, legal, housing, and vocational needs, which is the full biopsychosocial sweep. Where a distinction is drawn, the psychosocial assessment is the social work tradition's document, centered on functioning, relationships, and environment, while the biopsychosocial assessment gives the biological domain equal, explicit standing. Some settings extend it further to a biopsychosocial-spiritual assessment. What matters to a reviewer is whether the content your program requires is present, not the title. Is the biopsychosocial assessment billed under 90791? 90791 and 90792 describe a service, the psychiatric diagnostic evaluation, not a document, and Medicare contractor policy describes that service in this document's terms. Novitas policy states plainly: "The diagnostic evaluation is a biopsychosocial assessment." When the admission interview is billed as 90791, this document is the record that supports it. Frequency rules differ by contractor: Novitas caps coverage at three evaluations per year, per beneficiary, by the same provider or group, while another contractor instead allows repeats after an extended treatment hiatus, an inpatient admission, or a significant change in mental status. There is no national once-per-year rule; check your own MAC and payer contract. How long do biopsychosocial assessments have to be kept? HIPAA's six-year rule is about compliance documentation, not client charts: HHS says the Privacy Rule sets no medical-record retention period and state law governs. So the assessment follows your chart's rules: California requires 7 years after therapy terminates and New York at least 6, with longer for minors; Ontario's psychology college requires 10 years and British Columbia's health-professions college 16; Australia's Psychology Board code requires 7 years from the last entry, or until the 25th birthday for clients under 18. All of these run from the end of care or adulthood, not from the assessment date. How soon after admission must a biopsychosocial assessment be completed? The deadline belongs to the program, not the document. Opioid treatment programs: 14 calendar days, by federal law. Medicare-certified community mental health centers: an initial evaluation within 24 hours and the comprehensive assessment within 4 working days. CCBHCs: a comprehensive evaluation within 60 calendar days, and the criteria expressly allow treatment to begin during that window. California Medi-Cal replaced fixed general deadlines with a clinical-timeliness standard and pays for medically necessary services delivered while the assessment is still in progress. Private practice inherits none of these deadlines; timely completion and dated authentication are still the expectation everywhere. Who can write a biopsychosocial assessment, and does it need a co-signature? Authorship rules come from your program, payer, and licensing board, not from a national standard, and "QMHP" is not one national credential: states define the title differently. California's Medi-Cal rules show the common shape: licensed and non-licensed staff may contribute within their scopes, the diagnosis must be made or directed by an appropriately licensed practitioner, and an assessment completed by a registered counselor gets reviewed by a licensed practitioner who makes the initial diagnosis. No universal rule requires a supervisor to co-sign every pre-licensed clinician's assessment; Ontario's psychology college is unusually explicit that documents relied on for decisions affecting care need the responsible supervisor's signature. And no reviewed national rule requires the client to sign a biopsychosocial assessment: signature requirements attach to consents and treatment plans, not to this document's title. Can BastionGPT draft a biopsychosocial assessment from a transcript? Yes. Paste the admission interview transcript, dictate, or give it bullets, and it produces a structured draft with every domain and the integrated summary ready for your review. It can also merge referral packets, prior records, and screening scores into the right domains. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Intake Note Psychiatric Diagnostic Evaluation Trauma Assessment Case Formulation Note Treatment Plan Mental Status Exam (MSE) Note Substance Use Assessment Child & Adolescent Developmental Assessment Cultural Formulation Interview (CFI) Intake Aboriginal & Torres Strait Islander Culturally Safe Assessment Eating Disorder Assessment Indigenous & First Nations Wellness Assessment Addendum CAPS-5 Documentation CANS Assessment Documentation SASSI Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/intake-note All note templates # Intake Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A therapy intake note documents a new client's first clinical session: presenting problem, relevant history, a risk screen, mental status observations, an initial clinical impression, and the starting plan. Therapists, counselors, and psychologists write one at the start of each new episode of care, and the rest of the chart builds on it. A typical intake note runs 300 to 900 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, psychologists, clinical social workers Audience Downstream treating team, supervisors, payers, auditors Typical length 300 to 900 words · 20 to 40 minutes by hand (clinical team estimate) Format family Intake and assessment documentation (compare: biopsychosocial assessment, psychiatric diagnostic evaluation) When it's used At the first session with a new client, or at the start of a new episode of care Standards context No US, Canadian, or Australian law prescribes an intake form; rules attach to the record, facility, billed service, and payer On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an intake note? An intake note is the record of a client's first clinical session: why the person is seeking care, the history that bears on treatment, a risk screen, early observations, and the starting plan. Clinicians also call it an initial assessment, intake assessment, or initial evaluation. No single author or standards body created the format. It descends from the case-history tradition, with roots in Mary Richmond's 1917 Social Diagnosis, Lawrence Weed's 1968 problem-oriented medical record, and George Engel's 1977 biopsychosocial model, and every practice adapts it. No US, Canadian, or Australian rule prescribes a document called an intake note: requirements attach to the clinical record, the facility type, the billed service, and the payer program instead. Two boundaries matter more than the rest. First, the intake note is a document, while the psychiatric diagnostic evaluation (90791 or 90792) is a billable service: Medicare contractor policy describes that service as a biopsychosocial assessment, so when the first session is billed as one, the intake note is the record that has to support it. Second, an intake note is part of the standard clinical record, not a HIPAA psychotherapy note : the special protected category expressly excludes diagnosis, symptoms, functional status, treatment plans, prognosis, and progress, which is most of what an intake note contains. Write it knowing the client, other providers, and payers can read it. ## Who uses intake notes and when Every setting that accepts new clients writes intake documentation: private and group practices, community mental health centers, university counseling services, employee assistance programs, and telehealth practices. The intake note starts the episode of care, and its readers are often downstream: the clinician a client transfers to, the supervisor reviewing a pre-licensed clinician's work, the payer reviewing medical necessity, or you, a year later. After intake, ongoing sessions move to a progress note format and the treatment direction lives in the treatment plan . Specialized concerns get their own deeper workups, such as a trauma assessment , and agencies that need the full biological, psychological, and social picture use the longer biopsychosocial assessment . ## Intake note structure: what goes in each section Identifying information and referral. Client identifiers, date, clinician, referral source, service type, and start and stop times, together in one header block. Pitfall: scattering these details through the narrative; reviewers and records staff need them findable in seconds. Presenting problem. Why the client is here now, in their own words, plus onset, duration, severity, and what prompted the contact. Pitfall: recording only your clinical rephrasing; a short quote preserves the client's frame and anchors medical necessity. Relevant history. Psychiatric history and prior treatment, medical conditions and current medications, and family psychiatric history. Pitfall: writing "noncontributory" without asking; to an auditor it reads as not assessed. Social context. Living situation, relationships, work or school, legal involvement, and cultural or identity factors the client raises. Pitfall: overcollecting sensitive detail with no clinical use; everything here is part of the accessible record. Substance use. Current and past use, substances, frequency, and most recent use. Pitfall: skipping the screen because the presenting problem seems unrelated; intake is exactly where screening is expected. Risk screen. Current and historical suicidal ideation, self-harm, and thoughts of harming others, protective factors, and any action taken. Pitfall: leaving risk blank when findings are negative; write the negative finding so the record shows you asked. Mental status observations. A brief mental status exam: appearance, behavior, speech, mood, affect, thought process, insight, judgment. Pitfall: template MSE strings that contradict the narrative, such as "euthymic mood" above a tearful presentation. Initial impression and plan. Provisional diagnosis with rule-outs, initial goals, recommended modality and frequency, referrals, consent confirmation, and the next appointment. Pitfall: an impression that floats free; every diagnosis needs supporting symptoms documented above it. ### Blank template (copy and adapt) Client: [initials] DOB/Age: Date: Clinician: Referral source: Service: [intake / initial assessment] Start/stop time: PRESENTING PROBLEM In the client's words: Onset / duration / severity / precipitant: HISTORY Psychiatric (prior treatment, hospitalizations): Medical conditions + current medications: Family psychiatric history: SOCIAL CONTEXT (living, relationships, work/school, legal, cultural): SUBSTANCE USE (substances, frequency, last use): RISK SCREEN SI / self-harm / harm to others (current + history): Protective factors / action taken: MENTAL STATUS (appearance, speech, mood, affect, thought process, insight): STRENGTHS & SUPPORTS: MEASURES (e.g., PHQ-9, GAD-7 + scores): INITIAL IMPRESSION (provisional dx + rule-outs): INITIAL PLAN (goals, modality, frequency, referrals, next appt): Consent for treatment reviewed and signed: [Y/N] Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample intake note Scenario: adult client, self-referred for anxiety and sleep problems after a job change, first session in an outpatient private practice. All details are fictional. Client: R.L., 32 · Date: 07/09/2026 · Clinician: A. Morgan, LPC · Service: Intake / initial assessment, 58 min, in office · Referral: Self · Start/stop: 14:02 to 15:00 Presenting problem: Client sought care for "worry I can't switch off since the promotion," describing four months of escalating anxiety after moving into a supervisory role in March. Reports taking 60 to 90 minutes to fall asleep most nights, Sunday-evening dread, irritability at home, and two episodes of racing heart and shortness of breath during meetings, each resolving within minutes. Wants "tools to turn work off and be present with my family." History: No prior psychotherapy or psychiatric hospitalization; brief school counseling at 16 after parents' divorce. No psychiatric medications, current or past. Medical: unremarkable PCP visit January 2026 per client report; no current medications except occasional ibuprofen. Family history: mother treated for depression in her 40s, per client report. Social context: Lives with spouse and 4-year-old daughter; describes the marriage as supportive though "tense lately because I bring work home." Operations supervisor with 14 direct reports since March; denies legal involvement. Running was a primary coping outlet and has lapsed since the promotion. Substance use: Alcohol 2 to 3 drinks per week, unchanged in the past year; denies tobacco, cannabis, and other drug use; denies any history of problem use. Caffeine: 2 coffees daily. Risk screen: Denies current suicidal ideation, intent, or plan; denies history of suicidal behavior or self-harm; denies thoughts of harming others. Protective factors: engaged parenting role, supportive spouse, future orientation, help-seeking. No risk indicators observed in session. Routine monitoring is appropriate; no safety plan indicated at this time. Mental status: On time, groomed, cooperative. Speech normal in rate and volume. Mood "keyed up"; affect anxious, full range, congruent. Thought process linear and goal-directed. No perceptual disturbance reported or observed. Insight and judgment good. Measures: GAD-7 = 12 (moderate). PHQ-9 = 6 (mild); item 9 not endorsed. Strengths and supports: Motivated and specific about goals, stable employment, supportive family, prior success using structure and exercise as coping. Initial impression: Adjustment disorder with anxiety (F43.22), provisional; rule out generalized anxiety disorder given the four-month course and physical symptoms. Symptoms cause clinically significant distress and functional impact at work and home. Initial plan: Weekly individual CBT, 50-minute sessions, initial course of 8 to 12 sessions. Session 2: build the collaborative treatment plan with measurable goals. Begin a sleep log and scheduled worry time this week. Client will schedule a PCP follow-up to rule out medical contributors to the episodic palpitations. Re-administer GAD-7 and PHQ-9 at session 4. Limits of confidentiality, fees, and practice policies reviewed; informed consent signed and filed. Next appointment 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The risk screen is written out even though every finding is negative, so the record shows screening happened rather than silence. Baseline GAD-7 and PHQ-9 scores give every later progress note something measurable to move. The presenting problem keeps the client's own words, anchoring medical necessity in the client's report rather than clinician paraphrase. The provisional diagnosis traces to symptoms documented above it, with an explicit rule-out and a functional-impact statement. The plan is specific and dated, with frequency, modality, a re-screening point, and a next appointment, and consent is documented in the same note. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations An intake note is part of the designated record set: the client can request it, payers can audit it, and every later note builds on it. Keep private process reflections out of it; they belong in segregated psychotherapy notes with their own authorization rules. Two Ontario rules answer questions clinicians everywhere ask. A client cannot condition consent on keeping professionally required information out of the chart: Ontario's health privacy law makes such a condition ineffective where it "purports to prohibit or restrict any recording" required by law or established professional practice ( PHIPA s. 19(2) ). And rough interview jottings do not have to become part of the record: Ontario's psychotherapy college permits destroying rough notes the same day , provided they are used to complete the formal record first. Treat both as the shape of the rule and check your own jurisdiction. For a billed intake in the US, the medical-necessity chain starts here. Medicare contractor policy treats the diagnostic evaluation as a biopsychosocial assessment , and its documentation expectations reach the note: patient identification on every page, the legible signature of the clinician responsible for the care, and support for medical necessity. Frequency is contractor policy, not a single national rule. The same Novitas policy allows another evaluation for a new episode of illness, a readmission due to complications, or a new referral question, recognizes that children and geriatric patients may need more than one visit, and caps coverage at three evaluations per year, per beneficiary, by the same provider. The famous 24-hour intake deadline is also narrower than most template pages suggest: 42 CFR 485.914 gives Medicare-certified community mental health centers 24 hours for the initial evaluation and 4 working days for the comprehensive assessment; an ordinary private practice does not inherit those deadlines. In Australia, Better Access turns on a different document entirely: after each course of treatment the therapist must send the referrer a written report covering assessments, treatment, and recommendations, and a verbal report does not qualify for Medicare payment. Wherever you practice, the intake note feeds the treatment plan and sets the baseline that progress notes move against, so record baselines you intend to measure. ↑ Back to the template and downloads ## Common intake note errors auditors flag The audit record for intake documentation is largely the audit record for the services it supports. In Medicare's 2024 reporting-period data for outpatient psychiatry, the improper-payment rate was 16.1%, a projected $254.5 million, and insufficient documentation drove 78.3% of it, with another 17% having no documentation at all. A national OIG psychotherapy audit estimated $580 million improperly paid, with signatures missing on 31 of 216 sampled days. The same audit also rebuts one-size-fits-all advice: 35 sampled days had no treatment plan and were still allowable, because only three of the seven Medicare contractors involved required one. The BastionGPT Clinical Advisory Board sees the same errors most often in intake note reviews: History sections asserted rather than assessed. "Noncontributory" or "denies all" with nothing showing the question was asked. An auditor reads it as not assessed, and a successor clinician cannot rely on it. The questionnaire standing in for the assessment. The client's intake form is source material. Copying it forward without verification, synthesis, and the clinician's own conclusions leaves the note without professional judgment. A diagnosis without documented support. The impression names a disorder, but the symptoms, duration, and functional impact recorded above it do not add up to the criteria. That breaks medical necessity from day one of the episode. Missing authentication. No signature or credentials, an unsigned diagnostic conclusion, or a note completed and signed long after the session date. Risk silence at baseline. No statement about suicidal ideation, self-harm, or harm to others either way, so the whole chart inherits an intake with no documented starting risk status. ### Intake notes in the US, Canada, and Australia Aspect United States Canada Australia Status No federal intake form; content rules attach to facility type (the 24-hour rule is specific to Medicare-certified CMHCs) and to Medicare contractor policy Provincial college standards prescribe record contents (Ontario CRPO client profile, CPBAO record requirements, Quebec's psychologist regulation) Psychology Board Code of Conduct (December 2025) makes records mandatory but expressly leaves form and content to the service and setting Terminology Intake note, initial assessment, psychiatric diagnostic evaluation Client profile, initial assessment, clinical record Initial consultation, client records; Better Access adds the referrer's treatment plan and the therapist's course report What changes Payer and contractor documentation fields, medical-necessity support, authentication rules College-prescribed record contents and provincial privacy law (PHIPA and analogues) MBS referral and reporting chain, course caps, program eligibility Retention State law, not HIPAA: California 7 years after termination, New York at least 6 years, minors longer in both Province and college: Quebec 5 years, Ontario 10 years, British Columbia 16 years, each running from last contact or adulthood 7 years from the last entry for adults; until the 25th birthday for clients under 18 (Board code) The intake note itself travels well across all three countries; the rules around it do not. Content requirements come from your college, contractor, or program, and retention runs from the end of care or adulthood, not from the intake date. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on intake notes. Draft a complete intake note from a few bullets, a dictation, or a pasted first-session transcript. Summarize a referral packet or prior records into the history sections so intake starts informed. Check a finished note for a missing risk screen, baseline scores, or consent language before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should an intake note be? Most intake notes run 300 to 900 words and take 20 to 40 minutes by hand, longer than a progress note because they carry history and the baseline risk screen. No regulation in the US, Canada, or Australia sets a word count or a required intake format; length should come from what the note must support: continuity, safety decisions, and medical necessity, without irrelevant or stigmatizing detail. What is the difference between an intake note and a progress note? The intake note is the baseline: history, risk screen, initial impression, and starting plan, written once per episode of care. A progress note documents one session's work and what changed against that baseline. Progress notes should not keep reproducing the intake history; they move the story forward. Is an intake note the same as a biopsychosocial assessment? A biopsychosocial assessment is a content framework covering biological, psychological, and social domains. It can be the intake note, an attached structured assessment, or a separate, longer document; no rule requires them to be separate. Agencies and higher-acuity settings tend to use the fuller standalone version, while private practices often fold the domains into a leaner intake note. Is the intake note a psychotherapy note under HIPAA? No. HIPAA's psychotherapy-notes category expressly excludes diagnosis, symptoms, functional status, the treatment plan, prognosis, and progress, which is most of what an intake note contains. The intake note is part of the standard record and the designated record set, and labeling it a psychotherapy note does not change that. True psychotherapy notes must be kept separate and have their own authorization rules . Is the intake billed as 90791, and how often can that be billed? 90791 and 90792 describe a service, the psychiatric diagnostic evaluation, and Medicare contractor policy describes it as a biopsychosocial assessment. When that service is actually performed at the first session, the intake note is the record that supports it. Frequency is contractor policy rather than one national rule: Novitas policy, for example, allows another evaluation for a new episode, a readmission with complications, or a new referral question, recognizes that children and geriatric patients may need more than one visit, and caps coverage at three per year per beneficiary by the same provider. Check your own MAC and payer contract. How soon after the first session must the intake note be completed? There is no universal deadline. The often-quoted 24-hour rule belongs to Medicare-certified community mental health centers, where 42 CFR 485.914 requires an initial evaluation within 24 hours of admission and a comprehensive assessment within 4 working days; private practices do not inherit it. Australia's Psychology Board code says records are made at the time of events or as soon as possible afterward, and every jurisdiction expects timely, dated authentication. Does a supervisor have to co-sign a pre-licensed clinician's intake note? There is no blanket rule. Ontario's psychology college has a targeted version: documents containing conclusions, judgments, decisions, diagnoses, or recommendations need the responsible registrant's signature, and both sign when the author is not authorized for autonomous practice. US Medicare requires a supervising signature where the billing or supervision arrangement calls for one. What governs is your licensing board, the payer's rules, the facility policy, and the written supervision agreement. How long do intake records have to be kept? HIPAA sets no chart-retention period; state law governs in the US. California requires 7 years after therapy terminates, and New York at least 6 years, with longer periods for minors in both. Canada varies by more than threefold: Quebec psychologists 5 years, Ontario 10 years, British Columbia 16 years. Australia's Board code requires 7 years from the last entry, or until the 25th birthday for clients under 18. All of these run from the end of care or adulthood, not from the intake date. Can BastionGPT draft an intake note from a session transcript? Yes. Paste a first-session transcript, dictate, or give it bullets, and it produces a structured intake note draft for your review, including the sections payers expect. It can also summarize a referral packet into the history sections. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and the exclusions that keep ordinary intake content in the standard record. HHS, HIPAA FAQ 580 : the Privacy Rule sets no medical-record retention period; state law governs. 42 CFR § 485.914 : the 24-hour initial evaluation and 4-working-day comprehensive assessment for Medicare-certified community mental health centers. CMS Medicare Coverage Database, Novitas LCD L35101 : the diagnostic evaluation as a biopsychosocial assessment, re-evaluation circumstances, and frequency limits. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : $580 million estimated improper payments, missing signatures, and the contractor-by-contractor treatment-plan finding. CMS Medicare Learning Network, outpatient psychiatric care compliance tips : 2024 improper-payment rate and denial-reason shares. Ontario, Personal Health Information Protection Act, 2004, s. 19(2) : conditions on consent cannot suppress professionally required chart entries. College of Registered Psychotherapists of Ontario, Clinical Records standard : client profile contents, 10-year retention, and same-day destruction of rough notes. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : record contents, the dual-signature rule, and 10-year retention. Quebec, Regulation respecting record keeping by psychologists, C-26, r. 221 : required record contents and 5-year retention. College of Health and Care Professionals of BC, Records practice standard : 16-year retention under the harmonized standard. Psychology Board of Australia, Code of Conduct (effective December 2025): record standards, contemporaneity, and retention. Australian Government Department of Health, AskMBS advisory on Better Access allied mental health services : the written course-report requirement. ## Related templates Trauma Assessment Treatment Plan Psychotherapy Progress Note SOAP Note Case Formulation Note Psychiatric Diagnostic Evaluation Biopsychosocial Assessment Suicide Risk Assessment Safety Plan Discharge Summary Prior Authorization Support Letter Psychological Evaluation Report Release of Information (ROI) Authorization Mental Status Exam (MSE) Note Substance Use Assessment Child & Adolescent Developmental Assessment Cultural Formulation Interview (CFI) Intake Good Faith Estimate (No Surprises Act) Financial Policy & Fee Agreement Consent for Treatment of a Minor Aboriginal & Torres Strait Islander Culturally Safe Assessment General Risk Screening Note Eating Disorder Assessment Parent & Caregiver Collateral Note Indigenous & First Nations Wellness Assessment Addendum CANS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mental-status-exam All note templates # Mental Status Exam (MSE): Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A mental status exam (MSE) is the clinician's structured description of how a client presents at one point in time: appearance, behavior, speech, mood, affect, thought process, thought content, perception, cognition, insight, and judgment. Psychologists, psychiatrists, therapists, and medical teams record a complete MSE at intake or diagnostic evaluation and briefer focused versions afterward. Most written MSEs run 80 to 250 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, psychiatrists, therapists, psychiatric NPs, medical teams Audience Treating clinician, care team, supervisors, payers, auditors, courts Typical length 80 to 250 words · 2 to 6 minutes by hand (clinical team estimate) Format family Structured exam section (compare: intake note, psychiatric diagnostic evaluation) When it's used Complete at intake and diagnostic evaluations; focused versions in follow-up notes Standards context A clinical convention, not a mandated form; federal law requires a mental status record only in inpatient psychiatric hospitals On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a mental status exam? A mental status exam (MSE) is the clinician's structured description of how a client presents at one point in time, organized into domains: appearance, behavior, speech, mood, affect, thought process, thought content, perception, cognition, insight, and judgment. It is psychiatry's counterpart to the physical exam, and it records observation, not history. StatPearls credits Adolf Meyer with the first standardized outline in 1918, and the modern domain structure was codified by textbooks, most visibly Trzepacz and Baker's The Psychiatric Mental Status Examination (Oxford University Press, 1993), not by any regulator. Australian and UK sources write "mental state examination"; the exam is the same. It usually sits inside a larger document, an intake note , a psychiatric diagnostic evaluation , or a progress note, and it can also stand alone. Two clarifications do most of the work. First, the MSE is a convention, not a mandate: no US, Canadian, or Australian statute standardizes its content or domain count. The one federal rule that comes close, 42 CFR 482.61 , requires inpatient psychiatric hospitals to include "a record of mental status" in a psychiatric evaluation completed within 60 hours of admission, and it applies nowhere else. Second, the MSE is not the MMSE. The Mini-Mental State Examination is a scored 30-point cognitive screen, published by Folstein and colleagues in 1975, that has been copyrighted and actively licensed since 2000; the free-form MSE has no owner, no fee, and no fixed form. ## Who uses mental status exams and when Psychologists, psychiatrists, therapists, psychiatric NPs, and medical teams all record MSEs, with depth matched to the encounter. A complete exam belongs in the psychiatric diagnostic evaluation , where Medicare contractor policy expects it, and in most intake and biopsychosocial workups. Focused versions carry the observational load in progress notes, medication follow-ups, and crisis encounters , where the same contractor policy pairs a mental status exam with a disposition. Capacity and guardianship work leans on the MSE hardest: Washington's DSHS, for example, requires a detailed MSE with every mental incapacity evaluation. If a note has room for only one objective element, the MSE is usually it. ## Mental status exam structure: what goes in each section Presentation: appearance, behavior and attitude, speech. Emotion: mood in the client's words, affect as you observe it. Thinking: thought process, thought content with a stated SI/HI status, perception. Capacity: cognition, insight, judgment. The familiar ten-part list below is textbook convention, not a legal requirement, so adapt the headings to your setting. Each domain, with the pitfall that most often undermines it: Appearance. Observable presentation: grooming, dress, apparent versus stated age, notable physical features. Pitfall: evaluative shorthand such as "disheveled" with nothing observable behind it; write what a reviewer could have seen. Behavior and attitude. Eye contact, cooperation, rapport, and psychomotor activity such as agitation or slowing. Pitfall: "calm and cooperative" copied forward every visit; this is a point-in-time exam, and identical entries across sessions read as cloning. Speech. Delivery, not content: rate, volume, fluency, latency; pressured, slowed, monotone, halting. Pitfall: describing what the client said; that belongs under thought content. Mood. The client's own report of their internal state, ideally quoted. Pitfall: recording your inference as the client's mood; if the client says "fine," quote it and let affect carry your observation. Affect. What you observe: range, intensity, stability, and congruence with mood and content. Pitfall: mood and affect that contradict each other with no comment; they are different data sources, and reviewers notice silent mismatches. Thought process. Organization and flow: linear and goal-directed, circumstantial, tangential, loose. Pitfall: a label with no example; "tangential" earns its place with one sentence showing the tangent. Thought content. Preoccupations, obsessions, delusions, and an explicit suicidal and homicidal ideation status. Pitfall: silence on SI/HI in a session where risk came up; state it either way and keep it consistent with your suicide risk assessment . Perception. Hallucinations, illusions, and dissociative experiences, asked about or observed. Pitfall: writing "denies hallucinations" when you never asked; document only what you screened. Cognition. Orientation, attention, concentration, and memory, observed or briefly tested. Pitfall: pasting an MMSE or MoCA score as the whole entry; a screening score can inform this domain, it cannot replace the exam. Insight and judgment. Awareness of the condition, and the quality of recent decisions, each anchored to something the client actually said or did. Pitfall: bare "fair" or "poor" ratings with no example and no link to a treatment decision. ### Blank template (copy and adapt) Client: [initials] Date/time: Setting: Embedded in: [intake / diagnostic evaluation / progress note / standalone] Appearance: Behavior and attitude: Speech: Mood (client's words): Affect: Thought process: Thought content (SI/HI status): Perception: Cognition (orientation, attention, memory): Insight: Judgment: Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample mental status exam Scenario: adult client, initial diagnostic evaluation for depressive symptoms; this is the MSE section of that evaluation. All details are fictional. Client: R.L., 41 · Date: 07/14/2026 · Service: Initial diagnostic evaluation, MSE section · Clinician: Licensed psychologist Appearance: Arrived on time. Dressed appropriately for the weather, adequate grooming, appears stated age. Behavior and attitude: Cooperative and engaged. Intermittent eye contact, mild psychomotor slowing, no agitation observed. Speech: Normal rate and volume; noticeable pauses before answering. Mood: "Worn out, like I'm running on empty." Affect: Constricted and dysphoric, congruent with mood and content; brightened briefly when discussing family. Thought process: Linear and goal-directed. Thought content: Persistent worry about job performance. No delusions, obsessions, or phobias elicited. Denies current suicidal or homicidal ideation, plan, or intent; full risk screen documented in the evaluation's risk section. Perception: Denies hallucinations and illusions when asked; none observed. Cognition: Alert. Oriented to person, place, time, and situation. Attention intact through a 70-minute interview; registered 3 of 3 words and recalled 2 of 3 at five minutes, the third with a category cue. Insight: Good. Connects low mood to six months of poor sleep and work stress, and identifies what has helped before. Judgment: Intact. Sought evaluation independently and arranged work coverage to attend appointments. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Mood is quoted and affect is observed separately, keeping the two data sources distinct and visibly consistent. Thought content states SI/HI status either way and points to the fuller risk screen, so the record shows monitoring rather than silence. Nearly every label carries an observable: pauses before answering, 2 of 3 recall with a cue, work coverage arranged. That is what an auditor or a court can rely on. Cognition documents brief testing in plain observations without letting a screening score stand in for the exam. The header records which document the exam sits inside, so it stays findable in the clinical record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations MSE findings live in the standard clinical record, the designated record set, and are not HIPAA psychotherapy notes: 45 CFR 164.501 excludes any summary of diagnosis, symptoms, functional status, prognosis, and progress from that protected category. Clients can request the exam, a release of information can send it to third parties, and courts can subpoena it. Write each domain as observables a stranger could verify, and keep hypothesis-level process material in segregated psychotherapy notes with their own authorization rules . Where a complete MSE is expected is payer policy and convention, not statute. First Coast's LCD L33252 lists "a complete mental status exam" among the elements of a psychiatric diagnostic evaluation (90791, 90792) and pairs a mental status exam with a disposition in psychotherapy for crisis; other Medicare contractors phrase expectations differently, and no rule requires a full MSE in every routine psychotherapy note. Law enters in one narrow place: 42 CFR 482.61 requires inpatient psychiatric hospitals to complete a psychiatric evaluation containing "a record of mental status" within 60 hours of admission. Since January 1, 2021, office-visit E/M levels come from medical decision making or total time, so the exam documents clinical reality and medical necessity rather than setting the billing level ( CY2021 fee schedule final rule ). In Australia, MBS explanatory notes require "adequate and contemporaneous records" judged by Professional Services Review peers, with no MSE-specific content rule; in Canada, provincial college standards govern records and no statute names the exam. The format is a convention; the content is the requirement. ↑ Back to the template and downloads ## Common mental status exam errors auditors flag Domain completeness is rarely the direct denial trigger; behavioral health audits turn on time, signatures, and internal consistency, and the MSE is where inconsistency shows first. In a 2020 OIG audit of a psychiatric services provider , 99 of 100 sampled psychotherapy services failed Medicare requirements, time was undocumented in 82 of them, and the estimated $421,272 overpayment equaled 93 percent of everything paid. Exam quality is still measurable: one published audit cycle moved correctly recorded MSE parameters from 69 percent to 83 percent after a standardized format was introduced ( Psychiatric Bulletin ). The BastionGPT Clinical Advisory Board sees the same errors most often in mental status exam reviews: Mood and affect that conflict. A quoted mood of "terrible" beside "euthymic, full range," or an affect line the session narrative contradicts, with no comment on the mismatch. The cloned all-normal exam. An identical within-normal-limits block pasted across visits. Medicaid program guidance is blunt that documentation cloning is prohibited, and an exam that never changes stops being evidence of anything. A screening score standing in for the exam. An MMSE or MoCA number pasted where observations belong. Washington's DSHS guideline states it plainly: "A mini mental status exam (MMSE) cannot be substituted for a MSE." SI/HI silence in thought content. Risk discussed in session, assessed on a separate suicide risk assessment , and then absent from the MSE, leaving the two documents out of sync. Labels without observables. "Guarded," "tangential," or "poor judgment" with no example. Published reliability work found attitude, affect, and motor activity the most variably rated domains ( Nordic Journal of Psychiatry ), which is exactly where anchors matter most. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on mental status exam notes. Extract a complete MSE from a session transcript or dictation, with each observation sorted into the right domain. Check a drafted exam before you sign: missing domains, mood and affect mismatches, an unstated SI/HI status. Draft the full section from a few observation bullets, in narrative or list form to match your chart. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I need a full mental status exam in every progress note? No regulation requires it. Medicare contractor policy attaches "a complete mental status exam" to the psychiatric diagnostic evaluation (90791, 90792), and Medicare's Program Integrity Manual says progress notes "may be in any form or format." A brief, focused MSE in routine sessions is convention and good practice, not law: document what you observed and what changed. What are the standard components of a mental status exam? The textbook list runs about ten domains: appearance, behavior and attitude, speech, mood, affect, thought process, thought content, perception, cognition, insight, and judgment. No US, Canadian, or Australian regulation specifies a required set or count, so adapt the headings to your setting and keep the observations concrete. Are MSE findings protected as psychotherapy notes under HIPAA? No. HIPAA's psychotherapy-notes definition excludes any summary of diagnosis, symptoms, functional status, prognosis, and progress, which is what an MSE is, and anything kept in the ordinary chart falls outside the protected category anyway. Treat the exam as releasable and discoverable, and keep private process notes segregated under their own authorization rules . What is the difference between the MSE, the MMSE, and the MoCA? The MSE is a free-form clinical examination with no owner or fee. The MMSE is a scored 30-point cognitive screen, published in 1975 and enforced under copyright since 2000, and the MoCA requires training and certification for many of the clinicians who administer it. A score from either can inform the cognition domain; neither replaces the exam. Washington's DSHS guideline is explicit that an MMSE "cannot be substituted" for an MSE. Can I write "within normal limits" for the whole exam? For genuinely unremarkable domains it is defensible. Reviewers read positive descriptive findings as stronger evidence than a blanket WNL, so spend words where the findings are: mood, affect, thought content, and anything abnormal. An all-WNL exam repeated identically across visits reads as cloning. Where does suicide risk go in a mental status exam? In thought content, as an explicit SI/HI status recorded either way. A screening or full suicide risk assessment is its own document; the MSE line should match it. Mismatches between the two are among the first things reviewers and attorneys look for. How do I document a mental status exam in telehealth? Record what the modality lets you observe and name its limits: video hides gait and parts of appearance, audio-only hides everything visual. Note the modality, describe observable domains normally, and mark limited ones as not assessable rather than guessing. The telehealth therapy note page covers the session-level requirements. Is it "mental status examination" or "mental state examination"? The same exam. US sources, including the one federal rule that mentions it, say "mental status"; Australian and UK guidance, such as NSW Health's clinical documentation guidelines, says "mental state." Nothing regulatory follows from the name. Can BastionGPT draft a mental status exam? Yes. Give it a transcript, a dictation, or a few observation bullets and it drafts the MSE with each finding in the right domain, then checks for missing domains and internal consistency before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 42 CFR § 482.61 : inpatient psychiatric hospitals must complete a psychiatric evaluation containing "a record of mental status" within 60 hours of admission; the only US federal rule that names one. 45 CFR § 164.501 : HIPAA's psychotherapy-notes definition and the exclusions that keep MSE content in the standard record. First Coast Service Options, LCD L33252 : "a complete mental status exam" within the psychiatric diagnostic evaluation, and a mental status exam with disposition in psychotherapy for crisis. CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes "may be in any form or format." HHS Office of Inspector General, Grand Desert Psychiatric Services audit (2020) : 99 of 100 sampled psychotherapy services noncompliant, time undocumented in 82, $421,272 estimated overpayment. CMS, CY2021 Physician Fee Schedule final rule : history and exam no longer select office-visit E/M levels. StatPearls, Mental Status Examination (NBK546682) : Adolf Meyer's 1918 outline and the conventional domain structure. Blaabjerg et al., Nordic Journal of Psychiatry (2020) : inter-rater reliability of the MSE and its most variably rated domains. Kareem and Ashby, Psychiatric Bulletin (2000) : MSE recording quality across audit cycles with a standardized format. Washington State DSHS, Form 13-865 Addendum, Mental Status Exam Guidelines : required MSE detail for mental incapacity evaluations, and the MMSE substitution bar. NSW Health, Mental Health Clinical Documentation Guidelines (GL2014_002) : Australian "mental state examination" usage and documentation modules. MBS Online, explanatory note GN.15.39 : "adequate and contemporaneous records" and Professional Services Review peer judgment. Copyright at the Bedside (PMC4160306) : the MMSE copyright registration and licensing history. MoCA training and certification : current certification requirements for administering clinicians. ## Related templates Intake Note Psychiatric Diagnostic Evaluation Biopsychosocial Assessment Psychotherapy Progress Note SOAP Note RBANS Report Write-Up Substance Use Assessment Cultural Formulation Interview (CFI) Intake Hospital Consult & Liaison Note Medication Management Follow-Up Note Capacity & Competency Evaluation Aboriginal & Torres Strait Islander Culturally Safe Assessment MoCA Score Documentation MMSE Score Documentation Mini-Cog Score Documentation SLUMS Score Documentation CAPS-5 Documentation GAF Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychiatric-diagnostic-evaluation All note templates # Psychiatric Diagnostic Evaluation: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychiatric diagnostic evaluation is the comprehensive initial assessment that establishes a mental health diagnosis and treatment plan: psychiatric, medical, family, and social history, a mental status exam, and a risk assessment. Psychiatrists, psychologists, therapists, and psychiatric nurse practitioners complete one at the start of care; US payers bill it as CPT 90791, or 90792 when medical services are included. A typical evaluation runs 500 to 1,200 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychiatrists, psychiatric NPs and PAs, psychologists, therapists, counselors, clinical social workers Audience Treating clinician, treatment team, payers, auditors, referral sources Typical length 500 to 1,200 words · 30 to 60 minutes by hand (clinical team estimate) Format family Structured diagnostic assessment (compare: intake note, biopsychosocial assessment) When it's used At the start of care, or when a new episode or a significant change requires reassessment Standards context No law prescribes one form; US Medicare contractor policies list expected elements (90791/90792), and Australia's MBS item 291 codifies its own assessment pathway On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychiatric diagnostic evaluation? A psychiatric diagnostic evaluation is the comprehensive assessment that opens an episode of mental health care: it establishes the diagnosis, the medical necessity for treatment, and the initial plan. Medicare contractor policy describes the service as an "integrated biopsychosocial assessment" : a complete medical, psychiatric, family, and social history, a full mental status exam, a tentative diagnosis, and an evaluation of the patient's ability and willingness to participate in the proposed treatment. Clinicians and payers also call it an initial psychiatric evaluation, a diagnostic assessment, a psychiatric intake, or simply a psych eval; one Medicare contractor shortens it to PDE. The current structure dates to January 1, 2013, when the CPT restructure split the older diagnostic interview code 90801 into two services, 90791 without medical services and 90792 with them, and moved interactive complexity to the add-on code 90785. The codes standardize a service, not a form. No US authority publishes a mandatory evaluation template: content expectations come from your Medicare contractor, payer contracts, licensure, and setting, and they genuinely differ. Even the one federal deadline clinicians quote, 60 hours, belongs to a different setting entirely: 42 CFR 482.61(b) applies to Medicare-certified psychiatric hospitals, an inpatient condition of participation, not an office evaluation. Two boundaries keep the record clean. The evaluation is the billable service that a first-session intake note documents when a qualified clinician performs it. And it is not a HIPAA psychotherapy note: diagnoses, symptoms, treatment plans, and test results are ordinary record material that payers can request, while the protected category covers only separately kept process notes . ## Who uses psychiatric diagnostic evaluations and when Every discipline that opens mental health care performs this evaluation, and the split falls on the medical component. Psychologists, therapists, counselors, and clinical social workers furnish it without medical services (90791); psychiatrists, psychiatric nurse practitioners, and physician assistants can add prescribing, examination, and laboratory work (90792). Medicare has paid marriage and family therapists and mental health counselors directly for it since January 2024. Settings run from private practice and community mental health to hospital outpatient departments, integrated primary care, and telehealth. The evaluation belongs at the start of care, and again when something changes the diagnostic question: a new episode after a gap in treatment, an admission, or a referral that asks something new. Once treatment is underway it hands off to the psychotherapy progress note , and when a focused question surfaces it feeds deeper workups such as the trauma assessment or a full case formulation . ## Psychiatric diagnostic evaluation structure: what goes in each section No single form is mandated anywhere. The sections below are the elements Medicare contractor policies list and reviewers across the US, Canada, and Australia expect to find, each with the pitfall that most often undermines it. Identification and referral. Who the patient is, who referred them and why, and the chief complaint in the patient's own words. Pitfall: paraphrasing the complaint into clinician language; the patient's words anchor medical necessity. History of present illness. Onset, duration, course, severity, and what the symptoms prevent the patient from doing. Pitfall: symptoms listed without how long they have existed; contractor policy expects the length of existence of the problems. Past psychiatric history. Prior diagnoses, treatment episodes, hospitalizations, medication trials and response, and past suicide attempts or self-harm. Pitfall: skipping treatment response; the next clinician repeats what already failed. Medical history and current medications. Significant medical conditions, current medications with doses, and allergies, from every evaluator. Pitfall: therapists and psychologists leaving this section out because they do not prescribe; the expected-element list applies to the evaluation, not the credential. Substance use. Current and past use, amounts, route, last use, and prior treatment. Pitfall: a bare "denies substance use" with no substances named; an unspecified denial reads as an unasked question. Family and social history. Psychiatric conditions in the family; living situation, relationships, work or school, legal issues, and stressors. Pitfall: "noncontributory" one-liners; an absence with no content reads as not assessed. Mental status exam. Appearance, behavior, speech, mood, affect, thought process, thought content, perception, cognition, insight, and judgment. Pitfall: "MSE within normal limits" with no elements; a complete mental status exam is a listed element of the service, and its absence leaves the diagnosis unsupported. Risk assessment. Suicidal and homicidal ideation, self-harm, access to means, protective factors, and the response when risk is present. Pitfall: no explicit line when risk is negative; the record must show risk was assessed, not assumed absent. Strengths and liabilities. What the patient brings to treatment, such as supports, insight, and motivation, and what works against it. Pitfall: the element clinicians skip most; it appears on contractor expected-element lists and it is what makes the plan realistic. Diagnostic impression and treatment plan. The DSM-5-TR diagnosis with its ICD-10-CM code, rule-outs, and a plan with treatment methods, anticipated length where possible, measurable goals, and the patient's ability and willingness to participate. Pitfall: a diagnosis the documented findings do not support, or a plan with no measurable goals; either one breaks the medical-necessity chain every later session relies on. ### Blank template (copy and adapt) Patient: [initials] DOB/Age: Date: Clinician: Evaluation type: [90791 / 90792] Referral source & reason: Chief complaint (patient's words): History of present illness (onset / duration / course / severity / impact): Past psychiatric history (dx, treatment & response, hospitalizations, past SI or self-harm): Medical history & current medications (+ allergies): Substance use (current / past / last use / prior treatment): Family history (psychiatric & relevant medical): Social history (living situation / work or school / relationships / legal / stressors): Mental status exam: Appearance & behavior: Speech: Mood & affect: Thought process: Thought content: Perception: Cognition: Insight / judgment: Risk assessment (SI / HI / self-harm / access to means / protective factors): Strengths & liabilities: Diagnostic impression (DSM-5-TR / ICD-10-CM + rule-outs): Treatment plan (methods / frequency / anticipated length / measurable goals / referrals or labs): Patient's ability & willingness to participate: Clinician signature / credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychiatric diagnostic evaluation Scenario: adult client self-referred after a primary care visit, with three months of low mood and worry following a job loss; a licensed psychologist completes the evaluation (90791). All details are fictional. Patient: D.M., 33 · Date: 07/10/2026 · Type: Psychiatric diagnostic evaluation (90791) · Clinician: A. Reyes, PhD · Referral: self, after primary care visit Chief complaint: "I lost my job in April and I have not felt like myself since." History of present illness: About three months of depressed mood, loss of interest, initial insomnia, low energy, and concentration problems after a layoff in April 2026, with worry most days about finances and the job search. Symptoms are present most days and worse in the evenings; client has stopped going to the gym and withdrawn from friends. Passive thoughts of "what's the point" without intent. PHQ-9 today: 16 (moderately severe). GAD-7: 11 (moderate). Past psychiatric history: One prior depressive episode at age 21, treated with about ten sessions of CBT through college counseling, remitted. No psychiatric medications, no hospitalizations, no prior suicide attempts or self-harm. Medical history and current medications: Hypothyroidism, on levothyroxine 75 mcg daily; last TSH reported normal (02/2026). No other medications. No known drug allergies. Substance use: Alcohol two to three drinks per week without binge episodes. Denies tobacco, cannabis, and other drug use. No prior substance use treatment. Family history: Mother treated for depression. No known family history of suicide, bipolar disorder, or psychosis. Social history: Lives with partner; laid off from a logistics coordinator role in April 2026; bachelor's degree; no legal involvement. Primary stressor is financial strain. Supportive partner and one close friend. Mental status exam: Casually dressed with adequate grooming; cooperative. Speech normal in rate and volume. Mood "down"; affect constricted and congruent. Thought process linear and goal-directed. No delusions. Passive death wish without plan or intent; no hallucinations. Alert and oriented x4; attention and memory grossly intact. Insight good; judgment intact. Risk assessment: Passive ideation without plan, intent, or preparatory behavior; no access-to-means concerns raised. Denies homicidal ideation. Protective factors: supportive partner, treatment-seeking, future orientation with interviews scheduled. Risk judged low; safety discussion completed and documented; will reassess at each session. Strengths and liabilities: Psychologically minded, prior good response to CBT, stable housing and relationship. Working against treatment: financial stress, reduced activity, thin current social contact. Diagnostic impression: Major depressive disorder, recurrent, moderate (F33.1); rule out generalized anxiety disorder, with GAD-7 monitoring. The documented duration, prior episode, and functional impact support the diagnosis. Treatment plan: Weekly individual CBT with behavioral activation, 12 to 16 sessions anticipated. Goals: (1) PHQ-9 below 10 by session 8; (2) three scheduled activity blocks per week by session 4; (3) one social contact re-engaged per week by session 6. Re-administer PHQ-9 and GAD-7 every four weeks. Coordinate with the primary care physician on thyroid monitoring and discuss medication options if the PHQ-9 has not improved by session 8. Next appointment 07/17/2026. Participation: Patient understands the diagnosis and plan and is willing and able to participate in weekly therapy. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every element on the Medicare contractor's expected list is present, including medical history and current medications from a non-prescribing evaluator. The mental status exam is documented element by element rather than "within normal limits," so a reviewer can see what was actually examined. Risk gets an explicit finding with protective factors even though it is negative, and passive ideation is distinguished from plan and intent. The diagnosis follows the documented duration, prior episode, and functional impact, and each goal in the plan is measurable and time-anchored. The closing participation line ties the evaluation to the treatment plan, which is what the medical-necessity chain needs at every later session. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The evaluation sits in the designated record set: the patient can request it, payers can audit it, and every later note builds on it. Despite the specialty, it is not a protected HIPAA psychotherapy note; keep private process reflections in separately stored psychotherapy notes and write the evaluation knowing other clinicians, the patient, and reviewers will read it. On timing, there is no national 24-, 48-, or 72-hour completion rule for an outpatient evaluation: Medicare expects documentation at the time of service and accepts a late addition only when it is clearly marked as a delayed entry with its author and date (Program Integrity Manual, chapter 3). Retention is state and provincial law, not HIPAA: HHS states plainly that the Privacy Rule sets no medical-record retention period , while records supporting Medicare billing carry a seven-year duty of their own (42 CFR 424.516(f)). On the payer side, the evaluation opens the medical-necessity chain that every later claim leans on, and the sharpest rules are the ones template pages rarely state. Frequency is contractor policy, not a national rule, and the contractors genuinely differ: First Coast covers the evaluation once at the onset of an illness and again after roughly a six-month treatment hiatus, an inpatient admission, or a marked change; Novitas policy recognizes initial and periodic evaluations, notes that children and geriatric patients may need more than one visit to complete one, and caps coverage at three per year per beneficiary by the same provider; Noridian allows one per day, permits more than one per year with a separate evaluation, and adds two hard lines: never billed incident-to, and never on the same day as an E/M visit. Nationally, the 2026 NCCI manual bundles the evaluation with psychotherapy on the same date because "psychotherapy includes continuing psychiatric evaluation." The 90792 version needs documented medical work, such as examination elements, prescriptions, or labs, and the interactive complexity add-on 90785 needs a documented communication factor from the APA and AACAP guidance , not just a young patient or family in the room. Two current openings are worth knowing: the evaluation can serve as the initiating visit for principal illness navigation services, and it remains billable by telehealth , including from the patient's home, under flexibilities that run through December 31, 2027. ↑ Back to the template and downloads ## Common psychiatric diagnostic evaluation errors auditors flag The best public number is category-level, and it is not small: for the 2024 reporting period, Medicare's improper-payment rate for outpatient psychiatry was 16.1% , a projected $254.5 million, with insufficient documentation driving 78.3% of the improper payments and missing documentation another 17%. No published denial rate exists for 90791 or 90792 specifically, and pages that quote one are inventing it. The BastionGPT Clinical Advisory Board sees the same errors most often in psychiatric diagnostic evaluation reviews: A skeletal mental status exam. "MSE within normal limits" with no elements, or a template string that contradicts the narrative above it. The MSE is on every contractor's expected-element list, and without it the diagnosis floats unsupported. A checklist with no synthesis. Every history box filled, then no formulation, no diagnostic reasoning, and no conclusion. A cloned intake form does not show that a diagnostic evaluation actually happened, and that is the core denial theory reviewers apply. The medical version without medical work. 90792 billed on a record that documents only a psychosocial interview. The medical assessment, prescription work, or lab review has to be in the note, and holding a prescriber's license does not substitute for documenting it. A repeat evaluation with no new question. A second evaluation billed without a documented new episode, admission, marked change, or new referral question. Contractors read that as routine care relabeled as an intake. Risk and strengths left silent. No explicit risk statement even when findings are negative, and no strengths and liabilities line at all. Both appear on contractor element lists, and their absence reads as not assessed rather than not present. ### Psychiatric diagnostic evaluations in the US, Canada, and Australia Aspect United States Canada Australia Status No law mandates a form; CPT defines the service (90791/90792), Medicare contractor policies list expected elements and frequency, and national coding policy bundles same-day psychotherapy and E/M No national billing code or form exists; provincial fee schedules pay psychiatrist assessments and provincial colleges prescribe record content, with Quebec regulating psychologists' records by law MBS item 291 is a statutory pathway: a referred psychiatrist assessment over 45 minutes with a mental state examination, an outcome tool when appropriate, and a defined report-back; item 296 covers ordinary initial consultations Terminology Psychiatric diagnostic evaluation, initial psychiatric evaluation, diagnostic assessment, 90791/90792, PDE Initial psychiatric assessment, psychiatric consultation, comprehensive psychiatric assessment, diagnostic assessment Initial psychiatric consultation, comprehensive diagnostic assessment, assessment and management plan (item 291) What changes Which contractor's rules apply: First Coast covers one evaluation per illness with defined repeat triggers, Novitas caps three per year, Noridian bars incident-to billing Which province and college govern: record content, supervision co-signature rules, and retention all vary by regulator The report-back obligation: item 291 requires a written report to the referrer within two weeks containing the diagnostic assessment and a 12-month biopsychosocial management plan, and cannot repeat within 12 months Retention HIPAA sets no chart period; state law governs, and records supporting Medicare claims carry seven years (42 CFR 424.516(f)) Quebec sets five years by regulation, Ontario colleges use ten years or age 18 plus ten, and BC's harmonized standard runs 16 years Psychology Board code: seven years from last entry for adults, and past the 25th birthday for clients seen as minors; doctors follow state law The clinical skeleton, history, mental status exam, diagnosis, and plan, travels across all three countries. What changes is the surrounding machinery: a US contractor decides how often the service is payable, a Canadian college decides what the record must contain and for how long, and the Australian schedule turns one version of the assessment into a statutory report with a deadline. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric diagnostic evaluations. Draft a complete psychiatric diagnostic evaluation from your interview notes, dictation, or session transcript, with every expected element in place for your review. Pull histories, mental status findings, and risk statements out of a long first interview into the structured format, so nothing said in the room is lost to the note. Check a finished evaluation for the gaps auditors flag: a missing mental status exam, a diagnosis the documented findings do not support, or a plan without measurable goals. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What is the difference between 90791 and 90792? 90791 is the psychiatric diagnostic evaluation without medical services; 90792 includes them: clinically indicated examination elements, prescription work, and ordering or reviewing labs. Psychologists, therapists, counselors, and clinical social workers bill 90791. Physicians, psychiatric NPs, and PAs choose between 90792 and an E/M visit for the same encounter, and the medical work has to appear in the note either way. Medicare coding policy does not allow a separate E/M charge alongside either code on the same day. How often can a psychiatric diagnostic evaluation be repeated? There is no national once-a-year rule; frequency is Medicare contractor policy, and the contractors differ. First Coast covers one evaluation at the onset of an illness, with repeats after roughly a six-month treatment hiatus, an inpatient admission, or a significant change. Novitas policy recognizes initial and periodic evaluations for a new episode, a readmission, or a new referral question, and caps coverage at three per year per beneficiary by the same provider. Noridian allows one per day and more than one per year when each is a separate, medically necessary evaluation. If one evaluation takes two visits, which Novitas notes can happen with children and geriatric patients, it is still one service, not two. How long does a psychiatric diagnostic evaluation take, and is it time-based? 90791 and 90792 are not time-based codes: no national rule sets a minimum duration, a required 60 minutes, or a 16-to-90-minute window, despite what many template pages claim. Most clinicians block 45 to 90 minutes as a practical matter, and writing the note takes 30 to 60 minutes by hand (our clinical team's estimate; no published time study exists). The only federal completion deadline belongs to another setting: Medicare-certified psychiatric hospitals must complete the inpatient psychiatric evaluation within 60 hours of admission under 42 CFR 482.61(b) . That rule does not apply to an office evaluation. Can I bill 90791 and psychotherapy on the same day? Not to Medicare. The 2026 NCCI policy manual states that "psychotherapy includes continuing psychiatric evaluation," so 90791 and 90792 are not separately reportable with individual, group, family, crisis, or other psychotherapy on the same date. The same manual bars a separate E/M code with either evaluation code, and contractor policy excludes same-day crisis psychotherapy as well. If therapy genuinely begins the same day, the evaluation code carries the encounter; psychotherapy billing starts at the next visit. Who can bill 90791, and at what rate? Under Medicare: physicians, clinical psychologists, clinical social workers, clinical nurse specialists, NPs, PAs, and, since January 1, 2024, independently enrolled marriage and family therapists and mental health counselors. Payment differs by credential for the same service: physicians and clinical psychologists receive 100% of the fee schedule amount, while CSWs, MFTs, and MHCs receive 75% of the clinical psychologist rate . Noridian adds a rule many practices miss: the evaluation is never billed incident-to. 90792 requires a clinician qualified to furnish the medical work. Do I still need a five-axis diagnosis or a GAF score? No. DSM-5 retired the multiaxial system and the GAF score in 2013; the APA's own insurance guidance recommends separate severity and disability measures, with WHODAS 2.0 as the disability instrument. Some Medicare contractor policies still carry pre-2013 wording, such as "multi-axis diagnosis or diagnostic impression list": the second half of that phrase is the operative one. A single-axis DSM-5-TR diagnosis with its ICD-10-CM code, plus Z-codes for psychosocial factors, satisfies the intent. If a specific payer form still demands a GAF, that is the payer's requirement, not the DSM's. How is this different from an intake note or a biopsychosocial assessment? The psychiatric diagnostic evaluation is a billable service with payer-defined expectations. The intake note is the practice document that records a first session, and it supports the evaluation code when a qualified clinician actually performs the diagnostic work. A biopsychosocial assessment is a content framework covering biological, psychological, and social domains; it can supply most of an evaluation's content but is not itself a separate billable service. When standardized testing is needed, that is a different service family entirely, documented in a psychological evaluation report . Can a psychiatric diagnostic evaluation be done by telehealth or audio-only? Yes. 90791 and 90792 are on the Medicare telehealth list, and current flexibilities, extended through December 31, 2027, allow the patient to be at home. Video is the default; audio-only is billable with modifier 93 when the patient cannot or will not use video, and home-based services use place-of-service 10. Document the modality, the patient's location, and consent the same way you would for a telehealth therapy session . State licensure and payer-specific telehealth rules still apply. Can BastionGPT draft a psychiatric diagnostic evaluation? Yes. Give it your interview notes, a dictation, or a session transcript, and it drafts the full evaluation for your review: histories, mental status exam, risk statement, diagnostic impression, and a plan with measurable goals. It can also check a finished draft for the gaps auditors flag, like a missing MSE element or an unsupported diagnosis. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance and payer claims on this page trace to these authorities, last verified July 2026: CMS Medicare Coverage Database, First Coast LCD L33252 : the integrated biopsychosocial assessment definition, expected elements, and repeat-evaluation triggers. CMS Medicare Coverage Database, Novitas LCD L35101 and billing article A57130 : initial and periodic evaluations, multi-visit completion, and the three-per-year utilization parameter. CMS, NCCI Policy Manual 2026, Chapter 11 : same-day bundling of the diagnostic evaluation with psychotherapy and E/M services. CMS Medicare Learning Network, outpatient psychiatric care compliance tips : the 2024 improper-payment rate and denial-reason shares. Noridian Medicare, Mental Health billing guidance : one unit per day, repeat evaluations, and the incident-to and same-day E/M exclusions. CMS, Transmittal R2657CP : the 2013 implementation replacing 90801 with 90791 and 90792. CMS, MFT and MHC payment page : independent billing since January 1, 2024 at 75% of the clinical psychologist rate. 42 CFR 482.61 : the psychiatric hospital condition of participation, including the 60-hour evaluation deadline. HHS, HIPAA FAQ 580 : the Privacy Rule sets no medical-record retention period. American Psychiatric Association, Insurance Implications of DSM-5 : retirement of the multiaxial system and the GAF. APA and AACAP, Interactive Complexity guidance : the four communication factors that support 90785. Australian Government, Health Insurance (General Medical Services Table) Regulations, MBS item 291 : the statutory psychiatrist assessment pathway, two-week report, and 12-month management plan. Quebec, Regulation respecting record keeping by psychologists, C-26, r. 221 : record content and five-year retention. ## Related templates Intake Note Trauma Assessment Case Formulation Note Treatment Plan Psychotherapy Progress Note Biopsychosocial Assessment Suicide Risk Assessment Neuropsychological Evaluation Report Prior Authorization Support Letter Psychological Evaluation Report BASC-3 Report Write-Up WISC-V Report Write-Up WAIS-5 / WAIS-IV Report Write-Up Conners 4 Report Write-Up WIAT-4 Report Write-Up BRIEF-2 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up MMPI-3 Report Write-Up SRS-2 Report Write-Up CAARS-2 Report Write-Up Mental Status Exam (MSE) Note PAI Report Write-Up MCMI-IV Report Write-Up Substance Use Assessment Cultural Formulation Interview (CFI) Intake Hospital Consult & Liaison Note Consultation Response & Report Medication Management Follow-Up Note Forensic Psychological Evaluation Report Diagnosis & Problem List Update (ICD) CAPS-5 Documentation GAF Score Documentation WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/trauma-assessment All note templates # Trauma Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A trauma assessment is a structured clinical evaluation of a client's trauma history, current trauma- and stressor-related symptoms, dissociation, and immediate safety. Psychologists, therapists, and counselors complete one at intake, or when trauma surfaces during treatment, usually pairing a clinical interview with a validated screen such as the PCL-5 or PC-PTSD-5. A typical trauma assessment runs 300 to 900 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, therapists, counselors, clinical social workers Audience Treating clinician, downstream treatment team, supervisors, payers Typical length 300 to 900 words · 20 to 40 minutes by hand (clinical team estimate) Format family Structured intake assessment (compare: intake note, biopsychosocial assessment) When it's used At intake, or when trauma history emerges during treatment Standards context A clinical convention, not a mandated form: no US, Canadian, or Australian law requires a specific trauma assessment; content follows HIPAA and payer rules On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a trauma assessment? A trauma assessment is a structured clinical evaluation of a client's history of traumatic events and their current trauma- and stressor-related symptoms, dissociation, and immediate safety. It goes deeper than the trauma question on a general intake, and clinicians also call it a trauma history interview, a trauma screening, or simply the trauma part of the formulation . In DSM-5-TR it maps to the trauma- and stressor-related disorders (PTSD, acute stress disorder, and the adjustment disorders); ICD-11 adds a separate complex PTSD diagnosis for prolonged or repeated interpersonal trauma, which the DSM does not name. Two distinctions keep the assessment honest. First, exposure is not a diagnosis. DSM-5-TR Criterion A defines a traumatic event narrowly, as exposure to actual or threatened death, serious injury, or sexual violence, and many people with real, painful histories do not meet the full criteria for PTSD. The assessment documents what happened and what the symptoms are, then lets the diagnosis follow. Second, no law requires a particular trauma form or a specific questionnaire. SAMHSA's trauma-informed care model is a framework, not a mandate, and screening tools like the PCL-5 or the ACE questionnaire are professional convention rather than a statutory requirement. What payers and boards expect is that the record be accurate, complete, and tied to the care you provide. ## Who uses trauma assessments and when Trauma assessments are used across outpatient mental health, community behavioral health, integrated primary care, trauma-specialty and veterans' clinics, substance use programs (where trauma and substance use often co-occur), and child and adolescent services. A clinician reaches for a dedicated trauma assessment when trauma is the presenting concern, when a screen flags it, or when trauma history emerges mid-treatment and the brief trauma section inside a general case formulation or intake is no longer enough. It goes deeper than that section: a structured event screen, symptoms mapped to the trauma diagnoses, dissociation, and current safety, so the resulting treatment plan is built on the full picture. ## Trauma assessment structure: what goes in each section There is no single mandated trauma assessment form. The sections below are the ones auditors, supervisors, and the next clinician expect to see, each with the pitfall that most often undermines it. Referral and presenting concern. Why the assessment is happening now, the referral question, and the client's own words about what brings them in. Pitfall: moving into event details before establishing why the assessment is happening and what the client wants from it. Trauma history. A structured screen of potentially traumatic events across the lifespan (a tool such as the LEC-5 helps), then the index event or events in the client's words, with only the detail that is clinically necessary. Pitfall: pressing for a vivid narrative of every event; over-detailed disclosure at a first meeting can retraumatize and is rarely needed to reach a diagnosis or a plan. Current symptoms. Trauma- and stressor-related symptoms mapped to the four PTSD clusters (intrusion; avoidance; negative alterations in cognition and mood; arousal and reactivity), ideally anchored to a validated measure such as the PCL-5. Pitfall: recording a total score with no symptom detail, or symptom detail with no measure; the record wants both. Dissociation. Screen for depersonalization, derealization, and gaps in memory, since they change treatment pacing and can flag the dissociative subtype of PTSD. Pitfall: skipping dissociation altogether; it is one of the most under-documented sections and it directly shapes the plan. Safety and current risk. Current safety, suicidal and self-harm risk, and any ongoing threat such as active abuse or intimate-partner violence. Pitfall: documenting the trauma as historical while missing that the client is still in an unsafe situation now. Strengths, supports, and coping. Protective factors, social support, cultural and spiritual resources, and the coping strategies the client already uses. Pitfall: a deficit-only assessment; without strengths the record cannot support a realistic, trauma-informed plan. Functional impact. How symptoms affect work, relationships, parenting, sleep, and daily functioning. Pitfall: listing symptoms without tying them to impairment, which is what supports medical necessity. Clinical formulation and diagnostic impression. Your integrated understanding: whether criteria for PTSD, acute stress disorder, an adjustment disorder, or another condition are met, and the mechanisms that link history to presentation. Pitfall: treating “has a trauma history” as “has PTSD”; many trauma-exposed clients do not meet the criteria. Recommendations and plan. Trauma-informed next steps: modality, pacing, referrals, safety planning where indicated, and follow-up. Pitfall: recommending trauma-focused processing before the client has the stabilization and safety to tolerate it. ### Blank template (copy and adapt) Client: [initials] Date: Clinician: Referral / presenting concern: Trauma history (event screen, index event, age / onset): Current symptoms: Intrusion: Avoidance: Negative cognition / mood: Arousal / reactivity: Measure + score (e.g., PCL-5): Dissociation (depersonalization / derealization / memory gaps): Safety & current risk (safety now, SI / self-harm, ongoing threat): Strengths, supports & coping: Functional impact (work / relationships / sleep / daily life): Clinical formulation & diagnostic impression: Recommendations & plan (modality, pacing, referrals, follow-up): Clinician signature / credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample trauma assessment Scenario: adult client, single-incident trauma (a highway collision) four months ago, referred by primary care. All details are fictional. Client: R.O., 41 · Date: 07/09/2026 · Type: Trauma assessment (intake) · Referral: primary care Referral and presenting concern: Referred by primary care after four months of nightmares and feeling “on edge” since a highway collision. Wants to sleep through the night and drive on highways again. Trauma history: Life Events Checklist completed. Index event: a multi-vehicle highway collision (04/2026); client was the driver, treated in the emergency department and released the same day. No other events endorsed at a Criterion A level. No reported childhood abuse; a brief ACE screen was positive for one item (parental divorce). Current symptoms: Intrusion: near-nightly nightmares, intrusive images when merging into traffic. Avoidance: has not driven on highways since the collision; avoids news coverage of crashes. Negative cognition and mood: persistent belief that “the roads are not safe,” loss of interest in weekend cycling. Arousal: hypervigilance in traffic, exaggerated startle, initial insomnia. PCL-5 today: 42. Dissociation: Reports briefly “spacing out” at the moment of impact; denies current depersonalization, derealization, or memory gaps. Does not meet the threshold for the dissociative subtype. Safety and current risk: Denies suicidal ideation, self-harm, and homicidal ideation. No ongoing threat or unsafe situation. Drives local roads cautiously without impairment. Routine monitoring; no safety plan indicated at this time. Strengths, supports, and coping: Stable marriage and employment, supportive spouse, uses paced breathing, motivated and psychologically minded. Functional impact: Sleep reduced to about five hours nightly; declined a promotion that required a highway commute; increased irritability at home. Work performance otherwise maintained. Clinical formulation and diagnostic impression: Symptoms span all four clusters, have persisted more than one month, and cause functional impairment; the presentation is consistent with post-traumatic stress disorder following a single-incident adult trauma. No dissociative subtype. Short duration and strong protective factors suggest a good prognosis. Provisional: post-traumatic stress disorder. Recommendations and plan: Begin a trauma-focused therapy (for example prolonged exposure or cognitive processing therapy) after two stabilization sessions; add sleep and nightmare-focused strategies; re-administer the PCL-5 at session 4 and monthly; coordinate with the primary care physician. Next appointment 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The trauma history is screened with a structured tool and anchored to a dated index event, so the record shows method, not just a story. Symptoms are mapped to all four PTSD clusters and anchored to a PCL-5 score, which makes change measurable from one session to the next. Dissociation and safety each get an explicit line, even when the findings are negative, so the record shows they were assessed rather than skipped. The diagnostic impression follows from documented criteria and duration, not from trauma exposure alone. The plan sequences stabilization before trauma processing and sets a dated re-assessment, which supports a trauma-informed and defensible course of care. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A trauma assessment is part of the designated record set: the client can request it, payers can audit it, and the next clinician relies on it. Write it so all three can read it, and keep the level of trauma detail proportionate to what the diagnosis and plan actually require, rather than a verbatim narrative of every event. Auditors and payers expect the assessment to be thorough, dated, and signed, with a documented Criterion A event where PTSD is diagnosed, a mental status exam, an explicit risk assessment, and diagnoses supported by the findings. When a pre-licensed clinician conducts it, most boards expect a licensed supervisor to co-sign. Retention is set by state and payer rules, not by HIPAA, which does not dictate how long patient charts are kept; a common floor is at least seven years for adults and until the age of majority plus several years for minors, and Medicaid work often carries a longer window because the federal False Claims Act allows audits up to ten years back. On the payer side, the trauma assessment usually lives inside the diagnostic evaluation billed as 90791, or 90792 when a medical component is furnished, so the content, not the format, carries medical necessity: Medicare contractor guidance expects the record to support the diagnosis and the service billed. The trauma-informed framing is guidance, not a mandated form, and no regulation requires a specific questionnaire such as the ACE or the LEC-5; those are convention. Two content points do carry real weight. If the assessment is done inside a substance use treatment program, its contents are protected by 42 CFR Part 2 , a stricter federal standard than HIPAA that needs specific consent to disclose. And because trauma and acute risk often travel together, current safety and suicide or self-harm risk belong on every assessment, even when the finding is negative. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ## Common trauma assessment errors auditors flag The failures that get trauma assessments denied are documentation failures, not format choices. Payers deny the diagnostic evaluations these assessments support for insufficient documentation of medical necessity: an omitted mental status exam, a missing or unsigned risk assessment, or a diagnosis the record does not support. There is no claim-denial code specific to a “missing trauma assessment”; the denials read generically, as diagnosis not supported by the medical record, or documentation does not meet payer policy. The backdrop is a high-prevalence one, which is part of why the record matters: CDC data show 63.9% of US adults report at least one adverse childhood experience and 17.3% report four or more , and about 5% of US adults have PTSD in a given year. The BastionGPT Clinical Advisory Board sees the same errors most often in trauma assessment reviews: Exposure recorded as a diagnosis. The note documents a trauma history and jumps to PTSD without walking DSM-5-TR Criterion A and the symptom duration; not every trauma-exposed client meets the criteria. A score with no symptoms, or symptoms with no score. A PCL-5 total with no cluster detail, or a rich narrative with nothing measurable to track. Reviewers and the next clinician want both. Dissociation and current safety left blank. No line either way, so the record shows an absence rather than an assessment. Suicide and self-harm risk missing from a trauma assessment is a top audit finding. An over-detailed trauma narrative. More graphic event detail than the diagnosis or plan requires, which raises privacy exposure without adding clinical value. Trainee assessments without a supervisor co-signature. An intern or associate conducted the assessment and a licensed supervisor never signed, which many boards and payers treat as an incomplete record. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on trauma assessments. Draft a structured trauma assessment from your interview notes, a screening tool's results, or a dictation, with the trauma history in the client's frame. Redact or de-identify a trauma history before it goes to a referral or a report, so only what the recipient needs leaves your record. Check a finished assessment for the gaps supervisors and auditors flag: a symptom measure with no detail, a missing dissociation or safety section, or a diagnosis that outruns the findings. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How detailed does a trauma history need to be? As detailed as it takes to justify the diagnosis and the plan: which events meet DSM-5-TR Criterion A, roughly when they happened, and how the symptoms connect. Pair a screening score such as the PCL-5 or PC-PTSD-5 with a short narrative in the client's words. More graphic detail than the diagnosis and plan require adds privacy risk without adding clinical value. See the case formulation for how the history feeds your working hypothesis. Is any specific trauma assessment form required by law or by insurers? No. No US, Canadian, or Australian law mandates a particular trauma form, and no payer requires a specific questionnaire like the ACE or the LEC-5. What Medicare and other payers require is that the diagnostic evaluation be medically necessary and that the record support the diagnosis and the service billed. SAMHSA's trauma-informed care is a framework, not a mandated form. What is the difference between a trauma assessment and an intake or biopsychosocial? An intake note is a brief first-session record; a biopsychosocial, often billed as the 90791 diagnostic evaluation, is the broad, multi-domain assessment. A trauma assessment goes deep on one domain: the trauma history, symptoms, dissociation, and safety, and it usually lives inside that larger evaluation. Its output feeds the treatment plan . Does a positive trauma screen mean the client has PTSD? No. Exposure is not a diagnosis. DSM-5-TR Criterion A is specific, and many people with real trauma histories meet criteria for an adjustment disorder, for ICD-11 complex PTSD, or for no trauma disorder at all. The assessment documents the history and the symptoms; the diagnosis follows from the criteria and the duration. Which trauma screening tools are used most? The most common come from the VA's National Center for PTSD: the PCL-5 (a 20-item self-report), the PC-PTSD-5 (a 5-item primary-care screen), the LEC-5 (a life-events exposure checklist), and the clinician-administered CAPS-5. The ACE questionnaire covers childhood adversity. A PCL-5 score around 31 to 33 is widely used as a provisional PTSD cutoff, though no single cutpoint fits every setting or population. Do I have to screen for ACEs? Only if it fits your setting or population. ACE screening is not required by law anywhere. Some programs incentivize it (California's Medi-Cal ACEs Aware initiative pays trained providers) but do not mandate it, and beneficiaries cannot be forced to screen. If you use an ACE tool, document how you will act on a high score. How long do I keep a trauma assessment? There is no universal HIPAA rule for patient charts; retention comes from state and payer rules. A common floor is at least seven years for adults and until the age of majority plus several years for minors, with longer windows where Medicaid or litigation exposure applies. Check your jurisdiction and payer contracts. Can BastionGPT write a trauma assessment from my notes? Yes. Give it your interview notes, a screening tool's results, or a dictation, and it drafts a structured assessment for your review, keeping the trauma history in the client's frame. It can also check a draft for gaps such as a missing dissociation or safety section. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance and clinical claims on this page trace to these authorities, last verified July 2026: US Department of Veterans Affairs, National Center for PTSD, PTSD Checklist for DSM-5 (PCL-5) : the 20-item self-report and its provisional cutoff. National Center for PTSD, DSM-5 diagnostic criteria for PTSD : Criterion A and the symptom clusters. US Centers for Disease Control and Prevention, About Adverse Childhood Experiences : ACE prevalence and impact (the ACE study originated with Felitti and Anda, 1998). SAMHSA, Trauma and Violence : trauma-informed care as a framework, not a mandated form. 42 CFR Part 2 : confidentiality of substance use disorder patient records. CMS Medicare Coverage Database, Billing and Coding: Psychiatric Diagnostic Evaluation and Psychotherapy Services (A57520) : documentation expectations for the 90791/90792 evaluation. ## Related templates Case Formulation Note Treatment Plan Psychotherapy Progress Note SOAP Note DAP Note Crisis Psychotherapy Note Intake Note Psychiatric Diagnostic Evaluation Biopsychosocial Assessment Suicide Risk Assessment Safety Plan Neuropsychological Evaluation Report Psychological Evaluation Report Mandated Reporting & Duty-to-Protect Documentation Substance Use Assessment Self-Harm (NSSI) Assessment PCL-5 Score Documentation CAPS-5 Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/case-formulation All note templates # Case Formulation Note Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A case formulation note (also called a case conceptualization) is the working hypothesis of a case: how predisposing, precipitating, perpetuating, and protective factors fit together to produce the presenting problem, and what that means for treatment. It is written at intake and revised at major reviews rather than every session. Most run one to two pages, and no law or payer prescribes its format. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The assessing clinician: therapists, psychologists, counselors, prescribers; trainees write them for supervision and case reports Audience Mostly you, the care team, and supervisors; auditable and client-accessible like any record Typical length 1 to 2 pages · 20 to 40 minutes by hand (clinical team estimate) Format family Assessment and conceptualization document (also called case conceptualization, formulation, 4Ps or 5Ps) When it's used At intake after assessment, then revised at major reviews or when the case changes Standards context Professional convention everywhere; no US or Canadian law or payer mandates it, and Australia's GP plan form embeds it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a case formulation note? A case formulation note is the document where assessment turns into an explanation. Where an intake gathers history and a treatment plan commits to goals, the formulation sits between them and answers why this person, why this problem, why now : a hypothesis about how longstanding vulnerabilities, recent triggers, maintaining factors, and protective strengths fit together to produce what the client is experiencing. Clinicians also call it a case conceptualization, and payers rarely call it anything at all, which is the first thing worth knowing about it. No single body created or standardized the format. Australia's national Comorbidity Guidelines , funded by the Australian Government, say it plainly: "there is no standardised approach to case formulation," though a range of dimensions should be considered. The organizing frame most clinicians learn is the four Ps: predisposing, precipitating, perpetuating, and protective factors, often expanded to five Ps with presenting issues leading. That frame comes from training and registration standards, not statute: the Psychology Board of Australia's internship case-report guidelines assess whether a provisional psychologist "identifies the predisposing vulnerabilities, precipitating (triggers), perpetuating (maintaining) and protective factors" and integrates them into a formulation that accounts for the presenting problem. ## Who writes case formulation notes and when Anyone whose treatment follows from a theory of the case: therapists, psychologists, counselors, and prescribers write one at or shortly after intake, then revise it at major reviews, when treatment stalls, or when new information changes the picture. It is not a per-session document; the session record is a SOAP , DAP , or BIRP note that should trace back to the plan the formulation justifies. Trainees write formulations constantly: registration case reports are graded on them. And in one interesting corner of the world the formulation is embedded in a payer instrument: the RACGP template GPs use for Australia's GP Mental Health Treatment Plan (MBS items 2700 and 2701) contains a dedicated "Case formulation" section prompting the four Ps. ## How to write a case formulation note in five steps Anchor it in the presenting problem: the client's own words, baseline measures, and functional impact. Map the four Ps: predisposing vulnerabilities, precipitating triggers, perpetuating factors, protective strengths; note any you cannot yet fill. Write the working hypothesis: one paragraph tying the factors together in your theoretical frame. Draw the treatment implications: what the hypothesis says to target first, and with what approach. Date it, sign it, revisit it: record authorship like any entry and update the formulation when the case changes. Each element in more detail, with the pitfall to avoid: 1: Anchor it in the presenting problem. State the problem as the client describes it, with scores and real-life impact. Pitfall: a symptom list with no story; the formulation's job is to explain, not restate. 2: Map the four Ps. Longstanding vulnerabilities (developmental, biological, relational), the triggers behind this episode, what maintains the problem now, and the strengths that treatment can recruit. Pitfall: treating the Ps as a legal requirement; they come from training standards, and no regulation prescribes them. 3: Write the working hypothesis. One paragraph that makes a claim: what drives the problem and what keeps it going, framed in your model (CBT, psychodynamic, systemic, ACT). Pitfall: repeating the history; a hypothesis is testable and can be wrong. 4: Draw the treatment implications. If the hypothesis is right, what do you target first? These lines become the treatment plan's goals, which is the golden thread auditors trace. Pitfall: a formulation that never changes the plan. 5: Date it, sign it, revisit it. Date, time, and author on the entry, like any record. Pitfall: assuming the formulation is shielded as "psychotherapy notes"; one with diagnosis or plan content is part of the regular, auditable record. ### Blank template (copy and adapt) Client: [initials] Date: Diagnosis (ICD-10), if established: Presenting problem & context (client's words, baselines): Predisposing factors: Precipitating factors: Perpetuating factors: Protective factors: Working hypothesis (how the factors fit together): Treatment implications (what to target first): Formulation discussed with client: [ ] Review / update date: Clinician signature / credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with the four-Ps checklist; the DOCX is the blank formulation, ready to adapt. ## Sample case formulation note Scenario: adult client, social anxiety disorder, CBT frame, written after a two-session assessment. All details are fictional. Client: S.M., 29 · Date: 07/08/2026 · Diagnosis: Social anxiety disorder (F40.10) Presenting problem & context: "I freeze in meetings and I'm starting to dodge them." Marked fear of negative evaluation in work settings since a promotion five months ago; declining meeting attendance, camera-off defaults, and one avoided presentation. SPIN at intake: 41. Sleep loss the night before scheduled meetings. Predisposing: Longstanding shyness with peer teasing in adolescence; family norm of "don't draw attention"; no prior treatment. Precipitating: Promotion to a client-facing role requiring weekly presentations; a poorly received early presentation the client describes as "the disaster." Perpetuating: Avoidance and safety behaviors (camera off, scripting, delegating speaking parts) prevent disconfirmation; post-event rumination; partner increasingly handles social logistics. Protective: Strong partner support; valued long-term career goals; good response to structure; no substance use; physically active. Working hypothesis: Early evaluative experiences established a core belief of being "exposed as inadequate," activated by the promotion and the salient failure experience. Avoidance and safety behaviors reduce anxiety short-term but block corrective learning and maintain the fear cycle; rumination consolidates each episode as further evidence. Treatment that reduces safety behaviors while building graded exposure to evaluative situations should interrupt the maintenance loop. Treatment implications: CBT with graded exposure and behavioral experiments targeting safety-behavior reduction first (camera on, unscripted contributions), then presentation-specific exposures; cognitive work on post-event rumination; recruit partner support for approach rather than accommodation. Informs Goals 1 and 2 of the treatment plan dated 07/08/2026. Discussed with client: yes, collaboratively in session 2. Review: at session 8. Signature: treating clinician (credentials, dated). This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works It explains rather than restates: the hypothesis makes a testable claim about what maintains the problem, not a second copy of the history. All four Ps are addressed, each with case-specific content, and the client's own words anchor the presenting problem. The maintenance loop is specific (safety behaviors block disconfirmation), so the treatment implications follow logically instead of generically. Treatment implications name what to target first and point at numbered treatment plan goals, the golden thread auditors trace. Collaboration is documented: the formulation was discussed with the client, which the evidence base favors and some payers expect for plans. It is dated, signed, and scheduled for review, meeting the record-entry expectations that do apply to every document in the chart. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The first thing to know is what does not exist: no US federal statute, CMS rule, or Canadian law defines a case formulation note or prescribes its contents. Even the four-Ps frame is convention; Australia's government-funded Comorbidity Guidelines state there is "no standardised approach to case formulation." The structure clinicians treat as mandatory comes from training and registration standards, such as the Psychology Board of Australia's internship case-report guidelines , which grade provisional psychologists on identifying predisposing, precipitating, perpetuating, and protective factors. Templates that present the five Ps as a compliance requirement are conflating good pedagogy with law. Texas illustrates the actual legal posture: the LPC rules at 22 TAC §681.36 enumerate what a counseling record must contain (consent, intake assessment, dates, treatment methods, progress notes, treatment plan, billing), and a formulation is not on the list. The trap worth a paragraph of anyone's attention is HIPAA's "psychotherapy notes" category. Under 45 CFR §164.501 , that protected category expressly excludes "any summary of diagnosis, functional status, the treatment plan, symptoms, prognosis, and progress to date," which is precisely what a case formulation contains. A formulation is therefore part of the regular medical record: auditable, releasable, and accessible to the client. Medicare's Program Integrity Manual §3.3.2.6 tells review contractors they may never request true psychotherapy notes, and that mixing protected process notes into the same document does not extend protection to the rest; the provider is responsible for extracting what supports the claim. That instruction has been CMS policy since Change Request 3457 in 2005. Practical upshot: keep the formulation in the chart proper, and keep genuinely private process notes physically separate. Audit exposure comes from the general record rules, not formulation content. In the 2023 OIG psychotherapy audit , 128 of 216 sampled enrollee days failed Medicare requirements (psychotherapy time not documented, among others) and 54 more failed guidance such as missing signatures, driving an estimated $580 million of $1 billion in improper payments. A formulation never saves a note that lacks time, date, or signature; Ahpra's record guidance expects every entry to identify the date, time, and provider of the service. Retention follows the chart it lives in: state law in the US (Texas: seven years from termination, or five years past the age of majority), CMPA guidance of at least 10 years (16 in British Columbia) in Canada, and 7 years or until age 25 under the Psychology Board of Australia's code . ↑ Back to the template and downloads ## Common case formulation errors clinicians make No auditor scores a formulation as such, so the errors here are the ones that surface in supervision, case-report assessment, and the general audit rules that do apply. The BastionGPT Clinical Advisory Board sees these most often: Restating instead of explaining. A paragraph of history with "formulation" written above it. The hypothesis should make a claim about mechanism that could in principle be wrong. Treating the Ps as a legal checklist. The four Ps are a teaching frame from registration standards, not a mandate; forcing thin content into every box reads worse than noting "insufficient information yet" where true. A formulation that never touches the plan. If the treatment plan's goals do not follow from the hypothesis, the golden thread breaks and both documents look boilerplate. Mislabeling it "psychotherapy notes." Under 45 CFR §164.501 a document containing diagnosis or treatment-plan summary content is part of the regular record no matter what it is titled; assuming it is shielded leads to bad release and audit decisions. Never revisiting it. A formulation dated at intake and untouched through a stalled course of treatment is a missed instrument; revision when the case changes is the point. Missing the entry basics. Date, time, and author, the things record rules do require, are the omissions that actually get flagged. ### Case formulation notes in the US, Canada, and Australia Aspect United States Canada Australia Status CONVENTION; no statute or payer policy defines it, and record-content rules like Texas 22 TAC §681.36 do not list it CONVENTION; college record-keeping standards govern the chart, not the formulation CONVENTION, with one payer-form twist: the GP Mental Health Treatment Plan form (MBS 2700/2701) embeds a Case formulation section Terminology Case formulation, case conceptualization Case formulation, case conceptualization Case formulation; the four Ps are prompted verbatim on the MHTP form Where the structure comes from Training programs and supervision; payer rules are silent Training programs and supervision; colleges require accurate records generally Registration standards: internship case reports are assessed on the four Ps HIPAA/privacy posture Not "psychotherapy notes" under 45 CFR §164.501; part of the regular, releasable record Part of the clinical record under provincial health-information law Part of the health record; Ahpra expects date, time, and provider on every entry Retention State law governs (e.g., Texas: 7 years, or 5 past majority) CMPA recommends at least 10 years, 16 in British Columbia 7 years since last entry; until age 25 for clients under 18 The formulation's content is ungoverned everywhere; what travels is the chart rules around it: authorship, timing, access, and retention. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on case formulations. Draft a four-Ps or five-Ps formulation with a working hypothesis from your intake note, assessment bullets, or biopsychosocial history. Stress-test an existing formulation: does the hypothesis explain rather than restate, and do the treatment implications follow? Keep the thread intact: generate treatment plan goals from the formulation, and update the formulation when reviews change the picture. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a case formulation note required by law or insurance? No. No US federal rule, state record law we know of, Canadian statute, or payer policy mandates a standalone case formulation or prescribes its format; the structure comes from training standards. The closest thing to an exception is Australia, where the GP Mental Health Treatment Plan form used for MBS items 2700 and 2701 includes a Case formulation section prompting the four Ps, and that is a form design, not a statute. Are case formulation notes protected as psychotherapy notes under HIPAA? No, and this surprises many clinicians. HIPAA's protected "psychotherapy notes" category at 45 CFR §164.501 expressly excludes any summary of diagnosis, functional status, treatment plan, symptoms, prognosis, or progress, which is exactly what a formulation is. It is part of the regular record: releasable, auditable, and accessible to the client. Keep genuinely private process notes physically separate from the chart if you keep them at all. What are the 4 Ps (or 5 Ps) of case formulation? Predisposing factors (longstanding vulnerabilities), precipitating factors (what triggered this episode), perpetuating factors (what maintains it now), and protective factors (strengths treatment can recruit). The five-Ps version leads with Presenting issues. The frame comes from clinical training, and registration case-report standards in Australia assess it explicitly; no regulation anywhere requires it. What is the difference between a case formulation and a treatment plan? The formulation explains the case; the treatment plan commits to what happens next. The formulation's hypothesis says why the problem exists and persists; the plan turns that into measurable goals, objectives, interventions, and dose. A strong chart shows the thread: formulation, then plan goals that follow from it, then session notes that reference those goals. Where does the case formulation live in the chart? Wherever your record system puts assessment content, most commonly inside the intake or psychiatric diagnostic evaluation report, or as its own entry that the plan references. There is no billing code for a formulation; it supports the services you bill rather than being one. Like any entry it needs a date, time, and author. Can clients read their case formulation? Yes. Because it is part of the regular record rather than protected psychotherapy notes, it falls under normal access rights. Many clinicians treat that as a feature: collaborative formulation, where the hypothesis is built and shared with the client, is well supported clinically, and writing the document knowing the client may read it usually improves it. How is a case formulation different from a biopsychosocial assessment? The biopsychosocial assessment gathers and organizes information across biological, psychological, and social domains; the formulation is the step after, a hypothesis about how those facts fit together to produce this problem now. Many charts fold the formulation into the tail of the assessment report, which is fine; the failure mode is an assessment with no explanatory paragraph at all. How long does a case formulation take to write? There is no published time-to-complete research for this document type. Our clinical team's estimate is 20 to 40 minutes by hand after a completed assessment, and unlike a progress note it is written once and revised at reviews rather than repeated every session. Can BastionGPT draft a case formulation? Yes. Give it your intake note, biopsychosocial bullets, or assessment summary and your theoretical frame, and it drafts a four- or five-Ps formulation with a working hypothesis and treatment implications for your review and signature. It can also generate treatment plan goals from the formulation so the golden thread stays intact. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: Comorbidity Guidelines (Australian Government funded), An overview of case formulation : no standardised approach exists; the 5Ps framework and formulation-as-hypothesis. Psychology Board of Australia (Ahpra), Guidelines for the 4+2 internship program : case reports assessed on identifying predisposing, precipitating, perpetuating, and protective factors. RACGP, GP Mental Health Treatment Plan template (adult) : the Case formulation section prompting the four Ps on the MBS 2700/2701 form. MBS Online, MBS item 2700 : the GP Mental Health Treatment Plan item the template serves. eCFR, 45 CFR §164.501 : the psychotherapy-notes definition and its exclusions for diagnosis, treatment plan, symptoms, prognosis, and progress summaries. CMS, Medicare Program Integrity Manual, Chapter 3, §3.3.2.6 : contractors may never request psychotherapy notes; integrating excluded information does not protect it; providers extract what supports the claim. CMS, Transmittal R98PI, Change Request 3457 (2005) : the origin of the psychotherapy-notes review policy. HHS Office of Inspector General, 2023 psychotherapy audit : 128 of 216 sampled enrollee days failed requirements, 54 more failed guidance, an estimated $580 million of $1 billion improper. Texas Behavioral Health Executive Council, 22 TAC §681.36, Client Records : the enumerated record elements and the seven-year retention rule. Canadian Medical Protective Association, Managing medical records: retention and transfer : the 10-year (16 in British Columbia) retention recommendation. Ahpra, Managing health records : each entry identifies the date, time, and provider of the service. Psychology Board of Australia, Code of Conduct (effective 1 Dec 2025): record retention of 7 years, or until age 25 for clients who were minors. ## Related templates Treatment Plan SOAP Note Psychotherapy Progress Note DAP Note BIRP Note Psychotherapy Notes Authorization Trauma Assessment Care Coordination Note Intake Note Psychiatric Diagnostic Evaluation Biopsychosocial Assessment Neuropsychological Evaluation Report Psychological Evaluation Report Treatment Plan Review & Update Relapse Prevention Plan Outcome Measure Review Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/treatment-plan All note templates # Mental Health Treatment Plan Template: Free PDF & Completed Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A mental health treatment plan is the roadmap document for a course of therapy: diagnoses, presenting problems, measurable goals and objectives, the interventions that will address them, and how often care happens. Unlike a progress note, it is future-focused and spans many sessions. Most run one to three pages and are reviewed on a set schedule. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The treating clinician, developed with the client; prescribers and supervisors sign where rules require Audience Client, care team, supervisors, payers, auditors Typical length 1 to 3 pages · 30 to 60 minutes by hand (clinical team estimate) Format family Longitudinal care document (also called individualized treatment plan, care plan, service plan, recovery plan) When it's used At intake or early in care, then reviewed and updated on a schedule Standards context Convention in routine outpatient therapy; a hard requirement in specific programs, several US states, and Australia's Better Access On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a mental health treatment plan? A treatment plan is the forward-looking document that organizes a course of care: the diagnoses and presenting problems, the goals the client is working toward, measurable objectives under each goal, the interventions that will get there, and how often care happens. Where a progress note records what happened in one session, the plan spans the whole episode and gives every session something to point back at. You will also see it called an individualized treatment plan (ITP), care plan, service plan, or recovery plan; Alberta Health Services, for example, frames it as an evolving agreement between client and clinician rather than a static form. No single body invented the format; it grew up as a standard of care. Its legal status is more layered than most template pages admit. For routine office psychotherapy in the US, no blanket federal rule mandates a separate written plan, but the exceptions are exactly where audits happen: Medicare's Benefit Policy Manual requires an individualized written plan for outpatient hospital psychiatric services stating the type, amount, frequency, and duration of services plus diagnoses and anticipated goals; psychiatric diagnostic evaluations are expected to produce an initial plan; Medicare's outpatient-psych documentation expectations include the treatment plan, prognosis, and progress; several states write plan content, signatures, and deadlines directly into law; and in Australia a GP or psychiatrist plan is the gateway to Better Access rebates. ## Who writes treatment plans and when Every clinician whose care is goal-directed keeps one: therapists, psychologists, counselors, and psychiatric prescribers, usually written at or shortly after intake and developed with the client. Payers read it as the anchor of medical necessity: progress notes are audited against the plan (the "golden thread" from assessed need to goal to intervention to outcome), which is why our SOAP , DAP , and BIRP pages all tell you to tie each session to a numbered plan goal. In some settings the plan is not optional at all: US partial-hospitalization programs, state-licensed clinics in states like Maryland and Minnesota, and any Australian client claiming Better Access rebates. ## How to write a mental health treatment plan in five steps Ground it in the assessment: diagnoses, presenting problems in the client's own language, and the strengths you will build on. Set measurable goals: two to four goals the client helped choose, each with a number or observable state and a target date. Break goals into objectives: small, observable, dated steps that sessions can actually move. Specify interventions and dose: modality, session length, frequency, planned duration, and who provides it. Sign, share, schedule the review: document client involvement, collect the signatures your rules require, and put the review date on the calendar. Each element in more detail, with the pitfall auditors flag: 1: Ground it in the assessment. Diagnosis, presenting needs, and strengths, stated so a reviewer can see why treatment is medically necessary. Pitfall: a plan that could belong to any client; individualization is the point and the requirement. 2: Set measurable goals. "Reduce PHQ-9 from 16 to 9 or below by session 12" is auditable; "feel better" is not. Pitfall: goals with no measure or date; state rules like Maryland's expect measurable goals with target dates. 3: Break goals into objectives. Each goal gets one to three observable steps with their own dates. Pitfall: objectives that restate the goal instead of stepping toward it. 4: Specify interventions and dose. Medicare's formula is the safest checklist: type, amount, frequency, and duration of services. Pitfall: missing frequency or duration; the provider-level OIG audit flagged exactly that. 5: Sign, share, schedule the review. Client involvement documented, signatures collected (client or guardian where law requires, co-signatures for supervisees, prescriber where medications are involved), review date set. Pitfall: unsigned plans; one OIG audit found no physician signature on plans for 96 of 100 sampled days. ### Blank template (copy and adapt) Client: [initials] Date of plan: Diagnosis (ICD-10): Presenting problems / needs: Strengths & client priorities: Goal 1 (measurable, with target date): Objectives (observable, dated): Interventions / modality, frequency, duration: Goal 2 (measurable, with target date): Objectives (observable, dated): Interventions / modality, frequency, duration: Review date: Client involvement documented: [ ] Clinician signature / credentials: Date: Client (or guardian) signature, where required: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with the goal-and-signature checklist; the DOCX is the blank plan, ready to adapt. ## Sample mental health treatment plan Scenario: adult client, major depressive disorder, individual CBT with behavioral activation, 16-session episode. All details are fictional. Client: A.R., 32 · Date of plan: 07/06/2026 · Diagnosis: Major depressive disorder, single episode, moderate (F32.1) Presenting problems: Depressed mood most days, loss of interest in valued activities, social withdrawal, reduced work performance over the past four months. PHQ-9 at intake: 16. Strengths: motivated for treatment, supportive partner, prior positive response to structured therapy. Goal 1: Reduce depressive symptoms. PHQ-9 from 16 to 9 or below by session 12 (target 09/28/2026). Objectives: (a) complete a weekly activity schedule with at least five planned activities per week by week 4; (b) complete at least three thought records per week by week 6. Interventions: individual CBT with behavioral activation, 50-minute sessions, weekly, 16 sessions planned, delivered by the treating clinician. Goal 2: Restore occupational functioning. Return to a full-time work schedule by week 12 (target 09/28/2026). Objectives: (a) agree a graded return-to-work plan with employer by week 8; (b) maintain a consistent morning routine at least five days per week by week 6. Interventions: activity scheduling and problem-solving within weekly CBT sessions; coordination letter to primary care if requested. Review: at session 8 or within 90 days, whichever comes first; PHQ-9 re-administered at each review. Client involvement: goals set collaboratively in session 2; copy of this plan offered to the client. Signatures: treating clinician (credentials, dated); client signature per practice policy. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every goal carries a measure and a target date, so progress is auditable instead of aspirational. The plan states Medicare's exact formula: the type, amount, frequency, and duration of services, plus diagnoses and anticipated goals. Objectives are observable steps with their own dates, giving each progress note something concrete to reference, the golden thread payers trace. Client involvement is documented and a copy offered, which several jurisdictions expect and some require in writing. A review date is scheduled up front; unreviewed, outdated plans are a recurring audit finding. The signature block anticipates the rules that bite: clinician credentials and date, plus client or guardian signature where required. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Know when the plan stops being optional. Routine outpatient psychotherapy in the US has no blanket federal plan mandate, but Medicare's Benefit Policy Manual requires an individualized written plan for outpatient hospital psychiatric services, psychiatric diagnostic evaluations are expected to produce an initial plan, and Medicare's outpatient-psych documentation expectations include the treatment plan, prognosis, and progress. State law can be far more specific: Maryland's regulations prescribe plan contents with measurable short- and long-term goals and require signatures from the client or guardian, two licensed clinicians, and the prescriber when medication is involved, with review at least every six months; Minnesota's statute sets hard deadlines for completing the plan and requires client co-signature, including on changes. Audits treat the plan as the root of medical necessity. In the On-Site Psychological Services audit , noncompliant treatment plans appeared on 111 of 120 sampled claims, and in the New York provider audit plans were absent or missing frequency and duration, and 96 of 100 sampled days had no physician signature on the plan. The practical rule: the plan must match the services you actually bill, and every progress note should trace back to one of its numbered goals. In Australia the plan is the gateway, not just the roadmap: under Better Access , a GP's Mental Health Treatment Plan or a psychiatrist's assessment and management plan must exist before allied mental health rebates can be claimed, and MBS item 2715 requires the plan to be documented in writing with a copy offered to the patient, with a review not undertaken within four weeks of the plan claim, nor more than once in a three-month period, and most patients needing no more than two reviews a year. On retention, the plan is part of the chart: US state law governs, Canadian guidance from the CMPA recommends at least 10 years (16 in British Columbia), and Australia's Psychology Board code sets 7 years, or until age 25 for clients who were minors. ↑ Back to the template and downloads ## Common treatment plan errors auditors flag Treatment plans are the single most-named failure in psychotherapy audits. In the On-Site Psychological Services audit , noncompliant plans appeared on 111 of 120 sampled claims; in the New York provider audit , plans were absent or missing required elements such as frequency and duration, and 96 of 100 sampled days lacked the treating physician's signature on the plan. OIG's recommendations in both cases amounted to the same instruction: ensure every chart has a complete, current, signed plan. The BastionGPT Clinical Advisory Board sees the same patterns most often in plan reviews: Goals without measures or dates. "Improve mood" gives a reviewer nothing to verify; a score with a target date does. Type, amount, frequency, or duration missing. Medicare's four-part formula is the checklist; frequency and duration are the two that go missing. The plan does not match the billed services. Sessions billing a modality or cadence the plan never ordered break the golden thread. Signature gaps. Missing client or guardian signatures where law requires them, no co-signature for supervisee-authored plans, or no prescriber signature where medication is part of treatment. No review on schedule. Plans that were never updated as goals changed, in settings where six-month or program-specific review rules apply. ### Treatment plans in the US, Canada, and Australia Aspect United States Canada Australia Status Convention for routine outpatient therapy; LAW in outpatient hospital psychiatric programs and in states like Maryland and Minnesota Convention; provincial health authorities and colleges shape expectations (e.g. Alberta Health Services planning guidance) Payer gateway: a GP or psychiatrist plan is required before Better Access rebates can be claimed Terminology Treatment plan, individualized treatment plan (ITP), plan of care Care plan, service plan, recovery plan Mental Health Treatment Plan (MHTP); psychiatrist assessment and management plan What changes State content, signature, deadline, and review rules (Maryland: client + two clinicians + prescriber, 6-month review; Minnesota: completion deadlines, client co-signs changes) Client-collaboration emphasis; CMPA retention guidance shapes practice MBS item 2715: plan in writing, copy offered to the patient, no review within four weeks of a plan claim or more than once in three months, and most patients need no more than two reviews a year; plans do not expire Retention Part of the chart; state law governs CMPA recommends at least 10 years from last entry, 16 years in British Columbia, longer for minors 7 years since last entry; until age 25 for clients under 18 (Psychology Board Code) The plan's clinical anatomy travels everywhere: problems, measurable goals, objectives, interventions, dose, review. What changes is who must sign it, when it must exist, and how long it must be kept. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on mental health treatment plans. Draft a complete, individualized plan from your intake note or assessment, with measurable goals and dated objectives. Check an existing plan against the type-amount-frequency-duration formula and flag missing signatures or review dates. Keep the golden thread intact: align progress notes to numbered plan goals, and update the plan when goals change. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What is the difference between a treatment plan and a progress note? The treatment plan is the roadmap for the whole course of care: goals, objectives, interventions, and dose. A progress note records one session. The "P" in a SOAP note is the immediate next step after a session, not the treatment plan itself; the two documents reference each other but are not interchangeable. Does Medicare require a treatment plan for outpatient therapy? Not as a blanket rule for routine office psychotherapy, a nuance most template pages miss. But Medicare requires an individualized written plan for outpatient hospital psychiatric services, expects an initial plan from a psychiatric diagnostic evaluation, and its outpatient-psych documentation expectations include the treatment plan, prognosis, and progress. In practice, auditors treat a current, signed plan as the anchor of medical necessity everywhere. What must a treatment plan include? The safest core is Medicare's formula: the type, amount, frequency, and duration of services, plus diagnoses and anticipated goals. State rules can add more: Maryland, for example, requires presenting needs, how each service addresses them, and measurable short- and long-term goals with target dates. Add objectives under each goal and a scheduled review date. Does the client have to sign the treatment plan? Sometimes by law. Maryland requires the client's signature (or a parent or guardian's for minors) alongside two licensed clinicians and the prescriber when medications are involved; Minnesota requires client co-signature, including on plan changes; Australia's MBS rules expect the patient's agreement to be recorded and a copy offered. If a client declines to sign, document the involvement and the refusal. How often should a treatment plan be reviewed or updated? There is no universal federal interval, and the "every 30 days" claim that circulates online has no general basis. Where rules exist, follow them: Maryland requires review at least every six months, and in Australia a review should not fall within four weeks of a plan claim or occur more than once in a three-month period, with most patients needing no more than two reviews a year. Otherwise, review on a set schedule and whenever goals materially change. What is a Mental Health Treatment Plan in Australia? The Better Access gateway document. A GP prepares an MHTP (or a psychiatrist an assessment and management plan) before Medicare will rebate psychologist and allied mental health sessions. MBS item 2715 requires it to be documented in writing with a copy offered to the patient; it does not expire, and a review should not fall within four weeks of a plan claim or occur more than once in a three-month period, with most patients needing no more than two reviews a year. Do treatment plan goals have to be SMART? No law says "SMART," a distinction worth knowing. What regulations and payers do expect is that goals be measurable and dated: Maryland's rules require measurable goals with target dates, and Medicare expects anticipated goals in the plan. SMART is the clinical convention that reliably satisfies those expectations. How long do I need to keep treatment plans? As long as the rest of the chart, because the plan is part of it. US retention is set by state law. In Canada, the CMPA recommends at least 10 years from the last entry (16 years in British Columbia), longer for minors. Australia's Psychology Board code sets 7 years since the last entry, or until age 25 for clients who were under 18. Can BastionGPT draft a treatment plan? Yes. Give it your intake note, assessment, or bullets, and it drafts an individualized plan with measurable goals, dated objectives, and the type-amount-frequency-duration formula filled in, for your review and signature. It can also check an existing plan for missing elements and keep progress notes aligned to plan goals. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, Medicare Benefit Policy Manual, Chapter 6 : the individualized written plan of treatment for outpatient hospital psychiatric services, stating type, amount, frequency, duration, diagnoses, and anticipated goals. CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : treatment plan, prognosis, and progress among documentation expectations; improper-payment data. CMS Medicare Coverage Database, Billing and Coding Article A57520 : the initial plan of treatment expected with psychiatric diagnostic evaluations. Maryland, COMAR 10.21.20.07 : individualized treatment plan contents, measurable goals with target dates, and client, clinician, and physician signature requirements. Minnesota, Statutes § 245G.06 : treatment plan completion deadlines and client co-signature, including on plan changes. HHS Office of Inspector General, On-Site Psychological Services audit (2020) : noncompliant treatment plans on 111 of 120 sampled claims. HHS Office of Inspector General, New York provider psychotherapy audit (2022) : plans missing required elements and physician signatures on 96 of 100 sampled days. Alberta Health Services, Integrated treatment planning : the plan as an evolving client-clinician agreement with measurable goal-setting. Canadian Medical Protective Association, Managing medical records: retention and transfer : the 10-year (16 in British Columbia) retention recommendation. Services Australia, Allied health referrals for mental health treatment services : a Mental Health Treatment Plan or psychiatrist plan as the precondition for Better Access rebates. MBS Online, MBS item 2715 : the GP Mental Health Treatment Plan documented in writing with a copy offered to the patient. Psychology Board of Australia, Code of Conduct (effective 1 Dec 2025): record retention of 7 years, or until age 25 for clients who were minors. ## Related templates Psychotherapy Progress Note SOAP Note DAP Note BIRP Note Case Formulation Note Psychotherapy Notes Authorization Trauma Assessment Care Coordination Note PIE Note Crisis Psychotherapy Note Group Therapy Note Family & Couples Therapy Note Intake Note Psychiatric Diagnostic Evaluation Biopsychosocial Assessment Suicide Risk Assessment Safety Plan Neuropsychological Evaluation Report Discharge Summary Prior Authorization Support Letter Psychological Evaluation Report Claim-Support Billing Note Treatment Plan Review & Update Individual Therapy Session Note BASC-3 Report Write-Up Chronic Care Management (CCM) Note Diagnosis & Problem List Update (ICD) Prior Authorization Request Medical Necessity Letter Y-BOCS Score Documentation Insomnia Severity Index Documentation CAPS-5 Documentation CANS Assessment Documentation GAF Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/treatment-plan-review All note templates # Treatment Plan Review & Update: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A treatment plan review is the periodic note that measures progress against each goal in the active mental health treatment plan, records barriers and new clinical information, and ends in a decision: continue the plan unchanged or update it. Clinicians complete one on their program's cadence and whenever the clinical picture changes. Most run 150 to 500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The treating clinician; treatment teams in program settings; supervisors and physicians co-sign where rules require Audience Client, care team, supervisors, payers, auditors Typical length 150 to 500 words · 10 to 20 minutes by hand (clinical team estimate) Format family Longitudinal care document (also called treatment plan update, periodic review, ITP review) When it's used On the program's review cadence during care, and off-cycle whenever goals or the clinical picture change Standards context No universal review interval exists; programs, states, and payers set hard cadences from 30 days to annual On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a treatment plan review? A treatment plan review is the periodic note that tests whether the active treatment plan still fits. It compares current symptoms, functioning, and risk against each goal in the plan, records the response to each intervention, and ends in a decision: continue the plan unchanged or update it. The review is the decision point; the update is the revision it produces when something changes, and a compliant review can conclude that no change is needed. You will also see it called a treatment plan update, ITP review, plan-of-care review, periodic review, or master treatment plan update. The format descends from Lawrence Weed's problem-oriented medical record, published in the New England Journal of Medicine in 1968, and the earliest nationwide US rule requiring periodic review of an individualized psychiatric plan dates to 1978: the Medicaid rule for inpatient psychiatric care of people under 21, still codified at 42 CFR 441.155 . The fact most template pages miss: there is no universal "every 90 days" rule. The three-month Medicare requirement widely repeated online is a local coverage determination from one contractor jurisdiction, not national policy. Real cadences run from 30 calendar days in Medicare-certified community mental health centers, to 60-day intensive outpatient recertifications, to 90 days in Montana Medicaid, 180 days in Illinois Medicaid, annual formal reviews in New York clinics, and event-driven review with no fixed interval at all in opioid treatment programs. A review is also not a physician recertification: Medicare's partial hospitalization and intensive outpatient recertifications under 42 CFR 424.24 are payment attestations with their own deadlines and required signers, and they run alongside, not instead of, the treatment team's clinical review. The rule that governs your review is the one your program, state, and payer name, so identify it before you inherit someone else's calendar. ## Who writes treatment plan reviews and when Anyone who keeps a treatment plan eventually reviews one: therapists and psychologists in outpatient practice on whatever cadence their payer contracts or state rules set, treatment teams in community mental health centers and residential programs on fixed program clocks, and physicians in partial hospitalization and intensive outpatient programs, where recertification runs alongside the team's clinical review. Canadian regulators expect the function even without a form: Ontario's psychology standards require the record to show reviews of progress and the continued relevance of the service plan, and British Columbia expects the plan to be adapted, with the client's consent, when outcomes show it is not working. In Australia, review of a GP's Mental Health Treatment Plan belongs to the referring practitioner, while the allied mental health provider writes a course-end report back to the referrer; the two documents are routinely conflated. Off the clock, a review is due whenever the clinical picture moves: a goal is achieved, progress stalls, risk changes, or a new service starts. If progress notes are the running commentary on the plan and the discharge summary closes it out, the review is the checkpoint document in between. ## Treatment plan review structure: what goes in each section Header and plan linkage. Review date, the date or version of the plan under review, the period covered, and the program or setting. Pitfall: a review that never identifies which plan it reviews; the version trail from plan to review to next review is exactly what auditors follow. Participation. Who took part: the client, family or guardian where relevant, and team members, plus how the client's input was gathered. Pitfall: no documented client involvement in settings that require it; Medicare's community mental health center rule expects the record to show the client's understanding, involvement, and agreement. Progress toward each goal. Goal-by-goal status using the measure the plan itself set: scale scores, attendance, functional markers, observable behavior. Pitfall: one global "client is doing well" line; reviewers look for per-goal evidence against the plan's own numbers. Barriers and new clinical information. What interfered (symptoms, access, motivation, life events) and anything new since the last review: diagnosis changes, risk changes, medication changes. Pitfall: barriers listed with no response; every named barrier should meet a plan adjustment or a stated reason to stay the course. Continued medical necessity. Why treatment at this frequency and level remains necessary, or what steps down. Where maintenance is the goal, say what deterioration is being prevented; Medicare's outpatient psychiatric manual recognizes maintenance treatment where withdrawal would predictably lead to relapse. Pitfall: progress documented so thoroughly the note undercuts necessity, with no statement of remaining impairment. The decision: changes or a no-change rationale. Revised goals, objectives, target dates, or frequency; discontinued goals; movement toward discharge criteria. If nothing changes, record the clinical evidence that continuing unchanged is right. Pitfall: re-signing the same text each cycle; payers treat cloned reviews as no review at all. Signatures, dates, next review. Clinician with credentials, client where a program requires it, supervisor or physician where rules require, and the next scheduled review date. Pitfall: unsigned or undated reviews; signature gaps are among the most-cited plan findings in federal audits. ### Blank template (copy and adapt) Client: [initials] Review date: Review #: Plan under review (date/version): Period covered: Participants & client input: Progress by goal (measure + status): Goal 1: Status: [met / progressing / no change / regressed] Goal 2: Status: Barriers & new clinical information: Continued medical necessity: Plan changes (or rationale for no change): Next review date: Clinician signature/credentials: Date signed: Client signature (where required): Supervisor (where required): Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank review form, ready to adapt. ## Sample treatment plan review Scenario: adult client, recurrent major depressive disorder, individual CBT in private practice, first quarterly review of an April 2026 plan under the practice's payer-contract cadence. All details are fictional. Client: J.M., 41 · Review date: 07/14/2026 · Review #: 1 · Plan under review: dated 04/14/2026 · Period: 04/14/2026 to 07/14/2026 Participation: Review completed with the client in session on 07/14/2026. Client described progress and barriers in their own words; goal changes below were agreed together and a copy of the updated plan was offered. Diagnosis and attendance: Major depressive disorder, recurrent, moderate (F33.1), unchanged. 12 of 13 planned weekly CBT sessions attended; one cancellation, rescheduled within the week. Goal 1 (PHQ-9 from 17 to 9 or below by 10/14/2026): Progressing. PHQ-9 today 11, from 17 at plan start. Objective (a), behavioral activation schedule at least 5 days per week: met since week 3. Objective (b), thought records 3 times per week: partial, averaging 2 per week; reviewed obstacles and simplified the worksheet. Goal 2 (return to a four-day work week by 08/15/2026): No change. Barrier: employer restructuring has eliminated the additional hours; this is an external barrier, not clinical avoidance. Revised objective: client will request modified duties in writing by 08/30/2026; target date moved accordingly. New clinical information and risk: No new diagnoses this period. Sertraline 100 mg unchanged, prescribed and managed by primary care. Risk screened at each session; client denies suicidal ideation; no risk indicators observed this period. Continued medical necessity: Symptom reduction is partial (PHQ-9 11) and occupational impairment persists. Prior episode history indicates regression risk if weekly CBT is withdrawn now. Step-down to biweekly is planned when PHQ-9 remains below 10 for four consecutive weeks. Decision and next review: Goal 1 unchanged. Goal 2 objective and target date revised as above. Session frequency unchanged at weekly. Next review 10/13/2026, or sooner if the clinical picture changes. Signatures: treating clinician (LPC), dated 07/14/2026; client signature per payer contract, dated 07/14/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works It names the exact plan and version it reviews, so the record shows one continuous thread from plan to review. It states per-goal status using the plan's own measures, a PHQ-9 target and dated objectives, instead of a global impression. Every barrier meets a response: the stalled work goal gets a revised objective and target date, not silence. It says why continued care at this frequency is still medically necessary, and names the step-down criterion. Client participation and agreement are documented, with the signature lines the practice's rules ask for. The next review date is set, so the cadence is visible in the record itself. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A treatment plan review is part of the standard clinical record. HIPAA's psychotherapy-notes definition expressly excludes the treatment plan and progress from its protected category, so reviews sit in the designated record set: clients can request them, payers can audit them, and reviewers read them against the plan and the progress notes on either side. That cross-reading is where reviews fail. Arizona Medicaid's claims guidance states flatly that "Documentation cloning is strictly prohibited," and its claim edits deny services when no valid treatment plan covers the date of service, with plans valid for one year, so a plan that quietly ages out takes its claims with it. There is no national denial code that means "treatment plan review missing"; what a records request that goes unanswered draws instead is the generic combination of group code CO, reason code 50, and remark code M127 under CMS Transmittal 663 . Cadence is the compliance question that generates the most confusion, and the answer is: identify your rule, because there are at least seven. Medicare-certified community mental health centers must review, revise, and document the plan at least every 30 calendar days under 42 CFR 485.916 ; Medicaid inpatient psychiatric care for people under 21 reviews every 30 days; Medicare partial hospitalization recertifies on day 18 and then at least every 30 days, and intensive outpatient at least every 60, both physician functions under 42 CFR 424.24 ; First Coast's local coverage determination expects updates "generally at least every three months" in its jurisdiction, while other contractors expect a periodic update without naming a number; Montana Medicaid reviews at least every 90 days, Illinois Medicaid every 180, and New York clinics formally at least annually, with some changes allowed in progress notes. Medicaid home and community-based service plans are reviewed at least every 12 months, when circumstances change significantly, and at the person's request under 42 CFR 441.725 . Opioid treatment programs have no fixed interval at all: 42 CFR 8.12 ties updates to treatment response and life-context changes. Psychiatric hospitals set their own interval, and CMS survey guidance expects an off-cycle review when a goal is achieved, the patient regresses, progress is absent, or a new goal is needed. Participation and signatures deserve the same source-checking. Medicare's CMHC rule requires the plan to be established with the client and the record to show understanding, involvement, and agreement; New York requires participation documented by notation; Montana requires the member's signature and dating; Ontario's psychology standards require both a supervised author and the supervising registrant to sign documents that carry conclusions or recommendations; British Columbia requires client consent before the plan is modified. None of that adds up to a universal client-signature rule, so document participation the way your rule says, and where a client declines a required signature, record the request, the refusal, and any payer instruction. When a review supports a continued-care request to an insurer, it feeds the prior authorization letter ; when it concludes that goals are met, it starts the discharge summary . ↑ Back to the template and downloads ## Common treatment plan review errors auditors flag Federal auditors keep finding the same thing: plans that exist on paper but were never really reviewed. In the On-Site Psychological Services audit , 111 of 120 sampled claims had noncompliant treatment plans, part of an estimated $3.3 million in overpayments, and the same audit flagged records that did not support improvement or a reasonable expectation of it, the exact judgment a review exists to document. In the New York provider audit , every one of 100 sampled days failed at least one requirement and 96 lacked evidence of the required physician signature on the plan. Nationally, OIG estimated $580 million in improper psychotherapy payments , and in Medicare's 2024 reporting-period data for outpatient psychiatric services, insufficient documentation drove 78.3% of improper payments. A February 2025 OIG audit of a Florida mental health center shows the other side: the provider generally met Medicare billing requirements, with one miscoded claim line in a 100-line sample, so the bar is reachable. The BastionGPT Clinical Advisory Board sees the same errors most often in treatment plan review checks: Cloned reviews. The same text re-signed cycle after cycle. Copy-forward language that does not reflect the client's current condition reads to a payer as no review at all, and Arizona Medicaid's guidance prohibits documentation cloning in so many words. A stale or orphaned plan behind the review. The review points at no identifiable plan version, or the plan has aged past the program's limit; Arizona's MD466 edit denies claims with a missing or invalid treatment plan for the date of service, and its plans are valid for one year. Global progress statements. "Client is doing well" with no per-goal status against the plan's own measures, which leaves the improvement judgment auditors look for undocumented. Progress that undercuts necessity. Improvement documented with no statement of remaining impairment, regression risk, or maintenance rationale, which is how continued-care requests get denied. Signature and participation gaps. A missing member signature where the payer requires one, no physician signature in billing arrangements that demand it, no supervisor co-signature for supervised clinicians, or no documented client involvement at all. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on treatment plan reviews. Draft a complete review from your bullets: per-goal progress, barriers, the medical-necessity statement, and the change summary, in your organization's format. Pull goal-by-goal status from a stack of progress notes and outcome scores, ready for you to verify against the chart. Check a drafted review before you sign: it flags a missing necessity statement, an unidentified plan version, or an unchanged plan with no rationale. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How often does a treatment plan have to be reviewed? There is no universal interval, and "every 90 days" is not a national US rule. Federal cadences that do exist are program-specific: 30 calendar days in Medicare-certified community mental health centers, 30 days in Medicaid inpatient psychiatric care for people under 21, day 18 then every 30 days for partial hospitalization recertification, and at least every 60 days for intensive outpatient. The three-month rule most sites quote is a First Coast local coverage determination. States add their own: Montana at least every 90 days, Illinois every 180, New York clinics at least annually. In Australia a Mental Health Treatment Plan does not expire. Identify the rule for your program and payer, then put the next review date in the plan itself. What is the difference between a treatment plan review and a progress note? A progress note records one encounter; a review steps back over the whole period and tests every goal in the plan. The two can share space: New York's clinic regulation allows some plan changes to be recorded in progress notes, reserving a formal update for a new service or increased intensity, and CMS psychiatric-hospital survey guidance states that a combined treatment and progress note may be written. Where a rule requires a separately signed review or recertification, a progress note cannot substitute. Is there a billing code for a treatment plan review? There is no universal standalone code for reviewing a psychotherapy treatment plan; for most outpatient therapists the work rides inside the services you already bill. Some state Medicaid programs do pay for plan development or review separately under HCPCS code H0032 with state-specific rules and modifiers, so check your state's manual rather than a national code list. Australia is the clean contrast: the dedicated Medicare items for reviewing a GP's Mental Health Treatment Plan were repealed on November 1, 2025, and a review now runs through ordinary time-tiered attendance items. Never reach for a psychotherapy or evaluation code just because a review happened; the billed service has to satisfy its own definition. Does the client have to sign or participate in the review? Participation and signature are different requirements, and no rule makes either universal. Montana Medicaid requires the member to sign and date plans. Medicare's community mental health center rule requires the plan to be established with the client and the record to show understanding, involvement, and agreement, without demanding a signature. New York documents participation by notation. Ontario's psychotherapy regulator expects consent to be logged as therapy changes, and British Columbia's psychology standards require client consent before the plan is modified. If a required signature is refused, document the request, the refusal, and any payer instruction you receive. Can I re-sign the same plan if nothing changed? A review is allowed to conclude that the plan still fits: Montana's policy, for example, expects the review to either amend the plan or document its continued appropriateness. What it cannot be is the same paragraph re-signed each cycle. Show the evidence behind the no-change decision: current scores, per-goal status, and why frequency and level of care remain right. Copy-forward text that ignores the client's current condition is a named audit failure, and Arizona Medicaid's claims guidance prohibits documentation cloning outright. What triggers a review before the scheduled date? CMS survey guidance for psychiatric hospitals names four: a goal is achieved, the patient regresses, progress is absent, or a new goal is needed. New York adds a new service or increased intensity; British Columbia expects action when outcomes show an intervention is ineffective or harmful. In practice, add risk changes, diagnosis or medication changes, and payer reauthorization requests, which often arrive on the insurer's calendar rather than yours (see the prior authorization letter ). What do auditors flag in treatment plan reviews? Plans that were never meaningfully reviewed. In one OIG audit, 111 of 120 sampled claims had noncompliant treatment plans; in another, 96 of 100 sampled days lacked a required physician signature on the plan; nationally, OIG estimated $580 million in improper psychotherapy payments. The recurring review-specific failures: cloned text, a stale or unidentified plan version, global progress statements with no per-goal evidence, necessity undercut by progress with no remaining-impairment statement, and missing signatures or participation documentation. Are treatment plan reviews part of the record clients can access? Yes. HIPAA's psychotherapy-notes category expressly excludes the treatment plan and progress, so plans and their reviews sit in the standard record and the designated record set clients can request. Write each review knowing the client, a payer, and a future treating clinician may all read it. Private process reflections belong in segregated psychotherapy notes with their own authorization rules . Can BastionGPT draft a treatment plan review? Yes. Give it the active plan plus your progress notes or bullets, and it drafts the review: per-goal status against the plan's own measures, barriers with responses, the medical-necessity statement, and the change summary, ready for your review and signature. It can also check a drafted review for a missing necessity statement or an unidentified plan version. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 42 CFR § 485.916 : community mental health center plans established with the client and reviewed at least every 30 calendar days. 42 CFR § 441.155 : the 30-day plan review for Medicaid inpatient psychiatric care of people under 21, in force since 1978. 42 CFR § 424.24 : partial hospitalization recertification on day 18 and then at least every 30 days; intensive outpatient at least every 60 days. CMS, State Operations Manual, Appendix AA : psychiatric-hospital survey guidance on plan format, periodic review, and off-cycle triggers. 42 CFR § 8.12 : opioid treatment program care plans updated on treatment response and life-context changes, with no fixed cycle. First Coast Service Options, LCD L33252 : the local expectation that plans be updated about every three months. Montana DPHHS, Behavioral Health Policy 120 : 90-day review that amends the plan or documents continued appropriateness, with member signature. Illinois HFS, IM+CANS : treatment plan review and update at least every 180 days. New York OMH, 14 NYCRR Part 599 : annual formal review, update triggers, and participation documented by notation. HHS OIG, On-Site Psychological Services audit (2020) : noncompliant treatment plans on 111 of 120 sampled claims. HHS OIG, New York provider audit (2022) : plans missing elements and physician signatures on 96 of 100 sampled days. HHS OIG, national psychotherapy audit (2023) : an estimated $580 million in improper payments. CMS Medicare Learning Network, Outpatient psychiatric care compliance tips : 2024 reporting-period improper-payment data. Services Australia, Requirements of a Mental Health Treatment Plan : a plan does not expire, and review-claim spacing rules. Health Insurance Legislation Amendment (2025 Measures No. 3) Regulations 2025 : the November 1, 2025 repeal of the dedicated Mental Health Treatment Plan review items. College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct : reviews of progress in the record and supervisor co-signature rules. College of Health and Care Professionals of BC, Ethics and Practice Standards : consent before modifying a treatment plan and adapting it when outcomes fall short. ## Related templates Treatment Plan Psychotherapy Progress Note Case Formulation Note Discharge Summary Prior Authorization Support Letter Relapse Prevention Plan Outcome Measure Review Note Better Access Mental Health Treatment Plan Referral Multidisciplinary Team Meeting Note Case Conference Note Utilization Review Summary NDIS Plan Reassessment & Progress Report Workers' Compensation Progress Report Start-Stop Time Attestation DASS-21 Score Documentation Y-BOCS Score Documentation CANS Assessment Documentation MADRS Score Documentation WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mandated-report All note templates # Mandated Reporting & Duty-to-Protect Documentation: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 Mandated reporting and duty-to-protect documentation is the chart record a clinician creates when reporting suspected child or elder abuse, or acting on a client's threat of serious violence against an identifiable person. It captures the facts observed, the legal basis, who was notified and when, and the clinical reasoning. State law mandates the report itself; the note is the clinician's proof the duty was met. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, psychiatric prescribers, school and agency clinicians Audience The chart and treating team, supervisors, protective-services intake, law enforcement when warned, licensing boards, attorneys and courts Typical length 300 to 700 words · 15 to 30 minutes by hand (clinical team estimate) Format family Legal-protective documentation (compare: suicide risk assessment, safety plan, crisis note, incident report) When it's used When reasonable suspicion of child or elder abuse arises, when a client threatens an identifiable person, right after a report or warning is made, and when documenting a considered decision not to report Standards context State law mandates the report and its deadlines; no law prescribes the chart note, which is the clinician's evidence the duty was met On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is mandated reporting and duty-to-protect documentation? Mandated reporting and duty-to-protect documentation is the chart record a clinician creates when a legal duty to disclose overrides confidentiality: reporting suspected child abuse or neglect, reporting elder or dependent-adult abuse, or warning an identifiable person threatened by a client. The duties themselves have two distinct lineages. Child-abuse reporting laws spread state by state after the Child Abuse Prevention and Treatment Act of January 1974 conditioned federal grants on each state having a reporting law, with immunity for good-faith reporters ( 42 USC 5106a(b)(2)(B) ). The duty to protect comes from case law: California's Tarasoff decisions, a 1974 ruling framing a duty to warn and the 1976 rehearing that broadened it to a duty to protect, which states then adopted, modified, or rejected one by one. Clinicians call the resulting notes a CPS report note, an APS report note, a duty-to-warn or Tarasoff note, or protective disclosure documentation. Whatever the name, there are two artifacts, and they are not interchangeable: the report itself (the hotline call, the agency form, the warning delivered) belongs to the receiving agency, while the chart note documenting that report belongs to the clinical record. This page covers the chart note. The load-bearing distinction is what the law actually mandates. State law mandates the report: who must report, what triggers the duty, how fast, and in what form, with criminal penalties for failure. Almost nothing mandates the chart note that documents it, which is why the note is best understood as evidence: it is how a clinician later proves the duty was recognized, the threshold was met, and the report was made on time. Federal privacy law is the permission layer, not the barrier clinicians often assume: HIPAA permits disclosures required by law (45 CFR 164.512(a)), permits child-abuse reports to the authorities designated to receive them ( 45 CFR 164.512(b)(1)(ii) ), and permits disclosures to avert "a serious and imminent threat" when made to someone able to prevent it, including the target of the threat (45 CFR 164.512(j)(1)(i)). A mandated-report note is not a suicide risk assessment , which evaluates danger to self, and it is not a crisis note , which records an urgent encounter; it documents a specific legal duty and the actions that discharged it. ## Who uses mandated reporting and duty-to-protect documentation and when Counselors, therapists, and other mental health professionals are named mandated reporters for child abuse in 38 states, roughly 17 states require any person who suspects abuse to report, and every state permits anyone to report ( Child Welfare Information Gateway, statutes through May 2023 ), so as a practical matter this documentation belongs to therapists, psychologists, counselors, clinical social workers, and psychiatric prescribers in every setting: private practice, agencies, schools, hospitals, and telehealth. It gets written in four situations: when reasonable suspicion of abuse or neglect arises and a report is made; when a client communicates a serious threat against an identifiable person and the clinician warns or takes other protective action; when new information triggers the duty again in a case already reported; and, just as deliberately, when a clinician evaluates a possible report and concludes the threshold is not met. The neighboring documents divide the work: a suicide risk assessment handles danger to self, a safety plan equips the client, a crisis note records the urgent encounter that often surrounds these events, and a release of information covers disclosures the client authorizes. This note covers the disclosure the law compels. ## Mandated report and duty-to-protect note structure: what goes in each section No statute prescribes these headings. They mirror what a licensing board, a protective-services record request, or a courtroom will look for years later: what was learned, why it met or did not meet the legal threshold, what was done, and when. The through-line is proof: the note exists so the clinician can show the duty was recognized and discharged on time. Trigger and context. What was disclosed or observed, when, where, and by whom, in the source's words where possible: the client's statement, the child's disclosure, the observation that raised concern. Attribute every fact ("client stated," "this clinician observed"). Pitfall: recording the conclusion instead of the observation. "Child is being abused" is a finding the agency makes; "child described being struck and this clinician observed marks consistent with the description" is what belongs in a chart. Basis for the duty. Why the facts meet the standard, in the statute's own terms: reasonable suspicion for abuse reporting, or a serious threat against an identifiable person for protective action. State the standard, then connect the facts to it in one or two sentences. Pitfall: investigating instead of reporting. Reporters report suspicion; the agency investigates. Interview-style questioning of a child documented in the chart reads as contaminating the investigation, and delays the report you were required to make. Consultation. Who was consulted before or immediately after acting: supervisor, agency hotline staff, your board's practice consultant, your liability carrier, or an attorney, with names, times, and the substance of the advice. Pitfall: "consulted supervisor" with no name, time, or content. Consultation is the strongest evidence of reasonable professional judgment; an empty mention wastes it. Action taken. The operational core: which agency was called, at what time, by what method; the intake worker's name or ID; the report, reference, or confirmation number; the written follow-up form filed and its deadline; and for a duty-to-protect action, exactly what was communicated, to whom, and when, including law enforcement notification where the statute calls for it. Pitfall: no report number or agency identifier. Without it, the chart cannot prove the report happened, and "I called the hotline" becomes an assertion instead of a record. Client notification and response. Whether the client was told before or after the report, the clinical reasoning for the timing, and how the client responded. When the client is not the subject of the report, note any safety implications of disclosure for the victim. Pitfall: silence on notification. A chart that never says whether the client was told reads as concealment in a later dispute, whichever choice was actually made. Safety and follow-up. Immediate safety steps for the person at risk, changes to the treatment plan, the next scheduled contact, and a line acknowledging the ongoing duty: new information means a new report in most jurisdictions. Pitfall: treating the phone call as the end. The duty can recur with each new disclosure, and follow-up is where reviewers check whether the clinician stayed engaged. When you evaluate and do not report. The same rigor in reverse: the facts considered, the standard applied, why the threshold was not met, who was consulted, and the monitoring plan if concern resurfaces. Pitfall: an empty chart when the answer was no. The decision not to report is the one most likely to be second-guessed, and an undocumented one is indistinguishable from a missed one. ### Blank template (copy and adapt) Client: [initials] Date/time of note: Setting: Duty type: [ ] child abuse/neglect [ ] elder/dependent adult [ ] threat to identifiable person Trigger (what was disclosed/observed, when, by whom, in whose words): Basis for the duty (standard + facts that meet it): Consultation (who, when, substance of advice): Action taken: Agency/recipient: Date/time/method: Intake worker/ID: Report or reference #: Written follow-up form + deadline: For protective action: what was communicated, to whom, when (victim / law enforcement per your statute): Client notification ([ ] before [ ] after [ ] not informed + clinical rationale): Client response: Safety steps & follow-up (victim safety, plan changes, next contact, ongoing-duty acknowledgment): If NOT reporting: threshold analysis, consultation, monitoring plan: Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample mandated report note Scenario: a California therapist seeing an 8-year-old for individual therapy receives a disclosure in session and makes a child protective services report the same afternoon. All details are fictional, and the clinical content is deliberately generalized: it shows what a defensible note records, not how abuse presents. Client: J.R., 8 · Date of note: 07/21/2026, 3:40 pm · Setting: Outpatient office, scheduled individual session · Duty type: Suspected child abuse, report made Trigger and context: During today's 1:00 pm session, client spontaneously described being struck by an adult in the household during the past week and said it "happens when he gets mad." This clinician observed marks on the client's forearm consistent with the description. Client's statements are recorded here in the client's words as closely as possible; this clinician asked only the minimal clarifying questions needed to identify safety concerns and did not conduct an investigative interview. Basis for the duty: As a licensed marriage and family therapist, this clinician is a mandated reporter under California Penal Code 11165.7. The disclosure and observation described above constitute reasonable suspicion of physical abuse under Penal Code 11166(a). Certainty is not required and was not the standard applied. Consultation: Case reviewed by phone at 2:20 pm with clinical supervisor D.M., LMFT, who concurred that the reasonable-suspicion threshold was met and that the report should be made today. Action taken: Telephone report made to the county child protective services hotline today at 2:47 pm. Spoke with intake worker B. Alvarez, ID 4471; referral number 26-58214 was assigned and is recorded here. Written follow-up (form SS 8572) completed and submitted electronically at 3:25 pm today, within the 36-hour requirement. The information provided was limited to what the statute requires for the report. Client notification and response: Consistent with the hotline worker's guidance, the client's mother, A.R., who brought the client to session and is not the subject of the report, was informed in person at 3:10 pm that a report had been made and what to expect next. She was distressed but engaged, and safety options for the household were discussed with her. The client was told, in age-appropriate terms, that this clinician talked to people whose job is to help keep kids safe. Safety and follow-up: Mother confirmed the client will stay with her at the maternal grandmother's home tonight; she declined no offered support. Next session moved up to 07/23/2026, with a phone check-in scheduled for 07/22/2026. This clinician will cooperate with the agency's investigation as requested and understands the duty to report again if new or additional suspicion arises. Records note: This entry documents the report for the clinical record. The report itself belongs to the receiving agency; a copy of the submitted SS 8572 confirmation is retained per office policy. This sample is fictional and for educational purposes. It does not describe a real child, family, or report, and it intentionally omits the specifics a real record would contain. ↑ Back to the template and downloads ### Why this sample works It proves the timeline. The disclosure, the consultation, the phone report, and the written follow-up each carry a clock time, so the note itself demonstrates the immediate-report and 36-hour requirements were met, rather than asserting they were. The report is retrievable. Agency, intake worker, ID, and referral number are all recorded. If the agency's record ever cannot be located, the chart still proves the report happened, to whom, and when. Facts stay separated from conclusions. The note records what the child said and what the clinician observed, attributes each, and leaves the finding of abuse to the agency, while still stating plainly why the reasonable-suspicion standard was met. The notification decision is visible. Who was told, when, why, and how they responded are documented, so the choice can be defended instead of reconstructed from memory. The duty stays open. The ongoing-duty acknowledgment and the dated follow-up contacts show continued engagement, which is what reviewers check after the first report. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write this note knowing exactly who can read it. The agency's file is shielded by child-protection confidentiality statutes, and your identity as the reporter is protected there, but your chart note carries no such shield: it sits in the ordinary record, where the client, or a parent acting as a minor client's personal representative, generally holds HIPAA access rights, and where a subpoena or licensing-board review can reach it. That is why the note should carry facts, times, and reasoning rather than speculation, and why the report's own details beyond what the statute requires belong with the agency, not the chart. Federal privacy law adds one duty many clinicians have never heard of: when a practice discloses information about an adult it reasonably believes to be a victim of abuse, neglect, or domestic violence, HIPAA itself requires promptly informing that person a report was made, unless telling them would risk serious harm or would inform the personal representative believed responsible ( 45 CFR 164.512(c)(2) ). Child-abuse reports are expressly carved out of that pathway, so the informing duty does not attach to them. In a Part 2 substance-use program, the carve-out runs the other way: the regulations step aside for the initial child-abuse report itself, but keep protecting the underlying treatment records, including in any proceeding that grows out of the report (42 CFR 2.12(c)(6)). If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). Keep the legal layers straight, because they carry different weight. The report is LAW, with real clocks: California requires the phone report immediately or as soon as practicably possible and the written follow-up within 36 hours ( Penal Code 11166(a) ), New York requires the oral report immediately and the written report within 48 hours ( Social Services Law 415 ), and every state sets its own version. The chart note is CONVENTION: no statute reviewed for this page prescribes what the clinical note documenting a report must contain, which is exactly why it matters; it is the only evidence you control that the duty was met on time. Even the statute that comes closest, Queensland's child-protection act, which expressly lets a reporter confer with a colleague to "keep a record about giving a report," prescribes no fields for that record ( Child Protection Act 1999 s.13H ). Immunity is LAW everywhere: every state immunizes good-faith reports, a condition federal law has attached to child-welfare grants since CAPTA and broadened in 2019 to civil and criminal liability ( Gateway immunity compilation, January 2023 ), and California adds that a sued mandated reporter who prevails can claim up to $50,000 in attorney's fees from the state (Penal Code 11172). California's elder statute even carves a narrow psychotherapist exception when strict conditions are all met, including no independent corroboration of the client's statement, and expressly imposes no duty to investigate (Welfare and Institutions Code 15630(b)(3)). The duty to protect is the layer that varies most: a 2022 National Conference of State Legislatures review classifies roughly 29 jurisdictions as imposing a mandatory duty, 19 as permissive, and 3 as recognizing no duty ( NCSL ). California discharges it by "reasonable efforts to communicate the threat to the victim or victims and to a law enforcement agency" (Civil Code 43.92(b)); Washington extended it to foreseeable victims who were never named (Volk v. DeMeerleer, 2016); Texas imposes no duty at all and makes protective disclosure purely permissive (Thapar v. Zezulka, 1999; Health and Safety Code 611.004(a)(2)). The threshold also gets litigated on the record years later: in 2024 a California appellate court reversed board discipline against a psychologist after concluding the evidence showed no personal duty to warn existed at all ( Geffner v. Board of Psychology ), which is why the note's threshold analysis matters as much as the action taken. Document which duty you acted under and how your action tracked your state's formula. When a report, a warning, or a decision not to act touches your documentation, consult your attorney or board; state rules vary. Risk that reaches the client belongs in a suicide risk assessment and, where indicated, a safety plan alongside this note. ↑ Back to the template and downloads ## Common mandated report documentation errors boards flag The discipline data for this document comes from licensing boards and liability carriers rather than payers. In the 2024 CNA and HPSO counselor claim report, "reports to third parties" allegations made up 6.3 percent of closed professional-liability claims with paid indemnity, averaging $171,451 per claim, and 7.9 percent of board-defense matters; among the report's illustrative claim scenarios is a counselor who denied knowing of abuse concerns in a foster home but "did not have documentation to support this defense," a matter that closed with a total incurred over $900,000 ( CNA/HPSO, 2024 ). The knowledge gap is measured too: in the standard study of psychologists' duty-to-protect knowledge, 76 percent were wrong about their own state's law ( Pabian, Welfel, and Beebe, 2009 ). The BastionGPT Clinical Advisory Board sees the same errors most often in mandated report documentation reviews: A note that cannot prove the report. No clock time, no agency identifier, no intake worker, no reference number, no record the written follow-up went out inside its deadline. When the question is "did you report, and when," a note without those fields is an assertion, and the HPSO case study shows how that ends. Conclusions where observations belong. "Abuse confirmed" or "client is a danger" instead of what was said and seen, attributed and time-stamped. The finding belongs to the agency or the court; the chart's job is the observable basis for reasonable suspicion, in the source's words. Treating internal notification as the report. California's statute is explicit that the reporting duties "are individual," that no supervisor may impede them, and that internal policy cannot route the report through a supervisor (Penal Code 11166(i)); Ontario's act requires reporting "directly" and forbids relying on anyone else to do it (CYFSA s.125(3)). "I told my supervisor" is a consultation entry, not a report. An empty chart when the answer was no. Deciding the threshold was not met is legitimate clinical judgment, but reviewers cannot distinguish an undocumented decision from a missed duty. The facts weighed, the standard applied, the consultation, and the monitoring plan belong in the record with the same rigor as a report. Disclosing beyond what the permission covers. The federal permission runs to people "reasonably able to prevent or lessen the threat," including the target (45 CFR 164.512(j)), and Canada's Smith v. Jones limits disclosure to "only the information necessary to protect public safety." A warning that becomes a full history dump converts a protected disclosure into a privacy complaint. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on mandated reporting and duty-to-protect documentation. Draft the report note from your bullets or dictation, with the trigger, the basis for the duty, the consultation, the action taken, and the follow-up in the right places while the details are fresh. Check the note before you sign: flag a missing report number, an unrecorded time, an undocumented notification decision, a conclusion standing where an observation should be, or a not-reported decision with no threshold analysis. Prepare a minimum-necessary version when information leaves the chart, keeping a disclosure to an agency, attorney, or records request limited to what the recipient should see. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to be certain before making a report? No. The trigger everywhere is suspicion, not proof. California defines reasonable suspicion as what is "objectively reasonable for a person to entertain" based on the facts, drawing on training and experience, and states outright that it "does not require certainty" ( Penal Code 11166(a)(1) ); New York's standard is "reasonable cause to suspect" ( Social Services Law 413 ). Investigating to reach certainty is the agency's job, and waiting for it is how deadlines get missed. Document the facts that made the suspicion reasonable, and let the standard do its work. Should I tell the client before I make the report? For child-abuse reports, no statute reviewed for this page requires telling the client before or after; it is a clinical judgment about safety and the alliance, and informed-consent documents ideally covered the limits of confidentiality at intake. Do not delay the report to finish that conversation: the clocks run from when you learned the information, not from when the client is ready. Whatever you choose, document the choice and the reasoning. One exception is federal: when the disclosure concerns an adult victim of abuse, neglect, or domestic violence, HIPAA requires promptly informing that person a report was made unless doing so would risk serious harm or would inform a personal representative believed responsible (45 CFR 164.512(c)(2)). Child-abuse reports are expressly excluded from that informing duty. Can I be sued if the report turns out to be unfounded? Every state provides immunity for good-faith reports, a protection federal law has required as a condition of child-welfare funding and broadened in 2019 to cover civil and criminal liability ( Gateway compilation, January 2023 ). An unfounded report made in good faith is the system working as designed. California's immunity for mandated reporters is stronger still: no civil or criminal liability for any required or authorized report, plus up to $50,000 in attorney's fees from the state if you are sued and prevail (Penal Code 11172), and New York presumes your good faith (Social Services Law 419). The note that records your facts, timing, and consultation is what makes the immunity easy to apply. Is there a duty to warn in every state? No, and this is the most misunderstood fact in the field. A National Conference of State Legislatures review classifies roughly 29 jurisdictions as imposing a mandatory duty, 19 as permissive, and 3 as recognizing no duty ( NCSL, 2022 ). Texas imposes no duty at all and makes disclosure purely permissive (Thapar v. Zezulka, 1999), while Washington extended the duty to foreseeable victims the client never named (Volk v. DeMeerleer, 2016). In the standard study, 76 percent of surveyed psychologists were wrong about their own state's law ( Pabian, Welfel, and Beebe, 2009 ). Know your state's formula before the session where you need it, and document how your action tracked it. Does HIPAA allow me to report or warn without the client's consent? Yes. HIPAA permits disclosures required by law, permits child-abuse reports to the authorities designated to receive them, and permits disclosures the practice believes in good faith are "necessary to prevent or lessen a serious and imminent threat" when made to someone reasonably able to prevent it, including the threatened person (45 CFR 164.512(a), (b)(1)(ii), (j)). OCR's guidance confirms the threat permission covers even psychotherapy notes , and the rule presumes your good faith when the belief comes from your own knowledge of the client ( HHS FAQ 520 ). HIPAA is the permission layer; your state supplies the duty. What about clients in substance use treatment under 42 CFR Part 2? Part 2 steps aside for exactly one thing: the initial report of suspected child abuse or neglect to the appropriate authorities. The restrictions "continue to apply to the original substance use disorder patient records," including in any civil or criminal proceeding that arises out of the report ( 42 CFR 2.12(c)(6) ). Practically: a Part 2 program clinician makes the hotline report like any other mandated reporter, but the treatment records still need a court order or consent before they follow, and the chart note should reflect that boundary. An adult client discloses abuse they suffered as a child. Do I have to report it? Usually the duty is written around a child who is currently at risk: statutes like California's trigger when you know or reasonably suspect that "a child" has been abused, and an adult recounting their own history is no longer a child. The analysis changes when the person described still has access to children, which can create current reasonable suspicion about those children. State wording varies enough that this is a check-your-statute question, and a consultation with your board or attorney is worth documenting. Either way, record the analysis: the facts, the standard, and why the threshold was or was not met. A trauma assessment handles the clinical side of the history itself. What should I document when I decide not to report? The same elements as a report, in reverse: what was disclosed or observed, the statutory standard you applied, why the facts did not meet it, who you consulted, and the plan for monitoring if concern resurfaces. This is the decision most likely to be second-guessed later, and an undocumented no is indistinguishable from a missed duty. A dated consultation entry is the strongest single line: it converts a private judgment into a reasoned professional decision two people reached with the statute open. Can BastionGPT draft a mandated report note? Yes. Give it the timeline as bullets: what was disclosed, when you called, who took the report, the reference number, who you consulted, and who was told. It drafts the note with the trigger, basis, consultation, action, notification, and follow-up in the right places, and it can check a finished note for the missing element that matters, a report with no reference number, an unrecorded time, or a not-reported decision with no threshold analysis. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR 164.512 : the HIPAA permissions for required-by-law disclosures (a), child-abuse reports (b)(1)(ii), adult abuse-victim disclosures and the informing duty (c), and serious-and-imminent threats (j). HHS, HIPAA FAQ 520 : the serious-danger permission, including psychotherapy notes, and the good-faith presumption. 42 CFR 2.12(c)(6) : the Part 2 child-abuse-reporting exception and its limits. 42 USC 5106a(b)(2)(B) : CAPTA's state-plan conditions requiring reporting laws and good-faith immunity (Pub. L. 93-247, January 31, 1974; immunity broadened by Pub. L. 115-424, 2019). California Child Abuse and Neglect Reporting Act: Penal Code 11165.7 (mandated reporters), 11166 (trigger, timing, individual duty, penalty), and 11172 (immunity and attorney's fees). California Civil Code 43.92 : the duty-to-protect limitation and its discharge formula (renamed from duty to warn by SB 1134, effective 2013). California Welfare and Institutions Code 15630 : elder and dependent-adult reporting, as amended effective 2024, including the narrow psychotherapist exception at (b)(3). New York Social Services Law 413 , 415 , 419 , and 420 : reporters, procedure and the 48-hour written report, immunity, and penalties. New York Mental Hygiene Law 9.46 : the SAFE Act reporting duty and its firearms-licensing scope. Texas Health and Safety Code Chapter 611 and Thapar v. Zezulka, 994 S.W.2d 635 (Tex. 1999) : permissive disclosure, no duty to warn, and the 2021 no-independent-duty and immunity provisions. Tarasoff v. Regents, 17 Cal.3d 425 (1976) , Volk v. DeMeerleer, 187 Wn.2d 241 (2016) , and Geffner v. Board of Psychology (Cal. Ct. App. 2024) : the duty to protect, its Washington extension to foreseeable victims, and a 2024 reversal of board discipline on the evidence. Smith v. Jones, [1999] 1 S.C.R. 455 : Canada's public-safety exception and its clarity, seriousness, and imminence factors. Child Welfare Information Gateway compilations: mandatory reporting (May 2023), penalties for failure to report (February 2019), and reporter immunity (January 2023). HHS Administration for Children and Families, Child Maltreatment 2024 : report sources (professionals 70.9 percent; mental health personnel 6.4 percent) and national totals. NCSL, Mental Health Professionals' Duty to Warn (2022): the state-by-state mandatory, permissive, and no-duty classification. Johnson, Persad, and Sisti, The Tarasoff Rule, J Am Acad Psychiatry Law (2014) : interstate variation and the Pabian, Welfel, and Beebe (2009) knowledge findings. CNA and HPSO, Counselor Professional Liability Exposure Claim Report, 3rd edition (2024) : reporting-allegation frequency, severity, and the documentation case scenarios. Outside the US: Ontario's Child, Youth and Family Services Act, 2017, s. 125 ; Queensland's Child Protection Act 1999, ss. 13E-13H ; the AIFS mandatory reporting resource sheet (April 2026); and AHPRA's mandatory notifications guidance , a practitioner-conduct regime distinct from child protection. ## Related templates Suicide Risk Assessment Safety Plan Crisis Psychotherapy Note Trauma Assessment Release of Information (ROI) Authorization Violence & Homicide Risk Assessment Incident & Adverse Event Report Court Report & Treatment Progress Update Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/safety-plan All note templates # Safety Plan: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A safety plan is a brief document a clinician and client build together during or after a suicidal crisis: personal warning signs, coping strategies, people and places that help, professional and crisis contacts, and steps to make the environment safer. The client keeps a copy. Most follow the six-step Stanley-Brown format, often with an optional reasons-for-living step, and take 20 to 45 minutes to complete together. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, social workers, psychiatric clinicians, crisis and emergency department teams, with the client as co-author Audience The client, who keeps a copy; the treating team, crisis services, accreditation surveyors, and sometimes family or supports Typical length 150 to 400 words · 20 to 45 minutes with the client (clinical team estimate) Format family Risk mitigation document (compare: suicide risk assessment, crisis note, relapse prevention plan) When it's used After a positive suicide risk screen or crisis, at ED or inpatient discharge, and whenever risk or supports change Standards context No US law prescribes the form; accreditor expectations, one state discharge statute, and Medicare's G0560 rules shape what must be present On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a safety plan? A safety plan is a short, prioritized list of coping strategies and sources of support that a clinician and client build together during or after a suicidal crisis, for the client to use when warning signs return. The dominant format is the Safety Planning Intervention developed by psychologists Barbara Stanley (Columbia University) and Gregory Brown (University of Pennsylvania), first published as brief instructions for the VA in 2008 and described fully in 2012. Clinicians and payers also call it a Stanley-Brown safety plan, a crisis safety plan, a patient safety plan, or SPI; the closely related crisis response plan (CRP) grew out of military settings and covers the same ground with a slightly different structure. The developers are explicit that the intervention is more than the form: their protocol includes a risk assessment first, a rationale, collaborative drafting, a how-to-use conversation, and a follow-up review of whether the plan helped. Two boundaries keep getting blurred. First, a safety plan is a behavioral health document, distinct from the safety planning a domestic violence advocate does and from wellness tools like WRAP; if you searched "safety plan template," this page covers the suicide-prevention document. Second, a safety plan is what replaced the no-suicide contract: instead of asking the client to promise not to act, it gives the client specific things to do, and the evidence comparison has favored the planning approach (see the FAQs). It also sits apart from the suicide risk assessment that precedes it and the crisis note that records the encounter around it: the safety plan is the one document the client takes home. ## Who uses safety plans and when Safety plans show up wherever suicide risk is managed: outpatient therapy practices, community mental health, emergency departments, inpatient units at discharge, crisis teams, school-based clinicians, and the VA, where the format originated. The trigger is a positive risk screen or a suicide risk assessment that finds elevated risk, a crisis encounter documented in a crisis note , or a discharge from an ED or inpatient stay, the window accreditors single out for follow-up planning. Reach for a safety plan rather than its neighbors when the job is giving the client something to use between sessions: the risk assessment records your clinical judgment, the treatment plan sets the course of care, and the safety plan is the one document written for the client's own pocket. It gets reviewed and updated at follow-up visits, after any new crisis, and whenever contacts or circumstances change. ## Safety plan structure: what goes in each section The structure below follows the Stanley-Brown Safety Planning Intervention, the format the developers published for the VA in 2008 and the one most agencies and EHRs have adopted. The official form runs six steps; the optional reasons-for-living step used by the VA adaptation appears here as step 7. Two things make a safety plan different from every other document on this site: the client is the co-author, and the client keeps a copy. Write every entry in the client's own words. One licensing note: the official Stanley-Brown form permits individual clinical use but requires the authors' written permission for any changes or for use in an EHR, so the template below is our own wording of the standard components, free to adapt. Step 1: Warning signs. The personal thoughts, images, moods, situations, or behaviors that tell this client a crisis may be building. These cue the client to open the plan, so they come first. Pitfall: generic entries like "feeling sad." A warning sign only works if the client recognizes it as theirs; "lying awake replaying the layoff" cues action, "sadness" does not. Step 2: Internal coping strategies. Things the client can do alone to ride out the surge: a shower, a playlist, a walk, paced breathing. The point is to let time pass without contacting anyone, because crises crest and fall. Pitfall: strategies the client cannot actually reach at 2 am. Test each one: "could you do this alone, tonight, without leaving home?" Step 3: People and places that provide distraction. Social contacts and settings that take the client's mind off the crisis without requiring disclosure: a friend to text, a coffee shop, a group. The client does not have to say anything is wrong. Pitfall: collapsing this step into step 4. Distraction and asking for help are different jobs, and the developers kept them separate on purpose. Step 4: People I can ask for help. Family or friends the client is willing to tell "I am in crisis and need support," with phone numbers. Pitfall: names without numbers, or contacts the client would never actually call. Ask "who would you honestly reach out to?" and write down what they say. Step 5: Professionals and agencies to contact in a crisis. The treating clinician with after-hours instructions, crisis lines (988 in the US, 9-8-8 in Canada, Lifeline 13 11 14 in Australia), and the nearest emergency department with its address. Pitfall: listing only the therapist's office line. A plan used at midnight needs a number that answers at midnight. Step 6: Making the environment safer. The means-safety steps the client agreed to, stated as protective actions with who is helping and when it will be reviewed. Document the conversation even when the client declines a step. Pitfall: skipping or vaguing out this step. The developers call means safety a key component of the intervention, and it is the entry reviewers look for first. Step 7: Reasons for living (optional). What makes life worth living for this client, in their words. The official form has six steps; the VA adaptation adds this one, and many clinicians keep it because it personalizes the plan. Pitfall: filling it in for the client. A borrowed reason persuades nobody, least of all the person in crisis. Documentation wrapper. Around the plan itself, the chart should show the date, that the plan was built collaboratively, that a copy went home with the client and where they will keep it, who else received it with consent, and when it will be reviewed. Pitfall: a completed plan with no record that the client ever received a copy. The intervention assumes the client can reach the plan in a crisis; a copy that lives only in the EHR protects the file, not the person. ### Blank template (copy and adapt) Client: [initials] Date: Completed with: [clinician, credentials] Step 1. Warning signs a crisis may be building (my thoughts, moods, situations, behaviors): 1. 2. 3. Step 2. Internal coping strategies (what I can do on my own to let the surge pass): 1. 2. 3. Step 3. People and places that provide distraction (no need to disclose): Name/place: Contact: Name/place: Contact: Step 4. People I can ask for help (I will tell them I am in crisis): Name: Phone: Name: Phone: Step 5. Professionals and agencies I can contact in a crisis: Clinician + after-hours instructions: Crisis line: 988 (US) · 9-8-8 (Canada) · Lifeline 13 11 14 (Australia) Nearest emergency department + address: Step 6. Making my environment safer (agreed protective steps, who is helping, review date): 1. 2. Step 7. Reasons for living (optional): Copy given to client: [ ] yes Where I will keep it: Shared with (with consent): Review date: Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample safety plan Scenario: adult outpatient client with major depressive disorder, seen two days after an emergency department visit for suicidal ideation without attempt. The plan was completed collaboratively in a 40-minute session, and the entries are in the client's own words. All details are fictional. Client: M.T., 29 · Date: 07/16/2026 · Completed with: J. Alvarez, LCSW, in office, plan built collaboratively over 40 minutes Step 1, warning signs: Staying in bed past noon on days off. Turning my phone face down and letting texts pile up. The "everyone would be fine without me" loop starting at night. Step 2, internal coping: Hot shower, then the running playlist. Take Biscuit for a walk around the block. The 4-7-8 breathing we practiced until the wave passes. Step 3, people and places for distraction: Pickup basketball at the rec center (Tuesday and Thursday evenings). Texting my cousin D. about the game. The coffee shop on 5th where I do crosswords. Step 4, people I can ask for help: Sister, A.T. (phone in my contacts, she knows about this plan). Roommate, C., can sit with me and handles mornings well. Step 5, professionals and agencies: J. Alvarez, LCSW, office line with after-hours voicemail instructions. 988 by call or text, saved to favorites. St. Mary's Hospital emergency department, 1400 Oak Street, 10 minutes away. Step 6, making my environment safer: Medications moved to a lockbox that C. keeps; I hold a two-week supply. Agreed to keep alcohol out of the apartment for now. Review both steps with J. Alvarez on 07/30/2026. Step 7, reasons for living (optional): Biscuit. Watching my niece grow up. Finishing the welding program I started. Documentation wrapper (chart note): Plan completed collaboratively; client generated all entries and read the plan back. Printed copy given to client, who will keep it folded in their wallet; photo saved on their phone. Copy shared with sister A.T. with written consent. Plan to be reviewed at each session through August and after any crisis contact. Next session 07/23/2026. Signed: J. Alvarez, LCSW, 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every entry is in the client's own words, specific enough to act on at 2 am: named people, real places, saved numbers, not categories. The documentation wrapper records what reviewers actually look for: collaborative completion, a copy in the client's hands and where it lives, consented sharing, and a review plan. Step 6 documents means safety as agreed protective actions with a named helper and a dated review, without method detail that does not belong on a shareable document. Step 5 works after hours: a crisis line saved to the phone and a named emergency department with its address, not just the office line. The plan connects to care: a dated next session, review at each visit, and a clear link back to the treatment plan. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A safety plan lives in two places at once: the copy in the client's hands and the record in the chart, and the chart side is what surveyors and reviewers read. For Joint Commission-accredited organizations, NPSG 15.01.01 requires documenting each at-risk patient's overall risk level and "the plan to mitigate the risk for suicide" (EP 4), and requires written policies for counseling and follow-up care at discharge (EP 6), where the Commission's own rationale names developing a safety plan with the patient and providing crisis numbers as what works. One state goes further: for elevated-risk patients at psychiatric centers and OMH-licensed inpatient services, New York's Mental Hygiene Law 29.15 requires "an individualized community suicide safety plan completed before discharge," provided to the patient's aftercare providers. Chart the collaboration, not just the artifact: who participated, that the client kept a copy and where, who else received it with consent, and when it will be reviewed. When a safety plan follows a duty-to-warn situation or a mandated report, consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The payer layer is new and worth knowing. Since January 1, 2025, Medicare pays separately for safety planning interventions under G0560 : for patients in crisis, including suicidal ideation or risk of suicide or overdose, performed by the billing practitioner in a variety of settings and billed in 20-minute increments, so the time and the personalized work need to be visible in the note. The same rule created G0544, a monthly bundle of four follow-up contacts after an emergency department discharge for a crisis encounter, and safety planning joined the Medicare telehealth list on a permanent basis. Two details from the final rule matter for documentation: CMS finalized G0560 as a stand-alone service that the billing practitioner performs personally, so a template completed by other staff does not become billable through a co-signature, and CMS deliberately kept a crisis narrative and a specific written form out of the code definition, so the personalized elements themselves are what your note has to show. The form is a convention; the collaborative work, the time, and the copy in the client's hands are the requirement. When the plan changes what treatment looks like, close the loop in the treatment plan and the next progress note . ↑ Back to the template and downloads ## Common safety plan errors auditors flag The audit record for safety plans comes from the system that invented the format. The VA inspector general's December 2024 national review of 200 inpatient mental health discharges found no completed safety plan for 12 percent of patients, and among the completed plans that missed required elements, insufficient contact information was the leading defect. An earlier lethal-means review found that when patients with firearm-related suicidal behavior reported firearm access, 21 percent of safety-planning records had no documented safe-storage discussion. Medicare's G0560 is too new for any of this: as of July 2026 there were no published denial rates or utilization data for the code, so the payer-side audit record is still being written. The BastionGPT Clinical Advisory Board sees the same errors most often in safety plan reviews: Template entries, and contacts without numbers. "Call a friend" and "go for a walk" appear verbatim on thousands of plans, and a first name with no phone number fails at exactly the moment the plan exists for. The VA's 2024 review found insufficient social and professional contact information was the most common defect in completed plans. Strategies never checked against the client's own crisis story. In one VA inspector general case, a plan listed riding a wheelchair outside as a distraction strategy for a patient whose recent suicidal behavior was riding a wheelchair into traffic. The clinician described discussing the risk; the discussion was never documented, and to a reviewer an undocumented conversation did not happen. The means-safety step skipped or vague. "Discussed safety" does not show what was agreed. Record the protective steps, who is helping, and the review date, and record the conversation even when the client declines a step. No record that the client received a copy. The intervention assumes the client can reach the plan in a crisis. A plan that exists only in the EHR protects the file, not the person, and the chart should say where the client's copy lives. Completed once and never revisited. Contacts move, medications change, and a plan built during one crisis goes stale. A dated review at follow-ups is what shows ongoing risk management rather than a one-time form. A promise instead of a plan. Asking the client to sign a commitment not to act is a no-suicide contract wearing a safety plan's name. The plan works because it says what to do, not what not to do. ### Safety plans in the US, Canada, and Australia Aspect United States Canada Australia Status No federal law prescribes the form; Joint Commission NPSG 15.01.01 requires risk and mitigation documentation plus discharge policies, New York mandates a pre-discharge plan in covered psychiatric settings, and Medicare pays G0560 for the intervention Convention plus accreditation: the national suicide-prevention standard CAN/HSO 5064:2023 builds safety planning into accreditation programs; provincial college record standards govern the documentation Convention plus accreditation: NSQHS Comprehensive Care actions 5.31 and 5.32 require systems for identifying and responding to suicide risk, with no form prescribed Terminology Safety plan, Stanley-Brown safety plan, SPI; crisis response plan in military settings Safety plan, crisis plan Safety plan, crisis plan; part of "safety planning" in NSW policy What changes The payer layer: G0560, G0544, and the telehealth listing are Medicare-specific Crisis line is 9-8-8; no national billing code; Ontario supervisors co-sign plans relied on for care decisions Crisis line is Lifeline 13 11 14; NSW public services must give the person a copy and document one in the eMR; no MBS safety-planning item Retention Follows the chart, not the form: no universal HIPAA rule; Medicare hospital records five years, state law and payer contracts govern beyond that Provincial; Ontario colleges expect 10+ years from last interaction, or from the client's 18th birthday State rules; NSW requires 7 years for adults and until age 25 for minors The plan itself travels well: warning signs, coping steps, and contacts work the same way everywhere. What changes is the crisis number on step 5, the billing layer around the intervention, and the copy and retention rules for the chart. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on safety plans. Draft a structured safety plan from your session bullets, keeping every entry in the client's own words and flagging the steps you have not filled in yet. Rewrite a completed plan in plain language at the client's reading level, ready to print, send to the portal, or hand to a family member with consent. Check a plan before you sign: contacts without phone numbers, a skipped means-safety step, or a missing copy-given note in the chart documentation. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a safety plan required by law? No federal law prescribes a safety plan form. What exists is an accreditation and payment layer: Joint Commission-accredited organizations must document each at-risk patient's risk level and mitigation plan and follow written discharge follow-up policies under NPSG 15.01.01 , and Medicare pays for the intervention through G0560 without mandating it. One state statute does exist: New York's Mental Hygiene Law 29.15 requires an individualized community suicide safety plan before discharge for elevated-risk patients at OMH-licensed psychiatric settings. Everywhere else, the safety plan is how organizations satisfy broader expectations, which is different from the form itself being required. What is the difference between a safety plan and a no-suicide contract? A no-suicide contract asks the client to promise not to act; a safety plan gives the client specific things to do when warning signs appear. The only randomized trial to compare the approaches, Bryan and colleagues' 2017 study in US Army soldiers, found crisis response planning outperformed contracts for safety on subsequent attempts. Patient-safety guidance points the same way: the Pennsylvania Patient Safety Authority titled its advisory on the practice "Contracting for Safety: A Misused Tool." If a document asks for a signature on a promise, it is a contract, whatever the header says. Can I bill for creating a safety plan? Under Medicare, yes, since January 1, 2025. HCPCS G0560 pays for safety planning interventions for patients in crisis, including suicidal ideation or risk of suicide or overdose, performed by the billing practitioner in a variety of settings and billed in 20-minute increments, and the service sits on the Medicare telehealth list permanently. The same rule created G0544, a monthly bundle covering four follow-up contacts after an emergency department discharge for a crisis encounter. Document the time and the personalized, collaborative work; a photocopied form does not show a 20-minute intervention. Do safety plans actually work? The evidence is favorable with honest caveats. A 2021 meta-analysis of six studies with 3,536 participants found safety-planning-type interventions cut suicidal behavior nearly in half (relative risk 0.570, number needed to treat 16), with no measured effect on suicidal ideation. The best-known single study, the 2018 comparison across nine VA emergency departments, found 45 percent fewer suicidal behaviors (3.03 versus 5.29 percent over six months) and twice the odds of engaging in outpatient care; that result covers the plan plus follow-up calls, not the form alone. The 2024 VA/DoD guideline , which counts only randomized trials, graded the trial evidence insufficient while keeping safety planning as standard care in both systems and noting no evidence of harm. Does the client get a copy of the safety plan? Yes; the copy in the client's hands is the intervention. The developers' protocol includes discussing where the plan will be kept and the barriers to using it, and the Joint Commission's discharge rationale pairs the plan with crisis numbers the patient leaves with. Document that a copy was given, where the client will keep it, and who else received one with consent. A plan that exists only in the EHR protects the file, not the person. How long does a safety plan take to complete? Plan for 20 to 45 minutes of collaborative conversation, the range the developers themselves published; the VA/DoD guideline notes the intervention can be completed in a single session. The written product is short, usually 150 to 400 words across the six steps of the Stanley-Brown format, plus the optional reasons-for-living step the VA adaptation uses. The time goes into eliciting entries the client will actually recognize and use, which is also what G0560's 20-minute increments are paying for. What is the difference between a safety plan, a suicide risk assessment, and a crisis note? They answer different questions. The suicide risk assessment records how you evaluated risk and why the disposition was safe; it is written for the record. The crisis note documents a crisis encounter: what happened, what you did, and where the client went. The safety plan is the take-home product of that work, written with and for the client. A high-risk presentation often generates all three on the same day, each doing its own job. How often should a safety plan be reviewed or updated? No regulation sets a universal frequency. The 24-hour and 120-day windows clinicians sometimes cite come from MIPS Quality ID 504 , a reporting measure that only applies when a clinician reports it, not a chart deadline. The Joint Commission requires written policies that include guidelines for reassessment, and the developers' protocol ends with a follow-up review of whether the plan helped. In practice: review at each visit while risk is elevated, after any crisis contact, and whenever a contact, medication, or living situation on the plan changes. Date each review; a plan with a review trail reads as risk management, a plan without one reads as paperwork. Can BastionGPT draft a safety plan? Yes. Give it your session bullets and it drafts the plan with every entry kept in the client's own words, flags empty steps and contacts without numbers, and can rewrite the finished plan in plain language for the client's copy. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS, MM13887, CY 2025 Physician Fee Schedule final rule summary : G0560 safety planning interventions, G0544 post-discharge follow-up contacts, and the permanent telehealth listing. Joint Commission, R3 Report Issue 18: National Patient Safety Goal for suicide prevention : NPSG 15.01.01 elements of performance, including EP 4 documentation and EP 6 discharge requirements. Stanley B, Brown GK, et al., "Comparison of the Safety Planning Intervention With Follow-up vs Usual Care of Suicidal Patients Treated in the Emergency Department," JAMA Psychiatry (2018) : the 45 percent reduction in suicidal behavior and doubled treatment engagement. VA/DoD, Clinical Practice Guideline for the Assessment and Management of Patients at Risk for Suicide (2024) : safety planning as standard of care, the SPI/CRP comparison, and the evidence review. Stanley-Brown Safety Planning Intervention, official site : the six-step form, the optional reasons-for-living step, the intervention protocol, and the developers. Bryan CJ, et al., "Effect of crisis response planning vs. contracts for safety on suicide risk in U.S. Army Soldiers," Journal of Affective Disorders (2017) : the randomized comparison with no-suicide contracts. Nuij C, et al., "Safety planning-type interventions for suicide prevention: meta-analysis," British Journal of Psychiatry (2021) : the pooled effect on suicidal behavior and the null ideation finding. New York State Senate, Mental Hygiene Law 29.15 : the pre-discharge community suicide safety plan mandate for covered psychiatric settings. CMS, CY 2025 Physician Fee Schedule final rule : G0560 finalized as a stand-alone, personally performed service, and the decision to keep a crisis narrative and form language out of the code definition. VA Office of Inspector General, Deficiencies in Inpatient Mental Health Suicide Risk Assessment (December 2024) : the 12 percent missing-plan rate and contact-information findings. VA Office of Inspector General, Deficiencies in Lethal Means Safety Training, Firearms Access Assessment, and Safety Planning (2022) : the safe-storage documentation gap. Pennsylvania Patient Safety Authority, "Contracting for Safety: A Misused Tool" (2013) : patient-safety guidance on no-suicide contracts. CMS Quality Payment Program, MIPS Quality ID 504 specification (2026) : the source of the 24-hour and 120-day reporting windows. NSW Health, PD2022_043, Clinical Care of People Who May Be Suicidal (2022) : the patient-copy and eMR documentation rule. ## Related templates Suicide Risk Assessment Crisis Psychotherapy Note Trauma Assessment Treatment Plan Intake Note Psychotherapy Progress Note Discharge Summary Mandated Reporting & Duty-to-Protect Documentation Violence & Homicide Risk Assessment Self-Harm (NSSI) Assessment Patient Instructions & After-Visit Summary Appointment No-Show & Outreach Note Incident & Adverse Event Report Canada Telepsychology & Virtual Care Note General Risk Screening Note PHQ-9 Score Documentation C-SSRS Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/suicide-risk-assessment All note templates # Suicide Risk Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A suicide risk assessment is the clinical documentation of how a clinician evaluated a client's suicide risk: the inquiry into ideation, plan, intent, and access to means, the risk and protective factors weighed, a risk formulation, and the actions taken. Mental health professionals document one at intake, when risk indicators emerge, and after a crisis. Most run 200 to 800 words, and the formulation must support the disposition. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, psychiatrists, psychiatric NPs, crisis clinicians Audience Treating team, supervisors, crisis and emergency services, auditors, licensing boards, legal reviewers Typical length 200 to 800 words · 10 to 25 minutes by hand (clinical team estimate) Format family Risk assessment documentation (compare: safety plan, crisis note, self-harm assessment) When it's used At intake, when ideation or other risk indicators emerge, after a crisis or attempt, and at transitions in care Standards context No mandated format; Joint Commission standards define the assessment domains in accredited settings, the clinical standard of care everywhere else On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a suicide risk assessment? A suicide risk assessment is the clinical record of a suicide-specific evaluation: what the clinician asked and observed, how the findings were weighed, the risk formulation that resulted, and the actions taken in response. It is a younger document family than most note formats, and its structure comes from clinical frameworks rather than a single inventor. SAMHSA published SAFE-T, the five-step evaluation-and-triage framework conceived by Douglas Jacobs, in 2009, and documentation is its express fifth step ( SAMHSA, updated December 2024 ). The Columbia Protocol, or C-SSRS, was developed by teams at Columbia, the University of Pennsylvania, and the University of Pittsburgh with NIMH support, with its multisite validation published by Posner and colleagues in 2011 ( Columbia Lighthouse Project ). Clinicians and payers also say SRA, suicide assessment, suicide inquiry, brief suicide safety assessment, and suicide risk formulation. Whatever the name, the instrument is not the note: a completed screener can sit in the chart, but a score does not explain a disposition. Three distinctions carry most of the compliance weight. Screening is a brief case-finding step; assessment is the suicide-specific clinical inquiry a positive screen or disclosure triggers; formulation is the clinician's synthesis that connects the findings to the plan. No statute in the US, Canada, or Australia prescribes a universal template for any of them. The strongest explicit documentation requirements are accreditor standards, and the citation changed in 2026: Joint Commission's hospital requirements moved into National Performance Goal NPG.08.01.01 on January 1, 2026, while behavioral health organizations remain under NPSG.15.01.01 , so guidance citing one number for every setting is out of date. In covered settings the assessment must directly address suicidal ideation, plan, intent, suicidal or self-harm behaviors, risk factors, and protective factors, with the overall risk level and mitigation plan documented; a validated tool is encouraged but no tool is mandated by name ( Joint Commission FAQ, updated April 2026 ). Outside accredited organizations, licensing-board record standards and the clinical standard of care expect the same reconstruction, which is why this page treats the structure as one document type. It records the evaluation; the crisis note records an urgent encounter, and a safety plan is a separate intervention document the assessment often produces. ## Who uses suicide risk assessments and when Every licensed mental health discipline documents them: therapists, psychologists, counselors, clinical social workers, psychiatrists, and psychiatric NPs, along with ED and primary care teams that screen. In outpatient practice the trigger is usually a positive screen (an elevated PHQ-9 item 9, a C-SSRS screen) or an in-session disclosure; in hospitals and EDs, accreditation requirements build the assessment into the workflow for behavioral health presentations. The document also anchors specific moments in care: intake , a crisis encounter , a return from hospitalization, and other transitions. A risk assessment is the evaluation record; when the encounter itself was crisis psychotherapy, the crisis note carries the service, and the risk assessment reasoning lives inside or alongside it. ## Suicide risk assessment structure: what goes in each section No statute prescribes these headings. They mirror what reviewers, accreditors, and courts look for afterward: what prompted the assessment, what was asked and learned, how the clinician weighed it, and what was done about it. The through-line is that the formulation must explain the disposition. Reason for assessment and context. What triggered the assessment (a screening result, a disclosure, an intake, a transition in care) and the clinical context around it: precipitants, symptom course, substance use. Pitfall: a positive screen with no documented follow-up. An elevated PHQ-9 item 9 sitting in the chart with nothing after it is the single most indefensible gap in a later review. Suicide inquiry. Direct questioning documented in the client's own terms: ideation and its frequency, duration, and recency; any contemplated method; plan; intent; and the behavior history (prior attempts, aborted or interrupted attempts, preparatory acts, self-harm). Method, plan, and intent are separate findings, so record each rather than one collapsed "SI" line. Pitfall: "denies SI" standing alone. Record what was asked and what was endorsed, not only the denial; a later reader should see the inquiry, not just its bottom line. Access to means. The conversation about access to lethal means, asked about directly (firearms by name, plus anything this client's situation makes relevant), the restriction steps agreed, who will carry each step out, and what was actually verified. Pitfall: conclusions with no conversation shown. "No access to means," "means removed," and "means counseling provided" all read as unsupported unless the note records what was discussed, what was agreed, and what was confirmed. Risk and protective factors. Acute and chronic risk factors, and protective factors that are specific and current for this client, not a generic list. Pitfall: protective factors copied forward ("family support, future oriented") that appear in every note unchanged. Reviewers read that as boilerplate, and boilerplate cuts both ways in a courtroom. Instruments and scores. Any screener or scale used (C-SSRS, PHQ-9, ASQ) with the result and its date. Pitfall: treating the score as the assessment. An instrument structures the inquiry and triggers clinical judgment; it does not replace the formulation, and no score by itself supports a disposition. Risk formulation. The clinical synthesis: a risk characterization (many settings use low, moderate, or high for acute and chronic risk) with the rationale that connects the findings above to it. Pitfall: a level with no reasoning, or prediction language ("client will not act on thoughts"). The standard the note is held to is a reasonable assessment process, documented, never an accurate prediction. Interventions. What was done in response to the formulation: safety planning (documented as completed, with the client's copy noted), means-safety steps, supports involved with consent recorded, treatment changes, crisis resources provided. Pitfall: a "no-suicide contract" documented as the intervention. Contracts have no evidence of protective effect and read poorly in any later review; a collaborative safety plan is the defensible artifact. Disposition, follow-up, and consultation. The level of care decided, the dated next contact, escalation instructions given, any consultation or supervision obtained, and when risk will be reassessed. Pitfall: "will monitor" with no date, and no reassessment plan. Elevated risk documented once and never revisited reads as abandoned, not managed. ### Blank template (copy and adapt) Client: [initials] Date: Setting/context: Reason for assessment (screen result, disclosure, intake, transition): Suicide inquiry (ideation: frequency/duration/recency; plan; intent; behavior history: attempts, preparatory acts, self-harm): Access to means & means-safety steps agreed: Risk factors (acute / chronic): Protective factors (specific to this client, today): Instruments & scores (C-SSRS, PHQ-9, ASQ), with dates: Risk formulation (level + rationale connecting findings to disposition): Interventions (safety plan completed, means counseling, supports w/ consent, treatment changes, crisis resources given): Disposition & follow-up (level of care, dated next contact, escalation instructions): Consultation / supervision (who, when, outcome): Reassessment plan: Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample suicide risk assessment Scenario: adult client in weekly outpatient therapy for depression discloses new suicidal ideation during a scheduled session, and the clinician completes and documents a full risk assessment in the same visit. All details are fictional, and the risk content is deliberately generalized: it shows what a defensible note records, not how risk presents. Client: M.T., 29 · Date: 07/21/2026 · Setting: Outpatient office, scheduled session · Reason for assessment: PHQ-9 item 9 positive at check-in; ideation disclosed in session Context and precipitant: Client is in week 11 of weekly CBT for major depressive disorder. A four-year relationship ended three weeks ago. Since then client reports worsening sleep, social withdrawal, and drinking above baseline several evenings a week. PHQ-9 today 17 (moderately severe), up from 12 two weeks ago; item 9 scored 1. Suicide inquiry: Asked directly. Client reports wishing they "would not wake up" most evenings for about two weeks, and thoughts of ending their life on three or four occasions in the past week, lasting minutes, without method, plan, or intent. Most recent episode two days ago. Denies ideation in session today. Denies prior attempts, preparatory or rehearsal behavior, and any history of self-harm. C-SSRS screen: ideation level 2 (nonspecific active thoughts, past month); levels 3 to 5 negative; behavior scale negative. Access to means: Reviewed directly with client. No firearms in the home. One category of concern was identified, and a specific plan to restrict access was agreed with the client and recorded in the safety plan. Client agreed to a follow-up conversation about means safety at the next session. Risk factors: Acute: recent relationship loss, moderately severe depressive symptoms, disrupted sleep, increased alcohol use, new-onset ideation. Chronic: one prior depressive episode in early twenties, resolved with treatment; no attempt history; no family history of suicide reported. Protective factors: Engaged in weekly treatment and open in today's disclosure; close relationship with sister; stable employment; stated reasons for living, including care for their dog and a planned trip with friends in August; expressed clear ambivalence and willingness to plan for safety; no intoxication in session. Risk formulation: New passive and intermittent nonspecific active ideation without plan, intent, preparatory behavior, or attempt history, in the context of an acute loss, moderately severe depression, and increased alcohol use. Acute risk is judged low to moderate, manageable in outpatient care with the interventions below. Chronic risk is low. Rationale: ideation is recent, brief, and nonspecific; means access is addressed; protective factors are current and specific; the client's account was consistent and openly given; and the client engaged actively in safety planning. Foreseeable destabilizers (an escalation in drinking, renewed contact with the former partner) were named with the client and are covered by the plan below. Interventions: Collaborative safety plan completed in session (warning signs, internal coping steps, social contacts and settings, professional contacts, means safety), copy to client. 988 saved to client's phone. With client's written consent, sister was called during session and agreed to evening check-ins this week. Alcohol use discussed; client agreed to reduce while symptoms are acute and to review this next session. Disposition and follow-up: Continue outpatient care at increased frequency: sessions moved to twice weekly for two weeks, next appointment 07/24/2026, phone check-in 07/22/2026 at 5:00 pm. Referral placed for medication evaluation with practice's consulting psychiatric NP; client accepted. Client verbalized the safety plan steps and the instruction to call or text 988 or go to the nearest emergency department if thoughts escalate or intent develops. Consultation: Case reviewed same day with clinical supervisor, L.R., LCSW; supervisor concurred with formulation and outpatient disposition with increased contact. Reassessment plan: Ideation, means safety, and safety-plan use reviewed at every contact while ideation is active; full reassessment at the next session or with any change in presentation. This sample is fictional and for educational purposes. It does not describe a real patient, and it intentionally omits the clinical specifics a real record would contain. ↑ Back to the template and downloads ### Why this sample works The inquiry is shown, not summarized: what was asked, what the client endorsed, what they denied, and how recently, so "denies SI today" sits inside its real context instead of standing alone. Ideation is anchored to instruments (C-SSRS level, PHQ-9 trend) and to time (onset, frequency, duration, recency), which makes the next reassessment comparable. The formulation states a risk level and the rationale behind it, so the outpatient disposition is supported by reasoning a later reader can follow. Means access gets its own documented conversation with an agreed restriction step, rather than a bare "no access to means" conclusion. Every intervention is recorded as completed with dates and consent noted: the safety plan, the sister's involvement, the consult referral, and the same-day supervisor review. A reassessment plan closes the note, so elevated risk is tracked forward instead of assessed once and left behind. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A suicide risk assessment is written for more readers than any other clinical document: the treating team and supervisor first, then surveyors, auditors, licensing boards, and, after a bad outcome, coroners and courts. In accredited settings the documentation itself is the requirement: Joint Commission expects the overall risk level "clearly documented, with clinical justification, as well as the plans to mitigate" ( standards FAQ, April 2026 ), under NPG.08.01.01 for hospitals and NPSG.15.01.01 for behavioral health organizations. In private practice no accreditor applies, but board record standards and the clinical standard of care ask for the same reconstruction, and the note sits inside the ordinary record: clients generally hold HIPAA access rights to it, and it is reachable by subpoena, so write clinically necessary detail rather than speculation (the separately kept psychotherapy notes category never covers it). After an attempt or death, never revise the original entry. Federal hospital rules require records that are complete, dated, timed, authenticated, and protected from alteration ( 42 CFR 482.24 ), and colleges in all three countries expect corrections that preserve the audit trail: anything learned later goes in a clearly dated late entry or addendum. When a death, threat situation, or legal process touches your documentation, consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). On the payer side there is no suicide-specific code to defend: the assessment is reviewed as part of whatever service carried it, an intake, a psychiatric diagnostic evaluation , psychotherapy, or a crisis encounter , and no reviewed CMS or contractor policy names a required suicide instrument or publishes a suicide-note denial code. Exposure is generic: Medicare's 2024 outpatient psychiatric reporting put improper payments at 16.1 percent ($254.5 million), 78.3 percent of them for insufficient documentation ( CMS MLN ), so the risk work must be visible inside a note that also supports the billed service. The days after a level-of-care change deserve the most care: of 122 suicide deaths reported to Joint Commission as sentinel events in 2024, 75 percent occurred within seven days of discharge or during intensive behavioral health services ( sentinel event annual review ), which is why the dated follow-up contact, the escalation instructions, and the handoff you document are the parts most likely to be read again. Elevated risk should also flow into the treatment plan once the immediate response is documented. ↑ Back to the template and downloads ## Common suicide risk assessment errors reviewers flag The audit numbers for this document come from oversight bodies rather than payers. A 2024 VA Office of Inspector General national review found required suicide risk screening undocumented for 27 percent of sampled inpatients and safety plans missing for 12 percent of discharged patients ( VA OIG ). Joint Commission's 2024 sentinel event review identified 1,001 contributing factors across the 122 reported suicide deaths, with failure to follow screening, assessment, and reassessment policy among the leading categories. The BastionGPT Clinical Advisory Board sees the same errors most often in suicide risk assessment reviews: "Denies SI" as the entire assessment. A denial is one finding, not a formulation. It says nothing about past behavior, preparatory acts, means access, reliability, or why the clinician trusted it, and it is indefensible standing alone after a positive screen, a collateral warning, or a contradictory presentation. A positive screen with no documented follow-through. The screen sits in the chart with nothing after it. In covered settings a positive screen requires an evidence-based assessment, and everywhere else it is the first document a reviewer reads before asking what happened next. A risk level with no reasoning. "Low risk" without findings creates false precision. Joint Commission expects the record to show how the level was determined; the honest version pairs any label with an individualized formulation, because risk categories are weak predictors (pooled positive predictive value for suicide was 5.5 percent in a 2017 meta-analysis ). The note documents a reasonable process, never a prediction. Template text that contradicts the narrative. The history records active ideation while an auto-populated mental status line says "denies SI," or the narrative names a firearm while a checklist says "no access to means." Reviewers treat an internally contradictory record as unreliable everywhere, so reconcile the fields before signing. A no-suicide contract documented as protection. After an Ontario coroner review of nine hospital suicides, the hospital involved stopped using contracts for safety, citing "its lack of efficacy and effectiveness" ( CPSO Dialogue ), and a randomized trial favored crisis response planning over contracts. Document a collaborative safety plan and the assessment behind it instead. ### Suicide risk assessments in the US, Canada, and Australia Aspect United States Canada Australia Status No universal statute; Joint Commission defines the assessment domains in accredited settings (NPG.08.01.01 for hospitals since January 2026, NPSG.15.01.01 for behavioral health), while CMS requires safe-setting care and complete, authenticated records rather than named fields Accreditation Canada and HSO suicide-prevention requirements for participating organizations; college record standards enforceable under provincial law; provincial Mental Health Acts govern involuntary assessment separately NSQHS Comprehensive Care Standard Actions 5.31 and 5.32 require systems to identify and respond to self-harm and suicide risk; the Psychology Board Code of Conduct (December 2025) makes adequate records mandatory Terminology Suicide risk assessment, SRA, C-SSRS or Columbia protocol, SAFE-T Suicide risk assessment, risk formulation; Manitoba's 2026 provincial policy specifies field-level documentation, down to declined screens and the location of assessment "Risk of self-harm or suicide" in NSQHS wording; suicide risk assessment in practice What changes Which accreditor citation applies by setting, state training mandates (Washington's RCW 43.70.442 requires suicide assessment training, not a note format), and payer documentation exposure How specific the documentation expectations are, from general college record rules to Manitoba's itemized policy; involuntary-assessment forms are separate statutory documents State Mental Health Act forms for compulsory assessment; MBS adequate-and-contemporaneous record rules for billed services Retention HIPAA sets no clinical-record period; Medicare hospitals keep records at least 5 years, and state law often runs longer (California 7 years, New York 6) Ontario colleges require 10 years from last contact or from age 18, whichever is later 7 years from the last entry for adults; until the 25th birthday for clients seen as minors (Board code) The assessment itself travels well across all three countries: the same inquiry, formulation, and disposition reasoning is what every reviewer wants to reconstruct. What changes is who defines the fields, an accreditor, a college, a health system, or your own policy, and how long the record must survive. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on suicide risk assessments. Draft a structured risk assessment from your bullets or dictation, with the inquiry, risk and protective factors, formulation, interventions, and disposition in the right places. Check a finished note before you sign: flag a formulation with no rationale, an undocumented means conversation, a missing reassessment plan, or a disposition that nothing in the note supports. Summarize risk documentation across sessions for supervision, consultation, or a records request, and keep the summary to what the receiving reader should see. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a suicide risk assessment legally required, and is there a required format? No statute in the US, Canada, or Australia prescribes a suicide risk assessment template. The obligations are layered instead: Joint Commission standards define the assessment domains in accredited settings, licensing boards require adequate records everywhere, payers expect the record to support the billed service, and the clinical standard of care fills the rest. Most run 200 to 800 words, and the test any format is held to is whether the formulation supports the disposition. What changed with the Joint Commission suicide requirements in 2026? The citation moved. Joint Commission's National Performance Goals chapter replaced the National Patient Safety Goals chapter for hospitals and critical access hospitals on January 1, 2026, so the hospital suicide requirement is now NPG.08.01.01, while behavioral health organizations keep NPSG.15.01.01. The substance carried over: screen with a validated tool, assess positives with an evidence-based process, and document the overall risk level with the mitigation plan. Guidance citing NPSG.15.01.01 for hospitals is out of date. Do I have to use the C-SSRS or SAFE-T? No regulation reviewed for this page mandates a specific instrument in any of the three countries. Joint Commission encourages validated tools but accepts another evidence-based process when the organization can name its evidence base, the assessment directly covers ideation, plan, intent, behaviors, risk factors, and protective factors, and the record shows how the risk level was determined ( standards FAQ ). Your employer or health system can, and often does, require a named tool by policy; that requirement is the policy's, not the law's. How much should I document when a client denies suicidal ideation? Enough to show the inquiry, not just its bottom line: what you asked, what the client denied (passive thoughts, active thoughts, plan, intent), anything endorsed, and why the account seemed reliable. A routine denial in a stable client needs a sentence or two, not an essay. After a positive screen, a collateral warning, preparatory behavior, or a presentation that contradicts the denial, "denies SI" alone is the least defensible sentence a chart can carry. What is the difference between a method, a plan, and intent? A method is the general way a person has thought about dying. A plan adds specifics: place, timing, sequence, or preparation. Intent is the person's expectation or commitment to act, and preparatory behavior is action already taken. Document each separately, in the client's own words, because they carry different weight in the formulation. When the facts are ambiguous, record the ambiguity rather than forcing a label. Is a safety plan the same as a suicide risk assessment? No. The risk assessment is the clinician's record of the evaluation and reasoning; a safety plan is a client-facing intervention, the prioritized list of warning signs, coping steps, contacts, and means-safety actions a client uses as risk rises. One documents, the other equips, and the outcome evidence attaches to the intervention with follow-up ( Stanley and Brown, JAMA Psychiatry 2018 ). Document the safety plan as completed with a copy given, and keep the reasoning in the assessment. A crisis note is different again: it records an urgent encounter, which may contain a risk assessment. How accurate are risk levels like low, moderate, and high? Weak as predictions, which is exactly why the reasoning matters more than the label. A 2017 meta-analysis found a pooled positive predictive value of 5.5 percent for suicide among people classified high risk, and the 2024 VA/DoD guideline found insufficient evidence to prefer any stratification method. Accredited settings still require a documented overall level, so give the category your setting expects, pair it with the individualized formulation, and never write prediction language like "client will not act." What should I document after a client dies by suicide? Nothing in the original note: preserve the record and its audit trail exactly as they stood. Anything learned or done afterward belongs in a clearly dated late entry, addendum, or your organization's incident process, limited to facts such as notifications received and consultations held. Contact your risk management function, liability carrier, or attorney before writing more, and follow their guidance on what belongs in the chart versus a protected review process. Rewriting history is the one documentation act that makes everything else indefensible. Can BastionGPT draft a suicide risk assessment? Yes. Give it bullets or a dictation from the session, and it drafts a structured assessment with the inquiry, risk and protective factors, formulation, interventions, and disposition in the right places for your review. It can also check a finished note for a formulation with no rationale, an undocumented means conversation, or a missing reassessment plan before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: Joint Commission, National Performance Goals : the NPG chapter replacing the NPSG chapter for hospitals and critical access hospitals, effective January 1, 2026. Joint Commission, NPG #8: Reducing the Risk for Suicide : the hospital requirements, including documentation of the overall risk level and mitigation plan. Joint Commission, R3 Report Issue 18 (updated December 2025) : NPSG.15.01.01 elements of performance for behavioral health organizations. Joint Commission, standards FAQ on evidence-based assessment (updated April 2026) : no named tool required; required domains; documentation of how the risk level was determined. Joint Commission, Sentinel Event Data 2024 Annual Review : 122 reported suicide deaths, 75 percent within seven days of discharge or during intensive behavioral health services. VA Office of Inspector General, national review of inpatient mental health suicide risk practices (2024) : 27 percent missing required screening documentation, 12 percent missing safety plans at discharge. CMS Medicare Learning Network, outpatient psychiatric care compliance tips : 2024 improper-payment data, including the 78.3 percent insufficient-documentation share. 42 CFR 482.24 and 42 CFR 482.13 : hospital record integrity, retention, and the safe-setting right. HHS, HIPAA FAQ 580 : HIPAA sets no medical-record retention period. VA/DoD, Clinical Practice Guideline for Assessment and Management of Patients at Risk for Suicide (2024) : risk formulation guidance and the stratification-evidence finding. SAMHSA, SAFE-T (updated December 2024) and the Columbia Lighthouse Project, about the C-SSRS : framework and instrument lineage. Carter et al., British Journal of Psychiatry meta-analysis (2017) : positive predictive values of risk classification. Stanley and Brown et al., JAMA Psychiatry (2018) and Bryan et al., Journal of Affective Disorders (2017) : safety planning and crisis response planning evidence. CPSO Dialogue, coroner recommendations following hospital suicides (2023) : the Ontario coroner review and the end of contracts for safety. Australian Commission on Safety and Quality in Health Care, Comprehensive Care Standard (Actions 5.31 and 5.32), the Psychology Board of Australia Code of Conduct , and Manitoba Shared Health, provincial suicide prevention policy 600.105.118 (March 2026) : the accreditor, regulator, and health-system layers outside the US. ## Related templates Crisis Psychotherapy Note Trauma Assessment Intake Note Psychiatric Diagnostic Evaluation Treatment Plan Safety Plan Mandated Reporting & Duty-to-Protect Documentation Violence & Homicide Risk Assessment Self-Harm (NSSI) Assessment Incident & Adverse Event Report General Risk Screening Note PHQ-9 Score Documentation C-SSRS Documentation EPDS Score Documentation DASS-21 Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation MADRS Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/neuropsychological-report All note templates # Neuropsychological Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A neuropsychological evaluation report is the written product of neuropsychological testing: it integrates the referral question, history, behavioral observations, and standardized cognitive test scores into a domain-by-domain interpretation, a diagnosis, and functional recommendations. Neuropsychologists write one after evaluating conditions such as brain injury, dementia, ADHD, or epilepsy, addressed to the referring physician and increasingly read by the patient. Most clinical reports run 5 to 20 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinical neuropsychologists; postdoctoral fellows and psychometrists contribute under supervision Audience Referring neurologists and physicians, rehabilitation teams, the patient and family, schools, attorneys, insurers Typical length 2,500 to 8,000 words (5 to 20 pages) · 4 to 10 hours of scoring, interpretation, and writing (clinical team estimate) Format family Interpretive assessment report (compare: psychological and psychoeducational reports) When it's used After suspected or known brain injury, neurological disease, or cognitive change, when diagnosis or planning needs standardized measurement across cognitive domains Standards context No law prescribes a report format; AACN and NAN guidance plus Medicare contractor documentation lists shape what reviewers expect On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a neuropsychological evaluation report? A neuropsychological evaluation report is the written product of a neuropsychological assessment: it integrates the referral question, history, behavioral observations, and standardized cognitive test scores into a domain-by-domain interpretation, a diagnosis, and functional recommendations. Clinicians and payers also call it a neuropsychological assessment report, a neuropsych eval, or an NPE. No single body created the format. The specialty defined itself in stages: APA’s Division 40 published the first formal definition of a clinical neuropsychologist in 1989, the Houston Conference set the specialty’s training model in 1997, and the National Academy of Neuropsychology’s 2001 definition expanded it. The AACN’s 2007 practice guidelines remain the profession’s formal assessment guidance, and none of these documents prescribes a report template. Two boundaries define the document. First, it is the required evidence of a distinct, time-based evaluation service: since the 2019 restructure of the US testing codes, neuropsychological evaluation services (96132 and 96133) cover integrating data, interpreting results, clinical decision making, treatment planning, writing the report, and giving feedback, separate from test administration time. Second, it is not a HIPAA psychotherapy note : the report sits in the designated record set, so patients can request it, and information-blocking rules mean it often reaches the portal as soon as it is finalized. When the question is emotional, personality, or diagnostic without a cognitive-domain focus, its sibling document is the psychological evaluation report . ## Who uses neuropsychological evaluation reports and when Clinical neuropsychologists write them in academic medical centers, memory clinics, epilepsy and neurosurgical programs, rehabilitation hospitals, VA and military settings, and private practice; pediatric neuropsychologists write them in children’s hospitals and developmental clinics. The report is the deliverable the referrer is waiting on: a neurologist separating mild cognitive impairment from normal aging or depression, a rehabilitation team planning return to work after a brain injury, an epilepsy program mapping strengths before surgery, a psychiatrist untangling adult attention complaints from anxiety. It is the right tool when the question needs standardized, domain-by-domain measurement of cognition. When an interview can answer the question, the psychiatric diagnostic evaluation usually does it without a battery; when the question is personality or emotional functioning, the psychological evaluation report takes over; and a brief clinician-administered neurobehavioral status exam often precedes full testing to decide whether and what to test. ## Neuropsychological evaluation report structure: what goes in each section No authority mandates a single report format. The sections below are the consensus skeleton of the neuropsychological assessment tradition and the elements Medicare contractor policy lists for testing services, each with the pitfall that most often undermines it. Identifying information and referral question. Who the patient is, who referred them, and the specific question the evaluation should answer, stated in a sentence or two. Pitfall: restating a vague "memory problems, please evaluate" referral instead of sharpening it; when the question is not specific, the interpretation has nothing to aim at and the report reads as testing for its own sake. Notification and consent. What the patient was told about the purpose, the fees, the limits of confidentiality, and who will receive the report. Pitfall: skipping who-gets-the-report; access disputes start where the record is silent about the intended audience. Sources of information and tests administered. Every instrument by full name, the interview, the records and collateral informants used, who administered each test (evaluator or psychometrist), and the cumulative time for administration, scoring, and interpretation. Pitfall: no time documentation; the US testing codes are time based and cumulate across the whole episode of care, and a report that never states the hours cannot support the units billed. Relevant history. The medical and neurological, developmental, psychiatric, educational and occupational, substance, and family history that bears on the referral question. Pitfall: no premorbid anchor; without best-ever functioning (grades, degrees, job complexity), "below average" cannot be read as decline. Behavioral observations and performance validity. How the person presented and engaged, and an explicit statement on whether performance validity indicators support interpretation. Pitfall: no validity statement; in a discipline built on effort-dependent measures, a reviewer who cannot tell whether the data are valid cannot credit anything built on them. Test results by cognitive domain. Findings organized by domain rather than instrument: premorbid estimate, attention and processing speed, language, visuospatial, learning and memory, executive function, motor, and mood screening, with scores in context and integrated in prose. Pitfall: an instrument-by-instrument score dump; payers cover the evaluation service, and tables without integration document administration only. Summary and clinical impressions. The synthesis paragraphs that pull history, observations, and testing into one clinical picture, answer the referral question directly, and state the most likely etiology. Pitfall: the hedge-everything summary that answers nothing; the referrer needed a position, with its confidence stated honestly. Diagnostic impression. The DSM-5-TR diagnosis with its ICD-10-CM code (for example G31.84 for mild neurocognitive disorder), the conditions considered and ruled out, or a documented statement that no diagnosis is supported plus the suspected diagnosis that justified testing. Pitfall: omitting the basis-for-testing line when results come back normal; contractor policy expects the suspected diagnosis even when none is found. Functional implications and recommendations. What the findings mean for daily life: work, school, driving, medication management, finances, and independence, with numbered recommendations tied to findings, referrals, and a re-evaluation trigger. Pitfall: cognitive findings with no functional translation; the neurologist, the family, and the insurer all read this section first, and a domain profile without daily-life meaning wastes the evaluation. Signature and credentials. The evaluating neuropsychologist’s signature, credentials, and license, noting who administered testing, with the supervisor’s counter-signature where a trainee performed the work. Pitfall: technician-administered testing with no supervising psychologist’s signature anywhere in the record. ### Blank template (copy and adapt) Patient: [initials] DOB/Age: Dates of service: Report date: Evaluator [name / credentials / license]: Psychometrist (if any): Referral source & referral question: Notification & consent (purpose / fees / confidentiality limits / who receives the report): Sources of information (interview / records / collateral): Tests administered (full names; who administered each; administration, scoring & interpretation time): Relevant history: Medical / neurological / psychiatric: Developmental / educational / occupational (premorbid baseline): Substance / family: Behavioral observations & performance validity statement: Results by cognitive domain (in prose, anchored to the premorbid estimate): Premorbid estimate: Attention / processing speed: Language: Visuospatial: Learning & memory: Executive function: Motor: Mood / psychological screening: Summary & clinical impressions (answer the referral question; state the most likely etiology): Diagnostic impression (DSM-5-TR / ICD-10-CM + rule-outs, or "no diagnosis supported" + basis for testing): Functional implications & recommendations (numbered, tied to findings; work / school / driving / medications / independence; re-evaluation trigger): Evaluator signature / credentials: Date signed: Supervisor signature (if required): Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample neuropsychological evaluation report Scenario: a neurologist refers a 68-year-old for progressive memory complaints; the differential is mild cognitive impairment, normal aging, or a depressive contribution. A clinical neuropsychologist completes the evaluation with psychometrist support. The sample is condensed but structurally complete; a full report runs 5 to 20 pages. All details are fictional. Patient: E.M., 68 · Dates of service: 07/08/2026, 07/10/2026 · Report date: 07/16/2026 · Evaluator: L. Whitfield, PhD, ABPP-CN · Referred by: A. Nassar, MD (neurology) Referral question: Dr. Nassar asks whether E.M.'s reported memory decline reflects mild cognitive impairment, normal aging, or a depressive contribution, and what supports she needs day to day. Notification and consent: Purpose of the evaluation, fees, confidentiality limits, and recipients of the report (Dr. Nassar and E.M.) were reviewed; written consent signed 07/08/2026. A feedback session is scheduled. Sources of information and tests administered: Clinical interview with E.M. (60 minutes, 07/08/2026); collateral interview with her spouse (20 minutes, with consent); neurology consult note, recent laboratory results, and brain MRI report (mild age-consistent atrophy, no focal lesion). Tests: Test of Premorbid Functioning; WAIS-IV Digit Span and Coding; WMS-IV Logical Memory I and II; California Verbal Learning Test, Third Edition; Trail Making Test A and B; Boston Naming Test; letter and category fluency; Rey Complex Figure copy; Grooved Pegboard; Geriatric Depression Scale. Standalone and embedded performance validity indicators were included. Test administration and scoring by J. Rivera, psychometrist, under Dr. Whitfield's supervision, 3.5 hours; evaluator interview, integration, interpretation, and report preparation, 3.0 hours. Cumulative time documented for billing. Relevant history: Retired high-school science teacher (bachelor's degree, 34 years teaching). Two years of slowly increasing word-finding pauses and misplaced items; over the past eight months, repeated questions and two missed appointments noted by her spouse. She manages finances and medications independently and continues to drive, cook, and keep her routine, with growing reliance on written lists. Hypertension controlled on lisinopril; no stroke, head injury, or seizure history. No psychiatric treatment history; sleep adequate; alcohol about one glass of wine weekly. Her mother developed dementia in her eighties. Behavioral observations and performance validity: Arrived on time with her spouse, appropriately groomed, cooperative, with good persistence across both sessions; occasional word-finding pauses were evident in conversation. Standalone and embedded performance validity indicators were within normal limits. The results are considered a valid estimate of current functioning. Results by cognitive domain: Premorbid ability is estimated as high average, consistent with education and occupation. Attention and processing speed are average. Language shows mild low-average inefficiency on confrontation naming with intact comprehension; verbal fluency is average. Visuospatial construction is average. Learning and memory are the clearly affected domain: story recall after a delay fell at the 5th percentile, and list learning showed a flattened learning curve with delayed recall 1.5 to 2 standard deviations below age expectations and limited benefit from recognition cues, a storage pattern rather than retrieval inefficiency. Executive function is broadly average, though mildly below the premorbid estimate on set-shifting. Motor speed and dexterity are average bilaterally. On mood screening, the Geriatric Depression Scale score was 3 of 15, below the clinical cutoff, and E.M. denied persistent low mood or loss of interest. Summary and clinical impressions: Against a high-average baseline, E.M. shows an isolated amnestic pattern that exceeds normal aging: delayed recall is impaired with limited recognition benefit while other domains remain broadly preserved. The depression screen and interview do not support a depressive explanation. Daily instrumental activities remain independent by her account and her spouse's, so the picture does not meet criteria for dementia. The profile is most consistent with amnestic mild cognitive impairment. The storage-pattern memory profile and family history raise the possibility of an early neurodegenerative process; etiologic confirmation requires the medical workup and reassessment recommended below. This answers Dr. Nassar's question: mild cognitive impairment, not normal aging and not depression. Diagnostic impression: Mild neurocognitive disorder, amnestic presentation (G31.84). Major neurocognitive disorder not met: instrumental activities remain independent. Depressive disorder considered and not supported. Functional implications and recommendations: (1) Consolidate reminders into one calendar and list system, with the spouse-supported routine reviewed at feedback. (2) Medication management is currently accurate; use a weekly pillbox with a brief monthly check by her spouse. (3) Finances are currently managed well; simplify recurring bills to autopay and review annually. (4) No current indication to restrict driving; revisit if navigation lapses or new errors emerge. (5) Return to Dr. Nassar for the planned medical workup and monitoring discussion. (6) Repeat neuropsychological evaluation in 12 months, or sooner if functional changes appear (missed bills, medication errors, getting lost). (7) Feedback session 07/20/2026 with E.M. and her spouse to review these findings in plain language. Evaluator: L. Whitfield, PhD, ABPP-CN, Licensed Psychologist, License #3-7741 · Signed: 07/16/2026 This sample is fictional and for educational purposes. It does not describe a real patient, and the instruments named are examples of commonly used measures. ↑ Back to the template and downloads ### Why this sample works The referral question gets a direct answer in one findable place: the summary states what the memory complaint is and is not, with the confidence stated honestly. Performance validity is addressed before any interpretation, so every conclusion that follows has a stated foundation. Results are organized by cognitive domain and anchored to a premorbid estimate, in prose a referring physician can act on rather than a table dump. Tests, examiners (including the psychometrist), and cumulative time are documented, which is what the time-based US testing codes require the record to support. Recommendations are numbered, functional, and specific (medications, finances, driving, follow-up interval), and the report names what should trigger re-evaluation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The report lives in the designated record set, which shapes who sees it. Patients can request it under 45 CFR 164.524, and the federal information-blocking rules treat delay as suspect: the preventing-harm exception requires an individualized clinical determination, not a blanket hold, and neither rule contains a categorical test-security exception. In practice the report often reaches the patient portal the moment it is finalized, sometimes before the feedback session, so plan that conversation with the timing in mind. Raw scores and test forms follow different rules than the report itself: the APA ethics code treats test data as releasable to the patient with authorization while requiring psychologists to protect test materials such as stimuli and manuals. Canada writes that split into law from two directions: Ontario excludes raw standardized-test data from the provincial access right entirely, and British Columbia’s 2026 records standard expressly allows access limits where disclosure “would jeopardize the security and integrity of test materials, test data, or scoring keys.” Retention is jurisdictional: HIPAA sets no period, California writes seven years into statute, the APA record-keeping guidelines suggest seven years for adults as an aspirational standard, Medicare enrollment rules expect seven, and provincial rules run from five to sixteen years. The payer chain runs on medical necessity and time. Contractor policy covers testing that helps answer a diagnostic question an interview cannot, lists the elements it expects in the record (the tests performed, scoring and interpretation, and the time involved), and treats each test in the battery as something to justify individually; routine screening and identical batteries for every patient are what reviewers flag. The evaluation codes cumulate across the whole episode of care, the first-hour code is billed once per episode, and CMS confirms the professional time includes interpretation, integration, clinical decision making, and report preparation, reported when the episode is complete. Hour expectations are payer policy, not clinical rules: the Novitas billing article calls four to six hours typical and directs multi-day testing to be billed on the last date of service, a Texas state program caps testing at four hours per day and eight per calendar year absent documented necessity, and the APA’s 2026 testing guide advises documenting medical necessity past eight hours and expects roughly five hours of administration and scoring for every three hours of evaluation service unless the case justifies more. Psychological (96130, 96131) and neuropsychological (96132, 96133) evaluation services are never billed in the same episode. Supervision follows the current regulation, not the older FAQ: 42 CFR 410.32 now lets nurse practitioners, clinical nurse specialists, and physician assistants provide general supervision within state scope of practice alongside physicians and clinical psychologists, there is no incident-to route for diagnostic testing, technician administration is billable under general supervision, and unsupervised computer-administered tests are not billable administration time. ## Common neuropsychological evaluation report errors auditors flag Medicare’s most recent audit cycle measured a 6.55 percent improper-payment rate across all fee-for-service claims, about $28.83 billion, and no neuropsychology-specific denial rate has ever been published. Reviewers work from the contractor documentation lists instead, and starting with 2026 audit reports, records not produced within 60 days of a request count as errors on their own. The BastionGPT Clinical Advisory Board sees the same errors most often in neuropsychological evaluation report reviews: Scores without interpretation. Tables and percentiles with no integrative prose. Payers cover the evaluation service, and interpretation is the service; a score dump documents administration only. Missing time documentation. The testing codes are time based and cumulate across the whole episode of care; hours billed with no recorded administration, scoring, and interpretation time cannot survive review. No performance-validity statement. Nothing about effort or interpretability, which leaves every conclusion floating. No rule prescribes which validity measures to use; the statement that results are interpretable is what reviewers, referrers, and opposing experts look for first. No premorbid anchor. The report claims decline without establishing best-ever functioning from education, occupation, and history, so “below average” cannot be read as change. A battery no one justified. The same panel for every patient regardless of the question; contractor policy expects each test to be individually necessary, and a same-battery-every-time pattern is a flag. ### Neuropsychological evaluation reports in the US, Canada, and Australia Aspect United States Canada Australia Status No mandated form; CPT defines the timed testing services (2019 family), contractor articles list expected documentation, and access and information-blocking rules govern release No national rule; enforceable provincial college standards govern records and reports (Ontario 2024, BC 2026) First mandatory profession code effective 1 December 2025; no general MBS item funds a standalone adult neuropsychological assessment (one narrow under-25 neurodevelopmental group exists) Terminology Neuropsychological evaluation, neuropsych eval, NPE Neuropsychological assessment (report) Neuropsychological assessment, cognitive assessment What changes Which MAC’s article applies, and payer prior-authorization hour caps Province and college: Ontario requires the supervisor’s counter-signature for non-autonomous authors; BC allows test-security limits on access Funding stream (private, insurer, NDIS, state clinics) shapes audience and format; third-party assessments carry role-clarification duties Retention State law governs (California: seven years by statute); Medicare enrollment rules expect seven Ontario 10 years past age 18 or last contact; BC 16 years (2026); Quebec five years Seven years from the last entry; for minors, to the 25th birthday The domain-by-domain structure travels well across all three countries. What changes is the administrative layer around it, so adapt billing language, access and test-security handling, and retention practices to your jurisdiction. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on neuropsychological evaluation reports. Draft the full report from your test scores, history bullets, and behavioral observations, with every domain section in place for your review. Integrate scores from a full battery into domain-level interpretation that stays anchored to the premorbid estimate and the referral question. Translate findings into plain language for patients and families, and check a finished draft for the gaps reviewers flag: scores without interpretation, missing time documentation, or recommendations that never answer the functional question. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a legally required format for a neuropsychological report? No law in the US, Canada, or Australia prescribes a report format. What exists is payer policy and professional convention: Medicare contractor policy lists the elements it expects in the record, including the tests performed, scoring and interpretation, and the time involved, and profession-endorsed guidelines shape the conventional structure. The format is a convention; the content is the requirement. How long should a neuropsychological report be? No regulation specifies a length. In the 2018 stakeholders’ survey of neuropsychological report writing , referral sources valued the diagnosis and recommendations sections most, and the Inter Organizational Practice Committee summarizes the typical adult report at about 6 pages and 2 to 3 hours of writing, against about 11 pages and 5 to 10 hours for pediatric reports. The consistent pressure from referrers is toward shorter reports with findable answers. Length beyond what the referral question needs adds delay, not defensibility. How soon after testing should the report be finished? No US or Canadian rule sets a deadline, and Australia’s national code asks for reports to be signed within "a reasonable and justifiable timeframe." The clinical pressure is real: in the 2018 stakeholders’ survey, 73 percent of referral sources said slow turnaround had negatively affected patient care. Two operational clocks matter too: the billing episode for the evaluation services closes when the report is complete and feedback is given, and information-blocking rules push the signed report to the patient portal quickly, often before the feedback session unless you sequence it deliberately. Do I have to release the report and the raw test data to the patient? The report: yes. It is part of the designated record set, and 45 CFR 164.524 gives patients a right of access that information-blocking rules reinforce; neither contains a categorical test-security exception. Raw data and test materials follow different rules: the APA ethics code treats test data as releasable to the patient with authorization while requiring psychologists to protect test materials such as stimuli, manuals, and protocols. Canada diverges by statute: Ontario excludes raw standardized-test data from the provincial access right entirely, and British Columbia’s 2026 standard allows access limits where release would jeopardize test security. Know which regime you practice under before the request arrives. Can I bill for the time spent writing the report? Yes. CMS confirms the professional evaluation codes cover test interpretation, integration of data, clinical decision making, and report preparation, and the neuropsychological evaluation services (96132 and 96133) are billed on cumulative time across the whole episode of care, with the first-hour code once per episode and the total reported when the episode is complete. The APA’s 2026 testing guide adds a working threshold reviewers recognize: document medical necessity when a single evaluation runs past eight hours. Can a psychometrist administer the testing? Yes. Test administration and scoring by a trained technician under general supervision is billable with the technician codes (96138 and 96139), and psychologist-administered and technician-administered time can be reported together in the same evaluation. Three cautions. Diagnostic testing has no incident-to billing route, and Medicare has never defined "technician," so document who administered each test and under whose supervision. Medicare also distinguishes technicians from trainees: student or trainee services are not payable just because a supervisor signs. And California now registers psychological testing technicians (since January 2024) who may administer and score under direct supervision but may not select tests, interpret, write results, or give feedback. How is this different from a psychological evaluation report? The neuropsychological report answers brain-behavior questions with domain-by-domain cognitive measurement: memory, attention, language, executive function, and their functional implications. The psychological evaluation report answers emotional, personality, and diagnostic questions with instruments built for those constructs. The code families differ too, and the two evaluation services are never billed in the same episode of care; the predominant service decides which family applies. How is this different from a neurobehavioral status exam? The neurobehavioral status exam (96116 and 96121) is a briefer, clinician-administered assessment of cognitive status. It often precedes a full evaluation and guides which instruments to select, and the APA’s testing guide is direct about its limit: a neurobehavioral status exam alone is insufficient to assess mild cognitive impairment, which requires additional testing. Think of it as the scout, and the full neuropsychological evaluation with its report as the map. Can BastionGPT draft a neuropsychological evaluation report? Yes. Give it your test scores, history bullets, and behavioral observations, and it drafts the full report with every domain section in place for your review: premorbid anchoring, validity statement, domain-by-domain interpretation, and functional recommendations. It can also translate a finished report into plain language for the patient and family. Never paste client-identifying information into consumer AI tools; BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Psychiatric Diagnostic Evaluation Case Formulation Note Trauma Assessment Treatment Plan BASC-3 Report Write-Up WISC-V Report Write-Up WAIS-5 / WAIS-IV Report Write-Up Conners 4 Report Write-Up WIAT-4 Report Write-Up ADOS-2 Report Write-Up Vineland-3 Report Write-Up BRIEF-2 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up MMPI-3 Report Write-Up Psychoeducational Evaluation Report PAI Report Write-Up ABAS-3 Report Write-Up NEPSY-II Report Write-Up D-KEFS Report Write-Up RBANS Report Write-Up Psychological Testing Administration Note Psychological Testing Scoring Note Fitness-for-Duty Evaluation Capacity & Competency Evaluation Forensic Psychological Evaluation Report MoCA Score Documentation MMSE Score Documentation Mini-Cog Score Documentation SLUMS Score Documentation Stroop Test Score Documentation Trail Making Test Score Documentation Clock Drawing Test Score Documentation Boston Naming Test Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychoeducational-report All note templates # Psychoeducational Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychoeducational evaluation report is the written product of an assessment that integrates cognitive ability, academic achievement, and behavioral data to answer a question about learning. School psychologists write them to inform special-education eligibility and instruction planning; private evaluators write them as independent educational evaluations. Most run 4 to 15 pages, and parents receive a copy of a school evaluation report at no cost. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it School psychologists; licensed psychologists conducting independent educational evaluations; psychology trainees under supervision Audience Parents and the student, the school or IEP team, teachers, outside clinicians, colleges and testing agencies Typical length 2,000 to 6,000 words (4 to 15 pages) · 5 to 6 hours of testing, scoring, and writing (clinical team estimate) Format family Interpretive assessment report (compare: psychological and neuropsychological reports) When it's used When a learning question needs standardized cognitive and achievement measurement: suspected learning disability, eligibility, accommodations, or instruction planning Standards context No law mandates the format; IDEA and state education codes fix evaluation content and timelines, and assessment textbooks set the conventional structure On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychoeducational evaluation report? A psychoeducational evaluation report is the written product of an assessment built around a learning question: it integrates cognitive ability testing, academic achievement testing, and usually behavioral, social-emotional, and processing data with records, interviews, and observation to explain how a student learns and what should change. Clinicians and schools also call it a psychoed eval, a comprehensive psychoeducational assessment (the usual Canadian term), an LD evaluation, or, when a parent hires an outside evaluator, an independent educational evaluation (IEE) report. The document descends from the school-testing tradition that began with the Binet-Simon scale in 1905 and Terman's Stanford-Binet in 1916, and its modern legal engine is the special-education law enacted in 1975 that became the Individuals with Disabilities Education Act (IDEA). The familiar section order, referral question, background, observations, tests administered, results by domain, summary, recommendations, was standardized by assessment textbooks, Sattler's Assessment of Children and Schneider, Lichtenberger, Mather, and Kaufman's Essentials of Assessment Report Writing , not by any statute or agency. The sharpest boundary to understand is between the psychologist's report and the eligibility paperwork it feeds. The only US federal regulation that specifies written contents, 34 CFR 300.311 , governs the specific-learning-disability eligibility documentation produced by the team, not the psychologist's report, and its closest thing to a signature rule is that each group member must "certify in writing whether the report reflects the member's conclusion." IDEA itself never distinguishes a psychoeducational from a psychological evaluation; the statute speaks only of evaluation, so the label is a professional and payer distinction, not a legal one. In practice the psychological evaluation report is the broader diagnostic sibling written when the referral is about mental health rather than learning, and the neuropsychological evaluation report takes over when the question is brain-based. The psychoeducational report is the one written on the school's legal clock, and the parent is entitled to a copy of it at no cost. ## Who uses psychoeducational evaluation reports and when School psychologists write most of them: assessment sits above every other professional responsibility in the role, and the field's own membership survey found practitioners completing an average of 55 evaluations per year . The demand behind that number is structural: 7.5 million US students, 15 percent of public-school enrollment, receive services under IDEA, and the largest eligibility category, specific learning disabilities at 32 percent , is the one a psychoeducational battery exists to examine. Licensed psychologists in private practice write the same report as an independent educational evaluation when parents want a second opinion, sometimes at public expense, and clinical psychologists write it when a learning question surfaces inside a broader referral. The report is the right tool when the question is about learning: suspected learning disability, eligibility, accommodations, or instructional planning. When the referral is diagnostic and emotional-behavioral rather than academic, the psychological evaluation report fits better; when the question is brain-based, after injury, seizures, or a medical event, it becomes a neuropsychological evaluation report . The typical battery pairs a cognitive measure such as the WISC-V with an achievement measure such as the WIAT-4 or the Woodcock-Johnson , plus rating scales such as the BASC-3 , and the report is where those instruments stop being separate scores and become one explanation. ## Psychoeducational evaluation report structure: what goes in each section No authority mandates a single format. The skeleton below is textbook convention; what the law fixes is content: the evaluation must use a variety of technically sound measures, cover every area related to the suspected disability, and never rest on a single measure as the sole criterion ( 34 CFR 300.304 ). Some states go further and enumerate report contents in statute, California's Education Code section 56327 lists eight required elements, so check your state code before adapting the skeleton. Identifying information and referral question. Student, age, grade, school, dates of testing and report, who referred, and the specific learning question in one or two sentences. Pitfall: a generic "academic concerns" referral; when the question is vague the recommendations have nothing to answer, and reviewers read the question first. Background and educational history. Developmental, medical, family, and language history; instruction received; attendance; grades and state testing; prior evaluations; and what interventions were tried with what response, drawn from records, parent input, and teacher reports. Pitfall: omitting the intervention-response data. The eligibility team must rule out lack of appropriate instruction as the determinant factor, and a report silent on instruction and intervention response leaves the team's central question unanswered. Observation. The student in the learning environment, in the area of difficulty, plus how they presented and engaged during testing, with an explicit statement on effort and validity. Federal rule requires an observation in the classroom setting for SLD determinations ( 34 CFR 300.310 ), and pre-referral observations can count. Pitfall: no validity statement; a reviewer who cannot tell whether the scores are interpretable cannot credit anything built on them. Tests administered. Every instrument by full name and edition, with dates. Pitfall: undated instruments and aging norms; the regulation's standard is technically sound instruments, and an outdated edition invites the challenge. Cognitive results. Index-level scores with confidence intervals and percentiles, interpreted in prose. Pitfall: leading with a single IQ number as the gatekeeper. No federal regulation requires an IQ score, and the single-measure bar is explicit; index-level scatter is often the finding. Academic achievement results. Standard scores by domain tied back to the referral question, with the instructional meaning stated. Pitfall: scores without instructional meaning; a 76 in Calculation means little to a team until the report says what the student can and cannot yet do. Social-emotional and behavior results. Rating scales from more than one informant and setting, integrated with observation and history. Pitfall: single-informant conclusions; cross-informant agreement is modest, which is why the multi-source rule exists. Summary and integration. The paragraphs that pull cognition, achievement, behavior, instruction, and exclusionary factors into one explanation and answer the referral question directly. Pitfall: writing an eligibility verdict. Eligibility is the team's determination; the report informs it, and federal rule forbids a state from requiring the old severe-discrepancy test, so frame the pattern, not a formula. Recommendations. Numbered, specific, and traceable to findings: instruction and intervention, accommodations, progress-monitoring metrics, and what should trigger re-evaluation. Pitfall: boilerplate recommendations that could follow any battery; parents and hearing officers both notice when every report ends the same way. Signature and credentials. The evaluator's name, credentials, and license or certification, with the supervisor's counter-signature where required. Pitfall: assuming a signature rule exists federally; it is your state, board, or college that sets one, and supervised trainees' reports need the co-signature their jurisdiction requires. ### Blank template (copy and adapt) Student: [initials] Age/Grade: School: Dates of testing: Report date: Evaluator [name / credentials]: Referral source & referral question: Background & educational history: Developmental / medical / language: Instruction received & attendance: Interventions tried & response (data): Prior evaluations & records reviewed: Observation (classroom + testing session) & validity statement: Tests administered (full names, editions, dates): Cognitive results (indexes, CIs, percentiles, in prose): Academic achievement results (by domain, instructional meaning): Social-emotional / behavior results (multi-informant): Exclusionary factors considered (instruction, attendance, vision/hearing, language difference, other): Summary & integration (answer the referral question): Recommendations (numbered, specific, tied to findings): Instruction / intervention: Accommodations: Progress monitoring & re-evaluation trigger: Note: eligibility is determined by the team, not this report. Evaluator signature / credentials: Date signed: Supervisor signature (if required): Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychoeducational evaluation report Scenario: parents arrange a private psychoeducational evaluation over the summer for a rising seventh grader with persistent mathematics difficulty, after two years of school intervention with limited growth. A licensed psychologist completes the records review, testing, and report. The sample is condensed but structurally complete; a full report runs 4 to 15 pages. All details are fictional. Student: D.M., age 12, entering grade 7 · Dates of testing: 07/07/2026, 07/09/2026 · Report date: 07/16/2026 · Evaluator: L. Marsh, PhD, Licensed Psychologist · Requested by: Parents (independent evaluation) Referral question: D.M.'s parents ask why mathematics remains far harder for him than his other subjects despite two years of school support, whether the pattern reflects a specific learning disability in mathematics, and what instruction and accommodations would help as he enters middle school. Background and educational history: Developmental and medical history are unremarkable; vision and hearing were screened at school in January 2026 and passed. English is the home language. Attendance is strong (three absences in grade 6). Report cards show grade-level performance in English language arts, science, and social studies against persistent below-level marks in mathematics from grade 3 onward. D.M. received core instruction in the district's standard mathematics curriculum and, per district records, two intervention cycles: a grade 5 small-group cycle and a 16-week grade 6 Tier 2 program (30 minutes, twice weekly) targeting computation. Progress monitoring across the grade 6 cycle shows a gain of 2 digits correct on two-minute computation probes against an expected gain of roughly 8, a limited response to intervention delivered as designed. Prior evaluations: none. Records reviewed: report cards grades 3 to 6, state assessment results, intervention logs and probe data, January 2026 hearing and vision screening, and the district's March 2026 classroom observation summary. Observation and validity: Direct classroom observation was not possible in July; this report incorporates the district's March 2026 structured observation during the mathematics block, which recorded on-task behavior comparable to peers, frequent finger counting, and reliance on the teacher's worked example before starting problems, and notes the summer timing as a limitation. Across two testing sessions D.M. was cooperative and conversational, worked steadily with two short breaks, and persisted on difficult items; he counted on his fingers during mental arithmetic and erased frequently on paper-and-pencil calculation. Results are considered valid estimates of current functioning. Tests administered: Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V); Woodcock-Johnson IV Tests of Achievement, Form A (WJ IV ACH); Behavior Assessment System for Children, Third Edition (BASC-3), Parent Rating Scales completed 07/07/2026 and Teacher Rating Scales completed by his grade 6 mathematics teacher in the final week of the school year; parent interview; student interview. Cognitive results: WISC-V Full Scale IQ 99 (95% CI 94-104, 47th percentile, Average). Verbal Comprehension 103 (58th, Average), Visual Spatial 100 (50th, Average), Fluid Reasoning 92 (30th, Average), Working Memory 85 (16th, Low Average), Processing Speed 89 (23rd, Low Average). General ability is intact; holding and manipulating information in mind and clerical speed are relative weaknesses, a profile that commonly makes fact retrieval and multi-step computation effortful. Academic achievement results: WJ IV ACH reading and writing clusters fall in the Average range: Letter-Word Identification 98 (45th), Passage Comprehension 95 (37th), Spelling 97 (42nd), Writing Samples 94 (34th). Mathematics stands apart: Calculation 76 (5th percentile, Low), Math Facts Fluency 73 (4th percentile, Low), and Applied Problems 83 (13th percentile, Low Average). D.M. solves single-digit addition by counting, has not yet automatized multiplication facts, and loses steps in multi-digit procedures involving regrouping; word problems fare somewhat better when reading carries part of the load. The gap between intact reading and low mathematics is the profile the referral question predicted. Social-emotional and behavior results: BASC-3 composite scores from both informants fall in the Average range, with a single at-risk elevation on the teacher Attention Problems scale during mathematics instruction; the parent form shows no elevations. D.M. describes mathematics as "the class where I feel behind" and says he likes school otherwise. The behavioral picture is consistent with frustration localized to mathematics rather than a broad attention or emotional condition. Exclusionary factors: The pattern is not explained by lack of appropriate instruction (documented core curriculum plus two intervention cycles with progress data), attendance, sensory screening results, language difference, or an emotional or behavioral condition. Summary and integration: Across measures, records, and observation, D.M. shows average general cognitive ability and average reading and writing against persistently low mathematics calculation and fact fluency, with relative weaknesses in working memory and processing speed that fit the computational profile, and a documented limited response to two cycles of intervention. This pattern is consistent with a specific learning disability in mathematics affecting calculation and math fact fluency. Eligibility for special education is a determination for D.M.'s school team under IDEA and state criteria; this report is written to inform that determination and his parents' planning. Recommendations: 1. Explicit, systematic mathematics intervention delivered individually or in a small group at least four sessions weekly, following a concrete-to-representational-to-abstract sequence, with cumulative fact practice and worked examples faded to independent solving. 2. Weekly two-minute computation probes with a written aimline, reviewed every six weeks; an intervention change if four consecutive points fall below the aimline. 3. Classroom accommodations for the team's consideration: extended time on calculation-heavy tasks, a multiplication chart or fact reference where fluency is not the skill being measured, reduced copying from the board, and graph paper for multi-digit work. 4. Share this report with the school team; the family may request a team meeting to review it. 5. Re-evaluate in three years, or sooner if progress monitoring shows the gap widening. A plain-language summary for D.M. and his parents accompanies this report. Evaluator: L. Marsh, PhD, Licensed Psychologist, License #P-7241 · Signed: 07/16/2026 This sample is fictional and for educational purposes. It does not describe a real student, and the instruments named are examples of commonly used measures. ↑ Back to the template and downloads ### Why this sample works The referral question is answered in one findable place: the summary states what the mathematics difficulty is, what it is not, and what should happen next. No conclusion rests on a single measure: cognition, achievement, ratings from two informants, records, and observation each carry weight, which is what the evaluation regulation requires. Instruction and intervention response are documented with numbers, so the team can rule out lack of appropriate instruction instead of guessing. The observation requirement is handled honestly: the report incorporates the district's spring classroom observation and names the summer timing as a limitation rather than staying silent. Exclusionary factors get their own labeled section, mirroring the elements the eligibility team must document. Every recommendation is numbered, specific, and traceable to a finding, with a progress-monitoring metric and a defined trigger for changing course, and the report leaves eligibility where the law puts it: with the team. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Which privacy law governs the report depends on who wrote it, and the two regimes do not behave alike. Written by a school employee, the report is an education record: FERPA gives parents the right to inspect and review it within 45 days at most, and the school must respond to reasonable requests to explain and interpret it ( 34 CFR 99.10 ); HIPAA's definition of protected health information expressly excludes FERPA-covered records, so a school report never sits under both. IDEA adds its own guarantee: the agency provides a copy of the evaluation report to the parent at no cost ( 34 CFR 300.306 ). Written in private practice, the same report is a health record: HIPAA's access right reaches the report and the test data behind it, while APA Ethics Standards 9.04 and 9.11 draw the working line between releasable test data and protected test materials, and Ontario law runs the other way, excluding raw standardized test data from the patient access right. Retention belongs to your board or college, not to any education statute: California requires seven years, Ontario requires ten years after the client turns 18 or after last contact, BC's harmonized standard runs 16 years, and Australia's national code sets seven years from the last entry, or until a young client turns 25. Keep the strictest applicable rule. Money is where the psychoeducational label carries consequences. US commercial payers broadly exclude educational testing: Aetna's published policy states that testing performed for educational reasons "is not considered treatment of disease" ( CPB 0158 ), and Medicare contractor policy finds testing not medically necessary when administered for educational or vocational purposes that do not establish medical management ( L34520 ). Australia's Better Access items fund treatment of a clinically diagnosed mental disorder and are not a funding pathway for educational assessment, so school-question testing there is school-funded or private. The same hours framed around a clinical referral question, differential diagnosis, medical management, can be billable under the testing codes the psychological evaluation report page covers, which is why the psychoeducational-versus-psychological framing is a consequential coding decision and not a stylistic one. Families who disagree with a district's evaluation have a different funding lever entirely: an independent educational evaluation at public expense under 34 CFR 300.502 , which the district must consider though not adopt. Across all of it the through-line holds: the format is a convention; the content, access, and timeline rules are the requirement. ## Common psychoeducational evaluation report errors auditors flag No payer audit isolates psychoeducational reports, because most of this testing sits outside insurance; the scrutiny arrives through school-money audits and the courts instead, and it is not gentle. The HHS Office of Inspector General's 2024 Pennsylvania audit recommended returning $551.4 million in school-based Medicaid funds over unsupported time-study and cost-allocation documentation, with services by unlicensed providers among the findings. The courts police the evaluation itself: in Timothy O. v. Paso Robles (9th Cir. 2016), a district that relied on a staff member's informal observation instead of formally assessing a suspected disability was held to have denied a free appropriate public education; districts, the court wrote, "cannot circumvent that responsibility by way of informal observations." And the readers fail differently: research from Weddig's 1984 study onward finds traditional reports written above the reading level of most parents, while a 2026 meta-analysis of 14 experiments found accessible plain-language reports improved reader outcomes in every experiment. The BastionGPT Clinical Advisory Board sees the same errors most often in psychoeducational evaluation report reviews: An area of suspected disability never assessed. The referral hints at attention, language, or emotional factors and the battery never looks. The evaluation must cover all areas related to the suspected disability, and failure to assess is the error that loses hearings, as Timothy O. shows. Treating the severe-discrepancy formula as required. Federal rule says states "must not require the use of a severe discrepancy" ( 34 CFR 300.307 ), yet reports still lead with the ability-achievement subtraction as if it decided eligibility. Frame the pattern of data; let the team apply state criteria. Exclusionary factors skipped. Nothing on instruction received, intervention response, attendance, sensory screening, or language difference. The team must document these determinations, and a report silent on them forces the team to guess or re-evaluate. Boilerplate recommendations. The same closing list on every report, disconnected from the findings. Parents, teams, and hearing officers all read this section first, and generic advice wastes the evaluation. Written for the file, not the reader. A report parents cannot read is a finding they cannot use; the readability literature has said so for four decades, and the plain-language evidence now quantifies the fix. ### Psychoeducational evaluation reports in the US, Canada, and Australia Aspect United States Canada Australia Status IDEA fixes evaluation content, timelines, and the team's eligibility documentation, not the report's format; some state codes enumerate report contents (California EDC 56327 lists eight elements) and set their own timelines (Texas: 45 school days) No federal special-education statute; education is provincial. Provincial frameworks such as Ontario's IPRC identification process govern eligibility, and college standards with legal force govern the record The Disability Standards for Education require reasonable adjustments, and the national data collection counts students with a disability that is diagnosed or imputed; the current NCCD guidelines let a school team impute a disability when no diagnostic report is available Terminology Psychoeducational evaluation or report; full individual and initial evaluation (Texas); IEE when independently obtained Psychoeducational assessment, often "comprehensive"; the report feeds an IPRC or provincial designation Educational and developmental assessment; cognitive assessment; the report may support NCCD evidence or NDIS access but neither requires it What changes Eligibility categories and criteria are state-implemented: the federal floor is 60 days from consent to evaluation unless the state sets its own, and discrepancy cannot be required Waiting time is the defining constraint: Ontario's human-rights inquiry reported the Auditor General's finding that nearly a quarter of students with special-education needs in audited boards waited over a year for assessment, and 93% of elementary schools surveyed had assessment wait lists The adjustment framework runs without mandatory assessment: in 2025, 1,125,502 students, 27.0% of enrolments, received an adjustment due to disability, and Better Access funding does not extend to educational assessment Retention Your licensing board and the education records schedule, not IDEA: California's board rule is seven years (longer for minors), and districts may destroy records no longer needed subject to notice rules Ontario: 10 years after the client turns 18 or after last contact, whichever is later; BC: 16 years; Quebec: 5 years by regulation Psychology Board code: 7 years from the last entry; for clients seen as minors, until the 25th birthday The document is recognizably the same in all three countries: a learning question answered with cognitive and achievement data. The machinery around it is not. A US evaluator writes against federal and state education law with its timelines and team decisions, a Canadian evaluator against provincial frameworks where the wait list is often the story, and an Australian evaluator inside an adjustment system that, by design, never requires the report at all. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychoeducational evaluation reports. Draft the integrated report from your scores, observation notes, records review, and interview notes, with every section in place for your review, including the exclusionary-factors documentation teams need. Translate the finished report into a plain-language summary for parents and the student, the report style the reader-outcome evidence favors, without touching the clinical version. Check a draft before you sign: every area of suspected disability addressed, intervention response documented, observation present, and each recommendation traceable to a finding. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a legally required format for a psychoeducational evaluation report? No. The only US federal regulation that specifies written contents, 34 CFR 300.311 , governs the eligibility documentation the team produces for specific learning disability, not the psychologist's report, and no federal rule sets a section order, page count, IQ requirement, or signature requirement for the report itself. What exists instead is state law and convention: California's Education Code section 56327 enumerates eight required report elements, other states leave format alone, and the familiar section sequence comes from assessment textbooks. Treat the format as a convention and the content triggers, all areas of suspected disability, multiple measures, observation, timelines, as the law. Does the report have to include an IQ score or an ability-achievement discrepancy? Neither is federally required. No regulation requires an IQ score anywhere in the process, and the evaluation rule bars using any single measure as the sole criterion ( 34 CFR 300.304 ). On the discrepancy model the rule is stronger than most sample pages suggest: states "must not require the use of a severe discrepancy" between ability and achievement, must permit response-to-intervention approaches, and may permit other research-based procedures such as patterns of strengths and weaknesses ( 34 CFR 300.307 ). Cognitive testing usually still belongs in the battery, it explains how the student learns, but it is evidence, not a gate. Do parents automatically get a copy of the report? In a school evaluation, yes. IDEA requires the public agency to give the parent a copy of the evaluation report and the eligibility documentation at no cost ( 34 CFR 300.306 ). FERPA adds the general right to inspect and review education records within 45 days at most, plus a right to reasonable explanations and interpretations of them ( 34 CFR 99.10 ). A privately retained evaluation is governed by the service agreement and health-privacy law instead. Either way, plan the writing for parent readers: the readability research has flagged reports written above most parents' reading level since the 1980s. Will insurance pay for a psychoeducational evaluation? Usually not, and the exclusion is explicit rather than incidental. Aetna's published policy states that psychological and neuropsychological testing performed for educational reasons "is not considered treatment of disease" ( CPB 0158 ), and Medicare contractor policy finds testing not medically necessary when administered for educational or vocational purposes that do not establish medical management ( LCD L34520 ). Australia's Better Access items fund treatment of a clinically diagnosed mental disorder, not educational assessment. When a genuine clinical question drives the referral, differential diagnosis or medical management, the work may be billable under the psychological testing codes covered on the psychological evaluation report page; the framing decision belongs at intake, not at claim time. We disagree with the school's evaluation. What are our options? Parents who disagree with a district evaluation may request an independent educational evaluation at public expense; the district must then either fund it or, without unnecessary delay, defend its own evaluation at a due process hearing ( 34 CFR 300.502 ). The flip side matters as much: a privately obtained evaluation shared with the district must be considered, if it meets agency criteria, in decisions about the child, but the district does not have to adopt its conclusions or provide an IEP because of it. Evaluators writing IEE reports should know both halves, because families usually arrive believing the private report is binding. How long does the process take, and how long does the report take to write? Two clocks run. The legal clock: a US initial evaluation must be completed within 60 days of parental consent, unless the state sets its own timeframe, Texas allows 45 school days, and reevaluations occur at least every three years unless the parent and agency agree one is unnecessary. The workload clock: a published time study of school evaluations puts the typical case near five to six hours of administration, scoring, and report writing combined, and school psychologists average 55 evaluations per year . Systems without the legal clock show what demand does on its own: Ontario's Right to Read inquiry reported the Auditor General's finding that nearly a quarter of students with special-education needs in audited boards waited more than a year for assessment. How is this different from a psychological or neuropsychological evaluation report? By referral question, and the label carries billing weight. The psychoeducational report answers a learning question with a cognitive-plus-achievement battery, typically a WISC-V paired with a WIAT-4 or Woodcock-Johnson , and often a BASC-3 , on the school's legal timeline. The psychological evaluation report is the broader diagnostic document for emotional, behavioral, and personality questions, billed under the psychological testing codes when medically necessary. The neuropsychological evaluation report covers brain-based questions after injury, illness, or suspected decline under its own code family. IDEA never distinguishes the labels; payers do, which is why the same battery can be excluded as educational testing or covered as clinical testing depending on the question that launched it. How long do we keep the report, and how current must one be for accommodations? Retention follows your board or college, not IDEA: California's board rule is seven years (longer for minors), Ontario requires ten years after the client turns 18 or after last contact, BC's standard runs 16 years, and Australia's national code sets seven years from the last entry or until a young client turns 25. Schools follow their education-records schedules, and IDEA lets agencies destroy personally identifiable information no longer needed, subject to parent notice. Recency is a different question with a different answer: no federal rule sets one. Colleges, testing agencies, and exam boards each publish their own documentation windows as institutional policy, so an evaluation completed for grade school may need updating for university accommodations even though nothing in the law expired it. Can BastionGPT draft a psychoeducational evaluation report? Yes. Give it your scores, observation notes, records summary, and interview notes, and it drafts the integrated report for your review: referral question, background with intervention response, observation and validity, results by domain in prose, exclusionary factors, summary, and numbered recommendations tied to findings. It can also translate the finished report into a plain-language summary for parents and the student, the style the reader-outcome evidence favors, and check a draft for the gaps that lose hearings: an area of suspected disability never assessed, missing intervention data, or recommendations that connect to nothing. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance and legal claims on this page trace to these authorities, last verified July 2026: IDEA evaluation regulations, 34 CFR 300.301 , 300.303 , 300.304 , 300.306 , 300.307 , 300.310 , 300.311 , and 300.502 : evaluation timelines, multiple-measure and all-areas rules, the parent's copy, SLD criteria and the discrepancy bar, the observation requirement, the team's eligibility documentation, and independent educational evaluations. California Education Code section 56327 : a state statute enumerating written report contents. FERPA access rules, 34 CFR 99.10 , and the HIPAA definitions rule excluding FERPA records from protected health information, 45 CFR 160.103 . NCES, Condition of Education, Students With Disabilities : 7.5 million students served under IDEA, 15 percent of enrollment, specific learning disabilities 32 percent. NASP, 2020 Membership Survey Part 2 : an average of 55 evaluations per year. Aetna, Clinical Policy Bulletin 0158 , and First Coast Medicare LCD L34520 : the educational-testing exclusions. HHS OIG, audit A-02-21-01011 (2024) : the $551.4 million school-based Medicaid finding. Timothy O. v. Paso Robles Unified School District , 822 F.3d 1105 (9th Cir. 2016): failure to formally assess a suspected disability as a denial of FAPE. Weddig, 1984, Parental Interpretation of Psychoeducational Reports (Psychology in the Schools), and Hite, 2026, meta-analysis of accessible report writing : 14 experiments, 1,283 readers, Hedges' g = 0.83. Ontario Human Rights Commission, Right to Read inquiry, chapter 12 : the Auditor General's wait-time finding and the People for Education wait-list survey. ACARA, National Report on Schooling: school students with disability : 1,125,502 students, 27.0 percent of enrolments, 2025. NCCD, Guidelines effective 2026 onward : disability diagnosed or imputed, and the imputation rules. Retention and conduct: CPBAO Standards of Professional Conduct ; BC CHCPBC Records Practice Standard ; California Business and Professions Code section 2919 ; Psychology Board of Australia, Code of Conduct (December 2025) ; APA Ethics Code Standards 9.04 and 9.11. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report WISC-V Report Write-Up WIAT-4 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up BASC-3 Report Write-Up ABAS-3 Report Write-Up WPPSI-IV Report Write-Up Child & Adolescent Developmental Assessment Psychological Testing Administration Note Psychological Testing Scoring Note Forensic Psychological Evaluation Report Section 504 Clinician Statement IEP Psychoeducational Input School & Teacher Collateral Note Beery VMI Report Write-Up CELF-5 Report Write-Up GFTA-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychological-evaluation-report All note templates # Psychological Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychological evaluation report is the written product of psychological testing: it integrates the referral question, history, behavioral observations, and standardized test results into an interpretation, a diagnosis, and recommendations. Psychologists write one after an assessment battery, addressed to the referrer and increasingly read by the patient. Most clinical reports run 3 to 12 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Licensed psychologists: clinical, counseling, and school psychologists; psychology trainees under supervision Audience Referring clinicians, the patient and family, schools, courts, insurers, the treatment team Typical length 1,500 to 5,000 words (3 to 12 pages) · 2 to 6 hours of scoring, interpretation, and writing (clinical team estimate) Format family Interpretive assessment report (compare: neuropsychological and psychoeducational reports) When it's used After psychological testing, when a referral question needs more diagnostic clarity than an interview alone can give Standards context No law mandates a format; Medicare contractor policies list expected documentation elements, and APA assessment guidelines shape the conventional structure On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychological evaluation report? A psychological evaluation report is the written product of a psychological assessment: it takes a referral question in and sends an answered question back, integrating history, interview, behavioral observations, and standardized test results into an interpretation, a diagnostic conclusion, and recommendations. Clinicians and payers also call it a psychological assessment report, a psychological testing report, or simply a psych eval. The report is where testing becomes assessment: administering instruments produces scores, and the evaluation is the clinician's integration of those scores with everything else known about the person. The familiar section sequence descends from professional convention rather than statute. The technical standards trace from the APA's 1954 Technical Recommendations through the first joint AERA, APA, and NCME edition in 1966 to the 2014 edition current today, and the APA's 2020 Guidelines for Psychological Assessment and Evaluation are the most direct current guidance on how findings should be communicated. The US billing architecture is newer: on January 1, 2019, the CPT restructure separated the interpretive work, psychological testing evaluation services (96130 and 96131), from test administration and scoring (96136 and 96137), which made the thinking the report documents a separately identified, time-based service. The report standardizes a document, not a mandate. No statute, regulator, or payer in the US, Canada, or Australia prescribes a report title, section order, page count, or universal completion deadline; the enforceable rules attach to content, access, authentication, retention, and billing. Two boundaries keep the record clean. First, the report is not the billable service itself: the timed evaluation-services codes pay for the interpretive work, and the report is the evidence that work happened. Second, it is not a protected HIPAA psychotherapy note: the report, its scores, and its conclusions sit in the ordinary designated record set that patients can request and payers can audit. When the question is an initial diagnosis rather than standardized measurement, the interview-based psychiatric diagnostic evaluation answers it without a battery; Medicare coding policy treats the two as distinct same-day services and bars counting the same time in both. ## Who uses psychological evaluation reports and when Clinical and counseling psychologists write them in private practice, hospitals, community mental health, and integrated care; school psychologists write the psychoeducational variant inside special-education timelines, where federal law puts a copy of the evaluation report in the parent's hands at no cost. The report is the deliverable the referrer is waiting on: a psychiatrist clarifying a diagnosis before a medication change, a primary care physician untangling overlapping symptoms, a therapist asking whether personality structure explains a stalled treatment. It is the right tool when the question needs standardized measurement the interview cannot give: differential diagnosis, personality assessment, cognitive screening with follow-up questions. When the question is an initial diagnosis at the start of routine care, the interview-based psychiatric diagnostic evaluation usually answers it without testing, and when it concerns memory, attention, or other brain-based domains after injury or decline, the neuropsychological evaluation report takes over. Whoever writes it, the audience has widened: the referrer, the treatment team, schools, courts, insurers, and, through the patient portal, the person who was tested. ## Psychological evaluation report structure: what goes in each section No authority mandates a single report format. The sections below are the consensus skeleton of the APA assessment tradition and the elements Medicare contractor policy lists for psychological testing services, each with the pitfall that most often undermines it. Identifying information and referral question. Who the patient is, who referred them, and the specific question the evaluation should answer, stated in a sentence or two. Pitfall: restating a vague "rule out everything" referral instead of sharpening it; when the question is not specific, the interpretation has nothing to aim at and the report reads as testing for its own sake. Notification and consent. What the patient was told about the purpose, the fees, the limits of confidentiality, and who will receive the report. Pitfall: skipping who-gets-the-report; access disputes start where the record is silent about the intended audience. Sources of information and tests administered. Every instrument by full name, the interview, the records reviewed, and the dates and time involved in administration, scoring, and interpretation. Pitfall: no time documentation; the US testing codes are time based, and a report that never states the hours cannot support the units billed. Background history. The developmental, medical, psychiatric, educational or occupational, family, and social history that bears on the referral question. Pitfall: reprinting the entire intake; the report needs the history relevant to the question, not a second copy of the chart. Behavioral observations and validity. How the person presented and engaged during testing, and an explicit statement on effort and whether the results are interpretable. Pitfall: no validity statement; a reviewer who cannot tell whether the profile is valid cannot credit anything built on it. Test results and interpretation. Findings organized by domain or instrument, with scores placed in context and integrated in prose. Pitfall: pages of score tables with no interpretation; payers cover the evaluation service, and a score dump documents administration without evaluation. Summary and clinical impressions. The integrative paragraphs that pull interview, history, observations, and testing into one clinical picture and answer the referral question directly. Pitfall: introducing new findings here; the summary integrates what the body of the report already shows. Diagnostic impression. The DSM-5-TR diagnosis with its ICD-10-CM code, the conditions considered and ruled out, or a documented statement that no diagnosis is supported. Pitfall: leaving the diagnosis line out when testing is negative; contractor policy expects the suspected diagnosis that was the basis for testing even when no mental illness is found. Recommendations. Numbered, specific, and connected to the findings: treatment, school or workplace supports, referrals, and what should trigger re-evaluation. Pitfall: boilerplate recommendations that could follow any battery; the referrer and the family both read this section first, and generic advice wastes the evaluation. Signature and credentials. The evaluating psychologist's signature, credentials, and license, with the supervisor's counter-signature where a trainee performed the work. Pitfall: a trainee-authored report with no supervising psychologist's signature; unsigned supervised work is the simplest documentation failure a reviewer can find. ### Blank template (copy and adapt) Patient: [initials] DOB/Age: Dates of service: Report date: Evaluator [name / credentials / license]: Referral source & referral question: Notification & consent (purpose / fees / confidentiality limits / who receives the report): Sources of information (interview / records / collateral): Tests administered (full names + administration, scoring & interpretation time): Background history (relevant to the question): Developmental / medical / psychiatric: Educational / occupational: Family / social: Behavioral observations & validity statement: Results & interpretation (by domain or instrument, in prose): Summary & clinical impressions (answer the referral question): Diagnostic impression (DSM-5-TR / ICD-10-CM + rule-outs, or "no diagnosis supported" + basis for testing): Recommendations (numbered, specific, tied to findings): Re-evaluation trigger: Evaluator signature / credentials: Date signed: Supervisor signature (if required): Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychological evaluation report Scenario: a psychiatrist refers a 29-year-old client for diagnostic clarification of longstanding mood instability before finalizing a medication plan. A licensed psychologist completes the interview, testing battery, and report. The sample is condensed but structurally complete; a full report runs 3 to 12 pages. All details are fictional. Patient: R.T., 29 · Dates of service: 07/07/2026, 07/09/2026 · Report date: 07/14/2026 · Evaluator: M. Okafor, PsyD · Referred by: J. Patel, MD (psychiatry) Referral question: Dr. Patel asks whether R.T.'s mood instability reflects a bipolar spectrum disorder, recurrent depression, a personality pattern, or a combination, before finalizing a medication plan. Notification and consent: Purpose of the evaluation, the tests involved, fees, limits of confidentiality, and who receives the report (Dr. Patel and R.T.) were reviewed; written consent signed 07/07/2026. A feedback session is scheduled. Sources of information: Clinical interview (90 minutes, 07/07/2026); psychiatric intake note and primary care records; MMPI-3; PAI; PHQ-9; MDQ. Test administration 07/09/2026, 75 minutes; scoring, interpretation, and report preparation 2.5 hours, documented for billing. Background: Two prior depressive episodes (ages 19 and 24, each two to four months, untreated). Current symptoms began about ten weeks ago after a relationship ended: low mood, self-criticism, initial insomnia, and rapid mood shifts lasting hours, reliably cued by interpersonal events. No period of elevated or expansive mood lasting four days or more; sleep loss follows worry rather than a decreased need for sleep. Mother treated for depression. Alcohol two to three drinks weekly; no other substance use. Employed steadily as a marketing coordinator; no hospitalizations, no history of self-harm. Behavioral observations and validity: Arrived on time, casually dressed, cooperative and candid; affect ranged appropriately with brief tearfulness when describing the breakup. Attention and persistence were good across both sessions. Validity scales on the MMPI-3 and PAI were within acceptable limits; the profiles are considered interpretable and an accurate reflection of current functioning. Results and interpretation: PHQ-9 score of 14 indicates moderate depressive symptoms. MDQ screening was negative: two items endorsed, without co-occurrence or impairment. The MMPI-3 profile shows elevations on scales measuring demoralization and low positive emotions, consistent with an internalizing, depressive presentation; scales associated with hypomanic activation were not elevated. On the PAI, the depression scale was clinically elevated and the affective instability component was moderately elevated, while mania-related scales were unremarkable. Across instruments, the data converge on a depressive disorder with prominent affective reactivity: mood shifts are hours long and interpersonally cued, a pattern distinct from the sustained, episodic mood elevation of bipolar spectrum conditions. Summary and clinical impressions: Testing, history, and screening answer the referral question consistently: R.T.'s instability is best explained by a recurrent depressive disorder with trait-level affective reactivity. Nothing in the history, MDQ, or profile data supports a bipolar spectrum diagnosis. Borderline personality traits were considered; the affective instability finding is notable, but identity disturbance, abandonment-driven behavior, and self-harm are absent, and the diagnostic threshold is not met. Diagnostic impression: Major depressive disorder, recurrent, moderate (F33.1). Bipolar II disorder considered and not supported. Borderline personality traits noted, below diagnostic threshold; no personality disorder diagnosed. Recommendations: (1) Weekly individual psychotherapy with an emotion-regulation focus, such as DBT-informed skills training, targeting the interpersonally cued mood shifts. (2) Share this report with Dr. Patel for medication planning; the evaluation found no bipolar spectrum indicators, and emergent activation on antidepressant treatment should still be monitored. (3) Re-administer the PHQ-9 every four weeks, targeting a score below 10 within twelve weeks. (4) Re-evaluate diagnostically if R.T. experiences a distinct period of elevated mood or decreased need for sleep lasting four or more days. (5) Feedback session 07/17/2026 to review these findings with R.T. in plain language. Evaluator: M. Okafor, PsyD, Licensed Psychologist, License #4-8213 · Signed: 07/14/2026 This sample is fictional and for educational purposes. It does not describe a real patient, and the instruments named are examples of commonly used measures. ↑ Back to the template and downloads ### Why this sample works The referral question is answered in one findable place: the summary states what the mood instability is and is not, and the diagnosis follows from it. Validity is addressed explicitly, so every interpretation that follows has a stated foundation. Scores are interpreted in prose and integrated across instruments; no section is a bare table. Tests, dates, and administration and scoring time are documented, which is what the time-based US billing codes require the record to support. Each recommendation is numbered, specific, and traceable to a finding, and the report names what should trigger re-evaluation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the report knowing the patient will read it, probably before the feedback session. The report sits in the designated record set, so the patient holds a right of access under 45 CFR 164.524 , and the federal information-blocking rules make a finalized electronic report releasable by default: no rule demands instant portal posting, but holding every report until feedback requires an individualized judgment under the preventing-harm exception , not a standing office policy, and test security is not one of the named exceptions. The APA Ethics Code draws the release line clinicians need at request time: test data, meaning raw and scaled scores, the patient's responses, and your notes about their statements and behavior during the exam, goes to the patient or their designee with a release (Standard 9.04), while test materials, "manuals, instruments, protocols, and test questions or stimuli," stay protected ( Standard 9.11 ). Where the two masters meet, HIPAA is usually the stricter one: the Code permits withholding data to prevent misuse, but the regulation's only clinical denial ground is a professional judgment that access is "reasonably likely to endanger" life or physical safety, and the patient can have that denial reviewed. Since the reader is now often the patient, plain language is not a courtesy: a 2026 meta-analysis of 14 experiments (1,283 readers) found accessible, plain-language reports improved reader outcomes with a large effect, and every experiment favored the accessible version. Retention is state and provincial law, not HIPAA: HHS states plainly that the Privacy Rule sets no retention period , the APA's 2007 record-keeping guideline suggests seven years past the last service for adults as an aspirational floor, and records supporting Medicare billing carry a seven-year duty of their own (42 CFR 424.516(f)). On the payer side, the rules are specific and mostly absent from template pages. Medicare covers testing to clarify a suspected mental illness: contractor policy is blunt that screening without a suspected condition is not covered, that testing "administered for educational or vocational purposes" without a medical-management purpose is excluded, and that self-administered or self-scored inventories do not count as testing at all. The First Coast documentation list is short and auditable: reason for referral; tests administered with scoring, interpretation, and the time involved; the present evaluation; a diagnosis, or the suspected diagnosis that justified testing when none is found; recommendations; and the identity of the person performing the service, with patient identification and dates of service on every page. A standard battery is covered only when each component test is individually necessary , contractors describe four to six or four to eight hours as the typical range, testing beyond eight hours must be justified in the report itself, and multi-day testing is combined and reported on the last date of service. The 2026 NCCI manual adds the time discipline: the diagnostic interview, test administration, and evaluation services must be genuinely distinct same-date services, and minutes spent collecting or interpreting the same information cannot be reported twice. Who may bill is regulation, not convention: 42 CFR 410.32(b)(2)(iii) covers testing personally furnished by a clinical psychologist or independently practicing psychologist, or under the general supervision of a physician or clinical psychologist, with NPs, CNSs, and PAs acting within their own scope; general supervision does not require physical presence, though at least one contractor's article still says "direct supervision" for technician work, so check your MAC. These tests are never billed incident-to, an independently practicing psychologist needs a physician's or eligible NPP's order with the orderer's name on the claim, and clinical social workers, marriage and family therapists, and mental health counselors are absent from the who-may-bill list. The part clinicians most often underclaim: the current APA Services billing guide describes evaluation services as including data integration, interpretation, clinical decision making, treatment planning, report writing, and interactive feedback, with the first-hour code used once per assessment episode even when the work spans days. ↑ Back to the template and downloads ## Common psychological evaluation report errors auditors flag There is no current national audit number for this document type: no OIG or CERT publication isolates psychological testing evaluation services, and no testing-specific denial rate has ever been published, so a page that quotes one is inventing it. The audit that shaped today's rules is old and blunt: the HHS OIG's 2001 nursing-home review found 39 percent of psychological tests medically unnecessary, called testing "the most problematic of the nursing home services reviewed," and flagged self-administered questionnaires billed as testing, the direct ancestors of today's self-administration bar and each-test-necessary battery rule. The BastionGPT Clinical Advisory Board sees the same errors most often in psychological evaluation report reviews: Scores without interpretation. Tables, percentiles, and computer printouts with no integrative prose. Payers cover the evaluation service, and interpretation is the service; a score dump documents administration only. Missing time documentation. The testing codes are time based, and a report that never records administration, scoring, and interpretation time cannot support the units billed. The ceiling is real too: First Coast requires testing beyond eight hours to be justified in the report itself. The referral question never answered. A multi-page report that ends without stating what the referrer asked and what the answer is. The reviewer reads for the question; so do the referrer and the family. No validity statement. Nothing about effort, engagement, or whether the profiles are interpretable, which leaves every conclusion in the report floating. A battery nobody justified. The same instrument panel for every patient regardless of the question. Contractor policy covers a standard battery only when each individual test in it is medically necessary, and an identical-battery pattern across charts is exactly what claims data mining flags. ### Psychological evaluation reports in the US, Canada, and Australia Aspect United States Canada Australia Status No mandated form; CPT defines the timed testing-service family (2019), Medicare contractor policies list expected documentation and typical hours, and information-blocking rules govern release No national rule; provincial college standards carry legal force: Ontario prescribes record content and supervisory co-signing, Alberta requires countersigned provisional reports, BC's harmonized standard sets access and retention The Psychology Board's code of conduct (effective December 1, 2025) governs records and reports: true and objective, verified before signing, prepared "within a reasonable and justifiable timeframe"; MBS pays for testing only in narrow pathways Terminology Psychological evaluation report, psychological testing report, psych eval; "psychological testing evaluation services" on claims Psychological assessment report; psychoeducational assessment in schools Psychological assessment report; cognitive assessment What changes Which MAC's policy applies (hour ranges, supervision wording) and who may bill under 42 CFR 410.32 What the patient can access: Ontario law excludes raw standardized test data from its access right, Quebec releases raw data only to another psychologist, BC recognizes a test-integrity exception The funding stream defines the audience: the under-25 MBS assessment pathway requires a written report to the referrer but excludes report preparation from billable item time Retention State law governs; the APA's 2007 guideline suggests 7 years past last service (3 past majority for minors) as an aspirational floor; records supporting Medicare billing carry 7 years (42 CFR 424.516(f)) Ontario: 10 years after age 18 or last contact; Alberta: 10 years after last service; Quebec: 5 years by regulation; BC: 16 years after last encounter or majority Psychology Board code: 7 years from the last entry; for clients under 18, until the 25th birthday The report itself travels: referral question, methods, findings, interpretation, recommendations read the same in all three countries. The machinery around it does not. A US psychologist answers to a Medicare contractor's element list and the patient portal, a Canadian psychologist to a provincial college's record and co-signature standards, and an Australian psychologist to a national conduct code that regulates the report's honesty and timing without ever prescribing its shape. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological evaluation reports. Draft the full report from your test scores, interview notes, and behavioral observations, with every section in place for your review. Integrate results from multiple instruments into one interpretive summary that answers the referral question instead of restating score tables. Translate the clinical report into plain language for parents and patients, and check a finished draft for the gaps reviewers flag: scores without interpretation, missing time documentation, or recommendations that never connect to the findings. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a legally required format for a psychological evaluation report? No. No statute or regulator in the US, Canada, or Australia prescribes a report title, section order, page count, or universal completion deadline. What exists is payer policy and professional convention: Medicare contractor policy lists the elements the record must show (reason for referral; tests with scoring, interpretation, and time; the present evaluation; a diagnosis or the suspected diagnosis that justified testing; recommendations; who performed the service), and the APA assessment tradition supplies the familiar section sequence. Canadian colleges and Australia's Psychology Board regulate record content, accuracy, and retention rather than layout. Treat the format as a convention and the content as the requirement. Will the patient read the report before I can explain it? Plan for it. The report sits in the designated record set with a right of access under 45 CFR 164.524 , and the information-blocking rules make a finalized electronic report releasable by default. No rule requires instant portal posting, but delaying every report until after feedback needs an individualized preventing-harm analysis , not a standing policy, and test security is not a named exception. So write for two readers and schedule feedback quickly: a 2026 meta-analysis of 14 experiments found plain-language reports improved reader outcomes in every experiment, and the feedback conversation is itself part of the evaluation service. Do I have to release raw test data if the patient or their lawyer asks? In the US, usually yes, with a signed release. APA Ethics Standard 9.04 defines test data broadly: raw and scaled scores, the patient's responses, and your notes about their statements and behavior during the exam, including response-bearing protocol pages. Test materials, "manuals, instruments, protocols, and test questions or stimuli," stay protected under Standard 9.11. Where the Code would let you withhold data to prevent misuse, HIPAA usually overrides: its only clinical denial ground is danger to life or physical safety, and the patient can have a denial reviewed. Cross a border and the rule inverts: Ontario's health-privacy law excludes raw standardized test data from its access right, Quebec lets raw data go only to another psychologist, and BC's college standard recognizes a test-integrity exception. Testing ran across several days. Which date goes on the claim? For Medicare, combine it: First Coast and WPS both direct that testing performed over several days be combined and reported on the last date of service, and the first-hour evaluation code is used once per assessment episode even when the work spans days. Keep an internal log of the true dates, tasks, and minutes for every component: completion-date billing is payer methodology, not permission to rewrite when work happened. Commercial payers can run differently, so check the plan's policy before assuming the convention. Can I bill for the time spent writing the report? Yes, inside the evaluation-services codes. The 2019 restructure split administration and scoring (96136/96137 for a psychologist, 96138/96139 for a technician) from evaluation services (96130/96131), and the current APA Services billing guide describes evaluation services as including data integration, interpretation, clinical decision making, treatment planning, report writing, and interactive feedback. Two limits: the codes are timed, so the hours must appear in the record, and NCCI policy bars counting the same collection or interpretation time twice on one date, including against the diagnostic interview. Australia runs the opposite way: the MBS assessment pathway that requires a written report excludes report preparation from billable item time. Who can perform and bill psychological testing under Medicare? Under 42 CFR 410.32(b)(2)(iii) , testing is covered when personally furnished by a clinical psychologist or an independently practicing psychologist, or under the general supervision of a physician or clinical psychologist, or of an NP, CNS, PA, CRNA, or CNM within state scope of practice. General supervision means direction and control without required physical presence, though at least one contractor's article still says "direct supervision" for technician work, so check your MAC. Three traps: these tests are never billed incident-to, an independently practicing psychologist needs a physician's or eligible practitioner's order with the orderer's name on the claim, and clinical social workers, MFTs, and mental health counselors are absent from the who-may-bill list . Commercial payers can be stricter: one Blue plan's 2025 policy treats anyone working under a supervising psychologist, including trainees, as a technician for payment. Does my supervisor have to co-sign the report? It depends on where you practice, and the differences are sharp. Alberta's supervision standards require the supervisor to countersign every formal written report a registered provisional psychologist enters in the file. Ontario requires the supervisor's name on all reports about supervised services and a co-signature on any document that may reasonably be relied on for decisions affecting care, rights, or welfare. BC and Australia's national code set supervision duties without a blanket co-sign rule. In the US there is no single federal rule: state licensing law, training programs, and payer credentialing decide, and some payers will not pay for evaluation services performed by supervisees at all. How is this different from a psychiatric diagnostic evaluation or a neuropsychological report? The psychiatric diagnostic evaluation (90791/90792) is an interview-based service that opens an episode of care; testing enters when the question needs standardized measurement the interview cannot give, and Medicare treats them as distinct same-day services with no double-counted time. The neuropsychological evaluation report is the sibling document for brain-based questions, memory, attention, and executive function after injury, illness, or suspected decline, billed under its own evaluation codes (96132/96133); some payers require choosing the predominant service rather than billing both families for one episode. School psychoeducational reports answer eligibility questions on the school's legal timeline, and parents receive a copy of that report at no cost. Can BastionGPT draft a psychological evaluation report? Yes. Give it your test scores, interview notes, and behavioral observations, and it drafts the full report for your review: referral question, background, validity statement, integrated interpretation, diagnostic impression, and numbered recommendations. It can also translate a finished report into plain language for the feedback session, the report style the reader-outcome evidence favors, and check a draft for the gaps reviewers flag, like scores without interpretation or a missing time record. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance and payer claims on this page trace to these authorities, last verified July 2026: CMS Medicare Coverage Database, First Coast billing article A57780 and LCD L34520 : the documentation element list, every-page identification, four-to-six-hour range, the eight-hour justification rule, the battery rule, and last-date-of-service reporting. CMS Medicare Coverage Database, WPS article A57481 : the four-to-eight-hour typical range, multi-day combination, and technician supervision wording. 42 CFR 410.32(b)(2)(iii) : who may furnish and supervise diagnostic psychological and neuropsychological testing. CMS, NCCI Policy Manual 2026, Chapter 11 : same-date distinctness of the diagnostic interview, test administration, and evaluation services, and the bar on double-counted time. CMS, Medicare Benefit Policy Manual, ch. 15 §80.2 : no incident-to billing for these tests and the ordering rules for independently practicing psychologists. CMS Medicare Learning Network, Medicare & Mental Health Coverage (March 2026) : eligible professionals and the testing code family. APA Services, 2026 Psychological and Neuropsychological Testing Billing and Coding Guide : what evaluation services include, and the first-hour-once-per-episode rule. APA, Ethical Principles of Psychologists and Code of Conduct , Standards 9.04 and 9.11: test data versus test materials. APA, Guidelines for Psychological Assessment and Evaluation (2020) and the joint testing standards lineage . APA, Record Keeping Guidelines (2007) : the aspirational seven-year retention floor. 45 CFR 164.524 and 45 CFR 171.201 : the access right and the preventing-harm exception; HHS, HIPAA FAQ 580 on retention. HHS OIG, Medicare Payments for Psychiatric Services in Nursing Homes (2001) : the 39 percent finding. Ontario, PHIPA s. 51(1)(c) and CPBAO Standards of Professional Conduct ; Alberta CAP Standards for Supervision (2025) ; BC CHCPBC Records Practice Standard (2026) ; Quebec Code of ethics of psychologists and record-keeping regulation . Psychology Board of Australia, Code of Conduct (December 2025) , sections 8.5 and 8.9; Australian Government, MBS item 82000 . Hite, 2026, meta-analysis of accessible report writing : 14 experiments, 1,283 readers, Hedges' g = 0.83. ## Related templates Psychiatric Diagnostic Evaluation Trauma Assessment Case Formulation Note Intake Note Treatment Plan Neuropsychological Evaluation Report BASC-3 Report Write-Up WISC-V Report Write-Up WAIS-5 / WAIS-IV Report Write-Up Conners 4 Report Write-Up WIAT-4 Report Write-Up ADOS-2 Report Write-Up Vineland-3 Report Write-Up BRIEF-2 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up MMPI-3 Report Write-Up SRS-2 Report Write-Up Psychoeducational Evaluation Report CAARS-2 Report Write-Up PAI Report Write-Up ABAS-3 Report Write-Up WPPSI-IV Report Write-Up D-KEFS Report Write-Up RBANS Report Write-Up MCMI-IV Report Write-Up Consultation Response & Report Psychological Testing Administration Note Psychological Testing Scoring Note Fitness-for-Duty Evaluation NDIS Plan Reassessment & Progress Report Capacity & Competency Evaluation Forensic Psychological Evaluation Report Disability Evaluation & Medical Source Statement BDI-II Score Documentation PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/abas-3-report All note templates # ABAS-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The ABAS-3 (Adaptive Behavior Assessment System, Third Edition) is a norm-referenced measure of adaptive behavior, the everyday conceptual, social, and practical skills a person uses to function independently, rated entirely by forms: parent, teacher, and adult versions covering birth through age 89. Psychologists use it in intellectual-disability, autism, and disability evaluations. This page covers how to write up ABAS-3 results, with a fictional two-rater sample and a template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and developmental clinicians; publisher qualification level C Audience IEP and eligibility teams, physicians and therapists, disability programs (SSA, provincial services, NDIS), parents and caregivers Typical length 300 to 700 words for the results section · administration 15 to 20 minutes per form Format family Norm-referenced adaptive behavior rating scale When it's used Intellectual-disability and autism evaluations, developmental assessments, special-education eligibility, SSA and NDIS disability determinations, guardianship and capacity evaluations Standards context Published by Western Psychological Services (2015); current edition as of July 2026, no fourth edition announced; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the ABAS-3? The Adaptive Behavior Assessment System, Third Edition (ABAS-3) is a norm-referenced measure of adaptive behavior: the practical, everyday skills a person uses to care for themselves, relate to others, and manage the demands of home, school, work, and community life. Authored by Patti L. Harrison and Thomas Oakland and published in 2015, it is completed entirely by rating forms, five of them: a Parent/Primary Caregiver Form (birth to 5), a Teacher/Daycare Provider Form (ages 2 to 5), a Parent Form (ages 5 to 21), a Teacher Form (ages 5 to 21), and an Adult Form (ages 16 to 89) that can be completed either by someone who knows the adult well or as a self-report, which is why "five forms" and "six forms" both circulate. One publishing fact derails purchase orders and report citations alike: Western Psychological Services (WPS) is the publisher, copyright owner, and trademark holder, while Pearson and other houses act only as distributors. Pearson's Australian page states it plainly ("Pearson Clinical distributes but does not publish this product"), and Pearson Canada's authorization to sell WPS products ended on January 1, 2026, so Canadian evaluators now order directly from WPS. As of July 2026 the ABAS-3 is the current edition, with no fourth edition announced. The load-bearing comparison is with the Vineland-3 , the other major adaptive measure, and the two are not interchangeable. The ABAS-3 organizes its 11 skill areas into the Conceptual, Social, and Practical domains that DSM-5, DSM-5-TR, and the AAIDD definition system use, so a results narrative maps directly onto the framework that diagnostic severity is judged against; the Vineland-3 uses a Communication, Daily Living Skills, and Socialization architecture and adds a semi-structured interview option, where the ABAS-3 is rating-forms-only. The two instruments also produce different numbers in the same person: a peer-reviewed comparison found ABAS-3 composites ran higher than their Vineland-3 counterparts by as much as 16 standard-score points, with the ABAS-3 showing less impairment. Because a difference that size can move a person across a diagnostic threshold, a defensible write-up names the instrument and edition, keeps longitudinal files on one measure where possible, and treats any cross-instrument comparison as a stated limitation rather than an equivalence. ## Who uses ABAS-3 write-ups and when School psychologists write up ABAS-3 results inside special-education evaluations, where the federal intellectual-disability definition pairs subaverage intellectual functioning with concurrent adaptive-behavior deficits and the ABAS-3 supplies the adaptive evidence alongside cognitive testing in a psychoeducational report . Clinical psychologists use it in intellectual-disability and autism evaluations, where adaptive functioning sits beside direct-observation measures such as the ADOS-2 and carries the severity question under DSM-5-TR, and in the broader psychological evaluation report whenever a decision turns on what a person actually does day to day rather than what they can do in a testing room. The Adult Form, which runs to age 89 with self-report and rated-by-others versions, extends the same evidence into guardianship, supported-decision-making, and capacity questions documented in a neuropsychological report , and into SSA and provincial or NDIS disability processes. The ABAS-3 earns its place when multiple raters across settings matter and time is short: every form is a 15-to-20-minute rating scale, the birth-to-89 span covers the whole lifespan, and the Conceptual, Social, Practical structure speaks the same language as the diagnostic frameworks. Where the question narrows, neighbors carry the load: the Vineland-3 adds a clinician-guided interview option many evaluators prefer for early childhood and low-functioning precision, and AAIDD's DABS is built specifically for the near-cutoff adaptive determination in ages 4 to 21. The write-up itself has one job: named forms, named raters, one integrated account of adaptive functioning. ## How to structure an ABAS-3 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names every form and rater before interpreting anything, leads with the General Adaptive Composite before the domains, keeps the two ABAS-3 score metrics visibly separate, and treats informant disagreement as data about settings rather than a problem to average away. Each section carries the pitfall that most often undermines it. Measures, forms, raters, and method. Name the instrument and edition, which of the five forms were used, who completed each one and their relationship to the person, the dates, the administration method (paper or the publisher's online platform), and the norm set. Pitfall: "the ABAS-3 was administered," with no form or rater named. A rating-scale score means nothing without its source: a Teacher Form and a Parent Form sample different settings, and the Adult Form changes meaning entirely depending on whether it was self-report or rated by others. Rater context and validity of the ratings. For each rater, describe their opportunity to observe (relationship, duration and recency of contact, settings seen), note the count of estimated or guessed answers the form records, and state plainly whether each protocol could be fully scored and whether you consider it a valid estimate of current functioning. Pitfall: skipping the estimated-answer record. The ABAS-3 forms let raters flag answers they guessed at; publisher guidance treats heavily guessed skill areas with caution, and a 2025 psychometric study of the question found scores largely invariant to guessing, so the defensible move is to report the count and address it, not to hide it or to discard a protocol over it. Score summary, with the metrics labeled. Report the General Adaptive Composite (GAC) and the Conceptual, Social, and Practical domains as standard scores with a mean of 100 and a standard deviation of 15, each with its confidence interval and percentile rank; report the skill areas as scaled scores with a mean of 10 and a standard deviation of 3, labeled as a separate metric. Include test-age equivalents only when a school system requires them, and label them as descriptive rather than diagnostic. Pitfall: the metric trap. A skill-area scaled score of 7 is a modest weakness; a standard score of 70 is two standard deviations below the mean; a table that mixes the two without metric labels invites both misreadings at once. GAC first, then domains in prose. Lead with the GAC as the most complete estimate of adaptive functioning, then move through Conceptual, Social, and Practical in the same order the diagnostic frameworks use, spending the prose on what each result means for this person's daily life at home, at school, at work, and in the community, with skill-area strengths and weaknesses named inside their domains. Pitfall: the score dump. A paragraph per skill area in form order documents that rating happened; it does not answer the referral question. The publisher's own report samples interpret in functional prose, and so should the write-up. Cross-informant comparison. Where two or more raters completed forms, compare the profiles by setting: state where they agree, where they diverge, and what the divergence says about environmental demands and supports. Use the publisher's rater-comparison output where available, and do not average scores across raters. Pitfall: making the discrepancy disappear. Splitting the difference or crowning one rater correct throws away the setting information the disagreement carries; a parent's home-and-community window and a teacher's structured-classroom window are different data, not competing truths. Integration with cognitive results and severity language. Where the adaptive data feed an intellectual-disability question, pair them with an individually administered cognitive measure, state whether deficits are concurrent, and align any severity language with DSM-5-TR, which sets intellectual-disability severity by adaptive functioning rather than IQ. Remember that no single measure may serve as the sole eligibility criterion in US special education ( 34 CFR 300.304(b)(2) ). Pitfall: a severity verdict keyed to the IQ score, or an eligibility ruling announced from one GAC number. The results section supplies calibrated evidence; the diagnostic or eligibility team applies its criteria. Interpretive summary and limitations. Summarize the adaptive picture in two or three sentences, then state the limits: the 2015 United States norms, the settings the raters could and could not see, any estimated-answer concentrations, and, where a prior evaluation used a different adaptive instrument, the documented fact that the ABAS-3 and Vineland-3 yield different absolute scores in the same person. Pitfall: silent instrument switches. Comparing this year's ABAS-3 GAC against last cycle's Vineland composite as if they shared a ruler can manufacture progress or decline that is nothing but instrument difference. Recommendations linkage. Tie each recommended support to a specific skill-area or domain finding, and where you expect to re-measure, say so and name the instrument and interval. Pitfall: recommendations that could follow any profile. If a community-safety support does not trace back to the Health and Safety or Community Use findings, either the recommendation or the results section is incomplete. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: ABAS-3 Forms: [Parent / Teacher / Adult (self-report or rated-by-others) / Parent-Primary Caregiver / Teacher-Daycare] Norms: [US, 2015] Rater(s): [name / relationship / settings observed] Administration: [paper / publisher online platform] Date(s): RATER CONTEXT AND VALIDITY [Opportunity to observe, duration and recency of contact; estimated ("guessed") answer count per form; statement that each protocol was fully scorable and is a valid estimate of current functioning, or the stated limitation if not] SCORE SUMMARY (table, metrics labeled) [GAC + Conceptual, Social, Practical = STANDARD scores (M 100, SD 15) with 95% CI and percentile, per rater; skill areas = SCALED scores (M 10, SD 3), separate metric; test-age equivalents only if the receiving system needs them] GAC AND DOMAIN RESULTS (DSM order, functional prose) GAC first, then Conceptual / Social / Practical: [what each means for daily functioning; skill-area strengths and weaknesses named inside their domains] CROSS-INFORMANT COMPARISON (if more than one rater) [Agreements, divergences, and what the divergence says about settings; rater-comparison output used; scores never averaged] COGNITIVE INTEGRATION AND SEVERITY LANGUAGE [Concurrent deficits stated; severity language keyed to adaptive functioning per DSM-5-TR; single-measure limit acknowledged] INTERPRETIVE SUMMARY AND LIMITATIONS [Two-to-three-sentence adaptive picture; norms and settings limits; estimated-answer note; no silent instrument switches] RECOMMENDATIONS LINKAGE [Each support tied to a named skill-area or domain finding; re-measurement instrument and interval stated] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample ABAS-3 write-up (fictional) Scenario: a 9-year-old receiving special-education services is re-evaluated for continued eligibility in the intellectual-disability category. Adaptive functioning was rated on the ABAS-3 by his mother (Parent Form) and his classroom teacher (Teacher Form), alongside cognitive testing reported elsewhere. The profile shows coherent, significant adaptive deficits across both settings with a modest, setting-consistent parent-teacher difference in the Conceptual domain. This is the adaptive-behavior results section only, condensed but structurally complete. All details are fictional. Client: D.R., 9 (grade 4) · Referral: three-year re-evaluation, continued eligibility (intellectual disability) · Raters: mother (Parent Form); classroom teacher (Teacher Form) · Evaluator: S. Nguyen, PhD, NCSP, School Psychologist · Form dates: 07/08/2026 and 07/10/2026 · Report date: 07/17/2026 Measures and administration: Adaptive behavior was assessed with the Adaptive Behavior Assessment System, Third Edition (ABAS-3), scored against the instrument's United States normative sample. D.R.'s mother completed the Parent Form and his classroom teacher of ten months completed the Teacher Form, both on paper. The General Adaptive Composite (GAC) and the Conceptual, Social, and Practical domains are standard scores with a mean of 100 and a standard deviation of 15, each reported with its 95% confidence interval (CI) and percentile rank. Skill areas are scaled scores with a mean of 10 and a standard deviation of 3; these are a different metric from the domain standard scores and are not directly comparable to them. The same instrument was used at D.R.'s 2023 evaluation, so scores are compared without an instrument-change limitation. Rater context and validity: D.R.'s mother is his primary caregiver and observes him daily at home and in the community; she flagged two answers as estimates, both in the Community Use skill area. His teacher has taught him daily in a general-education classroom with resource support for ten months and flagged no estimated answers. Both protocols were complete and fully scorable, and both are considered valid estimates of D.R.'s current adaptive functioning in the settings each rater observes. GAC and domains by rater (standard scores, mean 100, SD 15) Domain Parent: standard score (95% CI) Parent: percentile Teacher: standard score (95% CI) Teacher: percentile Conceptual 70 (66-76) 2 66 (61-72) 1 Social 75 (69-81) 5 73 (67-80) 4 Practical 71 (66-78) 3 70 (65-77) 2 General Adaptive Composite 68 (65-71) 2 65 (62-69) 1 Parent Form skill areas (scaled scores, mean 10, SD 3) Domain Skill area Scaled score Conceptual Communication 4 Conceptual Functional Academics 4 Conceptual Self-Direction 5 Social Social 5 Social Leisure 6 Practical Self-Care 5 Practical Home Living 6 Practical Community Use 4 Practical Health and Safety 5 GAC and domain results: On the Parent Form, D.R.'s GAC of 68 (CI 65-71) falls in the Extremely Low range, as high as or higher than approximately 2% of children his age; the Teacher Form GAC of 65 is consistent, and the two confidence intervals overlap. Because the GAC is the most complete estimate of adaptive functioning, it anchors this section. In the Conceptual domain (parent 70; teacher 66), both raters describe short, concrete communication, early-primary academic skills used well below grade placement, and reliance on adult structure to start and finish multi-step tasks; the skill-area profile is uniform, with Communication and Functional Academics scaled scores of 4. In the Social domain (parent 75; teacher 73), D.R. is described as friendly and motivated toward peers, with play interests and social judgment resembling a younger child's; this is his relative strength, though still well below average. In the Practical domain (parent 71; teacher 70), he manages familiar self-care routines with reminders, and Community Use (scaled score 4) is the leading concern at home, where he does not yet cross streets or make small purchases independently. Cross-informant comparison: The parent and teacher profiles agree closely; the one notable difference is a slightly lower teacher Conceptual score, consistent with the heavier functional-academic and independent-work demands of the classroom relative to home. The profiles were not averaged; each is interpreted as a valid picture of the setting it samples, and together they establish that the deficits are cross-setting rather than situational. Cognitive integration and severity language: Cognitive testing reported earlier in this evaluation placed D.R.'s overall intellectual functioning well below the average range (composite in the low 60s), so significantly subaverage intellectual functioning and adaptive-behavior deficits are concurrent, and school records document these needs from preschool entry forward. Consistent with DSM-5-TR, any severity description in the diagnostic summary is anchored to D.R.'s adaptive functioning, the level of support he requires day to day, rather than to the IQ score. No single measure serves as the sole basis for eligibility; these adaptive findings join cognitive results, records review, and teacher and parent input before the team's determination. Summary and limitations: The ABAS-3 documents significant, cross-setting adaptive deficits (GAC 68 parent, 65 teacher, Extremely Low range) that are coherent across domains and consistent with the 2023 results, supporting continued intellectual-disability eligibility for the team's consideration. Scores reflect the 2015 United States norms, each rater's observational window, and two parent-flagged estimated answers in Community Use, none of which alters the interpretation. Linkage to recommendations: The Conceptual findings (Communication and Functional Academics scaled scores of 4) support recommendations 1 and 2, continued specialized academic instruction and structured task-initiation supports. The Community Use finding supports recommendation 3, a community-based instruction goal targeting street safety and purchasing skills. The Social relative strength supports recommendation 4, peer-mediated activities that build on his interest in other children. Re-measurement with the ABAS-3 at the next annual review will use the same forms and raters where possible. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The two metrics stay separate and labeled: the GAC and domain standard scores (mean 100, SD 15) never share a column with skill-area scaled scores (mean 10, SD 3), and every number carries its metric, confidence interval, and percentile. Both raters are named with their relationship, settings, and opportunity to observe, and the estimated-answer record is stated, so the validity of each profile can be judged rather than assumed. The parent and teacher profiles are reconciled by setting, not averaged: the write-up says what the difference means instead of erasing it. Every number matches its narrative, and the story leads with the GAC before the domains, in the same Conceptual, Social, Practical order the diagnostic frameworks use. Severity language is anchored to adaptive functioning rather than IQ, the single-measure limit is acknowledged, and the section states what the adaptive data do and do not establish. The edition, norm set, and administration dates are named, and every recommendation traces to a specific skill-area finding, which makes the section comparable at the next re-evaluation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decisions it will feed, and label each requirement by its strength. In US special education, adaptive-behavior evidence is genuinely required for an intellectual-disability identification: the federal definition pairs "significantly subaverage general intellectual functioning" with concurrent "deficits in adaptive behavior" ( 34 CFR 300.8(c)(6) ), and no single measure may be the sole criterion. That is LAW; the instruments, cutoffs, and number of deficit areas are set by each state, which is POLICY, so the write-up should not assert a single national number. In clinical diagnosis, DSM-5-TR keys intellectual-disability severity to adaptive functioning, the American Psychiatric Association's own summary putting it as "the severity of impairment based on adaptive functioning rather than IQ test scores alone," which is the prevailing CONVENTION and the reason the ABAS-3 narrative, not the IQ table, carries the severity specifier. AAIDD's current definition runs the same direction and now places onset "before the age of 22," matching the Social Security Administration's Listings 12.05 and 112.05, which require significant adaptive deficits with pre-age-22 onset; a widespread belief about SSA is simply wrong, because the agency states, "We do not require the results of an individually administered standardized test of adaptive functioning," and will consider such scores when the record already holds them. In Atkins-line capital litigation, Moore v. Texas (2017) directs courts to current clinical standards and notes that "the medical community focuses the adaptive-functioning inquiry on adaptive deficits," so a forensic ABAS-3 write-up that offsets deficits with unrelated strengths invites impeachment. Payer coverage is a separate layer: psychological and neuropsychological testing services bill under the 96130 through 96139 code families as payer policy, and the ABAS-3 is one instrument inside a medically necessary battery rather than a standalone service. Edition currency, access rules, and cross-border logistics round out defensibility. As of July 2026 the ABAS-3 (2015) is the current edition with no fourth edition announced, and its norms are a 4,500-person United States sample, which is a stated limitation when the instrument is used in Canada or Australia, where it is widely accepted but not separately normed. Purchasing runs through the publisher's qualification system: WPS lists the ABAS-3 at its Level C, and the publisher is explicit that it qualifies buyers rather than users, with test use governed by licensing boards and each professional's scope of practice. Distribution shifted recently and matters for Canadian evaluators: effective January 1, 2026, Pearson Canada is no longer authorized to sell or distribute WPS products, the ABAS-3 included, so current materials and online scoring come directly from WPS, a purchasing detail worth checking before a re-evaluation cycle. In Ontario, adult developmental-services eligibility under the Services and Supports to Promote the Social Inclusion of Persons with Developmental Disabilities Act turns on significant limitations in both cognitive and adaptive functioning with onset before age 18, confirmed by a psychologist or psychological associate registered with the College of Psychologists, so note the before-18 framing there against the before-22 language in AAIDD and SSA materials. In Australia, NDIS access decisions run on the agency's Applying to the NDIS operational guideline, and evaluators conventionally support intellectual-disability access requests with both cognitive and adaptive evidence from a registered psychologist. Whatever the jurisdiction, two conventions do steady work: document each rater's opportunity to observe and the estimated-answer record, and when raters disagree, interpret by setting with the publisher's comparison output rather than averaging, because the disagreement is usually the most clinically useful finding on the page. ABAS and ABAS-3 are trademarks of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common ABAS-3 write-up errors reviewers flag There is no payer audit series for adaptive-behavior write-ups; the accountability literature here is psychometric and legal, and it is specific. The ABAS-3 carries independent review in the Buros Center's 20th Mental Measurements Yearbook (2017). A 2025 appraisal of adaptive measures in Behavioral Sciences concluded that across instruments "the evidence for several key psychometric characteristics is either unavailable or suggests less-than-desirable properties," while noting the ABAS-3 manual reports inter-rater reliability in the good-to-excellent range. A peer-reviewed comparison in the Journal of Autism and Developmental Disorders found "the ABAS-3 revealed lower levels of impairment than the VABS-3," with composite differences reaching 16 standard-score points. And in Moore v. Texas the Supreme Court scrutinized exactly how adaptive-functioning evidence was written and weighed. Those findings predict the write-up failures. The BastionGPT Clinical Advisory Board sees the same errors most often in ABAS-3 reviews: The metric trap. A skill-area scaled score (mean 10, SD 3) is read against the standard-score metric (mean 100, SD 15), or the two share a table column without labels. A scaled score of 7 is a modest weakness; a standard score of 70 sits two standard deviations down; the write-up must name the metric beside every number. Informants averaged away. A parent-form and teacher-form disagreement gets split down the middle, or the more impaired profile is quietly preferred. The instrument is built for multiple raters and provides comparison output; the defensible sentence says what the setting difference means, and no arithmetic erases it. Instruments swapped as equivalents. This year's ABAS-3 is compared against last cycle's Vineland-3 as if they shared a ruler, when published comparisons show the ABAS-3 running higher in the same individuals, by as much as a full standard deviation on parallel composites. Progress and decline claims built on an instrument switch are artifacts unless the switch is named as a limitation. Severity keyed to the wrong number. The narrative assigns intellectual-disability severity from the IQ score, or announces eligibility from a single GAC value. DSM-5-TR keys severity to adaptive functioning, US special-education law bars any single measure as the sole criterion, and Moore directs the inquiry to adaptive deficits rather than offsetting strengths. Rater context and estimated answers undocumented. The report never says how well each rater knows the person, which settings they see, or how many answers the form records as guessed. Publisher guidance treats heavily guessed skill areas cautiously, a 2025 study found scores largely invariant to guessing, and opposing counsel and appeal reviewers read silence here as concealment; one documenting sentence per rater closes the gap. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (GAC, domains, skill areas, and the forms and raters used) and get a drafted results-section narrative organized GAC-first and domain by domain, with the standard-score and scaled-score metrics kept distinct and rater context and validity language in place for your review. Cross-check a finished draft for the gaps reviewers flag: a scaled score read against the wrong mean, numbers that disagree with their interpretation, informants averaged instead of reconciled, and an unnamed form, edition, or norm set. Translate the results section into a plain-language summary for parents, caregivers, and eligibility teams that explains adaptive functioning in everyday terms without leaking test content. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do ABAS-3 scores actually mean? Three levels, two metrics, and the metrics are not comparable. The General Adaptive Composite (GAC) summarizes overall adaptive functioning, and the three domains (Conceptual, Social, Practical) sit beneath it; all four are standard scores with a mean of 100 and a standard deviation of 15, each reported with a confidence interval and a percentile rank. The 11 skill areas beneath the domains (9 or 10 appear on any single form) are scaled scores with a mean of 10 and a standard deviation of 3. Because the metrics differ, a skill-area score of 7 is a modest weakness while a standard score of 70 is two standard deviations below the mean, so every number should carry its metric label. Scores also map to descriptive categories running from Extremely Low to High, and test-age equivalents exist for school systems that require them, though they are descriptive rather than diagnostic. ABAS-3 or Vineland-3, and can I switch between them? Choose by structure and setting, and then stay consistent. The ABAS-3 is rating-forms-only, spans birth to 89, and organizes results into the Conceptual, Social, and Practical domains that DSM-5-TR and AAIDD use, which makes the write-up map cleanly onto diagnostic language. The Vineland-3 adds a semi-structured interview option and is often preferred for early childhood and for precision in more severely affected people. They are not interchangeable: a peer-reviewed comparison found ABAS-3 composites running higher than their Vineland-3 counterparts, by as much as 16 standard-score points, enough to move a person across a diagnostic threshold. Keep longitudinal files on one instrument where possible, and when a prior evaluation used the other, name the switch as a limitation instead of comparing the numbers directly. Is the ABAS-3 current, or is an ABAS-4 coming? Keep testing on the ABAS-3. As of July 2026 the 2015 third edition is current, and no fourth edition is announced, released, or listed as coming soon on the publisher's product page . Publishers do not pre-announce far ahead, so treat edition status as a fact to re-verify before a high-stakes report. One purchasing fact changed recently and is worth more attention than the edition question: Western Psychological Services (WPS) publishes the ABAS-3, Pearson only distributes it, and Pearson Canada's authorization to sell WPS products ended January 1, 2026, so Canadian evaluators now order materials and online scoring directly from WPS. Whatever the storefront, the write-up rule is constant: name the edition and the norm set in the report. Is the ABAS-3 required for an intellectual-disability diagnosis, an IEP, or SSA disability? No decision-maker names the ABAS-3 specifically; what they require is adaptive-behavior evidence. For US special-education eligibility, the federal intellectual-disability definition requires concurrent deficits in adaptive behavior ( 34 CFR 300.8(c)(6) ), but instruments and cutoffs are set by each state, and no single measure may be the sole criterion. For SSA disability, Listing 12.05 requires significant adaptive deficits with onset before age 22, yet SSA states it does not require a standardized adaptive test and will consider one when the record already contains it. For clinical diagnosis, DSM-5-TR requires adaptive deficits and keys severity to adaptive functioning without naming an instrument. So the ABAS-3 is often the right evidence, and it is rarely, strictly speaking, mandatory. A rater guessed on a lot of items. Are the scores still valid? Usually yes, and the duty is to document rather than discard. ABAS-3 forms let raters flag answers they estimated instead of observed, and the publisher's guidance treats skill areas with several guessed answers cautiously. The best available evidence is reassuring: a 2025 study in the European Journal of Psychological Assessment tested protocols with and without guessed answers across the Parent, Teacher, and Adult forms and found the subscales measurement-invariant, meaning guessed answers did not meaningfully change what the scores measure. The defensible write-up states the estimated-answer count for each rater, notes where the guesses concentrate, and, when a skill area rests heavily on estimates, weighs whether a better-placed rater should complete an additional form. What undermines validity is not the guessing; it is a report that never mentions it. Can adults rate themselves, and is a self-report enough? The Adult Form (ages 16 to 89) comes in two versions, Rated by Others and Self-Report, and that pairing is the point. Self-report brings the person's own perspective and belongs in the record, but in intellectual-disability, guardianship, and capacity contexts it can overestimate independence because insight into one's own limitations is often part of what is impaired. Best practice collects the Self-Report alongside a Rated-by-Others form from someone with sustained, recent contact, reconciles the two in prose, and says which evidence anchors each conclusion. When the two diverge, the divergence itself is a finding worth a sentence, and decisions about supports should not rest on the self-report alone. The parent and teacher forms disagree. Which one is right? Usually both. The ABAS-3 is built for multiple raters precisely because adaptive behavior is setting-dependent: a parent sees home and community demands, a teacher sees functional academics and independent work in a structured classroom, and the publisher provides comparison output to examine the differences directly. The two moves that fail review are averaging the raters into a number nobody reported and quietly preferring whichever profile fits the referral. The move that survives review is a sentence that attributes the difference to settings and observation opportunity, plus a check that each rater actually knows the person well enough to rate them; a rater with thin contact is a validity problem, not a tiebreaker. Cross-setting agreement, where it exists, is powerful evidence that deficits are not situational. Can I see the actual ABAS-3 questions, or include them in a report? No. Items, record forms, and scoring materials are protected test content: WPS restricts the ABAS-3 to its qualification Level C and is explicit that it qualifies purchasers while test use is governed by licensing boards and each professional's scope of practice, and psychologists are ethically obligated to maintain test security under their ethics code and publisher agreements. Item exposure also damages the instrument itself, because norms assume the content has not been rehearsed. A report should carry scores, descriptors, and interpretation, never item text, and records requests that touch raw protocols route through your test-security obligations rather than reflexive printing. What a family can usefully receive is the scores with a plain-language account of what each domain means for daily life and what supports follow. Can BastionGPT draft my ABAS-3 results section? Yes. Paste your score summary (GAC, domains, skill areas, and the forms and raters used) and it drafts the results-section narrative for your review: GAC first, domains in prose, the standard-score and scaled-score metrics kept distinct, and rater context and validity language in place. It can also cross-check a draft you wrote for the errors reviewers flag, a scaled score read against the wrong mean, informants averaged instead of reconciled, or an unnamed form or norm set, and produce a plain-language summary of the findings for parents and eligibility teams. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from the scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Western Psychological Services, ABAS-3 product page : the 2015 publication, authorship, qualification Level C, the five forms and their age bands, the score metrics and descriptive categories, the 4,500-person US norm sample, and the absence of any fourth edition as of July 2026. Pearson Clinical Australia ABAS-3 page and Pearson Canada WPS notice : the distributor-not-publisher statement and the January 1, 2026 end of Pearson Canada's authorization to sell WPS products. 34 CFR 300.8(c)(6) and 34 CFR 300.304(b)(2) : the IDEA intellectual-disability definition requiring adaptive-behavior deficits, and the rule that no single measure may be the sole eligibility criterion. Social Security Administration, Listing 12.00/12.05 (adult) and childhood Listing 112.05: adaptive deficits with onset before age 22, and the provision that a standardized adaptive test is not required but will be considered if it already exists. US Supreme Court, Moore v. Texas , 581 U.S. 1 (2017), with Atkins v. Virginia (2002) and Hall v. Florida (2014): adaptive-functioning inquiry follows current clinical standards and focuses on deficits. Tamm, Day & Duncan, 2022, Journal of Autism and Developmental Disorders : the ABAS-3 showing less impairment than the Vineland-3 in the same individuals, with differences up to 16 standard-score points. Santisi, Ciancaleoni, Borghetto, Ferri, Landi & Lang, 2025, European Journal of Psychological Assessment : the guessed-answer indicator and the finding that ABAS-3 subscales are measurement-invariant with respect to guessed answers. Frazier, Youngstrom, Frazier & Uljarevic et al., 2025, Behavioral Sciences : the psychometric-evidence appraisal across adaptive measures and the manual-reported inter-rater reliability characterization. Buros Center for Testing, Tests reviewed in the Twentieth Mental Measurements Yearbook (2017): the independent review record for the ABAS-3. AAIDD, Defining Criteria for Intellectual Disability and Balboni, Tassé et al., 2014, Research in Developmental Disabilities : the conceptual, social, practical framework with onset before age 22, and the DABS diagnostic sensitivity and specificity figures. ## Related templates Vineland-3 Report Write-Up Psychoeducational Evaluation Report Psychological Evaluation Report ADOS-2 Report Write-Up Neuropsychological Evaluation Report Bayley-4 Report Write-Up Child & Adolescent Developmental Assessment NDIS Functional Capacity Assessment Report DAYC-2 Report Write-Up Battelle Developmental Inventory (BDI-3) Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ados-2-report All note templates # ADOS-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The ADOS-2 (Autism Diagnostic Observation Schedule, Second Edition) is a semi-structured observation used to assess social communication, play, and restricted or repetitive behaviors when autism is a question. Psychologists and other trained clinicians use it as one input to an autism evaluation, and it yields a classification, not a diagnosis. This page covers how to write up ADOS-2 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, developmental pediatricians, and clinicians trained on the ADOS-2; publisher qualification level C Audience Autism evaluation teams, physicians, IEP and early-intervention teams, payers such as TRICARE, NDIS reviewers, parents Typical length 300 to 600 words for the ADOS-2 section · administration 40 to 60 minutes per module Format family Semi-structured diagnostic observation When it's used Autism diagnostic evaluations across the lifespan, ages 12 months to adulthood; ABA authorization, school, and NDIS support requests Standards context Published by Western Psychological Services (2012); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the ADOS-2? The Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) is a semi-structured, standardized observation used to assess social communication, reciprocal interaction, play, and restricted or repetitive behaviors when autism is part of the question. Western Psychological Services published it in 2012, and it descends from the original ADOS (2001) and the earlier ADOS-G. It has five modules, the Toddler Module and Modules 1 through 4, chosen jointly by chronological age and expressive-language level, and it spans ages 12 months to adulthood, with each module taking about 40 to 60 minutes to administer. Modules 1 through 4 yield one of three classifications, autism, autism spectrum, or non-spectrum, while the Toddler Module yields ranges of concern rather than a classification. As of July 2026 the 2012 edition remains current, with no ADOS-3 or new norming announced. Beneath the classification, coded behaviors group into two domains, Social Affect and Restricted and Repetitive Behavior, that mirror the DSM-5 structure and sum to an overall total the algorithm compares against module-specific thresholds; a comparison score, also called the calibrated severity score, places severity on a 1 to 10 metric for Modules 1 through 3, with a Module 4 version added later by Hus and Lord (2014). This page reproduces none of those items, activities, or cutoff values. The load-bearing point for write-ups is simpler: an ADOS-2 result is a classification, not a diagnosis. WPS states plainly that it qualifies people to purchase the test, not to use it, and a defensible autism diagnosis under DSM-5-TR integrates the observation with developmental history (often the ADI-R), rating scales such as the SRS-2 or CARS-2, and cognitive and adaptive testing. The observation is evidence that feeds a whole evaluation , not the verdict itself. ## Who uses ADOS-2 write-ups and when Autism diagnostic clinics, developmental-behavioral pediatricians, child and school psychologists, and adult autism services use the ADOS-2 as the direct-observation component of an evaluation, choosing the module by the person's age and expressive-language level. It rarely stands alone: convention pairs it with a developmental-history interview (often the ADI-R), a trait or rating scale such as the SRS-2 or CARS-2, and cognitive and adaptive measures, and some teams add a qualitative, neurodiversity-affirming process such as MIGDAS-2 for subtle or high-masking presentations. Those cognitive and behavioral pieces are where the write-up connects to the rest of the battery: a cognitive measure and a behavior rating scale often sit alongside it, and the whole set anchors a psychological evaluation report or, where medical and neurologic questions are in play, a neuropsychological report . One boundary worth stating in the report itself: the ADOS-2 was not designed for video administration and is not validated remotely, so a remote observation belongs in a limitations line, not a silent footnote. ## How to structure an ADOS-2 results section No statute, payer, or publisher mandates a format for the ADOS-2 section of an autism evaluation. The sequence below is the convention experienced evaluators converge on because it survives review, and because it is built around one rule the top search results routinely break: report the classification and describe behavior in your own words, and leave the test's activities, item scores, algorithm totals, and cutoff numbers out. That omission is a test-security obligation under APA Ethics Standard 9.11 and WPS policy, not an oversight. Each section below carries the pitfall that most often undermines it. Instrument, module, and administration. Name the ADOS-2 and its 2012 edition, the module administered and why it was chosen (chronological age and expressive-language level), who administered it and their ADOS-2 training, and that it was one component of the evaluation. Pitfall: naming the ADOS-2 with no module, or writing it up as though the observation alone produced the diagnosis. Behavioral observations, in your own words. Describe what you saw: quality of eye contact, gesture, and shared attention; reciprocity and conversational give-and-take; play and imagination; restricted interests; and sensory or repetitive behavior. Pitfall: reproducing the standardized activities or presses. Describing the tasks themselves exposes protected content and erodes the norms, which assume the person has not seen them. Classification, reported not decoded. State the classification the module yielded, autism, autism spectrum, or non-spectrum, or a range of concern for the Toddler Module, and, if you use the comparison score, describe its severity band in words. Pitfall: printing algorithm totals, item-level scores, or cutoff values, or presenting the 1 to 10 comparison score as a fixed clinical truth. Integration with history and other measures. Place the observation next to developmental history, caregiver report, and any rating, cognitive, or adaptive data, and say where the sources agree and where they diverge. Pitfall: an ADOS-2 section that never connects to the rest of the battery, leaving the reader to guess whether the classification fit the wider picture. Diagnostic formulation under DSM-5-TR. State the diagnostic determination and the clinical judgment behind it, with the ADOS-2 named as one input among several. Pitfall: a diagnosis that rests on the classification alone. An autism classification is not a DSM-5-TR diagnosis, and a non-spectrum classification does not rule autism out. Limitations and validity of the session. Note anything that bears on interpretation: language, setting, in-person versus remote administration, masking or camouflage, and co-occurring conditions such as anxiety, trauma, or psychosis that can shift the picture. Pitfall: silence on limitations, especially for high-masking presentations in girls and women, where the observation can under-detect autism. Recommendations linkage. Tie each recommendation, whether supports, services, or further assessment, to a specific finding. Pitfall: recommendations that could follow any evaluation and trace back to nothing in the observation. ### Blank template (copy and adapt) ADOS-2 RESULTS SECTION Instrument/edition: ADOS-2 (2012) Module: [Toddler / 1 / 2 / 3 / 4] Module selected because: [chronological age + expressive-language level] Administered by: [name, credentials, ADOS-2 training] Date: [ ] BEHAVIORAL OBSERVATIONS (narrative, in your own words) [Social communication, reciprocity, eye contact, gesture, play and imagination, conversation, restricted interests, sensory and repetitive behavior; describe what was observed, not the tasks] CLASSIFICATION [Modules 1-4: autism / autism spectrum / non-spectrum] [Toddler Module: range of concern] [Comparison score band in words if used; no algorithm totals or cutoffs] INTEGRATION WITH HISTORY AND OTHER MEASURES [Developmental history, caregiver report, rating scales, cognitive and adaptive data; state where they converge or diverge] DSM-5-TR FORMULATION [The diagnostic determination and the clinical judgment behind it; the ADOS-2 is one input, not the diagnosis] LIMITATIONS AND VALIDITY OF THE SESSION [Language, setting, in-person vs remote, masking, co-occurring conditions that can affect the observation] RECOMMENDATIONS LINKAGE [Each recommendation tied to a specific finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample ADOS-2 write-up (fictional) Scenario: a 14-year-old is referred by their pediatrician for social difficulties and long-standing anxiety, with a history of early language on time but persistent trouble making and keeping friends. Module 3 was administered as the direct-observation component of a broader autism evaluation. This is the ADOS-2 section only, and it deliberately reports the classification and narrative behavior without item content, presses, or cutoff numbers, which are protected. All details are fictional. Client: J.T., 14 · Referral: pediatrician, social difficulty and anxiety · Evaluator: R. Alvarez, PhD, Licensed Psychologist (ADOS-2 trained) · Testing date: 07/10/2026 · Report date: 07/16/2026 Instrument, module, and administration: The Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) was administered as one component of an autism evaluation. Module 3 was selected because J.T. is a verbally fluent adolescent, and the module was chosen on the basis of chronological age and expressive-language level. Administration followed standardized procedures in person, and the session is considered a valid sample of current behavior. Behavioral observations: J.T. was cooperative and conversational throughout. Speech was fluent and vocabulary was strong, but conversation ran largely one way, with limited back-and-forth and few conversational bids offered to the examiner. Eye contact was present but not well coordinated with speech and gesture to manage the interaction. J.T. described an intense, detailed interest that recurred across topics, and reported discomfort with certain sounds and textures. Rapport was pleasant, and difficulties appeared in the quality of reciprocal social communication rather than in willingness to engage. Classification: On Module 3, J.T.'s pattern of social-affect and restricted, repetitive behaviors met the threshold for the autism spectrum classification. The comparison score placed severity in a moderate band relative to same-age peers with autism. Consistent with test-security convention, algorithm totals and cutoff values are not reproduced here. Integration with history and other measures: The observation converges with the developmental and rating-scale data. The caregiver developmental history documents early and persistent difficulty with peer relationships, and a completed social-communication rating scale fell in the elevated range. The reported anxiety is longstanding and real, but the social-communication pattern is present across settings and is not fully explained by anxiety alone. DSM-5-TR formulation: Integrating the ADOS-2 autism spectrum classification with the developmental history, the rating-scale results, and clinical judgment, J.T. meets DSM-5-TR criteria for autism spectrum disorder, requiring support (Level 1), with co-occurring anxiety documented as an associated condition rather than the primary explanation. The ADOS-2 is one input to this determination, not the determination itself. Limitations: The classification reflects behavior sampled in a single in-person session. Adolescents who have learned to compensate socially can present nearer the threshold, so the finding is read alongside history and self-report rather than in isolation. No remote administration was involved. Recommendations linkage: The reciprocal-communication findings support recommendation 1 (social-communication support at school with peer-mediated opportunities). The sensory findings support recommendation 2 (classroom sensory accommodations). The co-occurring anxiety supports recommendation 3 (referral for anxiety-focused therapy that accounts for autistic profile). The diagnosis supports recommendation 4 (eligibility and service discussions by the school and payer teams applying their own criteria). This sample is fictional and for educational purposes. It does not describe a real person or record, and the classification and details are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The classification is reported and its meaning explained, but no item content, presses, algorithm totals, or cutoff numbers appear, so the write-up respects test security and the norms. The module is named along with why it was selected, the person's age and expressive-language level, so a later reader can judge whether the right module was used. The observation is integrated with developmental history and a rating scale, and the summary states where the sources converge, which is what turns a classification into evidence. The diagnosis is attributed to DSM-5-TR criteria and clinical judgment, not to the ADOS-2 alone, and the co-occurring anxiety is named as a companion, not the whole explanation. Limitations are stated plainly, including the masking caveat, and every recommendation traces back to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the ADOS-2 section knowing which decision it will feed, and label the strength of each requirement honestly. No jurisdiction legally requires the ADOS-2 for an autism diagnosis. DSM-5-TR governs the diagnosis, and the American Academy of Pediatrics clinical report ( Hyman, Levy, and Myers, 2020 , reaffirmed October 2025) affirms that clinicians comfortable with DSM-5 criteria can diagnose without mandating any single instrument. What names the ADOS-2 is payer or program policy: TRICARE's Autism Care Demonstration lists it among validated assessment tools for ABA authorization, which is PAYER POLICY, not a clinical rule. Educational eligibility is separate again: under IDEA a school does not require a medical diagnosis or any specific instrument to identify autism, and the team must find that autism adversely affects educational performance, which is STATUTE and CONVENTION working together. In Canada, programs such as the Ontario Autism Program require a written DSM-5 diagnosis from a qualified professional but do not name the ADOS-2 in statute (PROGRAM POLICY). In Australia, the National Disability Insurance Scheme requires a DSM-5 diagnosis and, since the 2024 reforms, evidence of functional impact, and the Autism CRC National Guideline (2018, updated 2023 and NHMRC-approved) recommends standardized observation without mandating a specific tool (PROGRAM POLICY and CONVENTION). Testing and evaluation time is billed under the psychological and developmental testing code families by bare code number (for example 96112, 96113, 96130, 96131, 96132, 96133, 96136, 96137 in the United States, and the relevant MBS item numbers in Australia) as payer policy with plan-specific rules. The format is a convention; the classification, the narrative, and the integration are the requirement. Edition currency and the qualification boundary are the ADOS-2's live defensibility questions. As of July 2026 the ADOS-2, published in 2012, remains the current edition, with no ADOS-3 or new WPS norming announced; what is worth watching is the commercial and normative status of the BOSA and the evolving NDIS functional-evidence rules, not a new edition. Two boundaries belong in the write-up when they apply. First, administration mode: WPS and the ADOS-2 authors hold that the instrument was not designed for video administration and is not validated remotely, so a remote observation is a stated limitation, and the purpose-built adjacent tool, the BOSA, is preliminary. Second, who may use it: the ADOS-2 is qualification level C, and WPS states that it qualifies individuals for test purchase, not for test use, which is set by licensing boards and scope of practice; clinical-workshop training and the separate research-reliability standard are different things, and published sensitivity can exceed real-world clinic performance because of that gap. The ADOS-2 is also sold only as a whole kit, a reminder that its materials are controlled and must not be excerpted into a report. ADOS and ADOS-2 are trademarks of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common ADOS-2 write-up errors reviewers flag There is no payer audit series for autism observation write-ups; the accountability here is ethical and psychometric, and it is pointed. The ADOS-2 has never been reviewed in any Buros Mental Measurements Yearbook , and the widely cited two-author review is McCrimmon and Rostad (2014) in the Journal of Psychoeducational Assessment, a journal review rather than a Buros one, so do not label it as Buros. The peer-reviewed evidence is strong but bounded: a 2021 meta-analysis put ADOS-2 sensitivity at .89 to .92 and specificity at .81 to .85, with accuracy in clinical versus research settings described as mixed. The errors below are write-up errors, not administration errors. The BastionGPT Clinical Advisory Board sees the same ones most often in ADOS-2 report reviews: Classification written as diagnosis. Reporting an autism classification as the diagnosis itself, or a non-spectrum classification as ruling autism out. The ADOS-2 is one input; a defensible diagnosis integrates history, rating scales, and clinical judgment under DSM-5-TR. Reproducing protected content. Printing the standardized activities or presses, item scores, algorithm totals, or cutoff numbers. This breaches APA Ethics Standard 9.11 and WPS policy and degrades the norms; report the classification and a narrative description instead. Overstating the comparison score. Presenting the 1 to 10 calibrated severity score as fixed clinical truth. Recent peer-reviewed work reports it can attenuate the measurement signal and shows lower test-retest stability than raw scores, so it belongs in prose, cautiously, not as a verdict. Gold-standard framing that overclaims. Treating the ADOS-2 as definitive. Its sensitivity is lower for high-masking presentations, concentrated in girls and women, and generally lower in clinical than in research settings, so a non-spectrum result does not exclude autism. Module or training unstated, discordant results glossed. Not naming the module and why it was chosen, not noting the administrator's ADOS-2 training, or failing to explain a result that conflicts with the clinical picture, such as an elevated observation driven by anxiety or a non-elevated observation against a strong autistic history. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Draft a results-section narrative from a pasted summary of the module, the classification, and your behavioral observations, organized for your review, with no item content required. Cross-check a draft for the gaps reviewers flag: a classification written as a diagnosis, a non-spectrum result treated as ruling autism out, integration with history left unstated, or protected content that slipped in. Produce a plain-language summary of the findings for parents and referrers that keeps the classification as one input and the diagnosis as clinical judgment. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does an ADOS-2 classification actually mean? The ADOS-2 sorts observed behavior into two domains, Social Affect and Restricted and Repetitive Behavior, and the module algorithm compares the total against its thresholds to return a classification: autism , autism spectrum , or non-spectrum for Modules 1 through 4, or a range of concern for the Toddler Module. A separate comparison score, or calibrated severity score, places severity on a 1 to 10 scale across modules. The essential thing to write down, and to understand, is that a classification is not a diagnosis: it is one input a clinician integrates into a whole evaluation under DSM-5-TR. Report the classification and what you observed, and keep the algorithm totals and cutoff values out of the document. Is the ADOS-2 still current, or is there an ADOS-3? As of July 2026 the ADOS-2, published in 2012, is the current edition, and no ADOS-3 or new WPS norming has been announced. The one meaningful thing to watch is not a new edition but the BOSA (Brief Observation of Symptoms of Autism), a short observation adapted from ADOS-2 codes for situations where a full in-person ADOS-2 is not feasible; it is coded only by ADOS-2-trained clinicians and, as of its 2021 validation, was preliminary and not sold commercially. Test on the current edition, name it and the year in the report, and treat any future edition change as a stated limitation. Is the ADOS-2 required to diagnose autism? No law in the United States, Canada, or Australia requires the ADOS-2 for an autism diagnosis. DSM-5-TR governs, and the American Academy of Pediatrics clinical report (Hyman, Levy, and Myers, 2020, reaffirmed October 2025) affirms that clinicians comfortable with DSM-5 criteria can diagnose without mandating any single instrument. What names the ADOS-2 is payer or program policy : TRICARE's Autism Care Demonstration lists it among validated assessment tools for ABA authorization. Educational eligibility is separate again, since under IDEA a school does not require a medical diagnosis or any specific test. The honest sentence for a report is that the ADOS-2 is convention and, for some payers, policy, not a legal mandate. Does a non-spectrum ADOS-2 rule out autism? No. The ADOS-2 captures behavior in a single session, and people who have learned to compensate socially, a pattern documented especially in girls and women, can present below the threshold while still being autistic. A non-spectrum classification lowers the probability but does not exclude autism, and the report should say so when the developmental history, self-report, or collateral information points the other way. The mirror-image caution also holds: anxiety, trauma-related conditions, and psychosis can elevate an ADOS-2 in someone who is not autistic. Either way, the classification is weighed against the whole clinical picture, not read as a verdict. Can I include the actual ADOS-2 tasks or scores in a report? No. The activities, presses, protocols, and coding materials are protected test materials: psychologists are ethically obligated to maintain test security under APA Ethics Standard 9.11 , and WPS enforces the same through its purchase agreements and sells the ADOS-2 only as a whole kit. Reproducing the tasks also degrades the norms, which assume the person has not rehearsed them. A report carries the classification, your narrative description of behavior, and the interpretation, never the item content or cutoff values. Raw test data, as distinct from test materials, are releasable only under Standard 9.04 with an appropriate release, and still route through your test-security obligations. ADOS-2, ADI-R, SRS-2, or CARS-2, which do I use? They do different jobs and are strongest together. The ADOS-2 is a direct, semi-structured observation ; the ADI-R is a structured caregiver interview about developmental history; the SRS-2 is a quantitative trait scale that helps flag masking when observation is near the threshold; and the CARS-2 is a brief examiner rating scale useful for screening within a battery. A defensible autism evaluation usually pairs a direct observation with a developmental history and one or more rating scales, then integrates all of it under DSM-5-TR. Some teams add a qualitative process such as MIGDAS-2 for subtle presentations. Each is an input; none is the diagnosis. Is a telehealth ADOS-2 valid? No. WPS and the ADOS-2 authors hold that the instrument was not designed for video administration and is not validated remotely, so a remotely administered ADOS-2 should be described as a limitation rather than reported as a standard result. The purpose-built alternative for situations that call for it is the BOSA, a brief observation coded with ADOS-2 codes by ADOS-2-trained clinicians; as of its 2021 validation it was preliminary and not commercially sold, and its authors describe it as a supporting observation, not a standalone diagnostic. When a remote observation is unavoidable, name the method and its limits in the report. Can BastionGPT draft my ADOS-2 write-up? Yes. Paste a summary of the module you used, the classification, and your behavioral observations, and it drafts the ADOS-2 section for your review: the observation in narrative form, the classification stated plainly, and the integration with history and other measures framed so you can confirm it. It can also cross-check a draft you wrote for the gaps reviewers flag, a classification written as a diagnosis, a non-spectrum result treated as ruling autism out, or protected content that slipped in, and produce a plain-language summary for parents and referrers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and because you paste only your summary and observations, no item content leaves your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Western Psychological Services, ADOS-2 product page : 2012 publication, ages 12 months to adulthood, qualification level C, five modules at 40 to 60 minutes each, the three classifications and Toddler-Module ranges of concern, the purchase-not-use statement, and the whole-kit sale. American Psychological Association, Ethical Principles of Psychologists and Code of Conduct : Standard 9.11 on maintaining test security and Standard 9.04 on the release of test data. Lebersfeld, Swanson, Clesi, and O'Kelley, 2021, Journal of Autism and Developmental Disorders : meta-analytic ADOS-2 sensitivity of .89 to .92 and specificity of .81 to .85, with mixed clinical-versus-research accuracy. McCrimmon and Rostad, 2014, Journal of Psychoeducational Assessment (a journal test review, not a Buros review); Buros Center for Testing, Mental Measurements Yearbook : the ADOS-2 has not been reviewed in any MMY. American Academy of Pediatrics, Identification, Evaluation, and Management of Children With Autism Spectrum Disorder (Hyman, Levy, and Myers, 2020; reaffirmed October 2025): clinicians may diagnose under DSM-5 with no single instrument mandated. National Academies of Sciences, Engineering, and Medicine, The Comprehensive Autism Care Demonstration (2025): the TRICARE ACD recognizes the ADOS-2 among validated assessment tools for ABA authorization. Autism CRC, National Guideline for the Assessment and Diagnosis of Autism in Australia (2018; updated 2023, NHMRC-approved): standardized observation recommended, with no specific instrument mandated. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report BASC-3 Report Write-Up WISC-V Report Write-Up Conners 4 Report Write-Up Vineland-3 Report Write-Up SRS-2 Report Write-Up ABAS-3 Report Write-Up NEPSY-II Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up MIGDAS-2 Report Write-Up Autism Spectrum Evaluation Report M-CHAT-R/F Score Documentation Sensory Profile 2 Report Write-Up GARS-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/basc-3-report All note templates # BASC-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The BASC-3 (Behavior Assessment System for Children, Third Edition) is a multi-informant rating system measuring behavioral, emotional, and adaptive functioning from age 2 through college age, with teacher, parent, and self-report forms. Psychologists and school psychologists use it in ADHD, emotional, and autism-related evaluations. This page covers how to write up BASC-3 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and school psychologists; publisher qualification level B Audience IEP and Section 504 teams, referring physicians and therapists, parents and caregivers, payers Typical length 300 to 800 words for the results section · administration 10 to 30 minutes per form Format family Multi-informant behavior rating scale When it's used ADHD, emotional, behavioral, and autism-related referrals; special-education eligibility evaluations; treatment monitoring Standards context Published by Pearson (2015); BASC-4 released in 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the BASC-3? The Behavior Assessment System for Children, Third Edition (BASC-3) is a multi-informant, norm-referenced rating system by Cecil R. Reynolds and Randy W. Kamphaus, published by Pearson in 2015, that measures behavioral, emotional, and adaptive functioning from age 2 through college age. The core forms are the Teacher Rating Scales (TRS) and Parent Rating Scales (PRS), each spanning ages 2 to 21 across three levels, and the Self-Report of Personality (SRP), from age 6 through college; supporting components include a structured developmental history, a classroom observation system, the BESS screener, the Flex Monitor progress tool, and the Parenting Relationship Questionnaire. Results are T scores and percentiles against general or clinical norms, scored on Q-global or by hand . Edition status matters as of July 2026: the BASC-4 published in 2026 (same authors), Pearson's BASC-3 pages now carry a "BASC-4 is now available!" banner, and the BASC-3 remains on sale and supported during the transition. The load-bearing distinction for write-ups: the BASC-3 describes, it does not diagnose. Its scales quantify how often raters report behaviors relative to same-age norms, and its At-Risk and Clinically Significant classifications are bands of normative elevation, not clinical verdicts; the publisher's own generated reports state they "should not be the sole basis for making important diagnostic or treatment decisions." Rating scales are also a different kind of evidence from performance tests: a BASC-3 profile records how a child is perceived across settings, which is why it enters a psychological evaluation report or a diagnostic evaluation as converging context, never as the determinant. A results section that keeps those boundaries straight is most of what separates a defensible BASC-3 write-up from a score printout with sentences around it. ## Who uses BASC-3 write-ups and when School psychologists write up BASC-3 results inside psychoeducational and special-education eligibility evaluations; clinical and counseling psychologists use it for diagnostic clarification in ADHD, anxiety, mood, and autism-related referrals; pediatric neuropsychologists fold it into a neuropsychological report as the behavior-rating layer alongside performance testing; and prescribers and therapists order it when a diagnostic evaluation needs cross-setting measurement an interview cannot supply. It is the right tool when the question is broadband, what is going on across home and school, and which domains deserve a closer look. Narrowband instruments then carry the disorder-specific load: an ADHD-focused scale such as the Conners 4 for symptom-level detail, autism-specific measures for social communication, and adaptive-behavior scales such as the Vineland-3 when the question is daily-living skills rather than problem behavior. The BASC-3 write-up is where all of it gets anchored: multiple raters, one integrated account. ## How to structure a BASC-3 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it leads with what was administered and whether the data are interpretable, organizes findings by composite rather than by page of printout, and keeps the publisher's classification vocabulary (Average, At-Risk, Clinically Significant) doing exactly one job, describing normative elevation. Each section carries the pitfall that most often undermines it. Measures, raters, and norms. Name the instrument and edition, the forms and levels administered (TRS, PRS, SRP), each rater and their relationship and setting, administration dates, and the norm set used. Pitfall: writing "the BASC" with no edition or norm set. With the BASC-4 published in 2026 , an unnamed edition leaves reviewers unable to tell which norms produced your T scores. Validity indexes, first, for every rater. State each protocol's validity classifications and the interpretive consequence in plain language before reporting a single score. Pitfall: silently dropping a flagged protocol, or silently keeping one. An elevated F Index is a caution flag, not an automatic invalidation, and published evidence shows it can reflect genuine distress rather than exaggeration; either way the report must say how you handled it. Composite-level results, raters integrated. Move composite by composite (Externalizing Problems, Internalizing Problems, School Problems, Adaptive Skills, and the Behavioral Symptoms Index), comparing parent and teacher ratings within each domain rather than summarizing each rater's printout in sequence. Pitfall: the score dump. A paragraph per scale in test order documents administration, not evaluation, and buries the pattern the referral question needs. Self-report results. Give the SRP its own weight: composites including the Emotional Symptoms Index and Personal Adjustment, with scale-level elevations described in prose. Pitfall: treating the self-report as a footnote to the adult ratings. For internalizing concerns it is often the only informant with direct access to the symptoms. Content scales and probability indexes, if reported. These are an optional interpretive layer; name the ones you used and why. Pitfall: forgetting that the executive functioning indexes exist "available only via Q-global software reports," in the publisher's words, so a hand-scored protocol cannot produce them and a report citing them implies software scoring. Cross-informant integration. One paragraph that states where raters converge, where they diverge, and what the divergence means. Agreement between different informant types is modest by design: a meta-analysis of 341 studies found a mean cross-informant correlation of r = .28. Pitfall: averaging raters into a single verdict. A teacher-high, parent-average profile is usually information about settings, not evidence that one rater is wrong. Interpretive summary. Answer the referral question directly, using classification language as normative description with its plain-language meaning, and note that scores carry measurement error. Pitfall: letting At-Risk or Clinically Significant stand in for a diagnosis. The bands describe elevation relative to same-age peers; the publisher's own reports state they "should not be the sole basis for making important diagnostic or treatment decisions." Recommendations linkage. Tie every recommendation to a specific finding, and name the monitoring plan: who gets re-rated, with what, and when. Pitfall: recommendations that could follow any profile. If a recommendation does not trace back to an elevated or low scale, either the recommendation or the results section is incomplete. ### Blank template (copy and adapt) MEASURES AND RATERS Instrument/edition: BASC-3 Forms: [TRS / PRS / SRP + level] Raters: [name or role, relationship, setting, date completed] Norm set: [general combined / gender-specific / clinical] VALIDITY [Each protocol's validity classifications + what they mean for interpretation; how any flagged protocol was handled] RESULTS BY COMPOSITE (integrate raters within each domain) Externalizing Problems: [T scores, classification, convergence] Internalizing Problems: [T scores, classification, convergence] School Problems / Adaptive Skills: [T scores, classification] Behavioral Symptoms Index: [T scores, classification] SELF-REPORT (SRP) [Composites incl. Emotional Symptoms Index + Personal Adjustment; scale-level elevations in prose, with percentile context] CONTENT SCALES / INDEXES (if reported): [which, and why] CROSS-INFORMANT INTEGRATION [Where raters converge; what divergence means (setting, perspective, opportunity to observe); do not average it away] INTERPRETIVE SUMMARY (answer the referral question) [Classification language + plain-language meaning; measurement- error caveat; what the scores do NOT establish] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-rating plan + timeline] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample BASC-3 write-up (fictional) Scenario: a pediatrician refers a 14-year-old for evaluation after a routine visit surfaces low mood and a sharp drop in grades. Parent and teacher ratings come back unremarkable while the self-report is elevated, the profile shape that most often gets written up badly. This is the results section only, condensed but structurally complete. All details are fictional. Student: J.M., 14 (grade 9) · Referral: A. Osei, MD (pediatrics), low mood and academic decline · Evaluator: L. Navarro, PhD, Licensed Psychologist · Ratings completed: 07/08 to 07/10/2026 · Report date: 07/16/2026 Measures and raters: The Behavior Assessment System for Children, Third Edition (BASC-3) was administered as part of a broader evaluation: Parent Rating Scales, adolescent level (PRS-A), completed by J.M.'s mother; Teacher Rating Scales, adolescent level (TRS-A), completed by her English teacher, who has taught her daily since January; and the Self-Report of Personality, adolescent level (SRP-A). All forms were scored against the BASC-3 general combined-sex norms. On the clinical scales and composites, higher T scores mean more reported problems; on the adaptive scales and the SRP Personal Adjustment composite, lower scores mean more concern. Validity: Validity indexes for all three protocols, including the F Index, Consistency Index, and Response Pattern Index, fell in the Acceptable range. All three profiles are considered interpretable, and each rater had adequate recent opportunity to observe. Parent and teacher ratings: Across both adult raters, the Externalizing Problems, School Problems, and Behavioral Symptoms Index composites fell in the Average range, and no clinical scale reached the At-Risk band. Internalizing Problems composites were Average for both raters (mother T = 55; teacher T = 52), with the mother's Anxiety rating (T = 57) and the teacher's Withdrawal rating (T = 58) at the upper end of Average. Adaptive Skills composites were Average for both raters, with no adaptive scale in the At-Risk range. Self-report: J.M.'s SRP-A tells a different story. The Anxiety scale fell in the Clinically Significant range (T = 71, approximately the 98th percentile), with At-Risk elevations on Depression (T = 68), Social Stress (T = 66), and Sense of Inadequacy (T = 64). The Internalizing Problems composite (T = 67) and the Emotional Symptoms Index (T = 68), the SRP's most global indicator of emotional distress, were both At-Risk. The Personal Adjustment composite was low (T = 38, At-Risk), driven by low Self-Esteem (T = 39), indicating she reports fewer of the positive-adjustment experiences typical for her age. Cross-informant integration: The divergence between J.M.'s self-report and the adult ratings is a finding, not a contradiction to be averaged. Agreement between different types of informants is expected to be modest, and internal symptoms such as worry, self-criticism, and social distress are exactly the domain adults observe least directly. The pattern here, Average adult ratings alongside Clinically Significant and At-Risk self-report elevations, is most consistent with internal distress that has not yet become visible in classroom behavior or at home, an interpretation supported by her intact adaptive ratings. Summary: BASC-3 results indicate self-reported anxiety in the Clinically Significant range with At-Risk depressive symptoms, social stress, and low self-esteem, against Average parent and teacher ratings and intact adaptive functioning. These classifications describe elevation relative to same-age peers, are subject to measurement error, and are not by themselves a diagnosis. Taken with the interview and history reported elsewhere in this evaluation, they support an internalizing presentation currently carried privately, and answer the referral question: the academic decline is more plausibly downstream of anxiety and mood than of an attention or conduct problem, for which no rater produced elevations. Linkage to recommendations: The self-reported anxiety and depressive elevations support recommendations 1 and 2 (individual therapy with an anxiety focus; structured follow-up of mood symptoms with her pediatrician). The intact adaptive and school-problem ratings support recommendation 3 (no classroom placement change at this time). Recommendation 4 sets the monitoring plan: re-administer the SRP-A and both adult rating forms in 12 weeks to test whether treatment narrows the self-report elevations. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or protocol. ↑ Back to the template and downloads ### Why this sample works Every number matches its narrative: each T score appears with its classification, and no sentence upgrades or downgrades a band label along the way. Validity is addressed first for all three raters, so the interpretations that follow have a stated foundation. The edition, forms, raters, and norm set are named, which makes the T scores interpretable now and comparable at re-evaluation. Informant divergence is interpreted as setting and perspective information instead of being averaged, and the integration paragraph says so explicitly. The summary states what the scores do not establish (a diagnosis), and every recommendation traces to a specific finding, with a dated re-rating plan. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing what decisions it will feed. In US schools, BASC-3 data commonly support eligibility determinations for emotional disturbance, other health impairment, and autism under IDEA, and the law sets the frame: evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ), so the write-up should present the BASC-3 as one converging data source, never as the deciding instrument. The federal emotional-disturbance definition adds a trap the write-up has to navigate: it excludes children who are "socially maladjusted" without ever defining the term ( 34 CFR 300.8(c)(4)(ii) ), and Pearson's own webinar materials acknowledge that social maladjustment "has not been operationally defined by Federal IDEIA legislation" and that state practice varies, which is why its ED qualification scales are framed as aggregation aids, not eligibility answers; specific criteria are state law. Section 504 prescribes no test list at all. On the payer side, rating-scale work is billed inside the psychological-testing code families as payer policy, not law: evaluation services under 96130 and 96131, administration and scoring under 96136 through 96139, with 96146 for a single automated instrument, and the current APA Services billing guide describes evaluation services as including the integration and report writing this page covers. Medicare draws one bright line worth quoting in coverage disputes: the statute "does not extend coverage to screening procedures" ( WPS article A57481 ), and developmental and behavioral screening instruments have their own code family (96110, 96112, 96113, 96127) separate from testing, so a BASC-3 administered as universal screening is a different billing animal from one administered inside a medically necessary evaluation. In Canada, school assessment and identification run on provincial frameworks rather than a national statute, with a validated Canadian French adaptation of the BASC-3 available; in Australia, behaviour-rating results serve as supporting evidence in school-adjustment and disability processes by convention, while the enforceable record rules come from the Psychology Board's code of conduct : reports true and objective, records retained seven years, or until age 25 for child clients. Edition currency is now a defensibility question of its own. As of July 2026 the BASC-4 core forms and BESS have published (December 2025 pre-sales, 2026 publication, Reynolds and Kamphaus again), Pearson lists BASC-4 narrowband forms for ADHD, autism, and mood as coming soon, several BASC-3 components including the PRQ remain the current versions, and Pearson has announced no BASC-3 retirement date, with Q-global scoring still supported. During a transition, the defensible practices are boring and specific: name the edition and norm set in every report, finish an evaluation on the edition it started on, expect mixed-edition batteries across a district or clinic for a while, and treat cross-edition score comparisons (a BASC-3 baseline against a BASC-4 re-evaluation) as a limitation to state in text rather than a footnote to hope nobody notices. Access boundaries are also publisher-specific: BASC-3 protocols are qualification level B, purchase and administration are restricted accordingly, and Pearson's generated reports carry their own trade-secret handling notices, so route any records request that touches raw protocols through your test-security obligations rather than reflexively printing the packet. The write-up itself, your integration and conclusions, belongs in the ordinary record and should be written to be read there. BASC and Pearson are trademarks, in the US and other countries, of Pearson plc; the BASC-3 is published by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common BASC-3 write-up errors reviewers flag There is no payer audit series for behavior-rating write-ups; the accountability literature here is psychometric, and it is sharper than most template pages admit. The core BASC-3 was reviewed in the Buros Center's 20th Mental Measurements Yearbook , independent factor analyses of the teacher form found the composite scales may lack sufficient unique variance for confident stand-alone interpretation, the SRP-A F Index can misread genuinely depressed adolescents as over-reporters, and cross-informant agreement averages r = .28 across 341 studies . Those four findings predict almost every write-up failure. The BastionGPT Clinical Advisory Board sees the same errors most often in BASC-3 write-up reviews: Scale elevations written as diagnoses. "Hyperactivity is Clinically Significant" becomes "results indicate ADHD." The bands describe normative elevation; the publisher's own reports state the output "should not be the sole basis for making important diagnostic or treatment decisions," and the diagnostic conclusion belongs to the whole evaluation, not one instrument. Informant disagreement averaged away. The report splits the difference between a teacher-high and parent-average profile, or crowns one rater correct. Modest agreement is the empirical norm, and divergence usually carries setting information; the strong move is a sentence that says what the discrepancy means, not arithmetic that makes it disappear. Descriptor drift. The same T score is Average in one paragraph and "concerning" in the next, or At-Risk quietly hardens into a clinical finding by the summary. Reviewers read for classification consistency, and inconsistent band language is the fastest way to make an otherwise sound report look careless. Validity flags handled silently. A flagged F Index or Consistency Index protocol is either dropped without explanation or interpreted as if clean. The defensible pattern is three clauses: the index, its classification, and the interpretive consequence, remembering that an elevated F can reflect genuine severity rather than exaggeration. Edition and norms unnamed. "The BASC" with no edition, no form level, and no norm set. With two editions on the market as of 2026 and general, gender-specific, and clinical norms all in circulation, an unnamed norm context leaves the T scores uninterpretable and the re-evaluation uncomparable. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (scales, T scores, classifications, raters) and get a drafted results-section narrative organized by composite, with validity and cross-informant paragraphs in place for your review. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with their band labels, descriptor drift between sections, and validity indexes reported but never interpreted. Translate the results section into a plain-language summary for parents and teachers that keeps At-Risk and Clinically Significant framed as elevation bands, not diagnoses. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do BASC-3 T scores and percentiles actually mean? Every BASC-3 scale reports a T score, a standard score with a mean of 50 and a standard deviation of 10, plus a percentile against the norm group you selected ( general or clinical norms ). Direction depends on the scale family: on clinical scales and composites, higher scores mean more reported problems; on the adaptive scales and the SRP Personal Adjustment composite, lower scores mean more concern. The publisher's classification bands (Average, At-Risk, Clinically Significant) are normative descriptions: how elevated the rating is relative to same-age peers, subject to measurement error. Write them as bands, pair each with its plain-language meaning, and let the diagnosis, if any, come from the whole evaluation. Should I write my report on the BASC-3 or the BASC-4? As of July 2026, both are current. The BASC-4 published in 2026 (Reynolds and Kamphaus, qualification level B, same TRS/PRS/SRP architecture), while the BASC-3 remains on sale, scored on Q-global, with no announced retirement date, and Pearson lists BASC-4 narrowband ADHD, autism, and mood forms as coming soon. A report written on either edition is defensible if it names the edition and norm set, finishes on the edition it started on, and treats cross-edition comparisons as a stated limitation. What ages badly is a report that just says "the BASC": within a year or two, nobody will be able to tell which norms produced your T scores. Is the BASC-3 required for ADHD or special-education decisions? No authority names it. IDEA is instrument-neutral and points the other way: evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ), Section 504 prescribes no test list, and no DSM criterion requires any rating scale by name. The BASC-3's role is evidentiary convention: a norm-referenced, multi-informant way to document cross-setting patterns that eligibility categories and DSM criteria care about. The same logic covers the emotional-disturbance category's undefined social-maladjustment exclusion ( 34 CFR 300.8(c)(4) ): the scales inform the judgment, state criteria and the team make it. How many raters do I need for a defensible write-up? No publisher rule, regulation, or payer policy mandates a rater count, and "always parent plus teacher plus self" is convention, not law. Multi-informant, multi-setting coverage is best practice because agreement between different informant types averages only r = .28 across 341 studies , so each additional rater genuinely adds information rather than redundancy. A single-rater protocol can still be valid; it just limits cross-setting inference, and the write-up should say so explicitly. What reviewers actually flag is not the count but the handling: raters unnamed, settings unstated, or divergence between raters averaged away instead of interpreted. When do I use the BASC-3 versus the Conners 4 or the CBCL? By bandwidth. The BASC-3 is broadband: one administration surveys externalizing, internalizing, school, and adaptive domains across raters, which is what you want for cross-setting pervasiveness, comorbidity screening, and special-education questions. The Conners 4 is ADHD-specific and DSM-aligned, so it carries symptom-level detail in an ADHD work-up while the BASC-3 establishes the cross-setting and everything-else context; many evaluations use both. The Achenbach system (CBCL and relatives) is the closest broadband alternative, with DSM-oriented scales and the widest age span, while the BASC-3 counters with stronger adaptive scales and built-in validity indexes on informant forms. In autism referrals the same logic holds: BASC-3 findings are converging context, and the diagnostic weight sits with autism-specific measures and direct observation, inside a full psychological evaluation report . Can I include BASC-3 items or the Q-global narrative in my report? No to items, and quote the software sparingly if at all. Test items are protected test materials: reproducing them breaches the publisher's terms and the test-security obligations in APA Ethics Standard 9.11 , and Pearson's generated reports carry their own trade-secret handling notices. The interpretive narrative Q-global produces is copyrighted boilerplate anyway, and reviewers recognize it instantly; a results section assembled from pasted software paragraphs reads as unintegrated. The defensible pattern is the one this page teaches: report scores, classifications, and percentiles in your own prose, describe what elevated scales measure in your own words, and reserve quotation for your data, not the publisher's text. What do I do with an elevated F Index? Slow down, do not discard. The F Index is an infrequency scale flagging an inordinately negative depiction of the child. An elevation means interpret with caution against collateral data, not automatic invalidity, and the direction of the error matters: published evidence in depressed adolescents shows the SRP-A F Index can confound genuine psychopathology with intentional over-reporting, so a distressed teenager can look like an exaggerator. The write-up pattern reviewers accept has three clauses: name the index and its classification, state your interpretation of why it is elevated (severity, distortion, or unclear), and state the consequence for how much weight that protocol's scores carry. How do I explain At-Risk and Clinically Significant to parents? As comparison language, not verdicts. Something like: "Compared with other kids the same age, the ratings in this area are elevated enough that we should watch it" (At-Risk), and "elevated enough that we should act on it" (Clinically Significant), followed immediately by what you are recommending. Say plainly that no score here is a diagnosis, that scores carry measurement error, and that the publisher's own reports caution against using them as the sole basis for decisions. Parents hear "clinically significant" as "something is clinically wrong with my child"; the write-up and the feedback conversation both go better when the band label never appears without its plain-language meaning and its next step. Can BastionGPT draft my BASC-3 results section? Yes. Paste your score summary (forms, raters, T scores, classifications) and it drafts the results-section narrative for your review: organized by composite, validity addressed first, informants integrated rather than averaged, and classification language kept normative. It can also cross-check a draft you wrote for score-versus-narrative mismatches and descriptor drift, and produce a plain-language version of the findings for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, BASC-3 product page : authors, 2015 publication, age ranges, completion times, qualification level B, Q-global and manual scoring, norm options, and the BASC-4 availability banner. Pearson, BASC-4 product page and BASC-4 flyer : 2026 publication, same authorship, and narrowband ADHD, autism, and mood forms listed as coming soon. Pearson, BASC-3 Scales, Scores, and Indexes : composite and scale architecture, the Emotional Symptoms Index and Personal Adjustment composition and direction, and the executive functioning indexes' Q-global-only availability. Pearson Q-global sample reports ( rating scales ; multirater ): the F Index description, validity-index classifications, the p < .05 shading in multirater comparisons, and the sole-basis caution. Pearson, Clinical Applications of the BASC-3 Content Scales and Clinical Indexes webinar handout (2021) : the ED qualification scales and the statement that social maladjustment is not operationally defined in federal IDEIA legislation. 34 CFR 300.304 and 34 CFR 300.8 : the no-single-measure rule and the emotional-disturbance definition with its social-maladjustment exclusion. Buros Center for Testing, Tests reviewed in the Twentieth Mental Measurements Yearbook : independent review status of the core BASC-3 and the PRQ. De Los Reyes et al., 2015, Psychological Bulletin, the validity of the multi-informant approach (341 studies, mean cross-informant r = .28); Achenbach, McConaughy & Howell, 1987, Psychological Bulletin, the foundational cross-informant meta-analysis . Canivez, von der Embse & McGill, 2021, School Psychology, independent factor analyses of the BASC-3 TRS : composite scales may lack sufficient unique variance for stand-alone interpretation. Sonne et al., 2020, Psychological Assessment, interpretation problems with the BASC-3 SRP-A F Index in adolescents with depressive disorders. CMS Medicare Coverage Database, WPS article A57481 : the statutory screening exclusion and the separate screening code family; APA Services, 2026 testing billing and coding guide : what the evaluation-services codes include. Psychology Board of Australia, Code of Conduct (December 2025) : record accuracy and retention rules for Australian psychologists. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Psychiatric Diagnostic Evaluation Treatment Plan WISC-V Report Write-Up Conners 4 Report Write-Up WIAT-4 Report Write-Up ADOS-2 Report Write-Up Vineland-3 Report Write-Up BRIEF-2 Report Write-Up SRS-2 Report Write-Up Psychoeducational Evaluation Report Child & Adolescent Developmental Assessment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/brief-2-report All note templates # BRIEF-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The BRIEF-2 (Behavior Rating Inventory of Executive Function, Second Edition) is a rating scale of everyday executive function for ages 5 to 18, completed by parents, teachers, and adolescents themselves, scored as T scores across three indexes and a global composite. Psychologists use it in ADHD, autism, TBI, and special-education evaluations. This page covers how to write up BRIEF-2 results, with a fictional sample and a template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and neuropsychologists; publisher qualification level B (S for screening forms) Audience IEP and Section 504 teams, ADHD and autism evaluators, physicians, accommodations reviewers, parents and teachers Typical length 300 to 700 words for the results section · administration about 10 minutes per form Format family Multi-informant executive function rating scale When it's used ADHD, autism, and TBI evaluations, special-education and 504 assessments, treatment and medication-response monitoring, ages 5 to 18 Standards context Published by PAR (2015; updated 2026); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the BRIEF-2? The BRIEF-2 (Behavior Rating Inventory of Executive Function, Second Edition, styled BRIEF2 by its publisher) is a standardized rating scale of everyday executive function in children and adolescents, published by PAR in 2015 by Gerard Gioia, Peter Isquith, Steven Guy, and Lauren Kenworthy. Its ancestor, the original BRIEF (2000), was described in the Buros review as "the first published rating scale measuring self-regulatory abilities in children and adolescents" (McGill, 2021). Parents and teachers rate ages 5 to 18 on 63-item forms, and adolescents 11 to 18 complete a 55-item self-report; each form takes about 10 minutes. Nine clinical scales (seven on the self-report) roll up into three indexes, Behavior Regulation (BRI), Emotion Regulation (ERI), and Cognitive Regulation (CRI), plus an overall Global Executive Composite (GEC), with three embedded validity scales (Inconsistency, Negativity, Infrequency). All scores are T scores with a mean of 50 and a standard deviation of 10, and higher scores mean more reported difficulty. A 2026 update added age-based (combined-sex) norms as a second validated scoring option, folded the ADHD manual supplement into the core Professional Manual and reports, discontinued the standalone ADHD Form, and moved all scoring to the publisher's PARiConnect platform, without changing the items or the normative sample. Twelve-item screening forms exist for triage and require a separate qualification level. The load-bearing fact for write-ups is what a rating scale measures: reported everyday behavior, not tested performance. The largest analysis of the question found performance-based and rating measures of executive function correlate at a median of only r = .19, with just 24% of 286 correlations reaching significance ( Toplak, West & Stanovich, 2013 ), and the BRIEF-2 manual's own convergent data, as reported in the Buros review, put its Working Memory scale at r = -.08 to -.27 against the Wechsler working-memory indexes. The two method families capture, in Toplak's words, "the efficiency of cognitive abilities and success in goal pursuit," and a defensible report writes them as complementary levels of analysis, never as interchangeable measures of the same thing. Edition status also needs dating: as of July 2026 the BRIEF-2 is current for ages 5 to 18, the adult BRIEF2A (released November 12, 2024) replaced the two-index BRIEF-A and adopted the same three-index model, and the original BRIEF's two-index architecture (Behavioral Regulation plus Metacognition) survives only in old reports, where its vocabulary still leaks into new ones. ## Who uses BRIEF-2 write-ups and when School psychologists write up BRIEF-2 results inside special-education and Section 504 evaluations, where everyday executive function evidence speaks directly to adverse educational effect; clinical child psychologists use it in ADHD and autism evaluations alongside symptom-specific scales such as the Conners 4 and broadband measures such as the BASC-3 ; and pediatric neuropsychologists fold it into a neuropsychological report as the everyday-behavior counterpart to performance measures such as the D-KEFS , including in traumatic brain injury follow-up and concussion return-to-learn monitoring. The instrument boundaries do real work in a battery: the Conners 4 profiles ADHD symptoms against diagnostic criteria, the BASC-3 surveys behavior and emotion broadly, the CEFI offers a strengths-worded executive function rating with a single composite, and the BRIEF-2 is the one built to profile executive function domain by domain as it shows up at home and school. Age sets the family member: the BRIEF-P covers preschoolers, the BRIEF-2 runs 5 to 18 with a self-report from 11, and adult referrals use the BRIEF2A, whose informant form matters when self-awareness is itself in question. Rating evidence then feeds the parent document, usually a psychological evaluation report , a psychoeducational report, or an ADHD evaluation report, where reliable-change statistics also support re-administration to track treatment and medication response over time. ## How to structure a BRIEF-2 results section No statute, payer, or publisher mandates a results-section format. The sequence below follows the interpretive order the Professional Manual itself recommends, validity first, then indexes and scales, then profile shape, then cross-informant comparison, because a reviewer can follow it and because it makes the two genuinely contested moves in BRIEF-2 reporting, how to treat the GEC and how to reconcile raters who disagree, explicit stated decisions instead of silent ones. Each section carries the pitfall that most often undermines it. Measures, forms, informants, and norms. Name the instrument and edition, each form administered and who completed it (parent, teacher, self), the norm set used, and the scoring platform. Since the 2026 update there are two validated norm options, age-and-sex-based and age-based (combined-sex), so the choice is now a documentation item. Pitfall: "a BRIEF was completed" with no edition, rater roles, or norm set. The original BRIEF, the BRIEF-2, and the adult BRIEF2A score differently and even use different index architectures, so an unanchored reference is uninterpretable at re-evaluation. Validity scales before any scores. Report the Inconsistency, Negativity, and Infrequency results for every rater, then say what you did about any flag. Pitfall: treating clean validity scales as proof the profile is accurate. They screen response style, not truth, and independent validation of these scales is limited; a motivated or exhausted rater can still produce an interpretable-looking protocol. Index and composite results. Present BRI, ERI, and CRI (and the GEC) as T scores with percentiles, and describe elevations with the manual's vocabulary in prose: scores of 60 to 64 are described as mildly elevated, 65 to 69 as potentially clinically elevated, and 70 or above as clinically elevated. State the GEC convention you followed: the manual treats the GEC as most interpretable when the index scores hang together, and that heuristic itself has been questioned in the literature. Pitfall: leading with the GEC when the indexes diverge. A summary score that averages a clinically elevated CRI against an average ERI describes no ability the student actually shows. Scale-level findings, tightly bounded. Name the clinical scales driving each elevated index (for example Working Memory and Plan/Organize under the CRI) and give one functional example for each. The Buros review's central criticism is redundancy among the nine scale scores, so pick the drivers instead of narrating all nine. Pitfall: the nine-scale score dump. A paragraph per scale in form order documents that scoring occurred; it does not answer the referral question, and it overstates how much independent information the scales carry. Cross-informant comparison. Report each rater's results separately, then reconcile: agreement between informants is expected to be modest (a 341-study meta-analysis puts the cross-informant mean near r = .28), teachers rate against a classroom full of same-age peers while parents rate one child at home, and divergence is setting information, not error. Pitfall: averaging raters into one number or crowning one rater accurate. Both moves delete the data; the write-up's job is to say what the disagreement means for where support is needed. Integration with performance testing. When the battery includes performance measures of executive function, state the relationship honestly: ratings capture everyday goal pursuit, performance tests capture efficiency under structured conditions, and the two correlate weakly (median r = .19 across the literature). Divergence between an elevated BRIEF-2 and intact test performance is a finding to interpret, not a contradiction to explain away. Pitfall: writing "ecologically valid" as a trump card. The term is used inconsistently across the field, and executive function tests still account for only about a fifth of the variance in everyday executive behavior, so describe levels of analysis instead of declaring a winner. Functional-impairment translation. Convert elevations into observable, educationally relevant behavior: what starting multi-step assignments, holding instructions, or tracking materials actually looks like for this student. IDEA eligibility turns on adverse educational effect and Section 504 on a substantial limitation of a major life activity, and reviewers act on described behavior, not on scale labels. Pitfall: "elevated Working Memory scale" offered as the impairment. The scale name is the measure; the impairment is the behavior it summarizes. Interpretive summary and recommendations linkage. Answer the referral question, state plainly that the BRIEF-2 is not a standalone diagnostic measure and that no single measure may determine eligibility, and tie each recommendation to a specific elevated scale, which is also how the publisher's own intervention handouts are organized. Pitfall: the elevation-equals-ADHD leap. Executive function elevations occur in autism, anxiety, depression, traumatic brain injury, and sleep problems; the defensible sentence is "consistent with executive function difficulties," followed by what the whole evaluation shows. ### Blank template (copy and adapt) MEASURES AND INFORMANTS Instrument/edition: BRIEF-2 Norms: [age-and-sex / combined-sex] Forms and raters: [Parent: ___ Teacher: ___ Self-Report: ___] Scoring platform: [___] Dates completed: [___] VALIDITY SCALES (every rater, before any scores) [Inconsistency / Negativity / Infrequency status per rater; action taken on any flag] INDEX AND COMPOSITE RESULTS (T scores + percentiles) Behavior Regulation (BRI): [per rater + band in prose] Emotion Regulation (ERI): [same] Cognitive Regulation (CRI): [same] Global Executive Composite: [reported per the stated convention; note if indexes diverge] SCALE-LEVEL DRIVERS (only the scales doing the work) [Which clinical scales drive each elevated index + one functional example each] CROSS-INFORMANT COMPARISON [Each rater separately; reconciliation as setting information; no averaging, no crowned rater] INTEGRATION WITH PERFORMANCE TESTING (if administered) [Complementary levels of analysis; convergence and divergence interpreted, not explained away] INTERPRETIVE SUMMARY (answer the referral question) [Consistent-with language; not a standalone diagnostic measure; no single measure determines eligibility] RECOMMENDATIONS LINKAGE [Each recommendation tied to a specific elevated scale; monitoring plan and re-rating interval if any] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample BRIEF-2 write-up (fictional) Scenario: a pediatrician refers a 13-year-old for an ADHD evaluation after two years of incomplete assignments and lost materials despite strong grades on tests. Parent, teacher, and self-report forms were collected, and they disagree in the classic direction: both adults report clearly elevated cognitive regulation problems while the teen self-reports milder difficulty. This is the rating-scale section of the evaluation only, condensed but structurally complete. All details are fictional. Student: A.S., 13 (grade 8) · Referral: pediatrician, attention and work-completion concerns · Evaluator: J. Okafor, PhD, Licensed Psychologist · Forms completed: 07/06/2026 to 07/10/2026 · Report date: 07/15/2026 Measures and informants: The Behavior Rating Inventory of Executive Function, Second Edition (BRIEF-2) was administered as part of a broader ADHD evaluation. A.S.'s mother completed the Parent Form, his homeroom teacher (who teaches him for two academic subjects) completed the Teacher Form, and A.S. completed the Self-Report Form. All forms were scored digitally against the age-and-sex-based norms. BRIEF-2 scores are T scores with a mean of 50 and a standard deviation of 10; higher scores indicate more reported difficulty, and each score below is presented with its percentile. Validity of the ratings: The Inconsistency, Negativity, and Infrequency scales were within acceptable limits for all three raters, and no items were left unanswered. The ratings below are considered interpretable descriptions of each rater's observations in their own setting. Index Parent T (percentile) Teacher T (percentile) Self-Report T (percentile) Behavior Regulation (BRI) 62 (88th) 64 (92nd) 55 (69th) Emotion Regulation (ERI) 52 (58th) 54 (66th) 51 (54th) Cognitive Regulation (CRI) 73 (99th) 70 (98th) 61 (86th) Global Executive Composite (GEC) 68 (96th) 66 (95th) 57 (76th) Index results: The Cognitive Regulation Index is the elevated finding, and it is elevated for both adult raters: the parent rating (T = 73, 99th percentile) is clinically elevated and the teacher rating (T = 70, 98th percentile) reaches the same band, indicating that both settings see marked difficulty with the thinking-side work of getting tasks done. Behavior Regulation is mildly elevated for both adults (T = 62 and 64), driven by monitoring rather than impulse control, while Emotion Regulation is within typical limits for all three raters. Because the index scores diverge, the Global Executive Composite is reported for completeness but is not the interpretive anchor of this section; the profile, not the summary score, carries the meaning. Scale-level drivers: Within the Cognitive Regulation Index, the parent rating is driven by Working Memory (T = 74) and Plan/Organize (T = 71), both clinically elevated, with Initiate (T = 65) and Task-Monitor (T = 66) potentially clinically elevated and Organization of Materials mildly elevated (T = 62). The teacher pattern is parallel with slightly lower values, led by Working Memory (T = 72) and Plan/Organize (T = 68). Functionally, the raters describe the same student: instructions delivered in strings get lost, multi-step assignments stall at the starting step, long-term projects surface the night before they are due, and completed work is not turned in. These are the behaviors behind the scores, and they are the behaviors any support plan has to touch. Cross-informant comparison: A.S. rates himself consistently lower than the adults rate him (GEC T = 57 versus 68 and 66), with only the Cognitive Regulation Index mildly elevated in his self-report (T = 61). Modest informant agreement is the expected finding for behavior ratings, and the direction here, adolescents reporting fewer executive difficulties than the adults around them observe, is common in ADHD referrals. The adult ratings are treated as the primary description of day-to-day functioning, and the self-report as meaningful data about A.S.'s own experience: he reports effort ("I study more than my friends do") more than difficulty, which matters for how support is introduced to him. No rater is treated as wrong; home and school agree on the problem, and A.S.'s perspective shapes the intervention conversation. Relation to performance testing: Rating-scale results describe everyday executive behavior and correlate only weakly with performance tests of executive function, so the cognitive testing reported elsewhere in this evaluation is read as a complementary level of analysis rather than as confirmation or contradiction of these ratings. Summary: BRIEF-2 ratings from three informants converge on clinically significant everyday difficulty with working memory and planning, present across home and school, with behavior regulation mildly affected and emotion regulation intact. These results are consistent with the executive function profile commonly seen in ADHD, inattentive presentation, and they support that hypothesis within this evaluation; they are not by themselves diagnostic of any condition, and the diagnostic conclusion in this report rests on the full evaluation, including history, interview, symptom-specific scales, and classroom data. No single measure determines eligibility for services. Linkage to recommendations: The Working Memory elevations support recommendations 1 and 2 (instructions chunked, repeated, and paired with written steps). The Plan/Organize and Task-Monitor elevations support recommendation 3 (an assignment-tracking system checked daily by one adult at school and one at home, with long-term projects broken into dated steps). The self-report gap supports recommendation 5: supports should be framed with A.S. as effort-saving tools, matching his own account, rather than as remediation he does not believe he needs. Re-rating by the same informants after two school terms will use the publisher's reliable-change statistics to judge response. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, every form, every rater's role, and the norm set are named, so the scores are interpretable now and comparable at re-rating, even after the 2026 two-norm-option update. Validity scales are reported first for all three raters, every T score carries its percentile, the manual's band vocabulary is used consistently in prose, and every number in the narrative matches the table. The GEC is reported but explicitly declined as the interpretive anchor because the indexes diverge, with the convention stated instead of silently applied. Informant disagreement is reported as setting and self-awareness information with a stated reconciliation; no rater is averaged away or crowned accurate. Ratings are framed as everyday behavior complementary to performance testing, elevations are translated into observable classroom behaviors, the summary uses consistent-with language and the no-single-measure rule, and every recommendation traces to a named elevated scale. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label the strength of each requirement honestly. In US special education, federal law is instrument-neutral: no regulation names the BRIEF-2, and evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ). The categories where executive function evidence usually lands are Other Health Impairment, whose federal definition names attention deficit hyperactivity disorder and turns on adverse educational effect, and Traumatic Brain Injury, which is its own category with attention and reasoning in its definition, not a default into OHI ( 34 CFR 300.8(c)(9), (c)(12) ). Section 504 keys on an impairment that "substantially limits one or more major life activities" ( 34 CFR 104.3(j) ), which is why the write-up's functional-impairment translation matters more than any score. And no law, regulation, or PAR document sets a qualifying T score: the band vocabulary is interpretive convention from the manual, not an eligibility threshold, so "did not reach T = 65" is never a legal finding. For accommodations bodies the framing is supporting evidence: College Board's ADHD documentation guidelines require a request even when the student already has an IEP or 504 plan, state that "A medical note is usually not sufficient to support the need for accommodations," and ask for a comprehensive assessment with nationally normed measures, functional-limitation narrative, and current information, with teacher observations expressly welcomed; ACT's criteria likewise list "rating scale measures from multiple sources" among relevant testing and caution that unfinished timed tests cannot demonstrate impairment in isolation. On the payer side, the code families split by purpose as policy: brief emotional or behavioral screening instruments bill under 96127, Medicare's benefit excludes screening as such under Social Security Act 1862(a)(7) ( WPS article A57481 ), and the NCCI edits bundle 96127 into the 96130 to 96139 testing family on the same day, so a BRIEF-2 scored within a billed testing evaluation is part of that service, not a separate line. Edition and norms currency is now the BRIEF-2's live defensibility question, and the facts need dating. As of July 2026: the child and adolescent BRIEF-2 (2015) is current, and its 2026 update added age-based (combined-sex) norms as a second validated option, integrated the ADHD supplement into the Professional Manual and reports, discontinued the standalone ADHD Form, and moved all scoring to the publisher's platform with no manual lookup tables, while leaving items and the normative sample unchanged, so reports should now name which norm set was used. The adult BRIEF2A (released November 12, 2024) replaced the 2005 BRIEF-A, extended the range to age 99, renormed on the 2021 census, and adopted the three-index model, which retires the old two-index Behavioral Regulation and Metacognition vocabulary that still circulates in copied report shells. The band ladders also differ between the current child and adult instruments: on the BRIEF-2 the manual describes 60 to 64 as mildly elevated and 70 and above as clinically elevated, while the publisher's BRIEF2A reports describe 60 to 64 as within normal limits with possible subtle difficulties and reserve moderately and highly elevated for 70 to 74 and 75 and above, so carrying one form's vocabulary into the other's report misdescribes the same number. In Australia the BRIEF-2 is distributed by Psychological Assessments Australia , which notes the publisher is not reprinting paper forms (stock-only), and no Medicare item pays for a rating scale itself; the complex neurodevelopmental disorder assessment items (135, 137, 139, 289, patients under 25, once per lifetime) frame the evaluations these ratings inform, and the official item guidance notes 135 and 289 are not intended for standalone ADHD diagnosis. In Canada, use runs on provincial school-board practice and private assessment convention, with no pan-Canadian standard setting a BRIEF-2 threshold, and it is accurate to say so in a report. Purchase and use are qualification-gated by the publisher (level B for the core forms, S for the screening forms), and re-administration to track treatment response is supported by the publisher's reliable-change statistics, which is convention worth naming when a re-rating interval is part of the plan. BRIEF, BRIEF2, and BRIEF2A are trademarks or registered trademarks of PAR, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common BRIEF-2 write-up errors reviewers flag There is no payer audit series for rating-scale write-ups; the accountability literature here is psychometric, and it is specific. The BRIEF-2 was reviewed in the Buros Center's 21st Mental Measurements Yearbook , where the reviewer questioned how much unique information the nine scale scores add over one another, noted that the GEC was left out of the manual's factor models and that no correlations with performance-based executive tests were reported, and passed along the manual's own caution that the instrument is not a standalone diagnostic measure. The performance-versus-ratings literature is blunter still: a median correlation of r = .19 across 20 studies, with only 24% of correlations significant ( Toplak et al., 2013 ). The errors below are write-up errors, and they are yours. The BastionGPT Clinical Advisory Board sees the same ones most often in BRIEF-2 report reviews: An elevation written as a diagnosis. "The elevated Working Memory scale indicates ADHD." Executive function elevations occur in autism, anxiety, depression, traumatic brain injury, and sleep disruption, and the publisher's own interpretive reports state the instrument "should not be used in isolation as a diagnostic tool." Write "consistent with executive function difficulties," then let the whole evaluation carry the diagnosis. A T-score threshold treated as a rule. "Did not meet the clinical cutoff of 65" written as if some authority set one. No statute, agency, or PAR document makes any T score a qualifying line; the bands are interpretive vocabulary. Eligibility frameworks ask about adverse effect and substantial limitation, which live in the functional narrative, not at a score boundary. Informant disagreement averaged away or adjudicated. One number presented as "the" result, or the teacher dismissed because the parent score was higher. Cross-informant agreement is expected to be modest (mean r = .28 across a 341-study meta-analysis), teachers rate against a classroom of same-age peers, and divergence is setting information the report should interpret, not noise to delete. Ratings and performance tests written as the same evidence. An elevated BRIEF-2 described as "confirmed by testing," or intact test scores used to dismiss the ratings. The two families correlate weakly (median r = .19, and the manual's own working-memory convergence values are near zero), because they measure different levels: everyday goal pursuit versus efficiency under structured conditions. Report both, and interpret divergence instead of resolving it by fiat. Cross-edition vocabulary drift. "Metacognition Index" in a BRIEF-2 report (that architecture belongs to the original BRIEF and the retired BRIEF-A), child band labels applied to a BRIEF2A adult report even though the two ladders read the same T score differently, or a post-2026 report that never says which of the two validated norm sets was used. Reviewers read unanchored vocabulary as a copied shell, and it makes accurate scores look wrong. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your index and scale summary (T scores, percentiles, raters) and get a drafted results-section narrative organized validity-first and index-first, with the cross-informant reconciliation framed for your review and rationale. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with the table, band-vocabulary drift (including child and adult ladder mix-ups), and informant claims the scores do not support. Translate the results section into a plain-language summary for parents and teachers that turns elevations into everyday behaviors and keeps scores as estimates, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do BRIEF-2 T scores actually mean? Every BRIEF-2 scale, index, and composite score is a T score with a mean of 50 and a standard deviation of 10, reported with a percentile, and the direction is the opposite of an ability test: higher means more reported difficulty. The Buros review relays the manual's vocabulary: scores of 60 to 64 are described as mildly elevated, 65 to 69 as potentially clinically elevated, and 70 or above as clinically elevated, with T scores running from 35 to 90. Two cautions belong in the write-up: percentiles do not track T scores identically from scale to scale (the distributions are skewed), and the bands are interpretive convention, not qualifying thresholds. Write each score with its percentile, use one vocabulary consistently, and remember these are one rater's observations expressed as a number. Is the BRIEF-2 still current, and what changed in 2026? Yes. As of July 2026 the BRIEF-2 (2015) is the current child and adolescent edition, and its 2026 update changed the packaging rather than the test: age-based (combined-sex) norms were added as a second validated option, the ADHD supplement was folded into the Professional Manual and reports, the standalone ADHD Form was discontinued, and all scoring moved to the publisher's platform, with the items and normative sample unchanged. The practical consequence for reports is naming which norm set you used. For adults, the BRIEF2A (released November 12, 2024) replaced the 2005 BRIEF-A, runs to age 99, and uses the same three-index model as the BRIEF-2, so an adult evaluation still citing the BRIEF-A is citing a superseded edition. Preschoolers have their own form, the BRIEF-P. Do I need an elevated GEC, or a specific T score, to document executive function impairment? No, on both counts, and this is the most common false rule in circulation. No statute, agency, or PAR document requires a Global Executive Composite elevation or sets any T score as a qualifying cutoff. The manual's own convention runs the other way: the GEC is most interpretable when the index scores hang together, and when they diverge the indexes and their driving scales carry the interpretation, a heuristic the Buros reviewer notes has itself been questioned. What eligibility frameworks actually ask for is functional evidence: adverse educational effect for IDEA, substantial limitation of a major life activity for Section 504. A student with a clinically elevated Cognitive Regulation Index and an unremarkable GEC can be exactly the student who needs support, and the write-up's job is to describe the behavior behind that index, not to clear an invented score bar. Is the BRIEF-2 required for an ADHD diagnosis or an IEP? No requirement names it anywhere. Federal special-education law is instrument-neutral and forbids using any "single measure or assessment" as the sole criterion ( 34 CFR 300.304(b)(2) ); the OHI category names ADHD in its definition but prescribes no test, and TBI is a separate category with attention and reasoning in its own definition. Diagnosis is clinical: DSM-5-TR criteria plus history, interview, and multi-source evidence, in which the BRIEF-2 supplies the everyday executive function profile while symptom scales like the Conners 4 map the criteria themselves. Accommodations bodies treat it the same way: College Board asks for a comprehensive, nationally normed assessment with a functional-limitation narrative (a medical note alone is "usually not sufficient"), and ACT lists rating scales from multiple sources among relevant testing. So the BRIEF-2 is standard supporting evidence by convention, mandated by no one. Parent and teacher ratings disagree. Which one is right? Usually both. Modest cross-informant agreement is the most replicated finding in behavior-rating research (a 341-study meta-analysis puts the mean correlation near .28), the BRIEF-2 manual itself reports interrater agreement as low to moderate, and an early BRIEF clinical study found the same thing ( McCandless & O'Laughlin, 2007 ). The raters watch different demands: a teacher compares one student against a room of same-age peers under sustained task load, a parent sees one child across home routines, and adolescents commonly rate themselves lower than the adults around them. Report each rater separately, interpret the divergence as information about settings and self-awareness, and resist both bad fixes, averaging the raters into one number or declaring a winner. Where the elevation lives tells you where the support belongs. BRIEF-2 or a performance test like the D-KEFS: which should I trust? Neither over the other; they measure different things. Performance tests sample executive efficiency under structured, one-on-one conditions; ratings sample how often everyday self-regulation succeeds at home and school. The two families correlate at a median of only r = .19 ( Toplak et al., 2013 ), classic work found executive tests account for roughly 18 to 20% of the variance in everyday executive ability ( Chaytor et al., 2006 ), and even the label "ecologically valid" is used inconsistently across the field ( Suchy et al., 2024 ), so it settles nothing on its own. Administer both when the referral question spans settings and mechanisms, report both, and write divergence as a finding: intact test performance with elevated ratings often describes a student who can perform under structure and cannot self-organize without it. BRIEF-2, BASC-3, or Conners 4: which rating scale does what? They answer different questions, and batteries often carry two of them deliberately. The BASC-3 is the broadband instrument: it surveys behavior, emotion, and adaptive skills widely and flags executive function within a larger profile. The Conners 4 is the ADHD symptom instrument: it maps ratings onto diagnostic criteria when ADHD is the referral question. The BRIEF-2 is the executive function instrument: nine scales and three indexes profiling regulation domain by domain, which is what individualizes the support plan once a diagnosis is on the table. The CEFI is the closest alternative in the executive lane, with positively worded items and a single full-scale composite. In a typical ADHD evaluation the Conners 4 speaks to criteria while the BRIEF-2 specifies which regulation systems the plan must target; in an autism or TBI evaluation the BRIEF-2 pairs with broadband and performance measures instead. Can I see the actual BRIEF-2 items, or include them in a report? No. The items, forms, and scoring materials are protected test materials: PAR gates purchase by qualification level (B for the core forms, S for the screening forms), and psychologists are ethically obligated to maintain test security under APA Ethics Standard 9.11 and their publisher agreements. Item exposure also degrades the instrument itself, because norms assume raters have not rehearsed the content. A report should carry scores, percentiles, band language, and interpretation, never item text, and records requests touching protocols route through your test-security obligations. What families actually need travels fine without items: the publisher's own feedback materials and intervention handouts translate elevated scales into plain language and strategies, and a good parent summary does the same. Can BastionGPT draft my BRIEF-2 results section? Yes. Paste your index and scale summary (T scores, percentiles, and who rated) and it drafts the results-section narrative for your review: validity scales first, indexes before scales, the manual's band vocabulary applied consistently, and the cross-informant reconciliation framed with a stated rationale. It can also cross-check a draft you wrote for score-versus-narrative mismatches and band-vocabulary drift, and produce a plain-language summary that turns elevations into everyday behaviors for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: PAR, BRIEF2 product page : forms, ages, item counts, scales and indexes, validity scales, 10-minute administration, qualification levels B and S, standardization total of 3,603 ratings, and the 2026 update (combined-sex norms option, ADHD content integration, discontinued ADHD Form, all-digital scoring). PAR, BRIEF2A product page and release announcement : the November 12, 2024 release, ages 18 to 99, the three-index model, and the 2021-census normative samples. PAR, BRIEF2A sample interpretive report : the publisher's own statement that the instrument "should not be used in isolation as a diagnostic tool," the T-score conventions, and the adult band vocabulary. McGill, R. J., Test review of the BRIEF2 , in the 21st Mental Measurements Yearbook (pp. 81-83), author reprint; Buros Center for Testing, tests reviewed in the 21st MMY : the manual's band vocabulary, standardization detail, reliability ranges, interpretive sequence, and the redundancy, GEC, and performance-correlation critiques. Toplak, West & Stanovich, 2013, Journal of Child Psychology and Psychiatry : the median r = .19 and the 24% significance rate across 286 correlations. De Los Reyes et al., 2015, Psychological Bulletin : the cross-informant mean of r = .28; McCandless & O'Laughlin, 2007, Journal of Attention Disorders : low interrater reliability with acceptable convergent validity in an ADHD clinic sample; Halvorsen et al., 2019, Child Neuropsychology : independent confirmatory support for the three-factor model. Chaytor, Schmitter-Edgecombe & Burr, 2006, Archives of Clinical Neuropsychology : executive tests accounting for 18 to 20% of variance in everyday executive ability; Suchy et al., 2024, Journal of the International Neuropsychological Society : the systematic review of inconsistent "ecological validity" usage. 34 CFR 300.8 , 34 CFR 300.304 , and 34 CFR 104.3 : the OHI and TBI definitions, the no-single-measure rule, and the Section 504 substantial-limitation definition. College Board, ADHD documentation guidelines : the seven documentation criteria, the medical-note sentence, and the currency windows; ACT, criteria for diagnostic documentation : rating scales from multiple sources and the timed-test caution. WPS, Medicare article A57481 : the screening exclusion and the 96127 code family; CMS, NCCI procedure-to-procedure edits : the 96127 bundling into the 96130 to 96139 family. MBS Online, complex neurodevelopmental disorder items FAQ : items 135, 137, 139, and 289 and the standalone-ADHD note; Psychological Assessments Australia, BRIEF-2 product page : Australian distribution and the paper-form stock note. ## Related templates Neuropsychological Evaluation Report Psychological Evaluation Report BASC-3 Report Write-Up Conners 4 Report Write-Up Psychiatric Diagnostic Evaluation CAARS-2 Report Write-Up D-KEFS Report Write-Up ADHD Evaluation Report T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/caars-2-report All note templates # CAARS-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The CAARS-2 (Conners Adult ADHD Rating Scales, 2nd Edition) is a 2023 self-report and observer rating scale measuring ADHD symptoms, impairment, and response style in adults 18 and older. Psychologists and psychiatrists use it to quantify symptom severity inside adult ADHD evaluations. This page covers how to write up CAARS-2 results, with a fictional sample, reviewer-flagged errors, and free templates. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, psychiatrists, and other qualified mental health clinicians; publisher qualification level B Audience Prescribers and psychiatric teams, disability and testing accommodation offices, employers, payers, referring primary care Typical length 150 to 400 words for the CAARS-2 section · administration 10 to 20 minutes per full-length form Format family Multi-informant adult ADHD rating scale When it's used Adult ADHD diagnostic evaluations, first-time adult diagnosis, medication and treatment monitoring, accommodation documentation, disability and forensic contexts Standards context Published by MHS (2023); replaces the retired 1999 CAARS; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CAARS-2? The Conners Adult ADHD Rating Scales, Second Edition (CAARS-2, styled CAARS 2 by its publisher) is a norm-referenced ADHD rating-scale family by C. Keith Conners, Drew Erhardt, and Elizabeth P. Sparrow, published by Multi-Health Systems (MHS) in 2023 as the revision of the 1999 CAARS. Two rater perspectives cover every evaluation: a Self-Report completed by the adult being assessed and an Observer form completed by someone who knows them well, each in three lengths (full-length, 97 items; Short, 55 self-report and 52 observer; ADHD Index, 12 items), for ages 18 and older. The full-length forms yield T scores, standard scores with a mean of 50 and a standard deviation of 10, for five content scales (Inattention/Executive Dysfunction, Hyperactivity, Impulsivity, Emotional Dysregulation, Negative Self-Concept), three DSM symptom scales aligned to DSM-5-TR criteria, and the 12-item ADHD Index, plus a Response Style Analysis (a new Negative Impression Index, an updated Inconsistency Index, Omitted Items, and Pace for online administrations), Critical and Screening Items covering suicidality, self-harm, anxiety, and sadness, and item-level Impairment & Functional Outcome ratings. Norms come from 2,640 adults (1,320 per rater type, drawn from the US and Canada and stratified to census targets) across seven age bands reaching 70 and older, with General Population or ADHD Reference comparison samples and combined-gender or gender-specific options. Administration and scoring run digitally through MHS's online platform, on screen or by entering responses from a printed form; hand scoring does not exist for this edition. Edition status matters as of July 2026: MHS retired the original CAARS, ending online product sales on October 31, 2024 and platform scoring after May 31, 2025, so every new administration is a CAARS-2 administration. The load-bearing distinction for write-ups: CAARS-2 scores quantify how much ADHD-consistent difficulty an adult and the people around them report, and that is all they do. A diagnosis under DSM-5-TR needs evidence the ratings cannot supply alone, including symptoms present before age 12 and impairment across settings, so the results section reports elevations and lets the whole evaluation carry the conclusion. A second trap is specific to this instrument's transition: many sample reports and template pages online still describe the 1999 edition's scales (Inattention/Memory Problems, Hyperactivity/Restlessness, Impulsivity/Emotional Lability, Problems with Self-Concept). The CAARS-2 restructured that architecture, splitting the old Impulsivity/Emotional Lability scale into separate Impulsivity and Emotional Dysregulation scales, so the scale names in a results section quietly announce which edition, and which norms, produced the scores. A write-up that names the current scales and the norm reference, keeps validity language factual, and subordinates the instrument to the full evaluation, in the pattern of a psychological evaluation report or a psychiatric diagnostic evaluation , is most of what separates a defensible CAARS-2 section from a score printout with sentences around it. ## Who uses CAARS-2 write-ups and when Psychologists and neuropsychologists write up CAARS-2 results inside adult ADHD diagnostic evaluations, where the paired self and observer forms supply the quantified, cross-informant view of everyday functioning that an interview alone cannot; psychiatrists and psychiatric nurse practitioners use it to structure symptom severity and monitor treatment around medication decisions; college students and professionals carry CAARS-2-based reports to disability offices and testing agencies as accommodation documentation; and evaluators in disability and forensic contexts reach for it precisely because its Response Style Analysis gives validity a vocabulary. It is the right tool when the question is dimensional and adult: how much ADHD-consistent difficulty does this adult report, does someone who knows them well observe the same pattern, and how does severity change with treatment. The instruments around it carry their own loads: the Conners 4 covers ages 6 to 18 and hands off to the CAARS-2 at 18, the two built to be aligned for a lifespan approach; the ASRS is a free WHO screener that flags rather than quantifies; the BAARS-IV adds normed retrospective childhood forms, a strength for onset documentation; the DIVA-5 is a structured diagnostic interview that maps DSM criteria rather than producing norm-referenced scores; and when the referral question is executive function in daily life rather than ADHD symptom severity, an executive function rating such as the BRIEF-2 family (BRIEF2A for adults) carries that lane. The CAARS-2 write-up is where self-report, observer evidence, validity indicators, and impairment ratings get integrated into one adult ADHD account, clearly subordinated to the full evaluation. ## How to structure a CAARS-2 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names the edition and norms, reports response style before scores, keeps self and observer evidence distinct, and separates what the ratings show now from the childhood-onset evidence DSM-5-TR requires. Each section carries the pitfall that most often undermines it. Measures, forms, raters, and norms. Name the instrument and edition, the form length used (full-length, Short, or ADHD Index), each rater and their relationship to the client, completion dates, and the norm reference: General Population or ADHD Reference comparison, combined-gender or gender-specific norms. Pitfall: 1999-edition scale names in a CAARS-2 report. Writing "Inattention/Memory Problems" or "Impulsivity/Emotional Lability" announces that a retired instrument's template produced the section, and with the original CAARS unscoreable on the publisher's platform since May 2025, reviewers read old names as a currency problem. Response style and rating quality. Report the Response Style Analysis first and factually: Negative Impression Index, Inconsistency Index, Omitted Items, and Pace for online administrations, plus anything situational that could color the ratings. Keep the language non-accusatory in both directions. Pitfall: treating validity indicators as dispositive. The first independent CAARS-2 study found the Negative Impression Index missed 92 percent of coached simulators, so a clean index is supportive evidence about how the form was completed, never proof of accuracy, and an elevated one is a reason for caution, never a malingering verdict. Critical and Screening Items. State whether the Critical Items (suicidal thoughts or attempts, self-injury) or Screening Items (anxiety, sadness) were endorsed, and what follow-up occurred. One disposition sentence closes the loop. Pitfall: leaving a flagged item inside the score report. A reviewer who can see the flag but not your response reads a risk-assessment gap, not a formatting choice. Self-report results. Report the five content scales and the three DSM symptom scales with T scores and the manual's qualitative descriptors stated in words, keeping descriptor language consistent across scales. Pitfall: promoting a number to a diagnosis. No T score is a diagnostic threshold, and no authority publishes a cutoff that establishes ADHD. Observer results and cross-informant integration. Report the observer's scores in the same order, then integrate: where the raters converge, where they diverge, and what the divergence means. Moderate agreement is expected, the publisher's own inter-rater figures run r = .44 to .54, and each perspective contributes information the other cannot. Pitfall: averaging raters into one number or crowning one rater correct. Divergence is usually setting and perspective information, and longitudinal data show adults with childhood ADHD tend to under-report relative to informants. ADHD Index. Report it as what it is: a 12-item screening indicator built to differentiate adults with ADHD from the general population. Pitfall: the Index written as the verdict. A high Index score raises the prior probability that a full evaluation will support the diagnosis; it does not make the diagnosis, and coached simulators earn high Index scores easily. Impairment and functional outcomes. Summarize the Impairment & Functional Outcome item endorsements across broad domains (work, school, relationships) and specific tasks (money management, driving, sleep), and connect them to DSM-5-TR Criterion D and to any accommodation rationale. Pitfall: symptom counts with no functional story. Accommodation reviewers and payers act on documented limitation, not on elevation alone. Childhood-onset evidence. Give DSM-5-TR Criterion B its own paragraph built from retrospective history, school records, and collateral report, and say plainly how onset before age 12 was established. Pitfall: citing current CAARS-2 elevations as onset evidence. The instrument measures current functioning; it cannot testify about childhood, and reviewers know it. Interpretive summary and recommendations linkage. Answer the referral question by integrating ratings with interview, records, and any performance or validity testing, state what the scores do and do not establish, and tie each recommendation to a finding, including any re-rating plan on the same edition and norms. Pitfall: conclusions that outrun the data. "Results confirm ADHD" is a sentence no rating scale can carry; "results are consistent with, and corroborated by," survives review. ### Blank template (copy and adapt) MEASURES, FORMS, AND RATERS Instrument/edition: CAARS-2 (2023) Form length: [full-length / Short / ADHD Index] Raters: [self; observer name or role, relationship, time known] Norm reference: [General Population / ADHD Reference; combined-gender or gender-specific; age band] Dates completed: [self] [observer] RESPONSE STYLE AND RATING QUALITY [Negative Impression, Inconsistency, Omitted Items, Pace: stated factually; circumstances that could color ratings] CRITICAL AND SCREENING ITEMS [Endorsed / not endorsed; follow-up conducted and documented] SELF-REPORT RESULTS [Content scales + DSM symptom scales: T scores + descriptors in words; no score treated as a diagnostic cutoff] OBSERVER RESULTS AND CROSS-INFORMANT INTEGRATION [Each rater's scores; convergence and divergence interpreted, not averaged; what disagreement means here] ADHD INDEX [Screening indicator framing; never reported as a diagnosis] IMPAIRMENT AND FUNCTIONAL OUTCOMES [Domains and tasks endorsed; link to Criterion D and any accommodation rationale] CHILDHOOD-ONSET EVIDENCE (CRITERION B) [History, school records, collateral; kept separate from current ratings] INTERPRETIVE SUMMARY (answer the referral question) [What converges across sources; what the scores do NOT establish; differential considered] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-rating plan on the same edition and norms] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample CAARS-2 write-up (fictional) Scenario: a 34-year-old project manager self-referred for a first-time adult ADHD evaluation after repeated workplace difficulties, with her spouse completing the Observer form. Self and observer evidence converge, the internally consistent shape where the write-up's job is to integrate sources and still separate current ratings from childhood-onset proof. This is the CAARS-2 section of the larger evaluation report only, condensed but structurally complete. All details are fictional. Client: A.R., 34 · Referral: first-time adult ADHD evaluation, self-referred · Evaluator: S. Beaumont, PhD, Licensed Psychologist · Ratings completed: 07/09/2026 (self), 07/10/2026 (observer) · Report date: 07/17/2026 Measures and administration: The Conners Adult ADHD Rating Scales, Second Edition (CAARS-2) full-length Self-Report was completed by A.R. and the full-length Observer form by her spouse of nine years, both online, as part of an evaluation that also included a diagnostic interview keyed to DSM-5-TR criteria, developmental and academic history, review of school records, and a performance validity screen, reported in their own sections. Scores are T scores (mean 50, SD 10) referenced to the CAARS-2 General Population comparison sample with combined-gender norms for her age band. Severity language below follows the manual's qualitative descriptors, stated in words; no score is treated as a diagnostic threshold. Response style and rating quality: Both forms were fully completed, with no omitted items and typical response pace for online administration. The Negative Impression Index and Inconsistency Index were unremarkable for both raters. These indicators support, but cannot by themselves establish, that the forms were completed consistently and without exaggeration, and they are read here alongside the interview and record evidence rather than in place of it. No recent life disruption was identified that would be expected to color either rater's responses. Critical and Screening Items: Neither rater endorsed the Critical Items covering suicidal thoughts or attempts or self-injury. A.R. endorsed Screening Items for anxiety at low frequency; this was explored in the clinical interview and is addressed in the diagnostic formulation section of this report. CAARS-2 scale Self-Report T Observer T Inattention/Executive Dysfunction 74 68 Hyperactivity 58 55 Impulsivity 61 57 Emotional Dysregulation 68 64 Negative Self-Concept 66 59 DSM ADHD Inattentive Symptoms 75 69 DSM ADHD Hyperactive/Impulsive Symptoms 59 56 DSM Total ADHD Symptoms 69 64 Self-report results: A.R.'s ratings produced clear elevations on Inattention/Executive Dysfunction and DSM ADHD Inattentive Symptoms, in ranges the manual characterizes as markedly above average, indicating frequent difficulty sustaining attention, organizing work, and following tasks through. Emotional Dysregulation and Negative Self-Concept were moderately elevated, consistent with her interview account of frustration spikes and longstanding self-criticism. Hyperactivity and Impulsivity scales and the DSM Hyperactive/Impulsive scale were not meaningfully elevated. Observer results and cross-informant integration: Her spouse's ratings followed the same profile shape at slightly lower elevations, with Inattention/Executive Dysfunction and DSM ADHD Inattentive Symptoms clearly elevated and hyperactive/impulsive content unremarkable. The direction of the small self-observer difference is common in self-referred adults, and both raters' inattentive-domain scores are independently elevated, so the two perspectives corroborate rather than contradict each other. Neither rater is treated as more correct; the convergence across home and work descriptions strengthens the cross-setting evidence DSM-5-TR requires. ADHD Index: The CAARS-2 ADHD Index was elevated for both raters. The Index is a 12-item screening indicator built to differentiate adults with ADHD from the general population; it raises the probability that a full evaluation will support the diagnosis and is given no diagnostic weight of its own here. Impairment and functional outcomes: Both raters endorsed impairment items concentrated in work performance (missed deadlines, incomplete projects) and household management, with A.R. additionally endorsing difficulty with money management and task-switching. These endorsements, corroborated by the employer documentation she provided, satisfy the functional-impairment showing that DSM-5-TR Criterion D and any later accommodation request both depend on. Childhood-onset evidence: Current ratings cannot establish onset, so Criterion B rests on separate evidence: grade 3 through 6 report cards contain repeated teacher comments about incomplete work and inattention, and a structured collateral interview with A.R.'s mother described the same pattern before age 12, managed at the time with heavy parental scaffolding. That history, not the present CAARS-2 elevations, carries the onset requirement. Interpretive summary and recommendations linkage: Cross-informant CAARS-2 evidence, the diagnostic interview, school records, and collateral history converge on a presentation consistent with DSM-5-TR attention-deficit/hyperactivity disorder, predominantly inattentive presentation, with the diagnosis resting on the full evaluation rather than on any score. The inattentive and executive findings support the medication consultation and the workplace-strategy recommendations in the plan; the emotional dysregulation findings support the skills-focused psychotherapy referral. Re-rating with the same CAARS-2 forms, raters, edition, and norm reference in six months is recommended to measure treatment response on comparable scores. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works The edition, form length, each rater's relationship, and the norm reference (comparison sample and gender-norm choice) are named, which makes the T scores interpretable now and comparable at re-rating. Response style is reported first, factually, and in both directions: the indices are described as supportive evidence about how the forms were completed, never as proof of honesty or of exaggeration. Critical and Screening Items get an explicit disposition sentence, so a reviewer can see the flags were read and followed rather than left inside the score report. Childhood-onset evidence sits in its own paragraph, built from records and collateral history, and the current ratings are never cited as proof that symptoms began before age 12. Severity language stays qualitative and consistent, the ADHD Index is framed as a screening indicator, and every recommendation, including the re-rating plan on the same edition and norms, traces to a finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the CAARS-2 section knowing which decision it will feed, because the surrounding rules differ by lane. On the payer side, US coverage for adult ADHD testing is narrower than most referral sources expect: Aetna's clinical policy CPB 0158 states that neuropsychological or psychological testing is "rarely considered medically necessary for uncomplicated cases" of ADHD, and Blue Cross Blue Shield of Massachusetts policy 151 is blunter: "Neuropsychological testing for the routine diagnosis of ADHD is NOT MEDICALLY NECESSARY." Testing and rating-scale work is billed inside the psychological testing code families by bare code number (96130, 96131, 96136 through 96139 for testing services; 96127 for brief emotional or behavioral instruments), which is coding convention and payer policy, not law, and Medicare's billing articles A57481 and A57780 frame which services those codes describe, so verify coverage before promising a reimbursed battery. On the accommodations side, the expectations are about reasoning, not score dumps: ETS's documentation guidelines for ADHD (2026 edition) ask for a specific current DSM or ICD diagnosis with supporting data, rule-outs for alternative explanations, functional limitations across life contexts, and requested accommodations explicitly linked to those limitations, under the ADA Amendments Act of 2008 framework, and they restate the definitional point that ADHD "is exhibited in childhood or early adolescence", which is why the onset paragraph in your report matters to a testing agency. And on the prescribing side, the report often supports stimulant treatment, where the LAW currently carries an expiry date: the DEA and HHS Fourth Temporary Extension (Federal Register, December 31, 2025) allows telemedicine prescribing of controlled medications without a prior in-person exam through December 31, 2026, with no permanent rule finalized as of July 2026, so an evaluation written this year may be the document a prescriber relies on when those rules change. Edition currency is the second live question. MHS retired the original CAARS, ending online product sales on October 31, 2024, and after May 31, 2025 the original can no longer be administered or scored on the publisher's platform, so a defensible report names the CAARS-2, its form length, and the norm reference used (General Population or ADHD Reference; combined-gender or gender-specific), and treats any comparison against pre-2023 CAARS scores as a stated limitation rather than a trend line, because the content scales were restructured, the norms are new, and the DSM scales moved to DSM-5-TR. The qualification boundary is the publisher's: the CAARS-2 is qualification level B, purchase eligibility does not equal interpretive competence, and interpretation and write-up remain professional acts governed by licensing boards and scope of practice. Guideline context spans jurisdictions as of July 2026: the United States has no published adult ADHD practice guideline, with APSARD's first US guideline still forthcoming (publication has been expected as an open-access document during 2026), so US practice borrows structure from NICE guideline NG87 in the UK, whose recommendation 1.3.2 states a diagnosis should "not be made solely on the basis of rating scale or observational data". In Canada, CADDRA's Canadian ADHD Practice Guidelines (4.1 edition, 2020) frame diagnosis as clinical assessment, and psychological testing for adult ADHD is commonly arranged privately or through extended health benefits, with provincial coverage varying, a market reality rather than a rule. In Australia, the AADPA guideline (2022) recommends that a diagnosis "should not be made solely based on rating scales or observational data", and its adult rating-scale list names the Conners Adult ADHD Rating Scales as published before the CAARS-2 shipped; adult assessment typically routes through psychiatrist attendance items such as MBS item 291 or GP mental health treatment plans, while the complex-neurodevelopmental assessment items (135, 137, 139, 289) are once-per-lifetime items for patients under 25, with MBS guidance stating items 135 and 289 are "not intended for standalone diagnosis of Attention Deficit Hyperactivity Disorder", so CAARS-2 administration time for most Australian adults is not separately rebated; PAA distributes the instrument in Australia at the same qualification level. CAARS and CAARS 2 are trademarks of Multi-Health Systems Inc. (MHS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CAARS-2 write-up errors reviewers flag There is no payer audit series for rating-scale write-ups; the accountability record here is psychometric, unusually recent, and thin by design, because independent CAARS-2 research barely exists yet. As of July 2026 the standout independent peer-reviewed study is a caution, not a confirmation: in a simulation study of 222 adults whose feigning participants were coached, warned to avoid detection, and incentivized, the new Negative Impression Index showed "high specificity (92%) to honest examinees with ADHD, but low sensitivity to simulators (8%)", and most simulators stayed credible while still earning high ADHD Index scores (81 percent) and endorsing significant functional impairment (90 percent) (Weis and Sund, Journal of Attention Disorders, 2026). The one published CAARS-2 structural replication comes from the instrument's own author group with disclosed publisher royalties and employment, and the payers quoted above already treat testing for uncomplicated ADHD as rarely necessary. Those facts predict the failures below. The BastionGPT Clinical Advisory Board sees the same errors most often in CAARS-2 write-up reviews: A 1999-edition write-up wearing CAARS-2 clothes. The section reports "Inattention/Memory Problems" or "Impulsivity/Emotional Lability", scales that exist only on the retired original, or silently compares a 2019 CAARS baseline to current CAARS-2 scores. The architecture was restructured, the norms replaced, and the publisher's platform stopped scoring the original after May 2025; name the edition and treat cross-edition comparison as a limitation. The ADHD Index or a T score written as the diagnosis. "ADHD Index of [score] confirms ADHD" fails twice: the Index is a 12-item screening indicator, and no T score on any scale is a diagnostic cutoff. NICE guideline NG87 says a diagnosis should "not be made solely on the basis of rating scale or observational data", and Australia's AADPA guideline makes the same point; the DSM-5-TR diagnosis is clinical. Validity indices over-read in either direction. A clean Negative Impression Index gets written as proof the client answered honestly, or an elevated one as proof of malingering. The independent evidence supports neither reading: the index missed 92 percent of coached simulators, so describe response style factually, corroborate symptom claims with real-world functional evidence, and keep accusatory language out of the record. Current ratings offered as childhood-onset evidence. The report cites elevated CAARS-2 scores to satisfy Criterion B. The instrument measures current functioning; onset before age 12 needs history, records, or collateral report, and reviewers at testing agencies specifically look for that separation. Observer evidence mishandled at both extremes. The write-up either treats a missing observer form as invalidating, though no guideline mandates one and a documented absence with corroborating records is defensible, or it averages disagreeing raters into one number. Self and observer perspectives agree only moderately (the publisher reports r = .44 to .54), longitudinal cohorts show adults with childhood ADHD under-report relative to informants, and the write-up's job is to interpret the difference, not erase it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your score summary (forms, raters, T scores, response-style notes) and get a drafted CAARS-2 results-section narrative with the cross-informant integration, validity language, and onset-evidence separation scaffolded in place for your review. Cross-check a finished draft for the gaps reviewers flag: scores that disagree with their descriptor language, 1999-edition scale names in a CAARS-2 report, an ADHD Index written as a diagnosis, and current ratings cited as childhood-onset evidence. Translate the results section into a plain-language summary for the client, their prescriber, or a care team that keeps elevations framed as reported difficulty, not a diagnosis. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do CAARS-2 T-scores actually mean? Every CAARS-2 scale reports a T score, a standard score with a mean of 50 and a standard deviation of 10, referenced to the norm group you selected : the General Population sample or the ADHD Reference sample, with combined-gender or gender-specific options, across seven adult age bands. Higher scores mean the rater reports more ADHD-consistent difficulty than the comparison group. The manual characterizes elevations with qualitative descriptors, and the defensible write-up states those descriptors in words, names the norm reference, and treats no score as a diagnostic threshold: a T score quantifies reported symptoms, and the diagnosis belongs to the whole evaluation. What changed from the CAARS to the CAARS-2, and can I compare scores across them? The 2023 revision restructured the instrument: full-length forms grew from 66 to 97 items, the old Impulsivity/Emotional Lability scale was split into separate Impulsivity and Emotional Dysregulation scales, Inattention/Memory Problems became Inattention/Executive Dysfunction, DSM scales moved to DSM-5-TR, the ADHD Index was rebuilt with improved sensitivity and specificity, and a Response Style Analysis (Negative Impression Index, Inconsistency Index, Omitted Items, Pace), Critical and Screening Items, and item-level impairment ratings were added, all on new US-and-Canada norms reaching age 70 and older. MHS retired the original : online sales ended October 31, 2024, and platform scoring ended after May 31, 2025. Treat the two as different instruments; a pre-2023 CAARS baseline belongs in the text as history, with any comparison framed as a stated limitation, not a trend line. Is the CAARS-2 required to diagnose adult ADHD? No. No US, Canadian, or Australian authority requires the CAARS-2, any rating scale, an observer form, a continuous performance test, or any T-score cutoff for an adult ADHD diagnosis, and the guidelines that exist point the other way: NICE guideline NG87 (recommendation 1.3.2) states a diagnosis should "not be made solely on the basis of rating scale or observational data", and Australia's AADPA guideline (2022) says the same. The United States has no published adult guideline as of July 2026; APSARD's first US adult ADHD guideline remains forthcoming. The DSM-5-TR diagnosis is clinical: criteria mapping, onset before age 12, cross-setting impairment, and differential rule-outs. The CAARS-2 earns its place as corroborating, norm-referenced evidence inside that evaluation, which is exactly how the write-up should frame it. Do I need an Observer form, and what if there is no observer? No rule mandates an observer, and a missing Observer form does not invalidate an evaluation. It is still worth pursuing: self and observer ratings agree only moderately (the publisher reports inter-rater r = .44 to .54), which means each perspective adds information, and longitudinal cohorts show adults with childhood ADHD under-report relative to informants , with "Persistence rates were higher when using parent rather than self-reports" the MTA follow-up's finding (Sibley and colleagues, 2017). At the same time, adult self-report is not inherently untrustworthy: clinic-referred adults showed moderate to high self-other agreement on current functioning (Barkley, Knouse, and Murphy, 2011). Any adult who knows the person well can serve: partner, parent, sibling, close friend, sometimes a coworker. When none exists, the defensible write-up documents the absence, explains it, and leans on history, records, and interview instead. Can the CAARS-2 detect exaggeration or malingering? Only weakly, and the write-up should say so by design rather than by accident. The CAARS-2 added a Negative Impression Index precisely because the original CAARS had no over-reporting index, but the first independent test was sobering: in a 2026 simulation study of 222 adults , the index showed "high specificity (92%) to honest examinees with ADHD, but low sensitivity to simulators (8%)", and most coached simulators stayed credible while producing high ADHD Index scores and endorsing significant impairment (Weis and Sund, Journal of Attention Disorders, 2026). The authors' advice matches best practice: "continue to corroborate elevated CAARS-2 symptom scores with objective evidence of real-world functional limitations". So write validity indicators factually and non-accusatorily, treat a clean index as supportive rather than probative, pair rating scales with performance and symptom validity testing where stakes warrant it, and never let a single index carry a feigning conclusion. When do I use the CAARS-2 versus the Conners 4, ASRS, BAARS-IV, or DIVA-5? By age, depth, and job. The Conners 4 covers ages 6 to 18 and hands off to the CAARS-2 at 18; the publisher built the two to be aligned for lifespan continuity, which makes the pair the natural choice when a childhood Conners history exists. The ASRS is a free WHO screener: strong for flagging who needs evaluation, no validity scales, no diagnostic weight. The BAARS-IV is an inexpensive normed alternative whose retrospective childhood forms help document onset. The DIVA-5 is a structured diagnostic interview that maps DSM criteria conversationally rather than producing norm-referenced severity scores. And when the referral question is executive function in daily life rather than ADHD symptom severity, an executive function rating such as the BRIEF-2 family covers that lane. A defensible adult battery typically pairs a clinical interview and criterion mapping with one normed rating scale like the CAARS-2, collateral input, impairment documentation, and validity checks, with cognitive testing reserved for differential questions. Where can I find the CAARS-2 questions or a free scoring key? Not on this page, and not legitimately on any public page. The 97 items, scoring algorithms, and norm tables are protected test materials at MHS qualification level B , scoring exists only on the publisher's platform for this edition, and psychologists carry an ethical duty to maintain test security under APA Ethics Standard 9.11. Sites offering CAARS-2 item text or self-scoring undermine the controls that make the norms meaningful, and coaching research shows why that matters: simulators who know what scales look for evade detection easily. What a report can include: T scores, qualitative descriptors in words, and your own prose describing what the scales measure. What it cannot: item text, scoring rules, or reproduced norm tables. Will insurance cover CAARS-2 testing for adult ADHD? Verify before promising, because the default answer is narrower than referral sources expect. Major US commercial policies treat testing for uncomplicated ADHD as rarely or not medically necessary: Aetna CPB 0158 says "rarely considered medically necessary for uncomplicated cases", and BCBS of Massachusetts policy 151 states testing for routine ADHD diagnosis is not medically necessary; testing services bill under the 96130 through 96139 family and brief instruments under 96127, with Medicare's articles A57481 and A57780 framing those services. Coverage improves when complexity is documented: treatment that has not worked, neurological questions, or genuine differential complexity. In Canada, adult ADHD testing is commonly private-pay or covered through extended health benefits, varying by province. In Australia, assessment usually routes through psychiatrist attendance items such as item 291 or GP mental health treatment plans, and the under-25 complex-neurodevelopmental items are not intended for standalone ADHD diagnosis, so the rating scale itself is typically absorbed into consultation time rather than separately rebated. Can BastionGPT help with CAARS-2 write-ups? Yes. BastionGPT is trained and clinically tested on psychological and neuropsychological evaluation reports, the parent documents CAARS-2 sections live inside. Paste a score summary (forms, raters, T scores, response-style notes) and it drafts the results-section narrative with the cross-informant integration, factual validity language, and onset-evidence separation in place for your review; it can also check a finished section for score-descriptor mismatches, 1999-edition scale names, an ADHD Index written as a diagnosis, or a missing impairment linkage, and produce a plain-language summary for the client or their prescriber. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: MHS, CAARS 2 product page : 2023 publication, ages 18 and older, item counts (97/97 full-length; 55/52 Short; 12 ADHD Index), administration time ranges, qualification level B, translations, online administration and scoring, reference-sample and report options. MHS, CAARS 2 brochure : authorship, the five content scales, DSM-5-TR symptom scales, Response Style Analysis components, Critical and Screening Items, Impairment & Functional Outcome Items, normative sample composition (2,640; 1,320 per rater; US and Canada census stratification; seven age bands), reference samples and gender-norm options, reading levels, and the publisher-reported reliability, validity, and fairness figures cited as publisher-derived. MHS, CAARS 2 comparison with CAARS : the original's 66-item long forms versus 97-item CAARS 2, scale-architecture changes including the Impulsivity/Emotional Lability split, normative age-group expansion, administration formats including online email-link administration, and report types. MHS, CAARS 2 FAQ : digital-only scoring, paper administration with online entry, average administration times, Single-Rater Report contents, and the CAARS-to-CAARS-2 use conversion policy. Brainworx (authorized MHS distributor), CAARS discontinuation notices: online sales ending October 31, 2024 and platform use ending May 31, 2025 . Weis R, Sund AM, Susceptibility of the CAARS-2 Self-Report to Sophisticated Malingering , Journal of Attention Disorders, 2026;30(8):1080-1087: the Negative Impression Index specificity and sensitivity findings and the corroboration recommendation quoted on this page. Clark SV, Erhardt D, Sparrow EP, Solomon M, Is the structure of ADHD stable across adulthood? Insights from the CAARS 2 , Psychiatry Research, 2026;362:117194: five-factor measurement invariance across adult age bands, cited here with its disclosed publisher affiliations (author royalties and MHS employment). Suhr JA, Buelow M, Riddle T, Development of an Infrequency Index for the CAARS, Journal of Psychoeducational Assessment, 2011;29(2):160-170 ( doi:10.1177/0734282910380190 ): the research-derived over-reporting index for the original CAARS, which itself shipped with only an Inconsistency Index. Sibley MH et al., MTA Cooperative Group, Defining ADHD symptom persistence in adulthood , Journal of Child Psychology and Psychiatry, 2017;58(6):655-662, with Sibley et al., 2012 : informant-versus-self-report persistence gaps and adult under-reporting. Barkley RA, Knouse LE, Murphy KR, Correspondence and disparity in the self- and other ratings of current and childhood ADHD symptoms , Psychological Assessment, 2011;23(2):437-446: moderate to high self-other agreement on current functioning in adults. NICE, guideline NG87, Attention deficit hyperactivity disorder: diagnosis and management (2018, updated 2019), recommendation 1.3.2. AADPA, Australian Evidence-Based Clinical Practice ADHD Guideline, Summary of Recommendations (2022), recommendation 2.1.4, and CADDRA, Canadian ADHD Practice Guidelines (4.1 edition, 2020); APSARD, US guidelines for adults with ADHD (in development, unpublished as of July 2026). Aetna, clinical policy CPB 0158, Neuropsychological and Psychological Testing (last review March 2026), and Blue Cross Blue Shield of Massachusetts, medical policy 151, Neuropsychological and Psychological Testing : the ADHD medical-necessity positions quoted on this page. CMS Medicare Coverage Database, billing and coding articles A57481 and A57780 : the psychological testing and screening code families. DEA and HHS, Fourth Temporary Extension of COVID-19 Telemedicine Flexibilities for Prescription of Controlled Medications , Federal Register, December 31, 2025 (effective through December 31, 2026). ETS, Documentation Guidelines for Test Takers with ADHD (2026 edition): the diagnostic-report elements and functional-limitation linkage expectations. MBS Online, item 291 and the complex neurodevelopmental disorder items FAQ ; PAA, CAARS 2 Australian distribution . ## Related templates Conners 4 Report Write-Up BRIEF-2 Report Write-Up Psychological Evaluation Report Psychiatric Diagnostic Evaluation ADHD Evaluation Report ASRS v1.1 Score Documentation Vanderbilt ADHD Scale Documentation T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/conners-4-report All note templates # Conners 4 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The Conners 4 (Conners 4th Edition) is a multi-informant rating scale measuring ADHD symptoms, co-occurring problems, and functional impairment for ages 6 to 18, with parent, teacher, and self-report forms. Psychologists, school psychologists, and pediatric clinicians use it in ADHD evaluations, school assessments, and treatment monitoring. This page covers how to write up Conners 4 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and pediatric clinicians; publisher qualification level B Audience IEP and Section 504 teams, pediatricians and prescribers, payers, testing-accommodations reviewers, parents Typical length 300 to 700 words for the results section · administration about 12 to 15 minutes per full-length form Format family Multi-informant ADHD rating scale When it's used ADHD referrals and differential diagnosis, school and special-education evaluations, medication and treatment monitoring, accommodation applications Standards context Published by MHS (2022); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Conners 4? The Conners 4th Edition (Conners 4) is a multi-informant rating scale measuring ADHD symptoms, co-occurring problems, and functional impairment in young people ages 6 to 18, published by Multi-Health Systems (MHS) in 2022 under the authorship of C. Keith Conners, whose rating scales have anchored child ADHD assessment since the late 1960s. Parent and Teacher forms cover ages 6 through 18 and a Self-Report covers ages 8 through 18, each in three lengths: full-length, Short, and a 12-item ADHD Index. Administration and scoring are fully digital, on the MHS Online Assessment Center+ or Q-global through the Pearson distribution channel: paper responses can be entered by the examiner, but hand scoring is not possible, and results are T-scores and percentiles against combined United States and Canada norms. Edition status matters as of July 2026: MHS discontinued Conners 3 paper and hand-scored products at the end of 2023, so new Conners work runs on the Conners 4, with the Conners Early Childhood covering ages 2 to 6 and the CAARS 2 covering adults. The load-bearing rule for write-ups: the Conners 4 describes probability, not diagnosis. Its T-scores quantify how a young person's reported behaviors compare with same-age peers, and its ADHD Index expresses how closely a response pattern resembles those of youth already diagnosed with ADHD, useful evidence, never a verdict. A DSM-5-TR ADHD diagnosis requires symptoms in two or more settings, onset before age 12, and functional impairment, none of which a single rating form can establish. That is why Conners 4 results enter a psychological evaluation report or a diagnostic evaluation as one converging data source, and why the sentence that turns a score into a diagnosis is the one reviewers flag first. ## Who uses Conners 4 write-ups and when School psychologists write up Conners 4 results inside special-education and Section 504 evaluations; clinical child psychologists use it for diagnostic clarification when ADHD is the referral question; pediatricians, psychiatric nurse practitioners, and child psychiatrists use it to document symptom baselines and medication response; and pediatric neuropsychologists fold it into a neuropsychological report as the ADHD-specific rating layer alongside performance testing. It is the right tool when the question is ADHD-shaped and you need symptom-level, DSM-aligned detail with impairment ratings across home and school. When the question is broadband, what else might be going on, a system such as the BASC-3 surveys externalizing, internalizing, and adaptive domains in one pass, and many evaluations use both. In primary care, the free NICHQ Vanderbilt scales cover screening for ages 6 to 12, and at the transition to adulthood the CAARS 2 takes over from age 18. The Conners 4 write-up is where symptom detail, impairment, and informant perspective get integrated into one defensible account. ## How to structure a Conners 4 results section No statute, payer, or publisher mandates a results-section format. The sequence below follows the interpretation order the publisher's own materials teach, validity first, critical items next, then scales, then integration, because a write-up that mirrors it survives review: every score arrives with its foundation already stated. Each section carries the pitfall that most often undermines it. Measures, raters, and norms. Name the instrument and edition, the form length each rater completed (full-length, Short, or ADHD Index), each rater's relationship and setting, administration dates, and the norm set used (combined-gender by default, gender-specific if selected). Pitfall: writing "the Conners" with no edition or form length. The Conners 4 restructured scales and dropped others, so an unnamed edition leaves reviewers unable to tell which instrument, and which norms, produced your T-scores. Validity and response style, first. Before any score, state each protocol's Response Style Analysis results, Negative Impression Index, Inconsistency Index, Omitted Items, and Pace, and the interpretive consequence in plain language. Pitfall: importing Conners 3 validity vocabulary. The Positive Impression scale no longer exists, and a report that cites one announces that the template predates the instrument. Critical items and indicators. State that the Self-Harm and Severe Conduct critical items and the Sleep Problems Indicator were reviewed, whether anything was endorsed, and what action followed. If they were disabled at administration, say so. Pitfall: silence. A reviewer who cannot tell whether critical items were checked reads the omission as "not checked," and in a record that later matters, so may a board. DSM symptom scales, both metrics kept distinct. Report the ADHD Inattentive, ADHD Hyperactive/Impulsive, Total ADHD Symptoms, ODD, and CD scales with their T-scores, and the Symptom Counts alongside as absolute tallies. Write elevations as "consistent with" DSM-5-TR Criterion A, never as the diagnosis. Pitfall: conflating the metrics. A T-score is relative to norms; a Symptom Count is an absolute tally; one cannot stand in for the other, and the Total ADHD Symptoms scale has no count at all. Content scales. Move through the six full-length Content Scales, Inattention/Executive Dysfunction, Hyperactivity, Impulsivity, Emotional Dysregulation, Depressed Mood, Anxious Thoughts, integrating raters within each scale rather than summarizing printouts in sequence. Pitfall: citing scales the administered form does not contain. The Short form carries only the first four, and Conners 3 scales such as Learning Problems or the Global Index no longer exist. Impairment and functional outcome scales. Report Schoolwork, Peer Interactions, and Family Life, and connect them to the DSM impairment requirement: symptom elevation without documented functional impact supports nothing. Pitfall: expecting Family Life from a teacher protocol. The Teacher form does not include it; only parent and self-report do. ADHD Index, correctly framed. Give the probability band and its plain-language meaning: the score expresses how similar this profile is to those of young people already diagnosed with ADHD, useful screening weight and nothing more. Pitfall: the Index written as confirmation. "The ADHD Index confirms the diagnosis" is the single most flagged sentence in Conners write-ups. Cross-informant integration. One paragraph stating where raters converge, where they diverge, and what the divergence means. Modest agreement is the empirical norm, parent ratings typically run higher than teacher ratings, and divergence usually carries setting information. Pitfall: averaging raters into a verdict, or crowning one correct. A home-elevated, school-average profile is a finding about settings, not a scoring dispute. Interpretive summary and recommendations linkage. Answer the referral question directly, keep band language normative with its plain-language meaning, state what the scores do not establish, and tie every recommendation to a specific finding, ending with the re-rating plan: who gets re-rated, with which form length, and when. Pitfall: recommendations that could follow any profile. If nothing in the results section makes a recommendation necessary, either the recommendation or the results section is incomplete. ### Blank template (copy and adapt) MEASURES AND RATERS Instrument/edition: Conners 4 Form length: [full / Short / Index] Raters: [role, relationship, setting, form, date completed] Norm set: [combined-gender / gender-specific] VALIDITY / RESPONSE STYLE (every rater, before any score) [Negative Impression, Inconsistency, Omitted Items, Pace: classification + interpretive consequence; flagged protocols handled in writing, not silently] CRITICAL ITEMS AND INDICATORS [Self-Harm + Severe Conduct critical items, Sleep Problems Indicator: reviewed, endorsed or not, action taken] DSM SYMPTOM SCALES (per rater: T-score AND Symptom Count) Inattentive: [T / count] Hyperactive/Impulsive: [T / count] Total ADHD Symptoms: [T only] ODD / CD: [T / count] ["consistent with" DSM criteria, never "diagnostic of"] CONTENT SCALES (integrate raters within each scale) [Inattention/Exec Dysfunction, Hyperactivity, Impulsivity, Emotional Dysregulation, Depressed Mood, Anxious Thoughts; name only scales the administered form contains] IMPAIRMENT & FUNCTIONAL OUTCOME Schoolwork: [ ] Peer Interactions: [ ] Family Life (parent and self-report only): [ ] ADHD INDEX [Probability band + plain-language meaning; screening weight] CROSS-INFORMANT INTEGRATION [Convergence, divergence, what divergence means (settings, perspective); do not average it away] INTERPRETIVE SUMMARY (answer the referral question) [Bands as normative description; measurement error; what the scores do NOT establish on their own] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-rating plan: who, which form length, when] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample Conners 4 write-up (fictional) Scenario: a pediatrician refers an 11-year-old for evaluation after two years of escalating attention and behavior concerns at school and at home, the classic cross-setting ADHD referral. Parent, teacher, and self-report full-length forms were administered. This is the results section only, condensed but structurally complete. All details are fictional. Student: C.R., 11 (grade 5) · Referral: K. Iyer, MD (pediatrics), attention and behavior concerns at school and home · Evaluator: R. Vasquez, PsyD, Licensed Psychologist · Ratings completed: 07/06 to 07/09/2026 · Report date: 07/15/2026 Measures and raters: The Conners 4th Edition (Conners 4) was administered as part of a broader evaluation: full-length Parent form completed by C.R.'s mother; full-length Teacher form completed by his grade 5 classroom teacher, who taught him daily across the school year just ended; and the full-length Self-Report, completed by C.R. All forms were administered online and scored against the Conners 4 combined-gender norms. T-scores have a mean of 50 and a standard deviation of 10; higher scores mean more reported concerns. Validity and response style: On all three protocols, the Negative Impression Index, Inconsistency Index, and Omitted Items fell within acceptable limits, and response pacing was unremarkable. All three profiles are considered interpretable, and each rater had adequate recent opportunity to observe. Critical items and indicators: The Self-Harm and Severe Conduct critical items were reviewed for all raters; none were endorsed. The Sleep Problems Indicator was not elevated. No further action was indicated. DSM symptom scales: Parent and teacher ratings converge on inattentive and hyperactive/impulsive symptoms. On the ADHD Inattentive Symptoms scale, the parent T-score of 76 (Very Elevated, above the 99th percentile; Symptom Count 8 of 9) and teacher T-score of 70 (Very Elevated, approximately the 98th percentile; Symptom Count 7 of 9) indicate many more inattentive behaviors than are typical for his age. ADHD Hyperactive/Impulsive Symptoms were similarly Elevated to Very Elevated (parent T = 71, Symptom Count 7; teacher T = 65, Symptom Count 6). Total ADHD Symptoms T-scores were 75 (parent) and 69 (teacher). C.R.'s self-report was consistent in direction but more moderate (Inattentive T = 64, Slightly Elevated; Hyperactive/Impulsive T = 58, Average). Oppositional Defiant Disorder and Conduct Disorder symptom scales were Average for all raters, with Symptom Counts of one or zero. Symptom Counts are absolute tallies that contribute to, but do not establish, DSM-5-TR Criterion A. Content scales: The same pattern holds at the content level. Inattention/Executive Dysfunction was Very Elevated for the parent (T = 74) and Elevated for the teacher (T = 69) and C.R. (T = 66), with both adult raters describing difficulty sustaining focus, following multi-step directions, and finishing work. Hyperactivity (parent T = 69; teacher T = 66) and Impulsivity (parent T = 68; teacher T = 65) were Elevated across adult raters. Emotional Dysregulation, Depressed Mood, and Anxious Thoughts were Average for all raters, offering no rating-scale support for a mood or anxiety alternative to the referral question. Impairment and functional outcome: Impairment ratings track the symptom picture. Schoolwork impairment was Very Elevated for the parent (T = 77) and teacher (T = 72) and Slightly Elevated on self-report (T = 63). Peer Interactions ratings were Slightly Elevated for the parent (T = 64) and Average for the teacher (T = 58). On the two forms that assess it, the parent and self-report, Family Life impairment was Elevated for the parent (T = 66) and Slightly Elevated on self-report (T = 61), consistent with the homework and morning-routine conflict described at intake. The Teacher form does not include the Family Life scale. ADHD Index: The parent and teacher ADHD Index probabilities fell in the Very High band and the self-report in the High band, meaning the response patterns resemble those of young people diagnosed with ADHD more than those of the general population. The Index expresses similarity, not certainty; it carries screening weight and is not a diagnosis. Cross-informant integration: Agreement across the three raters is strong for this instrument class, where modest cross-informant correspondence is the norm. The parent's somewhat higher elevations than the teacher's follow the typical direction of informant differences and plausibly reflect the lower structure of the home setting, particularly around homework and routines; C.R.'s more moderate self-ratings are common for his age and do not undercut the adult consensus. No rater's data were discounted. Summary: Conners 4 results show convergent parent and teacher elevations on inattentive and hyperactive/impulsive symptom scales with corroborating self-report, High to Very High ADHD Index probabilities, documented impairment in schoolwork, peer, and family functioning, and no rating-scale support for a mood, anxiety, or conduct alternative. These findings are consistent with DSM-5-TR Criterion A symptom presence across home and school. They are rating-scale evidence: elevations relative to same-age peers, subject to measurement error, and not by themselves a diagnosis. Integrated with the developmental history, clinical interview, and records reviewed elsewhere in this evaluation, they support a diagnosis of ADHD, combined presentation, made on the full evaluation rather than on any score. Linkage to recommendations: The cross-setting symptom and Schoolwork impairment findings support recommendations 1 and 2 (school consultation regarding classroom supports, and discussion of evidence-based treatment options with the referring pediatrician). The Family Life impairment elevations support recommendation 3 (parent behavioral training with a homework-routine focus). Recommendation 4 sets the monitoring plan: re-rate with the parent and teacher Conners 4 Short forms in 10 to 12 weeks to measure response. This sample is fictional and for educational purposes. It does not describe a real child or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works Validity and critical items are addressed before any score is interpreted, so every finding that follows has a stated foundation. T-scores and Symptom Counts stay distinct: relative standing and absolute tallies each do their own evidentiary job, and no band label drifts between sections. The edition, form lengths, raters, settings, and norm set are named, which makes the scores interpretable now and comparable at re-rating. Informant differences are interpreted as setting information rather than averaged, and the ADHD Index is framed as probability, not proof. The summary states what the scores do not establish, and every recommendation traces to a specific finding, with a dated re-rating plan naming the form length. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision it will feed, and label the rules honestly. The diagnosis itself runs on DSM-5-TR criteria: several symptoms in two or more settings, several symptoms present before age 12, and clear evidence that symptoms interfere with functioning (summarized for clinicians on the CDC's clinical care page ); the AAP clinical practice guideline adds, as professional policy, that evaluation should gather rating-scale information from parents and school personnel across settings ( Pediatrics, 2019 ); choosing the Conners 4 as that instrument is convention, and no statute requires two informants or names any test. In US schools, ADHD is listed by name in the IDEA other-health-impairment definition, but eligibility turns on whether the condition "adversely affects a child's educational performance" ( 34 CFR 300.8(c)(9) ), no single measure may be the sole criterion ( 34 CFR 300.304(b)(2) ), and the Department of Education's 2016 ADHD guidance confirms a diagnosis alone does not finish the analysis: IDEA eligibility also requires that the student need special education, while Section 504 runs its own substantial-limitation standard with no test list. Payer-side rules are policy, not law: rating-scale work bills inside the psychological-testing code families (96130 and 96131 for evaluation services, 96136 through 96139 for administration and scoring, 96146 for a single automated instrument), brief standardized instruments have their own family (96110, 96112, 96113, 96127) under Medicare's statutory screening exclusion ( WPS article A57481 ), and current NCCI edits pair 96127 against the testing codes with no modifier bypass, so the brief-instrument code does not ride the same claim day as 96130-series testing. Stimulant prior-authorization checklists commonly expect documented DSM criteria, cross-informant scales, and impairment as plan policy, and testing-accommodation bodies run separate applications: College Board requires a request even with an IEP or 504 plan in place and says approval can take up to seven weeks , while ACT's documentation criteria ask for rating-scale measures from multiple sources and state that failure to finish timed tests cannot demonstrate impairment in isolation. Prescribers should date-stamp the telehealth context: DEA and HHS extended controlled-substance telemedicine flexibilities through December 31, 2026 ( Fourth Temporary Extension ) with the permanent special-registration framework still in rulemaking, and the enforcement backdrop, including Cerebral's 2024 non-prosecution agreement with a $3,652,000 forfeiture over its ADHD prescribing practices, has made documented, defensible assessment the industry's survival trait. In Canada, the CADDRA practice guidelines (4.1 edition, 2020) build ADHD assessment around a free toolkit of scales such as the SNAP-IV-26 and the Weiss functional impairment measures, so a commercial instrument such as the Conners 4 enters as clinician choice rather than toolkit default, carrying combined US-Canada norms that travel well. In Australia, the evidence-based ADHD guideline lists Conners rating scales among usable tools (its 2022 edition cites the Conners 3), recommends that a diagnosis never be made "solely based on rating scales or observational data," and its cited meta-analysis found no child ADHD screening tool met acceptable sensitivity and specificity, both above 0.8; the MBS complex-neurodevelopmental-disorder assessment items add their own scope rules, with items 135 and 289 explicitly not intended for standalone ADHD assessment without other neurodevelopmental comorbidity. Everywhere, rating-scale results support, and never substitute for, the clinical assessment. Edition currency is the second defensibility layer. The Conners 4 restructured the instrument, not just the norms: the Positive Impression scale is gone, the Response Style Analysis now runs on the Negative Impression Index, Inconsistency Index, Omitted Items, and Pace, separate Hyperactivity and Impulsivity content scales replaced the old combined scale, and Conners 3 furniture such as the Global Index and Learning Problems scale no longer exists. MHS discontinued Conners 3 paper and hand-scored products on December 31, 2023 and converts remaining online uses one for one, so treat Conners 3 baselines as a stated limitation and re-baseline on the Conners 4 rather than comparing T-scores across editions, whose norms and scales differ. Name the edition, form length, and norm set (combined-gender by default, gender-specific if selected) in every report; and note that scoring is digital-only on the publisher's and distributor's platforms, so a "hand-scored Conners 4" in a record is a red flag on its face. Access boundaries are publisher rules with professional-ethics backing: the Conners 4 is a qualification-level B instrument, purchase and administration are restricted accordingly, and item content is protected test material, so route records requests that touch protocols through your test-security obligations and put your integration and conclusions, the write-up itself, in the ordinary record where they belong. Conners is a registered trademark of Multi-Health Systems, Inc.; the Conners 4 is published by MHS and distributed by Pearson. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Conners 4 write-up errors reviewers flag There is no payer audit series for rating-scale write-ups; the accountability literature here is psychometric, and for the Conners 4 it is unusually one-sided. The instrument is reviewed in the Buros Center's 22nd Mental Measurements Yearbook , where the published review judges it well designed and well normed while flagging the May 2019 to March 2020 standardization window (pre-pandemic data, a caveat worth stating at re-evaluations), small samples in several reliability studies, and confirmatory-only analyses awaiting independent replication. Independent, non-publisher validation of the 2022 edition is essentially absent as of mid-2026, nearly every published figure traces to the manual, and at least one widely copied citation for the instrument's psychometrics, a supposed 2023 "Tatar et al." paper, appears in no database and its DOI resolves to nothing. Meanwhile cross-informant agreement in this instrument class averages r = .28 across 341 studies . Those findings predict the write-up failures. The BastionGPT Clinical Advisory Board sees the same errors most often in Conners 4 write-up reviews: Scores written as diagnoses. "Inattention is Very Elevated" becomes "the Conners confirms ADHD," or the ADHD Index is quoted as a verdict. Every scale is a probability statement relative to norms; the diagnosis belongs to the whole evaluation, with cross-setting evidence, onset, and impairment established outside any single form. Conners 3 furniture in a Conners 4 report. A Positive Impression score, a Global Index, a Learning Problems scale, or old response-anchor language in a report dated 2026 tells the reviewer the template predates the instrument, and puts every other sentence under suspicion. The Conners 4 changed the scale architecture, not just the norms. T-scores and Symptom Counts conflated. The count reported as if norm-referenced, the T-score treated as a symptom tally, or a count quietly attached to the Total ADHD Symptoms scale, which does not have one. The two metrics answer different questions, relative standing versus absolute presence, and the write-up must keep them doing separate work. Informant disagreement averaged away. Parent-high, teacher-moderate profiles get split down the middle, or one rater is crowned correct. Modest agreement is the empirical norm and parent ratings typically run higher; divergence carries setting information, and the defensible sentence says what the discrepancy means, not which rater won. Publisher psychometrics dressed up as independent evidence. Reliability and validity figures cited to secondary blogs, or to references that do not exist, when nearly every number available traces to the MHS manual. The defensible move is attribution: say the figures are publisher-reported, cite the review literature that exists, and let the thin independent base argue for, not against, your multi-source evaluation design. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (forms, raters, T-scores, Symptom Counts, classifications) and get a drafted results-section narrative with validity addressed first, both DSM metrics kept distinct, and informants integrated rather than averaged. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with their band labels, Conners 3 scale names left in a Conners 4 report, and an ADHD Index written as confirmation instead of probability. Translate the results section into a plain-language summary for parents and teachers that keeps elevations framed as comparisons with same-age peers, not diagnoses. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Conners 4 T-scores actually mean? Every Conners 4 scale reports a T-score, a standard score with a mean of 50 and a standard deviation of 10, plus a percentile against the combined US and Canada norms ; higher scores mean more reported concerns. The publisher's classification bands, running from Low through Average, Slightly Elevated, Elevated, and Very Elevated, are normative descriptions: how elevated the rating is relative to same-age peers, subject to measurement error. The DSM Symptom Scales add Symptom Counts, absolute tallies that are a different metric from the T-score. Write bands as bands, pair each with its plain-language meaning, and let the diagnosis, if any, come from the whole evaluation. Does a high ADHD Index score mean the client has ADHD? No. The ADHD Index expresses how closely a response pattern resembles those of young people already diagnosed with ADHD compared with the general population, reported as a probability with a qualitative band from Very Low to Very High. It is screening evidence: a high value says "this profile looks like the profiles of diagnosed youth," not "this child has ADHD," and a diagnosis still requires the DSM-5-TR package of cross-setting symptoms, onset before age 12, impairment, and a clinical evaluation. "The ADHD Index confirms the diagnosis" is the single most flagged sentence in Conners write-ups; the defensible framing is "consistent with," inside a full evaluation . Can I compare Conners 3 scores with Conners 4 scores? Not directly. The Conners 4 changed the scale architecture, the items, and the norms: separate Hyperactivity and Impulsivity scales, a new Total ADHD Symptoms scale, no Positive Impression scale, and combined US-Canada norms collected in 2019 and 2020. Published convergence between corresponding Conners 3 and Conners 4 scales varies widely, so a Conners 3 baseline is context, not a comparison point. MHS discontinued Conners 3 paper and hand-scored products at the end of 2023 and converts online uses one for one. The defensible pattern for monitoring: re-baseline on the Conners 4, name both editions in the record, and state the cross-edition limitation in text. Is the Conners 4 required for an ADHD diagnosis, an IEP, or a 504 plan? No authority names it. DSM-5-TR sets the diagnostic criteria, and the AAP guideline recommends rating-scale data from more than one setting as professional policy, without naming an instrument. For special education, ADHD is listed in the IDEA other-health-impairment definition, but eligibility requires adverse educational impact and forbids any single measure as the sole criterion ( 34 CFR 300.304(b)(2) ), and federal guidance is explicit that a medical diagnosis alone does not settle it; Section 504 runs its own substantial-limitation standard with no test list. The Conners 4's role is evidentiary convention: a norm-referenced, multi-informant way to document what the criteria and categories actually ask about. When do I use the Conners 4 versus the BASC-3, the Vanderbilt, or the CAARS 2? By bandwidth, budget, and age. The Conners 4 is the ADHD specialist: DSM-aligned symptom scales with Symptom Counts, ADHD-relevant content scales, impairment ratings, and an ADHD Index, for ages 6 to 18. The BASC-3 is broadband, surveying externalizing, internalizing, school, and adaptive domains when the differential is wide, and many evaluations use both. The free NICHQ Vanderbilt scales cover primary-care screening for ages 6 to 12, at zero cost but without the Conners 4's profile depth and impairment architecture. At age 18 the hand-off is the CAARS 2 , the adult Conners, published by MHS in 2023. The write-up rule is the same everywhere: name what you used and why it fit the referral question. What if I only have a parent rating and no teacher form? Document the second setting another way, and say so. DSM-5-TR requires evidence of symptoms in two or more settings; it does not require a teacher rating form, and no statute mandates two informants. When teacher forms do not come back, the accepted moves are school records and report cards, direct classroom observation, a school-staff interview, or ratings from another adult who sees the child in a structured setting, each named in the report as the cross-setting evidence. The write-up then states the limitation plainly: one informant rated, corroboration drawn from records, interpret accordingly. What reviewers flag is not the missing form; it is a report that claims cross-setting evidence it never documents. How often can I re-administer the Conners 4 to monitor treatment? No fixed interval exists in any rule. The publisher's own stability studies used re-rating windows of roughly two to four weeks, and the AAP guideline calls for structured follow-up of treatment response as policy, without prescribing a schedule. In practice, evaluators re-rate around meaningful change points, a medication start or adjustment, a new school term, a treatment-plan review, and use the Short form to cut respondent burden, since it keeps the core symptom scales while dropping two content scales and the Inconsistency Index. The write-up rule: name the interval, the form length, and the raters, and compare like with like, the same rater on the same form length, rather than mixing lengths across time points. Can I see the Conners 4 questions, or include items in my report? No. Conners 4 items are copyrighted, restricted test materials: the instrument sells only to qualified professionals at the publisher's qualification level B, and reproducing items in reports, advocacy documents, or web pages breaches the publisher's terms and the test-security obligations in APA Ethics Standard 9.11 . That is also why this page shows a results-section skeleton and a fictional write-up rather than any test content. In a report, describe what elevated scales measure in your own words and report the scores; for records requests that touch raw protocols, route the response through your test-security obligations rather than reflexively printing the packet. Parents who want to understand the instrument can be shown the publisher's public materials. What is the Conners 4 age range? The Conners 4 parent and teacher forms cover ages 6 through 18, and the self-report form covers ages 8 through 18, each in the same three lengths (full-length, Short, and the 12-item ADHD Index). The two-year gap is deliberate: self-report demands reading and self-observation skills a 6-year-old rarely has, so for ages 6 and 7 the Conners 4 runs on parent and teacher information alone. Outside the band, early-childhood questions belong to the Conners EC (ages 2 to 6), and adults belong to the CAARS-2 , whose observer and self-report forms start at 18. At exactly 18 the bands overlap: a high-school senior can sensibly stay on the Conners 4 so scores compare with earlier administrations, while an evaluation in an adult context (college, workplace) generally moves to the CAARS-2. Whichever form is used, the report names it, because each form carries its own norms. Can BastionGPT draft my Conners 4 results section? Yes. Paste your score summary (forms, raters, T-scores, Symptom Counts, classifications) and it drafts the results-section narrative for your review: validity and critical items addressed first, both DSM metrics kept distinct, informants integrated rather than averaged, and the ADHD Index framed as probability. It can also cross-check a draft you wrote for score-versus-narrative mismatches, band-label drift, and Conners 3 vocabulary left in a Conners 4 report, and produce a plain-language version of the findings for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: MHS, Conners 4 product page : publisher, 2022 publication, age ranges, form lengths, North American norms, digital administration and scoring, and qualification level B. Pearson, Conners 4 overview and distributor product page : scale architecture, response-style indexes, and the Conners trademark attribution to Multi-Health Systems, Inc. PAA, Conners 3 discontinuation notice : the December 31, 2023 discontinuation of Conners 3 paper and hand-scored products, with the online conversion to Conners 4 stated in the publisher's own FAQ. Buros Center for Testing, Tests reviewed in the Twenty-Second Mental Measurements Yearbook , and the Canivez review : independent review status, the standardization window, sample-size and replication critiques, and the parent-teacher agreement figures. 34 CFR 300.8 and 34 CFR 300.304 : ADHD in the other-health-impairment definition, the adverse-educational-impact requirement, and the no-single-measure rule. US Department of Education, 2016 Dear Colleague letter and resource guide on students with ADHD : a diagnosis alone does not determine IDEA eligibility, which also requires need for special education. Wolraich et al., AAP clinical practice guideline for ADHD (Pediatrics, 2019) : DSM-5 criteria plus rating-scale information from more than one setting. CDC, ADHD clinical care guidance : the DSM-5 diagnostic criteria summary (cross-setting symptoms, onset before age 12, functional interference). Fourth Temporary Extension of COVID-19 Telemedicine Flexibilities : controlled-substance telehealth prescribing flexibilities effective through December 31, 2026; US Attorney's Office EDNY, Cerebral non-prosecution agreement (November 2024) . CMS Medicare Coverage Database, WPS article A57481 : the statutory screening exclusion and the separate screening code family. De Los Reyes et al., 2015, Psychological Bulletin, the validity of the multi-informant approach : mean cross-informant r = .28 across 341 studies. AADPA, Australian evidence-based ADHD guideline and its recommendations summary ; Mulraney et al., JAACAP screening meta-analysis : no child ADHD screening tool met sensitivity and specificity both above 0.8; MBS Online, complex neurodevelopmental disorder items FAQ : the items 135 and 289 ADHD scope note. CADDRA, Canadian ADHD Practice Guidelines and free assessment toolkit : the Canadian guideline context and toolkit instruments. Psychology Board of Australia, Code of Conduct : record accuracy and retention rules for Australian psychologists. ## Related templates BASC-3 Report Write-Up Psychological Evaluation Report Neuropsychological Evaluation Report Psychiatric Diagnostic Evaluation ADOS-2 Report Write-Up BRIEF-2 Report Write-Up CAARS-2 Report Write-Up ADHD Evaluation Report ASRS v1.1 Score Documentation Vanderbilt ADHD Scale Documentation T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/d-kefs-report All note templates # D-KEFS Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The D-KEFS (Delis-Kaplan Executive Function System, 2001) is a battery of nine co-normed stand-alone tests of executive function for ages 8 to 89, with no overall composite by design. Neuropsychologists, clinical psychologists, and school psychologists select tests from it in ADHD, TBI, epilepsy, and dementia evaluations. This page covers how to write up D-KEFS results, with a fictional sample and free templates. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Neuropsychologists, clinical psychologists, and school psychologists; publisher qualification level C Audience Referring physicians and neurologists, IEP and Section 504 teams, rehabilitation teams, accommodation reviewers, courts and payers Typical length 300 to 700 words for the D-KEFS section · administration varies by tests selected, about 90 minutes for the full battery Format family Norm-referenced executive function battery When it's used Neuropsychological and psychoeducational referrals where executive function is in question: ADHD, TBI, epilepsy, dementia workups, rehabilitation planning Standards context Published by NCS Pearson (2001; all-digital D-KEFS Advanced released 2025); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the D-KEFS? The Delis-Kaplan Executive Function System (D-KEFS) is a battery of nine stand-alone tests of executive function published in 2001 by The Psychological Corporation, now NCS Pearson, and authored by Dean Delis, Edith Kaplan, and Joel Kramer. It was the first executive battery co-normed on a large national standardization sample, it spans ages 8 through 89, and it carries the authors' cognitive-process approach: each test samples a higher-level skill through layered conditions, so the examiner can watch how performance changes as executive demand is added to an intact lower-level skill. The nine tests are Trail Making, Verbal Fluency, Design Fluency, Color-Word Interference, Sorting, Twenty Questions, Word Context, Tower, and Proverb. Primary achievement measures are age-referenced scaled scores with a mean of 10 and a standard deviation of 3; contrast scores put the difference between two related conditions on that same metric. Since 2025 the name covers two current products, and sorting that out is now the first job of any write-up. The original 2001 battery remains in print, sold and supported, with no announced retirement. The D-KEFS Advanced, published in July 2025 and announced by Dean Delis that September, is administered and scored entirely on iPad through Q-interactive, carries six tests (updated Trail Making, Verbal Fluency, Color-Word Interference, and Tower, plus two new ones, the Social Sorting Test and the Risk-Reward Decision Test), spans ages 8 to 90, and is normed on a national sample of over 1,280 people matched to 2020 US Census figures. Pearson states that "The D-KEFS Advanced complements and does not replace the original D-KEFS", and Delis explained the naming directly: "That's also why we named the new edition the D-KEFS Advanced rather than the D-KEFS-2." A defensible report names the edition it used, because the two products share a name but not tests, format, or norms. The design choice that shapes every D-KEFS write-up is the absence of an overall composite score. The battery is built for profiling, not for a single executive number, so results are organized by construct rather than summed. It is also strictly a performance measure: it samples what a person can produce in a structured one-to-one session, which research shows is a different construct from the everyday executive behavior captured by rating scales such as the BRIEF-2 . In practice, a D-KEFS section lives inside a fuller neuropsychological report or psychological evaluation report rather than standing alone. ## Who uses the D-KEFS and when Neuropsychologists, clinical psychologists, and school psychologists administer the D-KEFS (Pearson qualification level C) whenever a referral turns on executive function: ADHD and learning evaluations, traumatic brain injury and rehabilitation planning, epilepsy workups, dementia and other adult neurocognitive questions, and psychiatric differentials where planning, inhibition, or flexibility is in doubt. Almost nobody gives all nine tests: Pearson's own product page invites examiners to administer any combination, and published clinical studies routinely use selected subtests, most often Trail Making, Verbal Fluency, Color-Word Interference, and Tower. Co-norming is the reason to select within the D-KEFS rather than assembling standalone instruments with different normative baselines. In a typical battery it runs alongside an intelligence measure, the WAIS for adults or the WISC-V for school-age clients, with a rating scale such as the BRIEF-2 covering the everyday-behavior dimension that performance testing does not reach; for young children who need a broader developmental battery, the NEPSY-II covers overlapping ground for ages 3 to 16. ## How to structure a D-KEFS results section No statute, payer, or publisher mandates a D-KEFS results-section format. The structure below is the convention experienced neuropsychologists converge on because it survives review: it names the edition and the tests given, organizes findings by executive construct instead of test by test, keeps primary scaled scores as the interpretive backbone, and quarantines process-level detail behind an explicit caution. Each section carries the pitfall that most often undermines it. Measures and edition statement. Name the battery and edition (original D-KEFS, 2001, or D-KEFS Advanced, 2025), list the tests administered, and record the format: paper, Q-interactive, or telepractice. Selective administration is standard practice, so the list is information, not a confession; add a clause on why these tests fit the referral question. Pitfall: writing "the D-KEFS" with no edition. Two current products share the name with different tests, formats, and norms, so an unnamed edition leaves the reader unable to place a single score. Behavioral observations during testing. Two or three sentences on engagement, pace, frustration tolerance, and strategy use, written to inform interpretation: whether instructions landed on first presentation, how the examinee handled timed switching demands, whether errors were noticed and self-corrected. Pitfall: observations that never connect to a score. If frustration on switching tasks was worth recording, it is worth referencing when the switching scores are interpreted. Results by executive construct. Group findings by what they measure across tests: inhibition (the Color-Word Interference inhibition condition), flexibility and switching (Trail Making number-letter switching, inhibition/switching, category switching), fluency and generativity (letter, category, and design fluency), planning and problem solving (Tower, Twenty Questions), and concept formation and abstraction (Sorting, Word Context, Proverb). Report primary achievement scaled scores (mean 10, SD 3) as the backbone, and use the intact lower-level baseline conditions to localize a weakness before calling it executive. Factor-analytic work supports construct-level grouping over test-level lists. Pitfall: the test-by-test laundry list. It buries the profile, invites over-reading of isolated scores, and reads like a score printout instead of an interpretation. Process and contrast observations. The battery's signature comparisons, such as switching versus baseline or category versus letter fluency, belong here as hypothesis-level detail: describe the pattern, say what it suggests, and state the caution in the section itself. Published reliability estimates for contrast measures are low, so no diagnostic conclusion should rest on one. Pitfall: a contrast score doing decisional work. Reviewers who know the psychometrics read that as a validity error, not a style choice. Integration with ratings, records, and history. Put the D-KEFS findings next to rating-scale results, school or work records, and interview history, and explain convergence and divergence rather than averaging them. Performance measures and rating scales sample different conditions and correlate weakly, so disagreement is an expected finding to interpret, not a problem to hide. Pitfall: declaring whichever source is inconvenient "invalid" instead of explaining what each one samples. Interpretive summary. A short paragraph stating the construct profile in plain language, tied to the referral question, with the limits of the data stated where a reviewer looks for them: the norms year, the selective battery, and the single-session sample. The D-KEFS has no overall composite, so the summary is a profile statement, never a summary score. Pitfall: inventing a composite by averaging scaled scores. The battery withholds that number on purpose. Recommendations linked to findings. Every recommendation traces to a documented strength or weakness: supports for switching demands if switching was weak, strategy instruction where planning broke down, re-evaluation timing that accounts for the battery's limited alternate forms. Pitfall: a boilerplate list that could sit under any profile. ### Blank template (copy and adapt) D-KEFS RESULTS SECTION SKELETON (adapt; delete guidance in parentheses before signing) MEASURES AND EDITION Battery and edition: original D-KEFS (2001) / D-KEFS Advanced (2025) Tests administered and why they fit the referral question: ____ Format (paper / Q-interactive / telepractice) and norms year stated: ____ BEHAVIORAL OBSERVATIONS DURING TESTING Engagement, pace, frustration tolerance, strategy use: ____ Observations that qualify specific scores: ____ RESULTS BY CONSTRUCT (primary scaled scores: mean 10, SD 3) Inhibition: ____ Flexibility and switching: ____ Fluency and generativity: ____ Planning and problem solving: ____ Concept formation and abstraction: ____ (Localize first: intact baseline conditions before any executive claim) PROCESS AND CONTRAST OBSERVATIONS (hypothesis level only) Pattern observed, what it suggests, and the reliability caution: ____ INTEGRATION WITH RATINGS, RECORDS, AND HISTORY Convergence and divergence, explained rather than averaged: ____ INTERPRETIVE SUMMARY (construct profile; the D-KEFS has no composite score) Profile statement tied to the referral question: ____ Limits of the data (norms year, selective battery, single session): ____ RECOMMENDATIONS LINKED TO FINDINGS 1. ____ 2. ____ 3. ____ Clinician signature, credentials, date: ____ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample D-KEFS write-up (fictional) Scenario: a 15-year-old tenth grader referred by his pediatrician for an ADHD evaluation after two years of declining grades and missed assignments. Four D-KEFS tests were administered inside a fuller battery that included the WISC-V and parent and teacher executive function ratings; this is the D-KEFS section of the larger report only, condensed but structurally complete. All details are fictional. Client: J.T., 15 · Referral: ADHD evaluation, referred by pediatrician · Evaluator: M. Okafor, PsyD, Licensed Psychologist · Testing date: 07/08/2026 · Report date: 07/15/2026 Measures and edition: Selected tests from the Delis-Kaplan Executive Function System (D-KEFS, 2001 print edition), administered on paper in one session: Trail Making, Verbal Fluency, Color-Word Interference, and Tower. Scores are age-referenced scaled scores (mean 10, SD 3) from the 2001 national norms; the age of those norms is noted under limitations below. Tests were selected to examine the inhibition, switching, and planning questions raised by the referral. Behavioral observations: J.T. engaged readily, understood each instruction on first presentation, and worked quickly and accurately on baseline conditions. On the switching conditions he slowed markedly, showed visible frustration, and caught most of his own errors, saying he could feel himself losing the rule. These observations inform the interpretation of the switching scores below. Test and condition Scaled score Trail Making: Visual Scanning 10 Trail Making: Number Sequencing 11 Trail Making: Number-Letter Switching 6 Verbal Fluency: Letter Fluency 11 Verbal Fluency: Category Fluency 12 Verbal Fluency: Category Switching 9 Color-Word Interference: Color Naming 10 Color-Word Interference: Word Reading 11 Color-Word Interference: Inhibition 6 Color-Word Interference: Inhibition/Switching 5 Tower: Total Achievement 9 Results by construct: Lower-level component skills are intact: visual scanning, number sequencing, color naming, and word reading all fell in the average range (scaled scores 10 to 11), which localizes the weaknesses that follow to executive demands rather than to speed, sequencing, or reading. Inhibition was a clear weakness: the Color-Word Interference inhibition condition (scaled score 6) fell at approximately the 9th percentile for age, and the combined inhibition/switching condition (5) near the 5th. Cognitive flexibility showed the same pattern on a second, independent test, with Trail Making number-letter switching (6) well below that test's intact baseline conditions. Fluency was average to high average (letter fluency 11, category fluency 12), with category switching (9) at the lower edge of the average range. Planning and rule-governed problem solving on the Tower test (9) was average. Process observations: Errors concentrated on the switching conditions and were mostly self-corrected, consistent with the observed loss of set under dual demands. Contrast comparisons pointed in the same direction and are noted only as supporting observation, given the published reliability limits of D-KEFS contrast measures; no conclusion in this report rests on a contrast score. Integration: Parent and teacher ratings on the BRIEF-2, reported earlier in this evaluation, were elevated for everyday inhibition, shifting, and task completion. The performance findings converge on inhibition and flexibility while the ratings describe the broader everyday picture, an expected relationship because structured one-to-one testing and daily classroom demands sample different conditions; the two sources are integrated here rather than averaged. School records and interview history, also reported earlier, describe longstanding inattentive and executive difficulties. Summary and recommendations: Across two independent tests, J.T. shows a specific weakness in inhibition and set-shifting against intact lower-level skills, average planning, and average to strong fluency. The D-KEFS yields no overall composite by design, so no summary score is reported; this construct profile, together with the rating-scale, record, and interview evidence elsewhere in this evaluation, supports the diagnostic formulation in that section. Limitations: a selective four-test battery in a single session, scored against the 2001 norms. Recommendations tied to these findings: classroom supports through his Section 504 plan that reduce rapid task-switching demands and chunk multi-step instructions, explicit strategy instruction in self-monitoring under time pressure, and re-evaluation planning that accounts for the D-KEFS having alternate forms for only three of its nine tests. This sample is fictional and for educational purposes. It does not describe a real client or record; the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, the four tests administered, the format, and the norms year are all named, so the scores stay interpretable now that two products share the D-KEFS name, and the selective battery is framed as a choice tied to the referral question. Results are organized by construct with primary scaled scores as the backbone, and the intact baseline conditions are used to localize each weakness before it is called executive. Contrast and process findings appear as supporting observation with the reliability caution stated inside the section, and no conclusion rests on them. Rating-scale divergence is explained by what each source samples, not averaged away and not treated as one instrument being wrong. The summary reports a construct profile rather than an invented composite, states the limitations where reviewers look for them, and every recommendation, including the alternate-forms note in the re-evaluation plan, traces to a documented finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations No law, payer policy, or education regulation names the D-KEFS; every requirement attaches to the evaluation around it, so the write-up's job is to satisfy the framework the report will be used in. Under IDEA, which is law, special-education eligibility rests on multiple sources: 34 CFR 300.304(b)(2) bars any single measure from serving as the sole criterion, and executive function evidence conventionally enters under categories such as traumatic brain injury or other health impairment as one contributor among several. Testing-accommodation documentation runs on agency policy layered over the ADA and Section 504: the ETS documentation guidelines (2026 edition), for example, expect a current comprehensive evaluation with each requested accommodation "explicitly linked to the individual's disability-related functional limitations". US payer coverage is policy about services, not instruments: neuropsychological testing bills under evaluation services 96132 and 96133 with test administration and scoring under 96136 and 96137 (96138 and 96139 by technician), Medicare billing articles A57481 and A57780 expect the record to show medical necessity, the tests administered, scoring and interpretation, and time, and psychological (96130, 96131) and neuropsychological evaluation services are not billed in the same episode; the neuropsychological report page covers that billing chain in depth. In Australia, a search of the MBS for neuropsychology returns no general item as of July 2026 (one narrow item exists for complex neurodevelopmental assessment under age 25), so funding typically runs through private, insurer, NDIS, or state-clinic routes; Canadian funding is a provincial and private mix. The defensibility question specific to the D-KEFS is edition and norms currency. Name the edition in every report. The original battery's norms date to 2001, a quarter century old as of July 2026, and that age is a disclosable limitation rather than a disqualification: the battery remains published, sold, and supported, with no announced retirement. The 2025 D-KEFS Advanced sharpens the question simply by existing, with norms collected against 2020 Census figures, an all-digital format, and a partly different test list, so edition choice is now a real clinical decision: weigh the constructs the referral needs (only the Advanced carries the new social-cognition and risk-decision tests; only the original carries Design Fluency, Sorting, Twenty Questions, Word Context, and Proverb), the format constraints of your setting, and the documentation-currency expectations of high-stakes reviewers. For serial assessment, alternate forms exist for only three tests of the original battery (Sorting, Verbal Fluency, and Twenty Questions), so practice effects belong in the interpretation of everything else. Record the administration format: Pearson has published digital-to-paper equivalence evidence for four subtests (Design Fluency, Verbal Fluency, Trail Making, and Color-Word Interference, in Q-interactive Technical Report 3) and maintains telepractice guidance for the battery. Purchase requires Pearson qualification level C; qualification to buy is not competence for every interpretive use, which remains a licensing and professional-standards matter. D-KEFS and D-KEFS Advanced are trademarks, in the US and other countries, of Pearson plc; both editions are published by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common D-KEFS write-up errors reviewers flag Two published findings explain most of the failures. Crawford, Sutherland, and Garthwaite (2008, Journal of the International Neuropsychological Society) estimated reliability for all 51 D-KEFS contrast measures: none exceeded 0.7, the mean was 0.27 and the median 0.30, standard errors of measurement often approached the scores' own standard deviations, and the authors concluded that contrast measures "should not be used in neuropsychological decision making". Toplak, West, and Stanovich (2013, Journal of Child Psychology and Psychiatry) reviewed 20 studies comparing performance-based executive measures with rating scales: 68 of 286 correlations (24 percent) reached significance, and the median correlation was .19. A write-up that ignores either finding reads as naive to the reviewers who know them. The BastionGPT Clinical Advisory Board sees the same errors most often in D-KEFS reviews: Reporting an overall executive composite. The D-KEFS provides none by design; averaging scaled scores manufactures a number the instrument deliberately withholds, and sample-report pages that teach "interpreting D-KEFS composite scores" are describing a battery that does not exist. Letting a contrast score decide. Contrast measures are hypothesis-level detail under the Crawford reliability findings, and the empirical warning is concrete: in a pediatric ADHD sample, groups separated on primary Color-Word Interference and Tower measures while no contrast score distinguished them (Wodka et al., 2008, Archives of Clinical Neuropsychology). Leaving the edition and norms unnamed. Two current products have shared the name since 2025, with different tests, formats, and norms; an unnamed edition on top of undisclosed 2001 norms is the combination accommodation reviewers now flag first. Writing test by test instead of by construct. A subtest laundry list buries the profile, over-reads isolated scores, and discards the organization that factor-analytic work actually supports: construct-level groupings such as inhibition, flexibility, and monitoring. Treating rating-scale disagreement as invalidity. Performance tests and questionnaires correlate weakly because they sample different constructs, not because one is broken; a report that dismisses the BRIEF-2 because the D-KEFS was average, or the reverse, has misread an expected divergence as a contradiction. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste a D-KEFS score summary (edition, tests given, scaled scores, observations) and BastionGPT drafts the construct-organized results narrative with the edition statement, prose percentile framing, and the contrast-score caution already in place for your review. It cross-checks a finished section for the reviewer flags on this page: an invented composite, score-versus-narrative mismatches, a missing edition or norms statement, or a contrast score doing decisional work. It turns finished findings into a plain-language summary for parents, clients, or the referring provider. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do D-KEFS scaled scores mean? Primary D-KEFS measures are age-referenced scaled scores with a mean of 10 and a standard deviation of 3, so each 3 points is one standard deviation: a score of 10 sits at the middle of the distribution for the person's age, a 7 falls about one standard deviation below it (roughly the 16th percentile), and a 4 about two below (near the 2nd percentile). Reports translate the numbers into qualitative descriptor words; whichever descriptor convention you use, apply it consistently and keep the number next to the word. Error scores and several optional process measures use other metrics, including cumulative percentile ranges, so a careful write-up says which score type each number is. There is no overall battery score to interpret: the D-KEFS reports no composite by design. What is the difference between the D-KEFS and the D-KEFS Advanced? They are two current products from the same author team and publisher. The original D-KEFS (2001) is the nine-test battery, print-based at its core, normed for ages 8 through 89. The D-KEFS Advanced (July 2025) is administered and scored entirely on iPad through Q-interactive, requires a Q-interactive Standard license or a Digital Assessment Library for Schools account, spans ages 8 to 90, was normed on over 1,280 people matched to 2020 US Census figures, and carries six tests: updated Trail Making, Verbal Fluency, Color-Word Interference, and Tower, plus the new Social Sorting Test and Risk-Reward Decision Test. Pearson frames the Advanced as a complement rather than a replacement, both stay on the market, and the deliberate naming (Advanced, not a second edition) signals exactly that. Name the edition in your report, and where currency matters, say which norms stand behind the scores. Do I have to administer all nine D-KEFS tests? No. No publisher, payer, or standard requires the full battery. Pearson's product page says "Administer any or all of the D-KEFS tests to customize your assessments", and published clinical research routinely uses selected subtests, most often Trail Making, Verbal Fluency, Color-Word Interference, and Tower. The reason to select within the D-KEFS rather than mixing standalone instruments is co-norming: every test is referenced to the same standardization sample. Document which tests you gave and why they fit the referral question, treat the constructs you did not test as untested rather than assumed intact, and remember when planning re-evaluation that alternate forms exist for only three tests: Sorting, Verbal Fluency, and Twenty Questions. Is the D-KEFS required for ADHD, IEP, or accommodation decisions? No authority names it. In schools, IDEA's rule is instrument-neutral: evaluators may "not use any single measure or assessment as the sole criterion" ( 34 CFR 300.304(b)(2) ), and executive function evidence enters eligibility categories such as traumatic brain injury or other health impairment by convention, not by mandate. Accommodation reviewers apply agency policy: the ETS documentation guidelines (2026 edition) ask for a current, comprehensive evaluation that ties each requested accommodation to documented functional limitations. Medicare pays for the testing service, not the instrument: the 96132 to 96139 code family under billing articles A57481 and A57780, none of which names the D-KEFS. In Australia, an MBS search for neuropsychology returns no general item as of July 2026. What every framework does require is documented reasoning: why this battery, what it showed, and how the findings connect to the decision; the neuropsychological report page covers those frameworks in depth. Why is there no D-KEFS composite or overall executive score? By design. The authors built the battery on a process approach in which executive function is multidimensional and the diagnostic meaning lives in the profile: comparing conditions within a test and constructs across tests. A single executive quotient would flatten exactly the variability the battery exists to reveal, so no composite is provided and none can be legitimately derived by averaging. Independent test reviews describe the same architecture: nine co-normed stand-alone tests interpreted through comparison (Homack, Lee, and Riccio, 2005, Journal of Clinical and Experimental Neuropsychology). In the write-up, the summary paragraph does the work a composite cannot: a construct profile in plain language, tied to the referral question. Pages offering to help you interpret your "D-KEFS composite score" are describing an instrument that does not exist. Should I report D-KEFS contrast or process scores? Report them as supporting observation, never as the basis of a decision. Crawford, Sutherland, and Garthwaite (2008) estimated reliability for all 51 contrast measures: none exceeded 0.7, the mean was 0.27, the median 0.30, and standard errors of measurement were large enough that the authors concluded contrast measures should stay out of neuropsychological decision making. The battery's authors defended the process approach in a published update (Delis, Kramer, Kaplan, and Holdnack, 2004, Journal of the International Neuropsychological Society), and the practical synthesis is stable: lead with primary achievement scaled scores, use contrast and error patterns to describe how a score was earned and to generate hypotheses, and state the caution inside the section. The empirical warning is concrete: in a pediatric ADHD sample, groups separated on primary measures while no contrast score distinguished them (Wodka et al., 2008). What if D-KEFS results disagree with BRIEF-2 ratings? Expect it and explain it. Across 20 studies, 68 of 286 correlations (24 percent) between performance-based executive measures and rating scales reached significance, with a median correlation of .19 (Toplak, West, and Stanovich, 2013), which is evidence that they measure related but different constructs, not evidence that one is wrong. The D-KEFS samples processing efficiency under optimal, structured, one-to-one conditions; the BRIEF-2 samples goal-directed behavior in everyday environments over months. A strong report states both findings, explains the difference in what was sampled, and uses the pattern diagnostically: average performance with elevated ratings often means the structure of the testing session supplied what the everyday environment does not. Averaging the two, or discarding one, is the error reviewers flag. Where can I find D-KEFS test questions, stimuli, or the scoring manual online? Not legitimately on any public page, including this one. Items, stimuli, record forms, norms tables, and the manuals are protected test materials sold under Pearson qualification level C, and psychologists carry an ethical duty to maintain test security under APA Ethics Standard 9.11. Sites circulating scoring-manual content undermine the norms every report depends on, and some of it is flatly wrong, like composite-scoring instructions for a battery that has no composite. What a report can include: scaled scores, your own prose describing what each test measures, and fictional illustrations like the sample on this page. What it cannot: item text, stimuli, scoring rules, or reproduced norm tables. If you need the materials themselves, the publisher's product page and platforms are the legitimate route. Can BastionGPT help with D-KEFS write-ups? Yes. BastionGPT is trained and clinically tested on psychological and neuropsychological evaluation reports, the parent documents D-KEFS sections live inside. Paste a score summary (edition, tests given, scaled scores, observations) and it drafts the construct-organized results narrative with the edition statement and contrast-score caution in place for your review; it can also check a finished section for an invented composite, score-versus-descriptor mismatches, or a missing norms disclosure, and produce a plain-language summary for the family or referring provider. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from a score summary you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: NCS Pearson, D-KEFS product page : 2001 publication, ages 8 through 89, variable completion time with the full battery at about 90 minutes, qualification level C, the nine tests, alternate record forms, and the administer-any-or-all flexibility line quoted on this page. NCS Pearson, D-KEFS Advanced product page : July 2025 publication, ages 8 to 90, Q-interactive-only administration and scoring, the six tests including Social Sorting and Risk-Reward Decision, the norm sample of over 1,280 matched to 2020 US Census figures, license requirements, and the complements-not-replaces statement quoted on this page. Pearson Assessments blog, Dean Delis, Going All-Digital with the D-KEFS Advanced (September 11, 2025): the launch announcement, the naming rationale quoted on this page, and the Advanced Trail Making switching conditions. Pearson, Q-interactive Technical Report 3 (Daniel, 2012): digital-to-paper equivalence evidence for Design Fluency, Verbal Fluency, Trail Making, and Color-Word Interference. Pearson, Telepractice and the D-KEFS : publisher guidance for remote administration. Crawford JR, Sutherland D, Garthwaite PH, On the reliability and standard errors of measurement of contrast measures from the D-KEFS , Journal of the International Neuropsychological Society, 2008;14(6):1069-1073: the 51 contrast-measure reliability estimates and the conclusion quoted on this page. Toplak ME, West RF, Stanovich KE, Practitioner review: do performance-based measures and ratings of executive function assess the same construct? , Journal of Child Psychology and Psychiatry, 2013;54(2):131-143: the 68 of 286 significant correlations and the median correlation of .19. Wodka EL, Loftis C, Mostofsky SH, Prahme C, Gidley Larson JC, Denckla MB, Mahone EM, Prediction of ADHD in Boys and Girls Using the D-KEFS , Archives of Clinical Neuropsychology, 2008;23(3):283-293: selected-subtest administration and the primary-versus-contrast-score findings. Latzman RD, Markon KE, The factor structure and age-related factorial invariance of the Delis-Kaplan Executive Function System , Assessment, 2010;17(2):172-184: the three-factor structure supporting construct-level organization. Karr JE, Hofer SM, Iverson GL, Garcia-Barrera MA, Examining the Latent Structure of the Delis-Kaplan Executive Function System , Archives of Clinical Neuropsychology, 2019;34(3):381-394: further factor-analytic evidence on construct grouping. Homack S, Lee D, Riccio CA, Test review: Delis-Kaplan Executive Function System , Journal of Clinical and Experimental Neuropsychology, 2005;27(5):599-609: the co-normed nine-test architecture and the process approach. Delis DC, Kramer JH, Kaplan E, Holdnack J, Reliability and validity of the Delis-Kaplan Executive Function System: an update , Journal of the International Neuropsychological Society, 2004;10(2):301-303: the authors' published response in the reliability debate. Buros Center for Testing, Tests reviewed in the Fifteenth Mental Measurements Yearbook : the D-KEFS's independent-review listing. 34 CFR 300.304 : IDEA evaluation procedures, including the no-single-measure rule quoted on this page. CMS Medicare Coverage Database, billing and coding articles A57481 and A57780 : psychological and neuropsychological testing service codes and documentation expectations. ETS Office of Disability Policy, Documentation guidelines for test takers with ADHD (2026 edition): the currency and functional-limitation expectations quoted on this page. ## Related templates Neuropsychological Evaluation Report BRIEF-2 Report Write-Up NEPSY-II Report Write-Up WAIS-5 / WAIS-IV Report Write-Up Psychological Evaluation Report RBANS Report Write-Up Stroop Test Score Documentation Trail Making Test Score Documentation Boston Naming Test Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ktea-3-report All note templates # KTEA-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The KTEA-3 (Kaufman Test of Educational Achievement, Third Edition) is an individually administered academic achievement battery for ages 4:0 through 25:11, measuring reading, math, written language, and oral language with standard scores set to a mean of 100. Psychologists and school psychologists use it in specific learning disability evaluations for its norm-referenced error analysis and parallel forms. This page covers how to write up KTEA-3 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and educational diagnosticians; publisher qualification level B Audience IEP and Section 504 teams, parents and teachers, dyslexia programs, accommodations reviewers, physicians Typical length 500 to 900 words for the results section · administration 15 to 85 minutes for the ASB Composite by grade Format family Norm-referenced academic achievement battery When it's used Specific learning disability and psychoeducational evaluations, dyslexia referrals, re-evaluations with parallel forms, ages 4 to 25 Standards context Published by NCS Pearson (2014); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the KTEA-3? The Kaufman Test of Educational Achievement, Third Edition (KTEA-3) is an individually administered academic achievement battery authored by Alan S. Kaufman and Nadeen L. Kaufman and published by NCS Pearson in 2014, with Pearson stating as of July 2026 that it remains the most current version of the KTEA. Nineteen subtests build fourteen composites across four academic domains, Reading, Math, Written Language, and Oral Language, with the Academic Skills Battery (ASB) Composite serving as the overall score and supplemental composites (Decoding, Sound-Symbol, Reading Fluency, Reading Understanding, Orthographic Processing, Oral Fluency, Comprehension, Expression, and Academic Fluency) sharpening specific referral questions. Every composite and every subtest reports on the same standard-score metric, a mean of 100 and a standard deviation of 15, with age-based norms for 4:0 through 25:11 and grade-based norms for pre-K through 12, plus percentile ranks, normal curve equivalents, stanines, cautioned age and grade equivalents, and Growth Scale Values for comparing performance across administrations. Two parallel forms (A and B) support retesting, administration runs 15 to 85 minutes for the ASB Composite depending on grade, and the battery runs on paper or Q-interactive with scoring by hand or in Q-global. Two satellites extend it: the KTEA-3 Brief Form (2015), a three-composite short form whose standard scores Pearson states may be used interchangeably with Comprehensive Form scores, and the KTEA-3 Dyslexia Index (2018, 2021), a 12-to-20-minute risk screener scored from a small set of Form A or Form B subtests. Two distinctions carry most KTEA-3 write-up mistakes. First, the metric: KTEA-3 subtests use standard scores with the same 100/15 scale as the composites, while the Wechsler cognitive batteries report subtest scaled scores with a mean of 10, so a psychoeducational report that mixes a WISC-V and a KTEA-3 must keep the two scales visibly separate or a subtest score of 9 and one of 90 will read as comparable. Second, the signature feature: norm-referenced error analysis, available for 10 of the 19 subtests, classifies what a student got wrong and compares the error count in each category with the average number of errors made by grade-level peers, which is what makes the KTEA-3 the achievement battery clinicians choose when the referral question is instructional. Against its nearest neighbors, the WIAT-4 offers newer norms (2020) and native Wechsler pairing and the Woodcock-Johnson offers a broad CHC cross-battery design, while the KTEA-3 competes on error analysis, true parallel forms, and two rapid automatized naming measures. ## Who uses KTEA-3 reports and when Psychologists, school psychologists, and educational diagnosticians interpret the KTEA-3, which Pearson lists at qualification level B, and their write-ups land in front of IEP and Section 504 teams, parents and teachers, dyslexia programs, postsecondary disability offices, and sometimes physicians and payers. The classic setting is a specific learning disability evaluation: the KTEA-3 supplies the achievement side of a school or clinic psychoeducational report , paired with a cognitive measure such as the WISC-V , for which Pearson's scoring platforms produce achievement and ability comparisons. It earns its seat over neighboring batteries in three situations: when the team needs error analysis translated into instructional targets, when a re-evaluation calls for a genuine parallel form so practice effects stay out of the comparison, and when a dyslexia referral wants rapid automatized naming measures and the Dyslexia Index screener inside the same kit. Age norms reaching 25:11 also make it a common choice for documenting academic skill deficits in college accommodation files. Upstream of it sit curriculum-based measures and universal screeners that flag risk within RTI or MTSS tiers; the KTEA-3 is the diagnostic, norm-referenced battery the evaluation reaches for once eligibility or diagnosis is on the table, and its results section reads differently from a WIAT-4 section mainly in how much instructional narrative the error analysis supports. ## How to structure a KTEA-3 results section No statute, payer, or publisher mandates a KTEA-3 results-section format. The structure below is the convention experienced evaluators converge on because it survives review: it names the form, norms, and mode before any number, organizes results by academic domain instead of printout order, turns error analysis into instructional narrative, and gives each composite exactly the scope it has. Each section carries the pitfall that most often undermines it. Measures and methods statement. Name the battery and edition (KTEA-3, 2014), the form administered (A or B), the norm set applied (age-based or grade-based, and which grade norms if applicable), the platform (paper, Q-interactive administration, Q-global or hand scoring), and the mode. If any part was administered via telepractice, Pearson's guidance directs the report to state that and briefly describe the method, because the norms were collected in person. Pitfall: writing "the KTEA-3 was administered" with nothing else. Form, norm type, and mode each change what a score means, and age-based and grade-based scores for the same raw performance can differ visibly. Behavioral observations during testing. Two or three sentences on engagement, attention, fatigue, sensory barriers, and how the examinee handled timed and open-ended tasks, closing with a statement of whether results are considered a valid estimate of current academic skills and any accommodation or deviation that occurred. Pitfall: boilerplate observations that could describe any child. Reviewers use this paragraph to judge whether the scores deserve their interpretation, and error analysis downstream depends on responses actually attempted. Domain-by-domain results. Report each relevant domain in a stable order (Reading, Math, Written Language, Oral Language): the domain composite first, then its subtests, each with standard score and percentile, then what the pattern within the domain shows. Describe timed against untimed contrasts (fluency subtests against their untimed counterparts) in prose. Pitfall: marching through the score report in printout order, or drifting between metrics. Every KTEA-3 subtest is a 100/15 standard score; if a Wechsler battery appears in the same report, keep the scales visibly separate and the descriptor system consistent. Error-analysis narrative. This is the KTEA-3's signature and the reason instructional readers keep the report. For the subtests analyzed (available on 10 of the 19), state which error categories were classified as weaknesses relative to the average number of errors grade peers make, then translate the categories into plain language about what the student can and cannot yet do, and connect each to an intervention target. Pitfall: reporting raw error counts without the normative comparison, or pasting Q-global's interpretive and intervention text. Pearson's report license limits excerpting to the minimum text needed for your conclusions, so the narrative must be yours. Composite handling. Report the ASB Composite as the overall academic summary while naming its scope: it is built from six reading, math, and written-language subtests and does not include the oral-language subtests. Give supplemental composites (Decoding, Sound-Symbol, Reading Fluency, Orthographic Processing, Academic Fluency) their referral-specific work, and report the Dyslexia Index as the risk screener it is. Pitfall: letting an average ASB imply the whole battery was average, or writing the Dyslexia Index as a diagnosis. Pearson states a single score is not sufficient to diagnose dyslexia. Cognitive linkage and eligibility framing. Where the evaluation pairs the KTEA-3 with a cognitive measure, report the achievement and ability comparison or pattern-of-strengths-and-weaknesses analysis under the identification model your state actually permits, and label it. Federal regulations bar states from requiring a severe ability-achievement discrepancy and require that an RTI-based process be permitted, so the defensible sentence names the model and the state rule rather than treating one model as universal. Pitfall: presenting a discrepancy table as "the" SLD test. Which models are permissible is state policy, and the eligibility decision belongs to the team, not to any one instrument. Summary and recommendations linkage. Close the section with the profile in two or three sentences (which domains are impaired, which are preserved, and how the error analysis explains the referral concern), then recommendations that trace to findings: intervention targets from the error categories, accommodations tied to documented fluency or transcription weaknesses, and a re-evaluation plan that names the alternate form and Growth Scale Values for measuring change. Pitfall: recommendations that ignore the error analysis, or a retest plan built on an invented waiting period. No authority sets a fixed retest interval; the parallel form exists so retesting is a design decision, not a countdown. ### Blank template (copy and adapt) KTEA-3 RESULTS SECTION SKELETON (adapt; delete guidance in parentheses before signing) MEASURES AND METHODS Edition (KTEA-3, 2014) and form (A / B): ____ Norm set (age-based / grade-based, fall or spring) and why: ____ Platform (paper / Q-interactive; Q-global or hand scoring) and mode (in person / telepractice, with method described): ____ BEHAVIORAL OBSERVATIONS Engagement, attention, timed-task behavior, accommodations: ____ Validity statement (results a valid estimate? deviations?): ____ DOMAIN-BY-DOMAIN RESULTS (composite first, then subtests; standard scores + percentiles) Reading (composite, subtests, timed vs untimed pattern): ____ Math (composite, concepts vs computation pattern): ____ Written Language (composite, spelling vs expression pattern): ____ Oral Language (composite, listening vs speaking pattern): ____ ERROR-ANALYSIS NARRATIVE (for analyzed subtests) Categories flagged as weaknesses vs grade peers: ____ What the student can / cannot yet do, in plain language: ____ Intervention targets tied to each flagged category: ____ COMPOSITE HANDLING ASB Composite with its scope stated (no oral-language subtests): ____ Supplemental composites used and why (Decoding, Sound-Symbol, Reading Fluency, Orthographic Processing, Academic Fluency): ____ Dyslexia Index reported as risk screening, not diagnosis: ____ COGNITIVE LINKAGE AND ELIGIBILITY FRAMING Ability-achievement / PSW analysis with the state model named: ____ SUMMARY AND RECOMMENDATIONS Profile summary tied to the referral question: ____ Recommendations traced to findings, incl. re-evaluation plan (alternate form, Growth Scale Values): ____ Examiner signature, credentials, date: ____ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample KTEA-3 write-up (fictional) Scenario: an 8-year-old third grader referred for an initial special education evaluation after two years of reading intervention with limited response. The KTEA-3 supplied the achievement component of a school psychoeducational evaluation; this is the KTEA-3 section of the resulting report, condensed but structurally complete. All details are fictional. Student: J.T., age 8:6, grade 3 · Referral: persistent reading difficulty despite tiered intervention; initial SLD evaluation · Examiner: D. Okafor, PsyD, School Psychologist · Testing date: 07/10/2026 · Report date: 07/16/2026 Measures and methods: Kaufman Test of Educational Achievement, Third Edition (KTEA-3), Form A, administered in person on paper and scored in Q-global. Age-based standard scores (mean 100, SD 15) are reported below; error analysis, which is normed on grade-level peers, was completed for the decoding, spelling, and phonological subtests given the referral question. The KTEA-3 was selected for this evaluation because its error analysis and naming-fluency measures speak directly to the instructional questions raised by two years of limited intervention response. Cognitive results (WISC-V) are reported in the preceding section. Behavioral observations: J. worked cooperatively through two sessions with a movement break, attempted every task presented, and stayed engaged with encouragement on reading tasks, where he read quietly and self-corrected often. He completed timed tasks without distress but visibly slowed on unfamiliar words, sounding them out letter by letter. Results are considered a valid estimate of his current academic skills; no deviations from standardized administration occurred. Results: Composite standard scores appear below; subtest scores follow in the domain narrative. Descriptive categories follow the score report's classification labels. Composite Standard score Percentile Descriptive category Academic Skills Battery (ASB) 84 14 Below average Reading 81 10 Below average Decoding 79 8 Low Sound-Symbol 77 6 Low Reading Fluency 76 5 Low Written Language 81 10 Below average Math 95 37 Average Oral Language 100 50 Average Reading. The Reading Composite of 81 (10th percentile) understates how uneven the underlying skills are. Letter & Word Recognition (82, 12th percentile) and Reading Comprehension (85, 16th percentile) sit at the low end of the Below average range, while the decoding machinery beneath them is weaker still: Nonsense Word Decoding 78 (7th percentile), Phonological Processing 80 (9th percentile), Word Recognition Fluency 76 (5th percentile), and Decoding Fluency 77 (6th percentile). The timed and untimed contrast matters: J. reads some real words accurately given unlimited time, but speed collapses when fluency is required, and pseudowords remove the sight-vocabulary support entirely. Rapid naming was low average to below average (Letter Naming Facility 84, 14th percentile; Object Naming Facility 88, 21st percentile), slower for letters than for objects. Error analysis (reading and spelling). On Nonsense Word Decoding, J.'s errors in vowel patterns, including short and long vowel forms and vowel teams, were classified as weaknesses, meaning he made more errors in those categories than the average third grader in the norm sample, while single-consonant errors fell in the average range. Spelling errors concentrated in the same place: plausible consonant frames with the vowel wrong (for example, spelling attempts preserving first and last sounds but not the medial vowel). Phonological Processing errors clustered on segmenting and blending items. In plain terms, J. hears and produces beginning and ending sounds reliably but cannot yet map vowels to print or pull apart the middle of words, and that single weakness explains most of what the classroom sees. Written language and math. The Written Language Composite of 81 (10th percentile) reflects Spelling at 79 (8th percentile) against stronger Written Expression at 86 (18th percentile): sentence-level ideas outrun transcription. Math is a relative strength (composite 95, 37th percentile; Math Concepts & Applications 96, Math Computation 94), with no error categories flagged. Oral language. The Oral Language Composite of 100 (50th percentile), with Listening Comprehension at 102 (55th percentile) and Oral Expression at 98 (45th percentile), is squarely average. The ASB Composite of 84 (14th percentile) summarizes the six reading, math, and written-language subtests and does not include these oral-language scores, so the overall number should not be read as contradicting the preserved listening and speaking skills. Summary for the team. Achievement testing shows a specific pattern: intact oral language and average math against weak decoding, spelling, and reading fluency, with error analysis locating the breakdown in vowel-pattern knowledge and phonemic segmentation. These achievement data feed the eligibility analysis in the Data Integration section under the state's criteria; no single score determines eligibility, and that decision belongs to the team. Recommendations linked to findings. (1) Structured literacy intervention targeting vowel patterns and phoneme segmentation and blending, the categories error analysis flagged, with decodable text practice to build fluency. (2) Classroom supports tied to the documented fluency weakness: extended time for reading-dependent tasks and no cold oral reading. (3) Spelling instruction aligned to the same vowel-pattern sequence rather than memorized lists. (4) At re-evaluation, administer KTEA-3 Form B and compare Growth Scale Values on the targeted subtests so practice effects and measurement error stay out of the progress judgment. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The methods statement is complete. Edition, form, norm type, platform, mode, and the reason this battery fits the referral are all named in one place, so any reviewer can reconstruct exactly what produced the numbers, including which scores are age-normed and which comparisons are grade-normed. Scores and narrative agree on one metric and one descriptor system. Every claim in the prose matches the table, all scores are 100/15 standard scores kept apart from the cognitive battery's scaled scores, and the descriptive categories follow the score report's own labels consistently. Error analysis does instructional work. Flagged categories are reported against the grade-norm comparison, translated into plain language about what the student can and cannot yet do, and carried straight into the recommendations, in the evaluator's own words rather than pasted platform text. Every composite gets exactly its scope. The ASB is summarized with its six-subtest coverage stated so it cannot contradict the intact oral-language findings, the supplemental composites answer the referral question, and the eligibility decision is routed to the team under the state's criteria. Recommendations and the retest plan trace to findings. Intervention targets come from the flagged error categories, accommodations tie to the documented fluency weakness, and re-evaluation names Form B with Growth Scale Values instead of an invented waiting period. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Label every eligibility and coverage claim by what it actually is, because the KTEA-3 lives in decision contexts with three different kinds of rules. Federal special education law is LAW: the IDEA regulations state that in identifying a specific learning disability a state "must not require the use of a severe discrepancy between intellectual ability and achievement", must permit a process based on the child's response to scientific, research-based intervention, and may permit other alternative research-based procedures, the clause under which pattern-of-strengths-and-weaknesses models operate, and the same regulations require a comprehensive evaluation in which no single measure or score decides anything. Which of those identification models your state actually permits is state POLICY, as are the dyslexia screening statutes that make the KTEA-3's phonological, naming-fluency, and decoding measures attractive in reading referrals. Payment is its own layer: school evaluations under IDEA are publicly funded rather than billed to health insurance, while clinic and private evaluations bill the psychological testing CPT family (evaluation services 96130 and 96131, administration and scoring 96136 through 96139), codes that APA Services notes appear in CPT Appendix P for audio-video and Appendix T for audio-only telemedicine and that are maintained on the CMS Medicare telehealth list for calendar year 2026, which is PAYER POLICY worth re-checking each year. In Canada there is no federal IDEA analogue: identification frameworks come from provincial ministries and school boards, and much psychoeducational testing is accessed privately or through school-board queues, which is CONVENTION rather than entitlement. In Australia, school-eligibility achievement testing is generally not a Medicare-rebatable service and runs through state education pathways or private assessment. In the UK, SASC guidance lists the KTEA-3 as suitable for higher-education specific-learning-difficulty assessment and remains on its current suitable-tests list, with the plain caution that "The tests are only normed up to age 25". The battery-specific duties sit in the methods and narrative. The KTEA-3's norms date to 2014 and Pearson states it remains the current edition as of July 2026, while the neighboring WIAT-4 (2020) and the fifth-edition Woodcock-Johnson carry newer standardizations, so a defensible report names the edition and norm vintage rather than assuming "current" is self-evident, states whether age-based or grade-based scores are reported (and that error analysis is normed on the grade sample), and dates any edition claim. The same paragraph should respect two publisher boundaries. Purchase requires Pearson qualification level B, and qualification to buy is not competence for every interpretive use, which remains a licensing and professional-standards matter. And the Q-global narrative is not free text: Pearson's own score report licenses excerpting "limited to the minimum text necessary to accurately describe their significant core conclusions", with no adaptations without written permission, so the results narrative and the error-analysis interpretation must be written in your own words with platform output treated as data. Telepractice administration carries a documentation duty of its own, since Pearson's guidance directs the report to state that testing occurred via telepractice and describe the method, the norms having been collected in person. Retention, access, and records handling follow the parent document: a KTEA-3 section inside a school or clinic psychoeducational report inherits that report's education-records and health-records obligations, including parent access and raw-data handling covered there. KTEA is a trademark of NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common KTEA-3 write-up errors reviewers flag The published record gives reviewers reasons to read KTEA-3 sections closely. A study of teacher trainees administering core KTEA-3 subtests found administration and clerical errors common, with Reading Comprehension the most error-prone subtest (Lockwood and colleagues, 2020), and an independent factor analysis in Contemporary School Psychology concluded that "the KTEA-3 is factorally complex", cautioning against over-reading structure into score patterns (Parkin and Frisby, 2019). Pearson's own report license, printed on every Q-global output, limits excerpting to the minimum text necessary for your conclusions. The BastionGPT Clinical Advisory Board sees the same errors most often in KTEA-3 write-up reviews: Metric drift in a mixed battery. KTEA-3 subtests are 100/15 standard scores; Wechsler subtests are 10/3 scaled scores. Reports that run both without labeling the scales, or that apply one battery's descriptor bands to the other's scores, misstate results even when every number is transcribed correctly. Name the metric once and keep one descriptor system throughout. Error analysis missing or mishandled. The instructional payload of the KTEA-3 is the norm-referenced error analysis, yet reviewers see sections that skip it, report raw error counts without the comparison to the average number of errors grade peers make, or paste the platform's interpretive and intervention text wholesale. The license allows minimum-text excerpting only; the translation into what the student can and cannot yet do is the evaluator's job. No form, norm-type, or edition statement. Form A and Form B are different item sets, age-based and grade-based scores answer different questions, and the 2014 norm vintage deserves naming while newer neighbor batteries exist. A score whose yardstick is unstated cannot be audited, and a re-evaluation that silently switched forms or norm sets cannot be compared to baseline. Grade equivalents doing placement work. Pearson lists age and grade equivalents among the scores with published cautions, and reviewers flag sentences that treat a grade equivalent of 1.8 as "reads at a first-grade level" or as an instructional placement. Keep interpretation on standard scores and percentiles, and let grade equivalents illustrate rather than decide. Composites given the wrong scope. The ASB Composite summarizes six reading, math, and written-language subtests and omits oral language, so an average ASB cannot clear the whole battery, and the Dyslexia Index is a risk screener whose own product page states a single score "is not sufficient to diagnose dyslexia". Reviewers also flag subtest-scatter stories built on small differences the battery's factor structure will not support. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste a KTEA-3 score summary (form, norm set, composites, subtests, flagged error categories) and BastionGPT drafts a domain-organized results narrative with the methods statement, percentile framing, and error-analysis-to-instruction translation set up for your review. It cross-checks a finished section for the reviewer flags on this page: numbers that disagree with the table, subtest standard scores drifting into scaled-score language, a missing form or norm-type statement, grade equivalents written as placements, or a Dyslexia Index phrased as a diagnosis. It turns finished findings into a plain-language summary for parents and teachers that keeps scores as estimates and explains what the eligibility team does next. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do KTEA-3 scores mean? Every KTEA-3 score, subtest and composite alike, is a standard score with a mean of 100 and a standard deviation of 15, so 100 sits at the middle of the distribution for same-age or same-grade peers, 85 falls about one standard deviation below it (roughly the 16th percentile), and 115 about one above (the 84th). Percentile ranks, normal curve equivalents, stanines, cautioned age and grade equivalents, and Growth Scale Values round out the score set. Keep the percentile next to the number, and if you use descriptor words, use one labeling system consistently; Pearson's score reports print descriptive categories such as Average and Below average, and mixing systems across a report is a reviewer flag. The error-analysis output is different in kind: it classifies what the student got wrong and compares each category's error count with the average number of errors made by grade-level peers. Is the KTEA-3 outdated? Is there a KTEA-4? No KTEA-4 exists: Pearson's product page states the KTEA-3 is the most current version of the KTEA, a statement verified in July 2026. The norms date to the 2014 publication, which matters because the neighboring WIAT-4 (2020) and the fifth-edition Woodcock-Johnson carry newer standardizations. That does not make KTEA-3 results indefensible; it makes edition and norm-vintage statements mandatory. Name the battery, year, and norm set in the methods sentence, date any "current edition" claim, and if an evaluation pairs the KTEA-3 with newer-normed instruments, keep the vintages visible rather than letting all the scores read as if they share one reference period. KTEA-3 or WIAT-4: which should I use? Both are individually administered, comprehensively normed Pearson achievement batteries at qualification level B, and no authority mandates either, so the choice is convention driven by the referral question. The KTEA-3's case: norm-referenced error analysis on ten subtests that converts directly into instructional targets, true parallel Forms A and B for re-evaluation, two rapid automatized naming measures, and age norms to 25:11. The WIAT-4 's case: a 2020 standardization and native pairing with the Wechsler cognitive scales. Teams often keep both on the shelf and pick per referral; what reviewers actually flag is a report that mixes the two batteries' scores without labeling metrics and norm vintages, or that switches batteries between baseline and re-evaluation without saying so. Is the KTEA-3 required for a specific learning disability evaluation? No instrument is. Federal law sets the frame: the IDEA regulations require a comprehensive evaluation using a variety of assessment tools, bar any single measure or score from being the sole criterion, state that states must not require a severe ability-achievement discrepancy for SLD identification, must permit a response-to-intervention process, and may permit alternative research-based procedures such as pattern-of-strengths-and-weaknesses analysis. Which models are actually permitted, and what a dyslexia evaluation must include, is state policy that varies. The KTEA-3 earns its slot as evidence, not as a requirement: it supplies the normed achievement data those models consume, and in reading referrals its phonological, decoding, and naming-fluency measures map onto what state dyslexia rules commonly ask evaluators to examine. The write-up convention that survives review names the state's model and routes the eligibility decision to the team documented in the psychoeducational report . Should I report age-based or grade-based scores? Report the set that answers the referral question, and say which one you used. Grade-based scores compare the student with others at the same point in schooling, which fits school eligibility and curriculum questions, and they are the reference frame the error-analysis norms are built on (the grade-norm sample). Age-based scores compare the student with same-age peers regardless of grade, which fits developmental and clinical questions and students markedly old or young for grade, and they are the frame that lines up with age-normed cognitive batteries in ability-achievement comparisons. The two sets can differ visibly for the same performance, especially for retained or accelerated students, so the methods sentence names the choice and the reason, and a re-evaluation uses the same frame as baseline or explains the switch. What is KTEA-3 error analysis and should the report include it? Error analysis is the KTEA-3's signature: for 10 of the 19 subtests, the system classifies each error by skill category, using examiner-judged classification on some subtests and automatic item-level classification on others, then compares the student's error count in each category with the average number of errors made by grade peers in the norm sample, labeling the category a strength, average, or weakness. That normative comparison is what separates it from informal error inspection, and the standardization research behind the 2017 special issue of the Journal of Psychoeducational Assessment found the error categories form interpretable structures. Include it whenever the referral is instructional: translated into plain language about what the student can and cannot yet do, it is the part of the report teachers and interventionists actually use. Two cautions: report the normative classification rather than raw counts, and write the interpretation yourself rather than pasting the scoring platform's text, which Pearson's license limits to minimum-text excerpting. How soon can the KTEA-3 be re-administered? No publisher or federal rule sets a fixed waiting period; the parallel forms exist precisely so retesting is a design decision rather than a countdown. The convention: administer the alternate form at re-evaluation (Form B where Form A was baseline, or the reverse), name in the report which form was given each time, keep the norm frame consistent with baseline, and describe change with Growth Scale Values, the Rasch-based scores Pearson provides for comparing performance across administrations, rather than eyeballing standard-score differences that mix real growth with measurement error and practice effects. Short intervals still deserve a stated rationale, since even alternate forms share format familiarity, and the Brief Form offers a lighter progress-monitoring option between full administrations, with Pearson stating Brief standard scores may be used interchangeably with Comprehensive Form scores. Where can I see the actual KTEA-3 questions or a filled-in record form? Not from evaluators, and not from this page. Test items, stimulus books, record forms, and scoring materials are copyrighted, sold under Pearson's qualification requirements, and protected by the test-security obligations in the APA Ethics Code, which directs psychologists to make reasonable efforts to maintain the integrity and security of test materials. Item exposure also invalidates the norms: a student coached on the items no longer has a meaningful score. What can be shared is everything this page contains, the structure of a results section, the score metrics, and a fully fictional sample write-up, and families are entitled to a clear explanation of results and, per the education-records rules, access to the reports written about their child. Evaluators handling records requests for raw test data have their own decision to document, balancing records access against test security and publisher agreements. Can BastionGPT help with KTEA-3 write-ups? Yes. BastionGPT is trained and clinically tested on psychological and psychoeducational evaluation reports, the parent documents KTEA-3 sections live inside. Paste a score summary (form, norm set, composites, subtests, flagged error categories) and it drafts a domain-organized results narrative with the methods statement, percentile framing, and error-analysis-to-instruction translation in place for your review; it can also check a finished section for metric drift, score-narrative mismatches, a missing form or norm statement, grade equivalents doing placement work, or a Dyslexia Index phrased as a diagnosis, and produce a plain-language summary for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from a score summary you paste means no item or record-form content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: NCS Pearson, KTEA-3 product page : 2014 publication, the most-current-version statement, ages 4:0 through 25:11, grades pre-K through 12, the 100/15 standard-score set with Growth Scale Values, qualification level B, 15 to 85 minute ASB completion time, Q-interactive and Q-global options, and the parallel-forms, RAN, and error-analysis positioning quoted on this page. Pearson, KTEA-3 error analysis : error analysis on 10 of the 19 subtests, the within-item and item-level classification methods, the comparison to the average number of errors made by the reference group, and the grade-norm error-norming (n = 2,600 for item-level subtests, a stratified subset for within-item subtests). Pearson, KTEA-3 sample score report (PDF) : the composite architecture and ASB composition, subtest standard scores, 90% confidence intervals, printed descriptive categories, and the report-excerpting license quoted on this page. Pearson, Telepractice and the KTEA-3 (PDF) : in-person normative data, telepractice as a deviation from standardized administration, and the requirement to state and describe telepractice in the written report. Pearson, KTEA-3 Brief Form and KTEA-3 Dyslexia Index product pages: the 2015 Brief Form's three composites and interchangeable-scores statement; the Dyslexia Index's 2018 and 2021 releases, ages 5 to 25, 12 to 20 minute completion time, and the not-sufficient-to-diagnose statement quoted on this page. IDEA regulations, 34 CFR 300.307 (govinfo PDF) : the severe-discrepancy prohibition, the RTI permission requirement, and the alternative research-based procedures clause; the pattern-of-strengths-and-weaknesses wording sits in 300.309. O'Brien, Pan, Courville, Bray, Breaux, Avitia & Choi, Exploratory factor analysis of reading, spelling, and math errors , Journal of Psychoeducational Assessment 35(1-2), 7-23 (2017), and Breaux, Bray, Root & Kaufman, Introduction to Special Issue and to KTEA-3 Error Analysis , JPA 35(1-2), 4-6 (2017): the empirical basis for treating error categories as interpretable. Lockwood, Sealander, Gross & Lanterman, Teacher trainees' administration and scoring errors on the KTEA-3 , Journal of Psychoeducational Assessment 38(5), 551-563 (2020): administration and clerical errors among 42 teacher trainees, Reading Comprehension the most error-prone subtest. Parkin & Frisby, Exploratory factor analysis of the KTEA-3 , Contemporary School Psychology 23(2), 138-151 (2019): the factorial-complexity finding quoted on this page. Breaux & Lichtenberger, Essentials of KTEA-3 and WIAT-III Assessment (Wiley, 2016): the interpretive tradition behind domain-organized reporting and the per-grade administration-time guidance. Buros Center, Tests reviewed in the Twentieth Mental Measurements Yearbook (2017): independent review listing for the KTEA-3. SASC, KTEA-3 guidance (June 2016, PDF) and the June 2025 suitable-tests list (PDF) : UK higher-education suitability, US norms, and the age-25 norming ceiling quoted on this page. APA Services, Telehealth reporting updates for 2026 , and CMS, Medicare telehealth services list : the psychological-testing CPT family's Appendix P and T listings and its place on the CY 2026 Medicare telehealth list. ## Related templates Psychoeducational Evaluation Report WIAT-4 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up WISC-V Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mmpi-3-report All note templates # MMPI-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The MMPI-3 (Minnesota Multiphasic Personality Inventory-3, 2020) is a 335-item true/false broadband personality inventory for adults 18 and older, published by the University of Minnesota Press and distributed by Pearson. Psychologists use it in clinical, forensic, public-safety, and presurgical evaluations to assess emotional, thought, and behavioral dysfunction. This page covers how to write up MMPI-3 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinical, forensic, and police and public safety psychologists; publisher qualification level C Audience Courts and attorneys, referring physicians and surgeons, employers after conditional offers, disability reviewers, treating clinicians Typical length 400 to 800 words for the results section · administration 25 to 50 minutes by format Format family Broadband self-report personality inventory When it's used Psychological and forensic evaluations, post-offer public-safety screening, presurgical psychological evaluations, disability and fitness-for-duty referrals, treatment planning Standards context Published by the University of Minnesota Press (2020), distributed by Pearson; described here for write-up purposes; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MMPI-3? The Minnesota Multiphasic Personality Inventory-3 (MMPI-3) is a 335-item true/false self-report inventory of personality and psychopathology for adults 18 and older, the 2020 edition in a lineage that runs from the original MMPI (1942) through the MMPI-2 (1989) and the MMPI-2-RF (2008). It was authored by Yossef Ben-Porath and Auke Tellegen, is published by the University of Minnesota Press, and is distributed by Pearson at qualification level C. Its 52 scales sit in a fixed hierarchy: ten validity scales in three families (inconsistent responding, over-reporting, under-reporting), three Higher-Order scales (Emotional/Internalizing, Thought, and Behavioral/Externalizing Dysfunction), eight Restructured Clinical scales, twenty-six Specific Problems scales, and five PSY-5 personality-pathology scales, with new scales measuring disordered eating, compulsivity, impulsivity, and self-importance. Scores are T scores (mean 50, SD 10) on non-gendered norms from 1,620 adults matched to 2020 US Census projections, joined for the first time by US Spanish-language norms from 550 Spanish speakers. Administration takes about 25 to 35 minutes by computer and 35 to 50 on paper, at a 4.5-grade reading level, on Q-global or Q Local or paper; the MMPI-3 is not a Q-interactive instrument. The load-bearing fact for write-ups, as of July 2026, is that three MMPIs are legitimately on the market and a report must say which one it used. The publisher's May 2026 update states there are "no plans at this time to discontinue" the MMPI-2 or MMPI-2-RF while calling the MMPI-3 the intended replacement, and reports current adult sales running 73% MMPI-3, 13% MMPI-2-RF, and 14% MMPI-2. The editions are not interchangeable on paper either: the MMPI-3 has no Clinical Scales and no code types, so write-up language built on the MMPI-2's 1-3 or 2-7 profile tradition, or on K-corrected scores, describes an instrument the MMPI-3 is not. Interpretation instead runs through the validity-first hierarchy above, guided by the MMPI-2-RF research base that the publisher's Technical Manual maps onto the new scales (Appendix E). An MMPI-3 profile is evidence, not a verdict: it answers no referral question by itself, and its results section feeds the psychological evaluation report or neuropsychological report that does. ## Who uses MMPI-3 reports and when Clinical psychologists administer the MMPI-3 inside psychological evaluations to clarify diagnosis and plan treatment, and neuropsychologists include it in neuropsychological batteries to characterize the emotional and personality context around cognitive findings; Medicare's own billing articles draw the boundary there, describing neuropsychological testing as the assessment that does not rely on self-report inventories like the MMPI. Forensic evaluators use it in criminal, civil, and family matters, where the MMPI family is among the most used and most litigated objective personality measures. Police and public-safety psychologists use it in post-offer screening with the publisher's Police, Correctional, Dispatcher, and Firefighter Candidate reports and setting-specific comparison groups, a use covered in the California POST Peace Officer Psychological Screening Manual. Health psychologists use it in presurgical psychological evaluations, including the spine-surgery and spinal cord stimulator pathway the November 2024 Spinal Procedure Candidate report was built for. It typically enters the chart alongside a psychiatric diagnostic evaluation and cognitive testing such as the WAIS , and it competes for the broadband-inventory slot with the PAI and, when the referral question is DSM personality disorder characterization, the MCMI-IV. Choose the MMPI-3 when the referral needs a broadband, validity-scale-anchored view of emotional, thought, behavioral, and interpersonal dysfunction; pair it with narrowband and performance validity measures rather than asking one inventory to carry the evaluation. ## How to structure an MMPI-3 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention the MMPI interpretive literature itself teaches: protocol validity before any substantive statement, findings organized by domain rather than scale by scale, and the computer narrative treated as a licensed input to paraphrase, never a report to paste. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and administration. Name the instrument and edition (MMPI-3), the language and norm set (English or US Spanish), the platform (Q-global, Q Local, or paper and pencil), the administration modality (in office, or remote via the publisher's On-Screen or Remote On-Screen Administration), and any setting-specific comparison group applied. Pitfall: writing "an MMPI" with no edition. Three MMPIs are on the market as of July 2026 and their scale sets differ; an edition-less score list is ambiguous at re-evaluation and easy to attack on review. Protocol validity, before anything else. Address the three validity families in order: response consistency (Cannot Say count, CRIN, VRIN, TRIN), then over-reporting (F, Fp, Fs, FBS, RBS), then under-reporting (L, K), and end the paragraph with a plain conclusion about whether the substantive profile is interpretable, fully or with stated limits. Pitfall: validity as an afterthought. A results section that presents substantive elevations first and mentions an elevated infrequency scale later has the logic backwards; if protocol validity is compromised, that finding is the finding. Substantive findings by domain, not by scale number. Work down the hierarchy the instrument is built on: the three Higher-Order scales set the frame, then somatic/cognitive, emotional-internalizing, thought, behavioral-externalizing, and interpersonal domains, drawing Restructured Clinical, Specific Problems, and PSY-5 scales into the domain they inform. State the elevation convention once (substantive T scores at or above 65 are read as clinically meaningful by convention in the MMPI-3 interpretive literature) and spend the prose on what elevated and what did not. Pitfall: the 52-scale inventory dump. A scale-by-scale recitation in test order documents that scoring occurred; it buries the clinical picture and invites over-interpretation of isolated scores. Comparison-group context, where one applies. In public-safety, presurgical, and similar settings, say which comparison group the report used and keep the two reference frames separate: T scores describe standing against the normative sample, while comparison-group data describe how candidates or patients in that setting typically respond. Pitfall: reading a guarded screening profile as affirmative evidence of health. Candidate profiles are routinely defensive; the write-up should say what a low-elevation, high-L/K protocol can and cannot support. The computer narrative, paraphrased and integrated. The Score and Clinical Interpretive Reports are inputs to your reasoning. The interpretive report's own license permits qualified users to excerpt only the minimum text needed to describe its core conclusions in a written evaluation, and the report states it is intended for a professional qualified to interpret the MMPI-3, considered against the test taker's background and circumstances. Paraphrase the narrative into your own case-anchored language, reconcile it with interview and history, and attribute the instrument where you rely on it. Pitfall: pasting canned paragraphs. Verbatim blocks of report output are a copyright problem, an APA Ethics 9.09 problem (the user remains responsible for interpretation), and a tell that reviewers and opposing experts search for. Interpretive summary against the referral question. State what the MMPI-3 findings suggest as hypotheses, corroborate or bound them with interview, history, records, and other measures, and answer the referral question in the evaluation's integration section rather than from the inventory alone. Pitfall: a verdict from a T score. No T score is a diagnosis or a legal threshold; the MMPI-3 answers no psycholegal or eligibility question directly, and a write-up that lets it reads as advocacy. Recommendations linkage. Tie each recommendation to a stated finding: treatment targets to the elevated internalizing scales, further assessment to the questions the profile raised but cannot settle, risk and safety planning to the evaluation's risk assessment rather than to the inventory. Pitfall: recommendations that could follow any profile. If the recommendation list would read the same with different scores, the linkage is missing. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: MMPI-3 Language/norms: [English / US Spanish] Platform: [Q-global / Q Local / paper] Modality: [in office / remote OSA-ROSA, documented] Comparison group: [none / named] PROTOCOL VALIDITY (before any substantive statement) Consistency: [CNS count; CRIN/VRIN/TRIN within limits or not] Over-reporting: [F-family, FBS, RBS findings in prose] Under-reporting: [L, K findings in prose] Conclusion: [profile interpretable / interpretable with stated limits / not interpretable, and what that means here] SUBSTANTIVE FINDINGS BY DOMAIN (state T>=65 convention once) Higher-Order frame: [EID / THD / BXD pattern in one sentence] Somatic/cognitive: [elevations and meaning, or "unremarkable"] Emotional-internalizing: [RC and SP scales woven into prose] Thought dysfunction: [same] Behavioral-externalizing: [same] Interpersonal: [same; PSY-5 where they add trait context] COMPARISON-GROUP CONTEXT (if applied) [Group used; what candidate/patient norms add; frames kept apart] INTERPRETIVE NARRATIVE INTEGRATION [Computer report paraphrased, minimum-necessary excerpts only, reconciled with interview, history, records, other measures] INTERPRETIVE SUMMARY (answer the referral question) [Hypotheses the profile supports; what it cannot establish; convergence or divergence with the rest of the evaluation] RECOMMENDATIONS LINKAGE [Each recommendation tied to a stated finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample MMPI-3 write-up (fictional) Scenario: a 34-year-old project manager, referred by her psychiatrist for psychological evaluation to clarify diagnosis and guide psychotherapy planning after a partial medication response for depressed mood and worry. The protocol is valid and cooperative with focal internalizing elevations, the treatment-planning shape where the write-up's job is to turn scale findings into corroborated therapy targets. This is the MMPI-3 results section only, condensed but structurally complete. All details are fictional. Client: A.V., 34 · Referral: diagnostic clarification and psychotherapy planning, referred by treating psychiatrist · Evaluator: S. Whitfield, PhD, Licensed Psychologist · Testing date: 07/09/2026 · Report date: 07/17/2026 Measures and administration: The Minnesota Multiphasic Personality Inventory-3 (MMPI-3, English) was administered in the office on Q-global as part of a psychological evaluation that also included a diagnostic interview, symptom rating scales, and record review, reported in their own sections. Scores are T scores (mean 50, SD 10) against the MMPI-3 normative sample on its non-gendered norms; no setting-specific comparison group was applied. By convention in the MMPI-3 interpretive literature, substantive-scale T scores at or above 65 are described below as clinically meaningful elevations. The interpretive report was reviewed and its findings are paraphrased and integrated here with interview and history rather than quoted. Protocol validity: A.V. answered every item. Response consistency was within acceptable limits (CRIN, VRIN, and TRIN all unelevated), and there was no indication of over-reporting on the infrequency or symptom-validity scales (F, Fp, Fs, FBS, RBS) or of a guarded, virtue-claiming presentation (L, K). The substantive profile is interpretable without qualification, and its moderate, focal pattern of elevations is itself consistent with an open, accurate self-description. Scale (family) Fictional T score Emotional/Internalizing Dysfunction, EID (Higher-Order) 72 Thought Dysfunction, THD (Higher-Order) 42 Behavioral/Externalizing Dysfunction, BXD (Higher-Order) 44 Demoralization, RCd (Restructured Clinical) 74 Low Positive Emotions, RC2 (Restructured Clinical) 68 Dysfunctional Negative Emotions, RC7 (Restructured Clinical) 71 Self-Doubt, SFD (Specific Problems) 70 Inefficacy, NFC (Specific Problems) 66 Worry, WRY (Specific Problems) 68 Negative Emotionality/Neuroticism, NEGE (PSY-5) 69 Introversion/Low Positive Emotionality, INTR (PSY-5) 63 Substantive findings: The Higher-Order pattern frames the profile: emotional-internalizing dysfunction is prominently elevated (EID 72) while the thought-dysfunction (THD 42) and behavioral-externalizing (BXD 44) domains are unremarkable, and the somatic/cognitive Specific Problems scales were likewise unelevated. Within the internalizing domain, A.V. reports pervasive demoralization (RCd 74), a marked loss of positive emotional experience and engagement (RC2 68), and high negative emotionality (RC7 71) that the Specific Problems scales localize as self-doubt (SFD 70), perceived inefficacy (NFC 66), and worry (WRY 68). The PSY-5 trait context matches: negative emotionality is elevated (NEGE 69) and introversion/low positive emotionality approaches but does not reach the elevation convention (INTR 63). Scales bearing on suicidal ideation were not elevated; risk was separately assessed by interview and is addressed in the risk section of this evaluation. Functionally, this is the self-report of a person who feels overwhelmed and ineffective, doubts her own judgment, worries persistently, and has lost access to pleasure and initiative, while thinking clearly and maintaining behavioral control. Integration and interpretive summary: The MMPI-3 findings converge with the diagnostic interview, the treating psychiatrist's observations, and A.V.'s elevated depression and worry ratings on the brief symptom measures reported elsewhere in this evaluation. Taken together, the pattern supports the working diagnostic impression of a moderate depressive episode with prominent generalized worry documented in the diagnostic section, and it argues against thought-disorder and externalizing explanations for her current difficulties. The profile generates hypotheses rather than verdicts: the demoralization and inefficacy findings suggest the depressive picture is amplified by a collapsing sense of agency at work, which the interview corroborated. Recommendations linkage: The elevated demoralization, low positive emotions, and self-doubt findings support a course of psychotherapy targeting behavioral activation and cognitive work on self-evaluative themes, coordinated with ongoing medication management; the worry findings support adding structured worry-management strategies. Re-administration of brief symptom measures is recommended to track response, with MMPI-3 retesting reserved for a genuinely new diagnostic question rather than routine monitoring. These recommendations follow from the scale findings above and from the corroborating interview and history, not from the inventory alone. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works Validity is settled before substance. The write-up disposes of consistency, over-reporting, and under-reporting in order and states a plain interpretability conclusion, so every later sentence stands on a declared foundation. Scores and narrative agree, with the convention stated once. Every substantive claim traces to a listed elevation, the T-at-or-above-65 reading is declared a convention rather than a rule, and the one near-threshold score (INTR 63) is described as approaching, not reaching, elevation. Domain organization, not a 52-scale dump. The Higher-Order frame comes first and the RC, Specific Problems, and PSY-5 scales are woven into the domains they inform, which is how the interpretive literature teaches the instrument. The computer narrative is integrated, not pasted. The report says the interpretive output was paraphrased and reconciled with interview and history, respecting the publisher's minimum-necessary excerpting license and APA Ethics 9.09 responsibility. Edition, platform, norms, and modality are named. MMPI-3, Q-global, in-office administration, non-gendered norms, no comparison group: one sentence that keeps the scores interpretable at any future re-evaluation. Findings become linked recommendations. Each treatment target tracks a named elevation corroborated outside the test, and the write-up says what the inventory cannot establish alone. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The decision contexts around the MMPI-3 mix law, agency policy, and convention, and the write-up should signal which is which. LAW (US employment): in Karraker v. Rent-A-Center , 411 F.3d 831 (7th Cir. 2005), the court held the MMPI is a medical examination under the ADA because it is "designed, at least in part, to reveal mental illness," so under EEOC guidance it belongs after a conditional offer, applied uniformly to entering employees in the job category. AGENCY POLICY: the FAA's Office of Aerospace Medicine still requires the MMPI-2 for pilots and air traffic control specialists and states the MMPI-2-RF and MMPI-3 "are not acceptable substitutes," while the test's authors answer that the FAA is an outlier among federal and public-safety agencies (their response, hosted by the publisher, discloses their financial interest); the California POST screening manual (revised February 2024) covers the MMPI-3 among its reviewed instruments as CONVENTION, not statute. PAYER POLICY (US): the professional work is billed under 96130/96131 with administration and scoring under 96136/96137, or 96138/96139 when a technician administers; Medicare contractor billing articles A57481 and A57780 govern documentation, and A57780 draws the psychological-versus-neuropsychological boundary by noting neuropsychological testing does not rely on self-report inventories like the MMPI. All six testing codes appear on the CMS List of Medicare Telehealth Services for CY 2026; remote administration itself runs through the publisher's Q-global On-Screen or Remote On-Screen pathways per Pearson's telepractice guidance , with the modality documented in the report (see the telehealth documentation page for the wider telehealth rules). Outside the US the funding frame changes: Australia's Better Access items (80000 to 80165) rebate psychological therapy services, and an MBS search returns no item for personality-inventory administration, so MMPI-3 assessment is typically privately, insurer, or medico-legally funded even though the publisher sells Australia and New Zealand community comparison groups (2022 manual supplement); Canadian practice likewise runs through provincial, insurer, employer, and medico-legal channels rather than a national fee schedule. Edition and qualification questions decide defensibility more often than any scoring detail. Three MMPIs remain on sale, and the publisher's May 2026 update pairs "no plans at this time to discontinue" with a clear statement that the MMPI-3 is the intended replacement, backed by current adult sales of 73% MMPI-3, 13% MMPI-2-RF, and 14% MMPI-2. The practical rule: choose the edition per the referral question, name it with its norms, and be ready to state the reason during the transition. The MMPI-3 brings the current, census-matched, non-gendered norms and the first US Spanish norms; the MMPI-2-RF carries the deeper independent research base and longer-established forensic comparison groups; the MMPI-2 remains the mandated instrument in the FAA's aerospace lane and the anchor of decades of case law. Cross-edition score comparisons are different measurements, not the same ruler, and the publisher's Appendix E correlate mapping is the stated bridge from MMPI-2-RF research to MMPI-3 interpretation, so cite it as the interpretation guide rather than treating the editions as interchangeable. Purchase and use sit at Pearson qualification level C, which is why the write-up names a qualified interpreter, and the interpretive report's license permits excerpting only the minimum text necessary for your core conclusions, which is why the narrative is paraphrased. Retention of the protocol and report follows the record-set rules on the psychological evaluation report page; raw item responses stay in the test file under APA Ethics 9.11 test security rather than in the released report. MMPI and Minnesota Multiphasic Personality Inventory are registered trademarks, and MMPI-3 and Minnesota Multiphasic Personality Inventory-3 are trademarks, of the Regents of the University of Minnesota. BastionGPT is not affiliated with, or endorsed by, the University of Minnesota Press or Pearson. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common MMPI-3 write-up errors reviewers flag There is no payer audit series for personality-test write-ups; the accountability pressure here comes from courts, opposing experts, licensing boards, and the publisher's own license terms, and it is well documented. In the largest study of psychological assessment evidence in court, Neal and colleagues (2019) found admissibility challenges were rare, raised against only 5.1% of assessment evidence, succeeding about a third of the time, which means the practical scrutiny falls on the written report itself. The MMPI-3 is independently reviewed in the Buros Center's Twenty-Second Mental Measurements Yearbook , its validity scales carry strong published feigning-detection evidence (Fp reached an effect size of g = 1.29 in the first peer-reviewed MMPI-3 over-reporting study), and its interpretive report prints a license limiting excerpts to "the minimum text necessary to accurately describe their significant core conclusions." The BastionGPT Clinical Advisory Board sees the same errors most often in MMPI-3 write-up reviews: The interpretive report pasted instead of paraphrased. Verbatim canned paragraphs violate the report's own excerpting license, hand your interpretive responsibility to software against APA Ethics 9.09, and read as boilerplate to any reviewer who has seen the same sentences before. Butcher and colleagues' foundational review of computer-based interpretation put it plainly: such reports are "adjuncts to, rather than substitutes for, clinical judgment." Substantive interpretation of a compromised protocol. If the consistency or over-reporting scales invalidate the protocol, that is the finding; presenting substantive elevations anyway, or burying the validity discussion after them, reverses the interpretive order the instrument is built on. In compensation-seeking contexts, the over-reporting caveat conditions every substantive sentence and belongs first. MMPI-2 vocabulary in an MMPI-3 report. Code types, Clinical Scale numbers, and K-corrected scores do not exist on the MMPI-3, and gendered norms are gone. Write-ups that carry them forward, and top-ranking sample reports online still do, tell a careful reader the narrative came from an old template rather than this administration. A T score treated as a diagnosis or a legal threshold. No authority sets a statutory MMPI-3 cutoff, elevation at T 65 is an interpretive convention, and the instrument, as the family-law literature notes, does not directly address any psycho-legal issue. Scale findings are hypotheses to corroborate with interview, records, and other measures, and the write-up should say so. Screening profiles over-read and reference frames mixed. In post-offer public-safety and presurgical contexts, a guarded, low-elevation protocol is common and mostly uninformative, not affirmative proof of health, and comparison-group data answer a different question than normative T scores. Keep the two frames separate, name the comparison group used, and state what a defensive protocol limits. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your scale summary (validity findings plus elevated scales with T scores) and get a drafted results-section narrative organized validity-first and by domain, with the elevation convention stated once and the referral question kept in frame. Cross-check a finished draft for the gaps reviewers flag: narrative claims with no corresponding elevation, MMPI-2 code-type or Clinical Scale vocabulary in an MMPI-3 report, validity findings buried after substantive ones, and unstated edition, platform, or modality. Translate the results section into a plain-language summary for the referral source, the treating clinician, or the client that keeps scale findings as corroborated hypotheses, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do MMPI-3 scores actually mean? MMPI-3 scales are reported as T scores with a mean of 50 and a standard deviation of 10, scored against a single set of non-gendered norms: 1,620 US adults selected to match 2020 Census projections, with separate US Spanish-language norms from 550 Spanish speakers. By convention in the MMPI-3 interpretive literature, substantive-scale T scores at or above 65 are described as clinically meaningful elevations; that is an interpretive convention, not a diagnostic or legal cutoff. No score means anything until the validity scales say the protocol is interpretable: consistency first (CNS, CRIN, VRIN, TRIN), then over-reporting (F, Fp, Fs, FBS, RBS), then under-reporting (L, K). An elevation is a hypothesis about how the person described themselves, to be corroborated with interview, history, and other measures. MMPI-3, MMPI-2-RF, or MMPI-2: which edition should I use? All three are on the market as of July 2026, and the choice is defensible either way if you can state your reason. The publisher's May 2026 update says there are "no plans at this time to discontinue" the older editions while calling the MMPI-3 the intended replacement, and reports adult sales running 73% MMPI-3, 13% MMPI-2-RF, 14% MMPI-2. The MMPI-3 brings current census-matched norms and the first US Spanish norms; the MMPI-2-RF carries the deeper independent research base and longer-established forensic comparison groups; and one agency lane is mandated: the FAA requires the MMPI-2 for pilots and air traffic controllers and rejects both newer versions, a position the test's authors have publicly answered as an outlier. Name the edition and norms in every report, and treat cross-edition comparisons as different measurements. How much of the computer interpretive report can I copy into my own report? Very little, and the boundary is printed on the report itself. The MMPI-3 Clinical Interpretive Report is copyrighted, states it is intended for a professional qualified to interpret the MMPI-3, and licenses qualified users to excerpt only "the minimum text necessary to accurately describe their significant core conclusions" in a written evaluation. Ethically the answer is the same: APA Standard 9.09 leaves interpretation your responsibility even when software generated the narrative, and the foundational review of computer-based interpretation called such reports "adjuncts to, rather than substitutes for, clinical judgment" ( Butcher, Perry & Atlis, 2000 ). Paraphrase the narrative into case-anchored language, reconcile it with interview and history, attribute the instrument, and keep verbatim use to that licensed minimum. Is the MMPI-3 required for police or pre-employment evaluations? No law names the MMPI-3 as required for any evaluation; requirements live in the ordering framework, not the instrument. What the law does regulate is timing: in Karraker v. Rent-A-Center (7th Cir. 2005) the MMPI was held a medical examination under the ADA because it is "designed, at least in part, to reveal mental illness," so under EEOC guidance it may be given only after a conditional offer. Agency policy then varies: the California POST screening manual covers the MMPI-3 among accepted instruments for peace-officer screening, many agencies choose it by convention, and the FAA mandates the MMPI-2 instead for its own aviation evaluations. Document what your jurisdiction and agency actually require, and state the post-offer posture in employment reports. Can the MMPI-3 be administered remotely, and how do I document that? Yes, through the publisher's own pathways: Pearson's telepractice guidance delivers the MMPI-3 by Q-global On-Screen Administration (OSA), Remote On-Screen Administration (ROSA), or Q-global Video Proctoring, and its guidance makes documentation of the procedures a professional responsibility, so the report should name the modality and conditions. The practice literature to cite is Corey and Ben-Porath (2020) on remote MMPI administration. On the payer side, the psychological testing codes (96130, 96131, 96136, 96137, 96138, 96139) all appear on the CMS List of Medicare Telehealth Services for CY 2026; the wider telehealth billing rules live on the telehealth documentation page . The MMPI-3 is not a Q-interactive instrument, so an iPad-app administration claim is a red flag in any report. When should I use the PAI or the MCMI-IV instead of the MMPI-3? All three are defensible broadband inventories; the referral question picks. The PAI is the MMPI-3's main competitor with a different design (4-point ratings rather than true/false, its own validity architecture) and its own strong symptom-validity research; the MCMI-IV is built for DSM personality-disorder characterization, so it leads when that is the referral question. The MMPI family remains the most used objective personality measure in forensic work: in Neal and Grisso's international survey it appeared in 15.2% of 868 forensic evaluations, against 9.6% for the PAI ( Neal & Grisso, 2014 ), and courts admit properly used MMPI-family testimony under Daubert and Frye, with the MMPI-3's legal-settings literature now published ( Ben-Porath, Heilbrun & Rizzo, 2022 ). Whichever you choose, pair the broadband inventory with narrowband and performance validity measures inside the full evaluation . What questions are on the MMPI-3, and can I see or share them? The items are protected test materials, so this page, and any competent report, reproduces none of them. Access is restricted to qualified professionals at Pearson qualification level C, psychologists are bound to maintain test security under APA Ethics Standard 9.11 and publisher agreements, and item exposure damages the instrument for everyone tested after. What can be said: there are 335 true/false statements at about a 4.5-grade reading level, taking 25 to 35 minutes by computer, and the validity scales are specifically built to detect coached, exaggerated, or minimized responding, with published detection effect sizes above g = 1.0 for the infrequency scales ( Reeves, Brown & Sellbom, 2022 ). The honest preparation is the one the instrument is designed for: answer candidly. Reports carry scores and interpretation, never item content, and raw responses stay in the test file. How is MMPI-3 testing billed? In the US, the professional evaluation work (integrating data, interpretation, report, feedback) bills under 96130 for the first hour and 96131 for each added hour; administration and scoring bill under 96136/96137 when the professional does it and 96138/96139 when a technician does. Medicare contractor billing articles A57481 and A57780 carry the documentation rules, and A57780 also draws the boundary that matters for code choice: neuropsychological testing is defined by cognitive assessment, not by self-report inventories like the MMPI, so an MMPI-3 inside a neuropsych battery is still psychological-testing work. In Australia there is no MBS item for personality-inventory administration (Better Access items 80000 to 80165 rebate therapy services), and Canadian funding runs through provincial, insurer, employer, and medico-legal channels, so assessment there is typically quoted privately. Can BastionGPT draft my MMPI-3 results section? Yes. Paste your scale summary (validity findings, elevated scales with T scores, the comparison group if any) and it drafts the results-section narrative for your review: validity first, findings organized by domain rather than scale number, the elevation convention stated once, and the interpretive-report material paraphrased rather than pasted. It can also cross-check a finished draft for score-versus-narrative mismatches and MMPI-2-era vocabulary, and produce a plain-language summary for the referral source or client. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: University of Minnesota Press, MMPI-3 test-division page : authors, publisher and distributor, 2020 publication, 335 items, administration time, reading level, age range, and the Appendix E interpretation-carryover statement; and the Press's Update on the MMPI-3 (May 2026) : the no-discontinuation and intended-replacement statements, the 73/13/14 sales split, the public-safety and spinal-procedure report suite, the 28,000-person development base, translations, and the Australia and New Zealand comparison-group supplement. Pearson, MMPI-3 product page and MMPI-3 brochure : qualification level C, completion times, the 52-scale architecture and scale families, the 1,620-person and 550-person normative samples, platforms (Q-global, Q Local, paper), and the trademark attribution. Pearson, sample MMPI-3 Clinical Interpretive Report : the printed excerpting license ("the minimum text necessary to accurately describe their significant core conclusions") and the qualified-professional statement; and Administering the MMPI-3 via telepractice : the Q-global OSA, ROSA, and Video Proctoring pathways and the documentation responsibility. Federal Aviation Administration, Office of Aerospace Medicine, Selecting the MMPI-2 versus the MMPI-3 (November 18, 2020): the MMPI-2 requirement and the not-acceptable-substitutes statement; and Ben-Porath & Corey, A Response to the FAA's MMPI-3 Policy (May 2021, with the authors' disclosure): the outlier characterization and the federal-agency usage list. Karraker v. Rent-A-Center, Inc., 411 F.3d 831 (7th Cir. 2005) : the ADA medical-examination holding; and EEOC, Enforcement Guidance on Preemployment Disability-Related Questions and Medical Examinations : the conditional-offer timing rule. CMS, billing and coding articles A57481 and A57780 (psychological and neuropsychological tests), and the List of Medicare Telehealth Services (CY 2026 final list): the testing-code family and telehealth status. California Commission on Peace Officer Standards and Training, Peace Officer Psychological Screening Manual (revised February 2024): MMPI-3 coverage in public-safety screening. Buros Center for Testing, Tests reviewed in the Twenty-Second Mental Measurements Yearbook : the MMPI-3's independent-review status. Reeves, Brown & Sellbom, Psychological Assessment (2022) : MMPI-3 over-reporting detection effect sizes; Whitman, Tylicki & Ben-Porath, Psychological Assessment (2021) : validity-scale simulation evidence; Whitman & Ben-Porath, Psychological Injury and Law (2024) : the Daubert-factors analysis. Neal & Grisso, Criminal Justice and Behavior (2014) : instrument use rates in 868 forensic evaluations; Neal, Slobogin, Saks, Faigman & Geisinger, Psychological Science in the Public Interest (2019) : admissibility-challenge rates; Ben-Porath, Heilbrun & Rizzo, Journal of Personality Assessment (2022) : using the MMPI-3 in legal settings; Mulchay, Journal of the American Academy of Matrimonial Lawyers (2022) : the psycho-legal-issue boundary and the Appendix E correlate count. Butcher, Perry & Atlis, Psychological Assessment (2000) : computer-based test interpretation as adjunct to clinical judgment; Corey & Ben-Porath, Professional Psychology: Research and Practice (2020) : remote MMPI administration guidance; Ben-Porath & Sellbom, Interpreting the MMPI-3 (University of Minnesota Press, 2023): the interpretive framework; Friedman & Nichols, The National Psychologist (2017, archived) : the published critique of the MMPI-3's lineage. MBS Online, Better Access psychological therapy items : the therapy-service framing and the absence of a personality-assessment item (search verified July 2026). ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report WAIS-5 / WAIS-IV Report Write-Up Psychiatric Diagnostic Evaluation PAI Report Write-Up MCMI-IV Report Write-Up Psychological Testing Scoring Note PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/nepsy-ii-report All note templates # NEPSY-II Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The NEPSY-II (2007) is a developmental neuropsychological battery for ages 3 to 16 that yields subtest-level scores across six domains, with no overall composite by design. Psychologists and school psychologists use it to profile attention, language, memory, sensorimotor, social perception, and visuospatial skills in pediatric evaluations. This page covers how to write up NEPSY-II results, with a fictional sample and free templates. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, pediatric neuropsychologists, and school psychologists; publisher qualification level C Audience IEP teams and school evaluation committees, referring pediatricians and pediatric specialists, intervention therapists, payers, parents Typical length 300 to 700 words for the NEPSY-II section · administration 45 minutes to 3 hours by battery Format family Norm-referenced developmental neuropsychological battery When it's used Pediatric neuropsychological and psychoeducational evaluations, autism and ADHD referrals, language and learning differentials, injury and medical follow-up Standards context Published by Pearson (2007, current edition as of July 2026); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the NEPSY-II? The NEPSY-II (NEPSY Second Edition, subtitled A Developmental Neuropsychological Assessment) is a norm-referenced pediatric neuropsychological battery by Marit Korkman, Ursula Kirk, and Sally Kemp, published in 2007 by Harcourt Assessment and sold today by Pearson . It revised the original 1998 NEPSY, extended the age range to 3 years 0 months through 16 years 11 months, and added a Social Perception domain built for autism-related referrals. The battery holds 32 stand-alone subtests and four delayed tasks organized into six theoretically derived domains: Attention and Executive Functioning, Language, Memory and Learning, Sensorimotor, Social Perception, and Visuospatial Processing. Clinicians rarely give all of it; the manual describes General, Diagnostic, Selective, and Full assessment approaches, plus eight referral batteries offered as guidelines for common questions such as ADHD, reading and language disorders, autism spectrum disorder, and mild intellectual disability. Two facts date every write-up: the normative data were collected in 2005 and 2006, stratified to October 2003 US census targets, and as of July 2026 no third edition has been announced, so the 2007 edition and its roughly two-decade-old norms remain the current published instrument. The load-bearing fact for write-ups: the NEPSY-II produces no overall composite and no domain scores, by design. The manual dropped domain-level scores "in favor of the more clinically sensitive subtest-level scores" and states plainly that domain scores are not derived, because subtests within a domain measure different constructs and a global number can mask the specific deficit a referral question is asking about. Any report, template, or scoring description that mentions a "NEPSY-II composite score" or averages subtests into a domain index is describing something the instrument does not produce. That makes the NEPSY-II the structural opposite of an IQ measure: it is built to profile specific neuropsychological processes, not to summarize general ability, and it is meant to supplement a cognitive measure such as the WISC-V rather than duplicate it. A defensible results section therefore stands on subtest-level reporting, grouped by domain and referral question, in the pattern of a neuropsychological report . ## Who uses NEPSY-II reports and when Psychologists, pediatric neuropsychologists, and school psychologists administer and interpret the NEPSY-II (Pearson sells it at qualification level C, its highest tier), and the write-up travels far beyond the testing room: IEP teams and school eligibility committees, referring pediatricians and pediatric neurologists, speech-language pathologists and occupational therapists planning intervention, payers reviewing medical necessity, and parents deciding what happens next. The battery appears wherever a referral question needs process-level detail rather than a single ability number: autism evaluations that pair Affect Recognition and Theory of Mind with the ADOS-2 , SRS-2 , and adaptive measures such as the Vineland-3 ; ADHD referrals weighted toward the Attention and Executive Functioning subtests; language-disorder differentials cross-referenced with speech-language testing; and medical or injury follow-up where sensorimotor and memory subtests document specific consequences. Its selective, referral-driven administration model is the draw: on Q-interactive, examiners can even combine selected NEPSY-II subtests with WISC-V subtests in one custom battery. For executive functioning in older adolescents and adults, the D-KEFS (ages 8 through 89) picks up where the NEPSY-II's age range ends; below age 8, the NEPSY-II's executive subtests are the option. The write-up conventions here sit inside a full neuropsychological report or psychoeducational evaluation, which carry the background, history, and diagnostic reasoning this instrument section feeds. ## How to structure a NEPSY-II results section Measures and battery rationale. Open by naming the instrument and edition (NEPSY-II, 2007), the assessment approach (General, Diagnostic, Selective, or Full), the subtests administered, and why those subtests answer the referral question; the manual's referral batteries are guidelines, not a fixed battery, so the rationale is yours to state. Anchor the norm base in one sentence: normative data collected 2005 to 2006. Pitfall: listing subtests with no stated rationale, which reads as reflexive full-battery testing and invites a medical-necessity question. Behavioral observations during testing. The NEPSY-II treats observed behaviors as scorable data recorded against the standardization sample, reported as cumulative percentages, so write observations as findings (what occurred, how often, and how common that is for age) rather than color commentary. Pitfall: anecdotal observations that never connect to the scores they explain. Subtest results, organized by domain and referral question. Report each administered subtest at the subtest level: scaled score (mean of 10, standard deviation of 3), percentile rank or percentile range where that is what the score yields, and the publisher's printed classification term. Group them under the six domains for readability, but interpret patterns across subtests, not domain averages. Pitfall: computing or implying a domain score or composite; the instrument does not produce one, and an averaged number has no norms behind it. Process, contrast, and error detail. Where they change the interpretation, add the process-level findings: which component of the task broke down, what the contrast score says about a higher-level skill once the lower-level skill is controlled, and what error patterns showed. This is the analytic layer the battery was built for. Pitfall: reporting a contrast score as if it were an ability score, or reading a percentile-range score as a precise percentile. Low-score framing. Before interpreting any low score as impairment, state the frame: having some low scores is common in healthy children, and the count that is unremarkable is higher on a longer battery, so the question is whether the pattern converges with history, observation, and other measures. Name the reliability caution where it applies, especially at the youngest and oldest age bands. Pitfall: a single Borderline score carrying a diagnostic conclusion on its own. Social Perception, cross-referenced. If Affect Recognition and Theory of Mind were administered, report them as two related but separate findings and integrate them with the autism-specific evidence (observation instruments, rating scales, adaptive and developmental history). Pitfall: writing the two subtests as one social composite or treating them as standalone autism markers; independent factor analyses could not recover a Social Perception dimension. Interpretive summary and recommendations linkage. Close the section by answering the referral question in plain language, integrating the NEPSY-II profile with the rest of the evaluation, and tie every recommendation to a documented finding, including any re-evaluation plan. State the norm-date limitation here or in a limitations paragraph. Pitfall: a summary that restates scores without saying what they mean for the referral question, and recommendations that appear from nowhere. ### Blank template (copy and adapt) MEASURES AND BATTERY RATIONALE Instrument/edition: NEPSY-II (2007) Norms collected: 2005-2006 Assessment approach: [General / Diagnostic / Selective / Full] Subtests administered and rationale (referral question): [subtest list + one line on why] BEHAVIORAL OBSERVATIONS DURING TESTING [Scorable observations: what occurred, how often, how common for age; connect to the scores they explain] SUBTEST RESULTS BY DOMAIN [Domain]: [subtest], scaled score [ ], percentile [rank/range], classification [publisher's printed term] Process / contrast / error detail where it changes interpretation: [ ] LOW-SCORE FRAMING [Base-rate statement; subtest reliability; whether the pattern converges with history, observation, and other measures] SOCIAL PERCEPTION (IF ADMINISTERED) [Affect Recognition and Theory of Mind reported separately; cross-referenced with observation, rating-scale, and adaptive evidence; no standalone diagnostic claims] INTERPRETIVE SUMMARY [Answer the referral question; integrate with the rest of the evaluation; no composite language] RECOMMENDATIONS LINKAGE [Each recommendation tied to a documented finding; re-evaluation plan if indicated] LIMITATIONS [Norm date 2005-2006; age-band reliability cautions; battery coverage] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample NEPSY-II write-up (fictional) Scenario: an 8-year-old boy referred by his pediatrician for social-communication concerns, evaluated as part of a broader autism evaluation that also included the ADOS-2, SRS-2, Vineland-3, and WISC-V. A Selective NEPSY-II battery targets the social-perception, language, and memory questions, the profile shape where subtest-level reporting and base-rate framing do the defensible work. This is the NEPSY-II section of the larger evaluation report only, condensed but structurally complete. All details are fictional. Client: L.M., 8 · Referral: social-communication concerns, autism evaluation · Evaluator: J. Whitfield, PhD, Licensed Psychologist · Testing dates: 07/08/2026 and 07/10/2026 · Report date: 07/16/2026 Measures and battery rationale: The NEPSY-II (2007) was administered using a Selective assessment approach chosen for the referral question. Subtests sampled language (Comprehension of Instructions, Speeded Naming, Word Generation), attention and executive functioning (Auditory Attention and Response Set, Inhibition), memory and learning (Narrative Memory, Memory for Faces with delayed condition), visuospatial processing (Design Copying), and social perception (Affect Recognition, Theory of Mind). The NEPSY-II yields subtest-level scores only; no domain or composite scores exist for this instrument, and none are reported. Normative data were collected in 2005 and 2006, which is noted as a limitation below. General cognitive ability (WISC-V), autism-specific observation (ADOS-2), informant ratings (SRS-2), and adaptive functioning (Vineland-3) are reported in their own sections of this evaluation. Behavioral observations during testing: L.M. separated readily, worked cooperatively across both sessions, and responded well to visual structure. He asked for instruction repetition on two language subtests, looked away from the examiner during conversational openings between tasks, and rarely referenced the examiner's face when uncertain, a pattern the examiner recorded as infrequent among same-age children. Observed effort was adequate on all subtests administered, and results are considered a valid estimate of current functioning. Language, attention, and executive functioning: Language subtests fell At Expected Level: Comprehension of Instructions scaled score 10 (50th percentile), Speeded Naming 9 (37th percentile), and Word Generation 11 (63rd percentile). Auditory Attention scaled score 10 and the Response Set condition 9 indicate age-appropriate selective and shifting auditory attention, and the Inhibition combined scaled score of 9 reflects age-typical inhibitory control with self-corrections noted on the switching condition. Structural language and regulated attention are not the locus of the referral concern. Memory, learning, and visuospatial processing: Narrative Memory fell At Expected Level (scaled score 10, 50th percentile) and Design Copying At Expected Level (scaled score 11, 63rd percentile). Memory for Faces fell in the Borderline range (scaled score 6, 9th percentile) with the delayed condition also Borderline (scaled score 6), and errors were distributed across immediate and delayed conditions rather than concentrated after delay, indicating weak initial encoding of faces rather than forgetting. Social perception: Affect Recognition fell Below Expected Level (scaled score 4, 2nd percentile), with errors spread across several emotion categories rather than confined to one. Theory of Mind also fell Below Expected Level (scaled score 5, 5th percentile), with errors on items requiring inference about others' beliefs and intentions. These are reported as two separate findings: the NEPSY-II provides no social-perception composite, and independent factor-analytic work has not supported treating these subtests as one dimension. NEPSY-II subtest (fictional scores) Scaled score Classification Comprehension of Instructions 10 At Expected Level Speeded Naming 9 At Expected Level Word Generation 11 At Expected Level Narrative Memory 10 At Expected Level Memory for Faces 6 Borderline Affect Recognition 4 Below Expected Level Theory of Mind 5 Below Expected Level Low-score framing and integration: Some low scores are common in healthy children, so isolated low results are interpreted cautiously. Here the three low scores are not isolated: they cluster in socially loaded tasks (face memory, affect recognition, social inference), match the behavioral observations recorded during testing, and converge with the ADOS-2 observation findings, elevated SRS-2 parent and teacher ratings, and the relative socialization weakness on the Vineland-3 reported elsewhere in this evaluation. Diagnostic conclusions rest on that full pattern, not on any NEPSY-II subtest alone. Summary and recommendations linkage: The NEPSY-II profile shows age-appropriate language, attention, executive, and visuospatial functioning alongside specific weaknesses in face memory, affect recognition, and social inference. Each recommendation traces to a documented finding: social-communication intervention targets (Affect Recognition and Theory of Mind findings, with ADOS-2 and SRS-2 convergence) offered as input to the school evaluation team; speech-language pathology consultation focused on pragmatic language (behavioral observations and Theory of Mind errors); and structured teaching of face-name associations with visual supports (Memory for Faces findings). Limitations: NEPSY-II normative data were collected in 2005 and 2006, and results are interpreted at the subtest level against that norm base alongside more recently normed measures in this battery. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The measures paragraph names the edition, the assessment approach, the subtests chosen, and why they were chosen, and it states the norm collection date (2005 to 2006) as a limitation up front, so nothing in the section rests on an unstated battery or an unstated norm base. Every score stays at the subtest level: no domain average, no composite, no language that implies one, which is exactly what the instrument's own design requires. Low scores are interpreted against the base-rate reality that having some low scores is common in healthy children, and the two social-perception results are described as related but separate findings, not a single social index. Behavioral observations and process detail appear as scorable findings that explain how the child earned the scores, not as anecdotes, and the classification language uses the publisher's printed descriptor terms consistently. The interpretive summary answers the referral question by integrating NEPSY-II results with the ADOS-2, SRS-2, and adaptive data, and every recommendation traces to a documented finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the NEPSY-II section knowing which decision it will feed, because the surrounding rules differ by lane. In US schools, the LAW is IDEA's evaluation framework: under 34 CFR 300.304 , a team must "not use any single measure or assessment as the sole criterion" for eligibility, so NEPSY-II findings contribute evidence toward categories such as autism, speech or language impairment, traumatic brain injury, or other health impairment, and the write-up's job is to show how the subtest profile converges with everything else. On the payer side, the POLICY is coding convention: neuropsychological testing evaluation is billed under 96132 and 96133, psychological testing evaluation under 96130 and 96131, administration and scoring under 96136 through 96139, single automated instruments under 96146, and developmental testing under 96112 and 96113. Medicare's billing article A57780 describes the neuropsychological family as testing intended to "diagnose and characterize the neurocognitive effects of medical disorders", which is why the developmental-versus-neuropsychological code choice matters and why the report should document the medical or neurodevelopmental question the testing answered. Whether a school district, hospital program, or private payer funds the evaluation at all is CONVENTION and local policy, not statute, so state the referral question and medical necessity chain explicitly. Edition and norm currency is the second live question. The NEPSY-II's normative data were collected in 2005 and 2006, the edition itself dates to 2007, and as of July 2026 no third edition has been announced, so the norms are roughly two decades old and there is nothing newer to switch to. The manual itself notes that seven subtests (Design Fluency, Imitating Hand Positions, List Memory, Manual Motor Sequences, Oromotor Sequences, Repetition of Nonsense Words, and Route Finding) were carried over unchanged from the 1998 NEPSY with the original norms reprinted, which pushes their norm base back further still. Defensible practice, especially in medico-legal, eligibility, and re-evaluation contexts, is to name the edition and norm date in the report, use the battery for the process-level questions it answers well, and cross-reference measures with more current norms where general ability or achievement is at issue. The qualification boundary is the publisher's: Pearson sells the NEPSY-II at qualification level C, and purchase eligibility does not equal interpretive competence; interpretation remains a professional act governed by licensing boards. Outside the US, funding pathways change the write-up's audience. In Canada, there is no fee-for-service testing pathway equivalent to CPT billing; pediatric neuropsychological assessment typically runs through hospital and provincial programs, school boards, or private pay, and school-eligibility expectations vary by province and board, so confirm what the receiving team needs before finalizing. In Australia, no MBS item rebates standardized test administration itself; the relevant LAW-adjacent structure is the complex neurodevelopmental disorder pathway, where a consultant paediatrician (item 135) or psychiatrist (item 289) refers patients under 25, each item claimable once per lifetime, and the MBS explanatory notes cap the allied-health side at "a total of 8 allied health assessment services per patient per lifetime", with a practitioner able to provide up to 4 services before a review by the referring practitioner and up to 20 treatment services per lifetime after diagnosis, per the official MBS guidance . A NEPSY-II write-up feeding that pathway should make the diagnostic contribution of each assessment service explicit. NEPSY and NEPSY-II are trademarks, in the US and other countries, of Pearson plc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common NEPSY-II write-up errors reviewers flag There is no payer audit series for instrument write-ups; the accountability record here is psychometric, and for the NEPSY-II it cuts unusually deep. The instrument's own standardization data show that having some low subtest scores is common in healthy children, and the count that qualifies as uncommon is higher on a longer battery (Brooks, Sherman, and Iverson, Archives of Clinical Neuropsychology, 2010), so an isolated low score is an expected event, not automatic evidence of impairment. Two decades after publication, independent factor analyses of the norming sample arrived at the same caution from another direction: a 2024 exploratory analysis reported that only about 80 percent of reliability estimates exceed the 0.70 level considered marginal for a clinical instrument, could not locate a Social Perception dimension at all, and warned against using those subtests for "speculation about Autism" (McGill and colleagues, Journal of Pediatric Neuropsychology, 2024), while a 2025 confirmatory analysis of 600 children ages 7 to 12 supported domain-organized subtest reporting but found most general-factor loadings below .50 (Singh and colleagues, Child Neuropsychology, 2025). Those facts predict the failures below. The BastionGPT Clinical Advisory Board sees the same errors most often in NEPSY-II write-up reviews: A composite the instrument does not produce. The section reports a "NEPSY-II composite score", a domain average, or a full-scale equivalent. None exist: the manual dropped domain scores "in favor of the more clinically sensitive subtest-level scores". Averaging subtests invents a number with no norms behind it; report and interpret at the subtest level. One low subtest written as impairment. A single Borderline or Below Expected Level score in an otherwise average profile becomes "significant deficit" in the summary. Base-rate data from the standardization sample say having some low scores is common in healthy children; state the profile pattern, the subtest's reliability, and the converging evidence before calling anything a deficit. Social Perception written as an autism test. Affect Recognition and Theory of Mind get reported as one "social composite" or cited as diagnostic evidence on their own. They are two findings, not one index: the latent dimension has not been recoverable in independent factor analyses, and the diagnostic weight in an autism evaluation belongs to tools like the ADOS-2 and SRS-2 plus adaptive and developmental history. No edition or norm anchor. The report never states that the scores come from the 2007 NEPSY-II with norms collected in 2005 and 2006. Reviewers, opposing experts, and re-evaluators all read norm currency as a defensibility issue; name the edition and norm date and treat norm age as a stated limitation. Descriptor and score-type drift. The write-up mixes another instrument's descriptor system into NEPSY-II classifications, reads a percentile-range score as a precise percentile, or reports a contrast score as if it were an ability score. Use the publisher's printed classification terms consistently, and describe process, contrast, and behavioral-observation scores as what they are. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your score summary (subtests, scaled scores, percentiles, process and behavioral observations) and get a drafted NEPSY-II results-section narrative organized by domain and referral question, with the base-rate framing and norm-date limitation scaffolded in for your review. Cross-check a finished draft for the gaps reviewers flag: scores that disagree with their descriptor language, composite or domain-score wording the instrument cannot support, and social-perception findings written as standalone diagnostic evidence. Translate the results section into a plain-language summary for parents, teachers, or the IEP team that keeps subtest findings accurate without jargon. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do NEPSY-II scaled scores and percentile ranks mean? Most NEPSY-II primary and process scores are scaled scores with a mean of 10 and a standard deviation of 3, or percentile ranks. Some scores report as percentile ranges or cumulative percentages instead, because error scores and behavioral observations have skewed distributions where a precise percentile would overstate precision. The battery also yields process scores (which component of a task broke down), contrast scores (scaled comparisons of two abilities within or between subtests), and behavioral observations recorded against the standardization sample. Pearson's score reports print classification terms such as At Expected Level, Borderline, and Below Expected Level, and a defensible write-up uses those printed terms consistently. There is deliberately no composite or domain score to report. Does the NEPSY-II have a composite, index, or full-scale score? No, and this is the most common error in NEPSY-II write-ups. The manual dropped domain-level scores "in favor of the more clinically sensitive subtest-level scores" and states that domain scores are not derived, because subtests within a domain measure different constructs and a global number can mask the specific deficit the referral question asks about. Descriptions that mention a "NEPSY-II composite score", including some research summaries and third-party scoring guides, describe something the instrument does not produce. Report and interpret at the subtest level, grouped by domain and referral question; a 2025 confirmatory factor analysis of the norming sample likewise supported the authors' emphasis on subtest-level rather than composite-level interpretation. How many low NEPSY-II scores are normal for a healthy child? More than most reports acknowledge. Brooks, Sherman, and Iverson analyzed the 1,200-child NEPSY-II standardization sample (Archives of Clinical Neuropsychology, 2010) using scores at or below the 10th percentile as "low", and found that having some low scores is common in healthy children ages 3 to 16. In their analysis, what counted as uncommon was three or more low scores for ages 3 to 6, and for ages 7 to 16 four or more on a one-hour battery or five or more on a two-hour battery, so the longer the battery, the more low scores remain unremarkable. The write-up rule: interpret a low score against that base-rate reality and the subtest's reliability, and let patterns plus converging evidence, not one number, carry any conclusion. Are the NEPSY-II norms too old to use in 2026? They are old, and the defensible move is to say so in the report rather than hope nobody asks. The normative data were collected in 2005 and 2006 for the 2007 edition, and seven subtests (Design Fluency, Imitating Hand Positions, List Memory, Manual Motor Sequences, Oromotor Sequences, Repetition of Nonsense Words, and Route Finding) carry reprinted 1998 NEPSY norms. As of July 2026 no third edition has been announced, so there is no newer NEPSY to switch to. Reliability also varies by subtest and age band: independent analysis found only about 80 percent of reliability estimates exceed .70, a level "considered marginal for a clinical instrument" ( McGill and colleagues, 2024 ). Defensible practice: name the edition and norm dates, use the battery for the process-level questions it answers well, and cross-reference measures with more current norms for ability and achievement. Is the NEPSY-II required for school eligibility or an autism diagnosis? No law or diagnostic system names it. In US schools the governing rule runs the other way: under IDEA's evaluation procedures , teams must not use any single measure as the sole criterion for eligibility, so the NEPSY-II can contribute evidence toward categories such as autism, speech or language impairment, or traumatic brain injury, but never establishes one alone. Payer coverage is policy, not statute, and turns on the medical or neurodevelopmental question the testing answers. For autism specifically, the diagnostic weight sits with observation and interview instruments, informant ratings, and adaptive measures, the territory of the ADOS-2 , SRS-2 , and Vineland-3 , with the NEPSY-II adding process-level cognitive detail. Which battery to use is professional convention, chosen by referral question. Can the Affect Recognition and Theory of Mind subtests diagnose autism? No. The Social Perception domain was created for the NEPSY-II to strengthen assessment in autism-related referrals, but the current independent evidence says to keep the claims modest: factor analyses of the norming sample could not locate a Social Perception latent dimension in either analyzable age group, and McGill and colleagues explicitly caution against using these subtests for "speculation about Autism" ( Journal of Pediatric Neuropsychology, 2024 ). Report Affect Recognition and Theory of Mind as two separate findings, not a social composite, and let the ADOS-2 , SRS-2 , developmental history, and adaptive data carry the diagnostic conclusion. Low scores on these subtests are converging evidence at most. When should I use the NEPSY-II instead of the D-KEFS or WISC-V? Choose by age and construct, not habit. The D-KEFS (2001) is an executive-function battery of nine stand-alone tests for ages 8 through 89, so for executive questions below age 8 the NEPSY-II's Attention and Executive Functioning subtests are the option, while older adolescents needing deeper executive profiling can move to the D-KEFS. The WISC-V measures general cognitive ability; the NEPSY-II deliberately is not an IQ test and covers what the WISC-V does not (language processing, sensorimotor skills, social perception, and memory and executive process scores), so use it to supplement rather than duplicate. On Q-interactive, examiners can combine selected NEPSY-II subtests with WISC-V subtests in one custom battery, which is the practical answer to redundancy: administer the minimum set that answers the referral question. Where can I see the actual NEPSY-II questions or a real score report? Not here, and not on any legitimate page. Test items, stimuli, record forms, scoring keys, and norms tables are publisher-controlled materials sold at qualification level C, and professional ethics codes (including APA Ethics Code Standard 9.11 on maintaining test security) obligate psychologists to protect them; reproducing them would also compromise the validity of every future administration. Pearson posts official sample score reports on its product page, which show report formats without exposing test content. This page describes structure, score types, and write-up conventions only, and reproduces no items, stimuli, norms, or scoring materials. Can BastionGPT write my NEPSY-II results section? Yes, as a drafting and checking partner with you as the clinician of record. Paste your subtest-level score summary and BastionGPT drafts the results-section narrative organized by domain and referral question, with the no-composite discipline, base-rate framing, and norm-date limitation built into the draft. It can also audit a finished section for the errors reviewers flag, including composite language, score-versus-descriptor mismatches, and social-perception over-reach, and translate findings into a parent-friendly summary. It is clinically tested on psychological and neuropsychological evaluation reports, and HIPAA-compliant, so score data stays protected. You review, edit, and sign; see the template library for the full set of report guides. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, NEPSY-II product page : 2007 publication, ages 3 through 16, qualification level C, two record forms, administration times by assessment approach, Q-interactive subtest availability, official sample reports. Korkman, Kirk, and Kemp, NEPSY-II manual Chapter 2, Design and Purpose (Pearson-hosted PDF): the no-composite design rationale, six domains and 32 subtests plus four delayed tasks, score types, normative data collected 2005 to 2006 stratified to October 2003 US census targets, the seven subtests carried over with reprinted 1998 norms, and the eight referral batteries. Pearson, NEPSY-II brochure (PDF): spec strip, administration options, and referral-battery purposes. Pearson, NEPSY-II clinician sample report (PDF): the publisher's printed score types and classification vocabulary used in the fictional sample. Brooks, Sherman, and Iverson (2010), Healthy children get low scores too: prevalence of low scores on the NEPSY-II , Archives of Clinical Neuropsychology, 25(3), 182-190: the 1,200-child base-rate analysis behind the low-score framing. Brooks, Sherman, and Strauss (2010), NEPSY-II test review , Child Neuropsychology, 16(1), 80-101: the standing independent review of the instrument. McGill, Beaujean, Benson, Dombrowski, and Canivez (2024), Exploratory factor analysis of the NEPSY-II conceptual template , Journal of Pediatric Neuropsychology, 10(2), 107-119: reliability counts, stability range, and the Social Perception caution. Singh, Floyd, Reynolds, Pike, and Huenergarde (2025), What does the NEPSY-II measure in children ages 7 to 12? , Child Neuropsychology, 31(2), 197-228: the confirmatory factor analysis supporting subtest-level interpretation. 34 CFR 300.304 : IDEA evaluation procedures, including the no-single-measure rule. CMS Medicare Coverage Database, billing article A57780 : the psychological and neuropsychological testing code families. MBS Online, item 82000 explanatory notes and the complex neurodevelopmental disorder items FAQ : the Australian assessment and treatment service caps and referral pathway. Pearson, D-KEFS product page : the age range and test count used in the instrument-boundary comparison. ## Related templates Neuropsychological Evaluation Report WISC-V Report Write-Up ADOS-2 Report Write-Up SRS-2 Report Write-Up Vineland-3 Report Write-Up D-KEFS Report Write-Up Stroop Test Score Documentation Trail Making Test Score Documentation Boston Naming Test Score Documentation Beery VMI Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pai-report All note templates # PAI Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The PAI (Personality Assessment Inventory, 1991) is a 344-item self-report inventory of adult personality and psychopathology for ages 18 to 89, authored by Leslie C. Morey and published by PAR. Psychologists use it in clinical, forensic, correctional, and post-offer public safety evaluations, valued for its treatment planning scales and fourth-grade reading level. This page covers how to write up PAI results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinical, forensic, and police and public safety psychologists; publisher qualification level C Audience Courts and attorneys, referring physicians and psychiatrists, employers after conditional offers, disability and injury reviewers, treating clinicians Typical length 400 to 800 words for the results section · administration 25 to 55 minutes Format family Broadband self-report personality inventory When it's used Psychological and forensic evaluations, treatment planning, post-offer public safety screening, personal-injury and disability evaluations, correctional assessment Standards context Published by PAR (1991; Professional Manual 2nd ed. 2007); described here for write-up purposes; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PAI? The Personality Assessment Inventory (PAI) is a 344-item self-report inventory of adult personality and psychopathology for ages 18 to 89, authored by Leslie C. Morey and published by PAR in 1991, with a second edition of the Professional Manual in 2007. Its 22 scales are nonoverlapping, no item scores on two scales, and they sit in four clusters: 4 validity scales, 11 clinical scales spanning the neurotic spectrum, the psychotic spectrum, and behavior and impulse-control problems, 5 treatment consideration scales, and 2 interpersonal scales measuring warmth and dominance; 10 scales carry conceptually derived subscales, and a separate Critical Items form flags 27 items across nine content areas that may need immediate attention. Scores are linear T scores (mean 50, SD 10) referenced to a census-matched community sample of 1,000 adults, with companion clinical (1,265 patients across 69 sites) and college (1,051 students) samples that let one profile be read against both community and clinical reference points. Items sit at a fourth-grade reading level on a four-point response format, administration runs 25 to 55 minutes on PARiConnect, on paper, or through the PAI-SP desktop software, with an audio administration option, and hand scoring takes 15 to 20 minutes with no scoring keys. An adolescent version, the PAI-A, covers ages 12 to 18 with the same 22-scale structure across 264 items, and two separate Spanish products exist: a revised translation scored on the English-language norms and a European Spanish edition normed primarily in Spain. The load-bearing fact for write-ups, as of July 2026, is that the PAI's items and norms date to its original standardization while its interpretive layer has kept moving. The 2007 second edition revised the Professional Manual, not the test: the item booklet still carries its original 1989, 1990, and 1991 copyrights, and PAI Plus, the current interpretive expansion, overlays new distortion and random-responding indices, an optional DSM-5 Alternative Model for Personality Disorders profile, and context-specific comparison groups (child custody, chronic pain, bariatric surgery, college students, deployed military) on the same 344 items. PAR's own public-safety report documentation describes the community sample as 1,000 adults matched to the 1995 census, and a peer-reviewed study published in November 2025 reported that roughly 95% of a 204-student Canadian sample elevated at least one of the 22 scales, concluding that "the PAI US norms are no longer appropriate for high-stakes assessments" (Uttl, Sikma & Tat, PeerJ). No re-norming has been announced as of July 2026. None of that removes the PAI from practice: it means the write-up must name the norms and form it used and corroborate what they suggest. A PAI profile is evidence, not a verdict: it answers no referral question by itself, and its results section feeds the psychological evaluation report or forensic report that does, often alongside the MMPI-3 it competes with. ## Who uses PAI reports and when Clinical psychologists administer the PAI inside psychological evaluations to clarify diagnosis and plan treatment, where its five treatment consideration scales, two harm indicators, two measures of the client's environment, and one measure of treatment motivation, give it a planning vocabulary the MMPI family does not carry. It typically enters the chart alongside a psychiatric diagnostic evaluation , and neuropsychologists include it in neuropsychological batteries to characterize the emotional context around cognitive findings; Medicare's billing articles draw that boundary themselves, describing neuropsychological testing as the assessment that does not rely on self-report inventories. Forensic evaluators use it across criminal, civil, and family matters: in the largest international snapshot of forensic practice, the PAI appeared in 9.6% of the evaluations examiners described, second only to the MMPI among personality instruments, and a 2022 case-law review found its admissibility rarely challenged. Police and public safety psychologists administer it after a conditional offer through the PAI Law Enforcement, Corrections, and Public Safety Selection Report, built on an applicant normative sample of 17,757 public safety candidates and covered in the California POST screening manual. Personal-injury and disability evaluators lean on its negative-distortion indicators, and correctional systems have a dedicated interpretive report. The PAI competes for the broadband slot with the MMPI-3 , which brings 2020 norms and the larger validity-scale literature, and with the MCMI-IV when the referral question is DSM personality-disorder characterization. Choose the PAI when treatment planning, a fourth-grade reading level, or nonoverlapping content-driven scales matter most; whichever inventory you choose, pair it with interview, records, and targeted measures rather than asking one instrument to carry the evaluation. ## How to structure a PAI results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention the PAI interpretive literature teaches: protocol validity before any substantive statement, clinical findings organized by domain, the treatment consideration scales given their own section because they are the reason many evaluators chose this instrument, and the computer narrative treated as an input to paraphrase, never a report to paste. Each section carries the pitfall that most often undermines it. Measures, form, norms, and administration. Name the instrument (PAI), the form (full 344-item form, or the 160-item short form with the reason for using it), the language product (English; the Spanish revised translation scored on English-language norms; or the separately normed European Spanish edition), the platform (PARiConnect, paper, or PAI-SP), the modality, and any overlay applied (a PAI Plus context-specific comparison group, or public-safety applicant norms). Pitfall: an unnamed form or Spanish product. The two Spanish PAIs have different normative bases and the short form is literally the first half of the test; a score list that does not say which was given is ambiguous at re-evaluation and easy to attack on review. Protocol validity, before anything else. Address response consistency first (the inconsistency and infrequency scales, ICN and INF), then negative distortion (NIM and the supplemental over-reporting indicators), then positive distortion (PIM and the defensiveness indicators, including the discriminant function from Cashel and colleagues), and end with a plain conclusion about whether the substantive profile is interpretable, fully or with stated limits. Pitfall: listing validity T scores without a conclusion, or leaning on the Rogers Discriminant Function in a real-world forensic matter: the current meta-analysis states that RDF "does not effectively measure over-reporting in criterion groups" (Herring et al., 2025), and elevated baseline ICN or INF also predicts unstable retest results. Clinical scale findings by domain, not by scale number. Work through the three classes the instrument's own structure defines, the neurotic spectrum, the psychotic spectrum, and behavior and impulse-control problems, weaving subscales into the domain they inform rather than reciting all 22 scales in test order. State once, in prose, which reference point each statement uses: the community sample, the clinical sample the profile form also displays, or both. Pitfall: the 22-scale inventory dump, and unlabeled reference frames. A statement that is true against community norms may be unremarkable against the clinical sample; a write-up that never says which comparison it is making invites cross-examination. Treatment consideration scales, as their own section. This is the PAI's distinctive contribution: two harm indicators (suicidal ideation and aggression), two environment measures (stress and nonsupport), and one treatment-motivation measure (treatment rejection). Route the harm indicators to the evaluation's interview-based risk assessment, tie the environment findings to concrete stressors and supports, and weigh treatment rejection against the overall distress level when you write the engagement plan. Pitfall: omitting these scales from a treatment-planning report, or reporting the suicide and violence indicators as predictions. They are hypothesis flags that require interview corroboration, not actuarial verdicts. Interpersonal functioning. Describe the warmth and dominance dimensions as style context, affiliative versus cold and controlling versus submissive, and connect them to alliance and treatment-format implications where relevant. Pitfall: pathologizing normal-range interpersonal scores; these two scales describe style, not disorder. Comparison-group context, where one applies. In post-offer public safety screening, say which applicant or incumbent comparison the report used and keep the frames separate: public safety applicants respond so uniformly that unremarkable community-referenced profiles are the norm, which is exactly why the selection report exists. In clinical contexts using PAI Plus overlays, name the context group. Pitfall: reading a guarded screening profile as affirmative evidence of health, or quoting applicant-norm findings without saying they are applicant-referenced. Interpretive summary against the referral question. State what the PAI findings suggest as hypotheses, corroborate or bound them with interview, history, records, and other measures, and answer the referral question in the evaluation's integration section rather than from the inventory alone. Pitfall: a verdict from a T score. No PAI score is a diagnosis, a risk prediction, or a legal threshold, and in high-stakes matters the write-up should also state the norms' vintage and how the findings were corroborated. Recommendations linkage. Tie each recommendation to a stated finding: engagement strategy to the treatment-motivation finding, environmental interventions to the stress and support findings, further assessment to what the profile raised but cannot settle, safety planning to the interview-based risk assessment. Pitfall: recommendations that could follow any profile. If the list would read the same with different scores, the linkage is missing. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/form: PAI [full 344-item / 160-item short form + reason] Language/norms: [English / Spanish revised translation, English norms / European Spanish edition, its own norms] Platform: [PARiConnect / paper / PAI-SP] Modality: [in office / remote, documented] Overlay: [none / PAI Plus context group / public-safety applicant norms] PROTOCOL VALIDITY (before any substantive statement) Consistency: [ICN / INF findings in prose] Negative distortion: [NIM + supplemental indicators in prose] Positive distortion: [PIM + defensiveness indicators in prose] Conclusion: [interpretable / interpretable with stated limits / not interpretable, and what that means here] CLINICAL SCALE FINDINGS (reference population named: community, clinical, or both; convention stated once) Neurotic spectrum: [elevations and meaning, or "unremarkable"] Psychotic spectrum: [same] Behavior / impulse control: [same; subscales woven into prose] TREATMENT CONSIDERATION SCALES Harm indicators: [suicidal ideation / aggression findings, routed to the interview-based risk assessment] Environment: [stress / nonsupport findings, tied to specifics] Treatment motivation: [treatment rejection, weighed against distress level, tied to the engagement plan] INTERPERSONAL FUNCTIONING [Warmth and dominance pattern in one or two sentences] COMPARISON-GROUP CONTEXT (if applied) [Applicant, incumbent, or PAI Plus group; frames kept separate] INTERPRETIVE SUMMARY (answer the referral question) [Hypotheses supported; what the inventory cannot establish; convergence with interview, history, records; norms vintage stated in high-stakes matters] RECOMMENDATIONS LINKAGE [Each recommendation tied to a stated finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample PAI write-up (fictional) Scenario: a 41-year-old maintenance planner, referred through his employee assistance program for diagnostic clarification and treatment planning after several months of low mood, worry, and declining work performance; he agreed to the evaluation at his wife's urging. The protocol is valid with focal depressive and anxious elevations and a treatment-motivation finding that shapes the engagement plan, the treatment-planning shape where the PAI's consideration scales earn their keep. This is the PAI results section only, condensed but structurally complete. All details are fictional. Client: D.M., 41 · Referral: diagnostic clarification and treatment planning, employee assistance program referral · Evaluator: R. Calloway, PhD, Licensed Psychologist · Testing date: 07/08/2026 · Report date: 07/16/2026 Measures and administration: The Personality Assessment Inventory (PAI, full 344-item form, English) was administered in the office on PARiConnect as part of an evaluation that also included a diagnostic interview, collateral input from his wife with consent, brief symptom ratings, and record review, reported in their own sections. Scores below are linear T scores (mean 50, SD 10) referenced to the PAI community normative sample; where a statement draws on the clinical-sample comparison the profile form also displays, it says so. Elevations are characterized per the conventions of the PAI interpretive literature, and the interpretive report was reviewed, paraphrased, and reconciled with interview and history rather than quoted. Protocol validity: D.M. completed every item. Response consistency was acceptable (ICN and INF within normal limits), and neither negative nor positive distortion was indicated: NIM was not elevated, PIM was low, consistent with a person describing problems candidly rather than minimizing them, and the supplemental distortion indicators raised no concern. The substantive profile is interpretable without qualification, and its focal pattern is itself consistent with an open self-description rather than a cry for help. Scale (cluster) Fictional T score Inconsistency, ICN (validity) 52 Infrequency, INF (validity) 48 Negative Impression, NIM (validity) 55 Positive Impression, PIM (validity) 44 Depression, DEP (clinical) 76 Anxiety, ANX (clinical) 71 Anxiety-Related Disorders, ARD (clinical) 63 Stress, STR (treatment consideration) 68 Nonsupport, NON (treatment consideration) 64 Treatment Rejection, RXR (treatment consideration) 64 Warmth, WRM (interpersonal) 41 Clinical scale findings: The profile is a focal neurotic-spectrum picture. Depression is markedly elevated (DEP 76), with the subscale pattern weighted toward cognitive and affective features, self-criticism, hopelessness about improvement, and loss of interest, more than physiological ones. Anxiety is also prominent (ANX 71), expressed mainly as tension and worry, and the anxiety-related disorders scale approaches but does not reach a meaningful elevation (ARD 63), with no indication of trauma-specific distress on its subscales. The psychotic-spectrum and behavior-domain scales, including the substance scales, were unremarkable against both community and clinical reference points, which argues against thought-disorder and externalizing explanations for his decline at work. Treatment considerations: The harm indicators were not elevated: the suicidal ideation scale was within normal limits, and risk was separately assessed by interview and is addressed in the risk section of this evaluation. The environment measures localize the pressure: perceived stress is elevated (STR 68) around a restructuring at his plant, and perceived support is thin (NON 64), with his wife described as supportive but his wider network as sparse. Treatment rejection (RXR 64) is the finding that most shapes the plan: an elevation of this size is notable in a person reporting this much distress, and it converges with his statement that he came "because my wife asked me to." He acknowledges the problems but doubts that talking about them will help. Interpersonal functioning: The interpersonal scales describe a reserved, deferential style (WRM 41, DOM 44): D.M. is slow to disclose and unlikely to challenge a provider openly, which matters for alliance-building and for how disagreement with the treatment plan will surface. Integration and interpretive summary: The PAI findings converge with the diagnostic interview, his wife's collateral description, and the elevated depression and worry ratings on the brief measures reported elsewhere in this evaluation. Taken together they support the working impression of a moderate depressive episode with prominent generalized worry, documented in the diagnostic section, in a man under real occupational strain with limited support and low confidence in treatment. The profile generates hypotheses rather than verdicts: the treatment-rejection and support findings suggest the main early risk is disengagement, not diagnostic ambiguity, and the interview corroborated that reading. Recommendations linkage: The depression and anxiety findings support a course of psychotherapy with behavioral activation and worry-management components, coordinated with his primary care physician regarding a medication consult. The treatment-rejection finding drives the engagement plan: a motivational-interviewing opening phase, explicit early wins tied to his stated goal of keeping his job performance, and a check-in on treatment fit by session four rather than waiting for dropout. The stress and nonsupport findings support one concrete environmental target, negotiating his role in the restructuring, and a supported invitation to include his wife in one early session. These recommendations follow from the scale findings above and from the corroborating interview and history, not from the inventory alone. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works Validity is settled before substance. The write-up disposes of consistency, negative distortion, and positive distortion in order and states a plain interpretability conclusion, so every later sentence stands on a declared foundation. Scores and narrative agree, with reference points named. Every substantive claim traces to a listed score, the one near-threshold scale (ARD 63) is described as approaching, not reaching, elevation, and statements that draw on the clinical-sample comparison say so instead of mixing frames silently. The treatment consideration scales drive the plan. Treatment rejection is weighed against the distress level rather than read in isolation, and the stress and support findings become concrete targets, which is the specific value the PAI adds over other broadband inventories. Harm indicators are routed, not headlined. The unelevated suicidal ideation finding is stated and referred to the interview-based risk assessment, keeping actuarial-sounding predictions out of a self-report write-up. The computer narrative is integrated, not pasted. The report says the interpretive output was paraphrased and reconciled with interview and history, honoring APA Ethics 9.09 responsibility for interpretation. Form, language product, platform, and norms are named. Full 344-item form, English norms, PARiConnect, community reference: one sentence that keeps the scores interpretable at any future re-evaluation, which the PAI's norms question makes more important, not less. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The decision contexts around the PAI mix law, agency policy, and convention, and the write-up should signal which is which. LAW (US employment): the ADA restricts medical examinations to the post-offer stage, and PAR's public-safety documentation states it plainly: because the PAI measures psychopathology, it is administered to job applicants only after a conditional offer of employment, consistent with EEOC guidance applied uniformly to entering employees in the category. AGENCY POLICY: the IACP pre-employment guidelines recommend the evaluation itself be conducted post-offer, and the California POST screening manual (revised February 2024) covers the PAI and its Police and Public Safety Selection Report among reviewed instruments; POST's own documents note that in 2009 the screening dimension shifted from detecting job-relevant psychopathology to determining suitability, that a psychopathology test is still required and paired with a normal-range personality measure, and that roughly a quarter of candidates have consulted a mental health professional at some point while pass rates for those with and without prior contact are "not appreciably different." PAYER POLICY (US): the professional work is billed under 96130/96131 with administration and scoring under 96136/96137, or 96138/96139 when a technician administers; Medicare contractor billing articles A57481 and A57780 govern documentation, all six testing codes appear on the CMS List of Medicare Telehealth Services for CY 2026, and employment screening and most medico-legal work is retained or agency-funded rather than billed to health plans (see the telehealth documentation page for the wider remote-care rules). Outside the US the funding frame changes: Australia's Better Access items rebate psychological therapy services and an MBS search returns no item for personality-inventory administration, so PAI assessment there is typically privately, insurer, or medico-legally funded through distributors PAA and ACER; Canadian practice runs through provincial, insurer, employer, and medico-legal channels, with SIGMA Assessment Systems distributing alongside PAR, and the instrument has never been normed on a Canadian population. Edition and norms questions decide defensibility more often than any scoring detail, and the PAI's version of the question is vintage, not competing editions. The item booklet carries its original 1989, 1990, and 1991 copyrights, the 2007 second edition revised the Professional Manual rather than the test, PAI Plus overlays new indices and context comparisons on the same items, and PAR's public-safety documentation describes the community sample as matched to the 1995 census. A November 2025 peer-reviewed study ( Uttl, Sikma & Tat, PeerJ ) reported that roughly 95% of a 204-student Canadian sample elevated at least one of the 22 scales and concluded "the PAI US norms are no longer appropriate for high-stakes assessments"; no re-norming has been announced as of July 2026. The practical rule for reports: name the form, language product, and norms used, state the vintage in high-stakes matters before opposing counsel does, and corroborate every scale-based hypothesis with interview, records, and other measures. Purchase and use sit at PAR qualification level C, which is why the write-up names a qualified interpreter. Automated narratives get the same treatment as any consultant: PARiConnect and PAI-SP interpretive reports run 10 to 15 pages, and under APA Ethics 9.09 psychologists "retain responsibility for the appropriate application, interpretation, and use of assessment instruments," so paraphrase the narrative into case-anchored language rather than pasting it. One PAI-specific trap: PAR's Item and Response report prints item text with the client's answers, which is test material under APA Ethics 9.11 and belongs in the secure test file, never appended to the released report; retention of the protocol and report follows the record-set rules on the psychological evaluation report page. PAI and Personality Assessment Inventory are trademarks of PAR, Inc. (Psychological Assessment Resources). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common PAI write-up errors reviewers flag There is no payer audit series for personality-test write-ups; the accountability pressure comes from courts, opposing experts, licensing boards, and the publisher's license terms, and it is well documented. A 2022 review of published US case law found "the admissibility of the instrument was rarely challenged" (Meaux and colleagues, Journal of Personality Assessment), which means the practical scrutiny lands on the written report itself. The validity scales carry strong published detection evidence, the current meta-analysis of 43 articles and 6,451 participants reports standard over-reporting scales at g = .99 to 1.50 (Herring et al., 2025), and the base rates are not hypothetical: in an Australian workers-compensation sample, "233 claimants (32.1% of the total sample) elevated one validity index" (Yoxall, Bahr & O'Neill, 2017), with smaller groups elevating two or all three. The original PAI is independently reviewed in the Buros Center's Twelfth Mental Measurements Yearbook . The BastionGPT Clinical Advisory Board sees the same errors most often in PAI write-up reviews: Substance before validity, or validity without a verdict. A results section that presents clinical elevations first and mentions the validity scales later has the logic backwards, and a validity paragraph that lists T scores without concluding whether the profile is interpretable has not done its job. The stakes are empirical: elevated baseline inconsistency and infrequency scales also predict unstable results on retest, so an unaddressed consistency problem contaminates every downstream sentence. The Rogers Discriminant Function carried into real-world decisions. The original meta-analysis found over-reporting effects "were larger in simulation than criterion groups studies" (Hawes & Boccaccini, 2009), and the 2025 update is blunter: RDF "does not effectively measure over-reporting in criterion groups" and should not be used for those decisions at present (Herring et al.). A forensic write-up that rests a malingering opinion on RDF hands opposing counsel its cross-examination. Reference frames mixed without labels. The PAI profile supports community and clinical-sample comparisons, and public-safety reports add applicant and incumbent norms; a statement that is striking against one frame can be unremarkable against another. Write-ups that never say which comparison they are making, especially screening reports quoting applicant-norm findings as if they were community T scores, read as either confused or advocacy. The interpretive report pasted, or the item report appended. PARiConnect and PAI-SP narratives run 10 to 15 pages of hypothesis text; pasting them hands your interpretive responsibility to software against APA Ethics 9.09, which says psychologists "retain responsibility for the appropriate application, interpretation, and use of assessment instruments." Worse, PAR's Item and Response report prints item text with the client's answers: appending it to a released report exposes test material that APA Ethics 9.11 obligates you to protect. The norms question ducked and the form unnamed. After the November 2025 obsolete-norms study, high-stakes PAI reports that never state the norms' vintage invite the question from someone less friendly; and a report that does not say whether it used the full form, the 160-item short form ("developed for situations in which respondents complete only the 1st half of the test," Frazier et al., 2006), or which of the two Spanish editions, leaves its own scores ambiguous. Name the form, name the norms, and say how the findings were corroborated. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your scale summary (validity findings plus elevated scales with T scores) and get a drafted results-section narrative organized validity-first and by domain, with the treatment consideration scales given their own section and the referral question kept in frame. Cross-check a finished draft for the gaps reviewers flag: narrative claims with no corresponding score, treatment-motivation findings missing from the engagement plan, validity discussed after substance, unlabeled reference populations, and an unnamed form or Spanish edition. Translate the results section into a plain-language summary for the referral source, the treating clinician, or the client that keeps scale findings as corroborated hypotheses, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do PAI scores actually mean? PAI scales are reported as linear T scores with a mean of 50 and a standard deviation of 10, referenced to a census-matched community sample of 1,000 US adults; companion clinical (1,265 patients) and college samples let the same profile be read against community and clinical reference points, and the profile form displays both. No score means anything until the validity scales say the protocol is interpretable: consistency first (ICN, INF), then negative distortion (NIM and the supplemental over-reporting indicators), then positive distortion (PIM and the defensiveness indicators). Elevations are then characterized per the conventions of the PAI interpretive literature and read as hypotheses about how the person described themselves during that administration, to be corroborated with interview, history, and records; no PAI score is a diagnosis, a risk prediction, or a legal threshold. Because the scales are content-driven and nonoverlapping, the subscale pattern often matters more than the full-scale number, and the results section feeds the psychological evaluation report that answers the referral question. Are the PAI's norms too old to rely on? The norms date to the instrument's original standardization: the item booklet carries 1989, 1990, and 1991 copyrights, the 2007 second edition revised the Professional Manual rather than the test, and PAI Plus adds interpretive overlays on the same items. A November 2025 peer-reviewed study reported that roughly 95% of a 204-student Canadian sample elevated at least one of the 22 scales and concluded "the PAI US norms are no longer appropriate for high-stakes assessments" ( Uttl, Sikma & Tat, PeerJ ); no re-norming has been announced as of July 2026. That is an argument to manage, not an automatic disqualifier: the case-law record finds PAI admissibility rarely challenged, but the study hands cross-examiners fresh material. In high-stakes reports, state the norms' vintage yourself, corroborate every scale-based hypothesis, and consider whether the referral question is better served by an instrument with current norms, such as the MMPI-3 . Should I use the PAI or the MMPI-3? They compete for the same broadband slot, and the honest answer is fit by referral question. The MMPI-3 brings norms from 2020, US Spanish-language norms, and the largest validity-scale research base, which now includes the norm-currency argument in its favor. The PAI answers with five treatment consideration scales the MMPI family does not carry, nonoverlapping content-driven scales, a fourth-grade reading level with an audio option, and an established public-safety selection report. For treatment planning and lower-reading-level clients, the PAI earns its slot; for high-stakes forensic matters where norm vintage will be litigated, the MMPI-3 removes an argument; when the referral question is DSM personality-disorder characterization, the MCMI-IV is the usual third candidate. Whichever you choose, name the instrument, its norms, and your reason in the report. Is the PAI required for police pre-employment screening? No jurisdiction mandates the PAI itself; the requirements stack differently. LAW: under the ADA, tests that assess psychopathology are medical examinations, so the PAI is administered only after a conditional offer of employment, per EEOC guidance . POLICY: the IACP guidelines recommend conducting the evaluation post-offer, and California POST requires a test of psychopathology paired with a normal-range personality measure, naming the PAI and MMPI instruments as qualifying examples; the PAI Law Enforcement, Corrections, and Public Safety Selection Report exists for exactly this use, built on an applicant normative sample of 17,757 public safety candidates. CONVENTION: outside such frameworks, instrument choice is agency and evaluator convention. Worth telling candidates: POST materials report roughly a quarter of applicants have consulted a mental health professional at some point, and pass rates with and without prior contact are "not appreciably different," so prior counseling is not the disqualifier applicants fear. How should I report the PAI's suicide and violence indicators? As routed hypotheses, never as standalone predictions. The suicidal ideation and aggression scales, and the supplemental suicide and violence potential indexes, flag content and configurations that warrant follow-up; the write-up states the finding, then hands it to the evaluation's interview-based risk assessment, where history, protective factors, and collateral information live. The separate Critical Items form flags 27 items across nine content areas chosen for crisis relevance and very low endorsement by normal individuals, and the convention is the same: reviewed and dispositioned by interview, documented in the risk section, not reported as an actuarial verdict. A self-report inventory can tell you what a client was willing to endorse that day; it cannot carry a risk opinion alone, and a write-up that lets it will read as indefensible precisely when the stakes are highest. Is the 160-item PAI short form acceptable? It exists and is validated for what it is: the short form is the first 160 items of the full test, "developed for situations in which respondents complete only the 1st half of the test" (Frazier et al., 2006). Published comparisons report short-to-full clinical-scale correlations from 0.85 to 0.95 (Ward et al., 2018), but the same literature cautions that shortening brings "diminished content coverage, and altered validity detection" (Frazier et al.). The working rule: the full form is the default, and the short form is a documented fallback for interrupted or capacity-limited administrations, not a time-saver; several subscales and supplemental indexes lose coverage on it. If you report short-form results, say so by name, say why, and note the limits. An unnamed short form is one of the quiet errors reviewers catch by counting scales. Can I see the PAI's questions, or include them in a report? No. The items are copyrighted test material sold at PAR qualification level C, and protecting them is an ethics obligation (APA Ethics 9.11, maintaining test security) as well as a license term; item exposure and coaching are exactly the scenarios the over-reporting detection literature tests. For test takers, the legitimate preview is what this page describes: format, length, reading level, and response style, not item content. For report writers, the same rule covers PAR's Item and Response report, which prints item text with the client's answers: it belongs in the secure test file, and the released report describes findings at the scale and subscale level. Third-party demands for test materials are typically resolved by producing them to another qualified professional under protective conditions rather than into the open record; involve counsel and your board's guidance before releasing anything item-level. How is PAI administration and scoring billed? In the US, psychological testing bills under 96130/96131 for the evaluation work and 96136/96137 (professional) or 96138/96139 (technician) for administration and scoring; Medicare contractor articles A57481 and A57780 carry the documentation rules, and all six codes appear on the CMS telehealth list for CY 2026. Employment screening and most medico-legal PAI work is retained or agency-funded rather than billed to health plans. In Australia there is no MBS item for personality-inventory administration (an MBS search returns no results), so assessment is privately, insurer, or medico-legally funded; in Canada, coverage runs through provincial, insurer, employer, and medico-legal channels rather than a national fee schedule. State the jurisdiction when you make a coverage claim; the billing frame changes at every border. Can BastionGPT draft my PAI results section? Yes. Paste your scale summary (validity findings, elevated scales with T scores, any comparison group) and it drafts the results-section narrative for your review: validity first, findings organized by domain, the treatment consideration scales given their own section, and interpretive-report material paraphrased rather than pasted. It can also cross-check a finished draft for score-versus-narrative mismatches, unlabeled reference frames, and an unnamed form or edition, and produce a plain-language summary for the referral source or client. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: PAR, PAI product page : author, 344 items, ages 18 to 89, administration time, qualification level C, fourth-grade reading level, the community/clinical/college normative samples, scale architecture, the Critical Items form, Spanish editions, the PAS screener, and the publisher's own citation of the Maffly-Kipp and Morey stability study; PAI Plus (interpretive overlays on the existing items; Professional Manual 2nd Ed.); PAI-A (ages 12 to 18, 264 items); and the Law Enforcement, Corrections, and Public Safety Selection Report page with its product-resource PDF : the approximately 18,000-applicant comparison, four incumbent job classifications, and the item-booklet copyright line (1989, 1990, 1991). California POST, Peace Officer Psychological Screening Manual (revised February 2024): the PAI and JR&A report descriptions, the psychopathology-plus-normal-range pairing, and the mental-health-contact and pass-rate findings; and The PAI Police and Public Safety Selection Report : JR&A authorship (1995), the post-conditional-offer ADA statement, the 1995-census description of the community sample, the 17,757-applicant demographics, and the 2009 dimension revision from stability to suitability. EEOC enforcement guidance on pre-employment disability-related questions and medical examinations, and the IACP Pre-Employment Psychological Evaluation Guidelines (2020 revision): the post-offer rule and recommendation. CMS, billing articles A57481 and A57780 and the CY 2026 List of Medicare Telehealth Services : the testing-code framework and telehealth status. Peer-reviewed literature: Uttl, Sikma & Tat, PeerJ 13:e20340 (2025) , the obsolete-norms study; Hawes & Boccaccini, Psychological Assessment (2009) , the original over-reporting meta-analysis; Herring, Albertorio, Diehl & Ingram, Journal of Psychopathology and Behavioral Assessment 47:61 (2025) , the updated meta-analysis and RDF criterion-group finding; Meaux, Cox, Edens, DeMatteo, Martinez & Bownes, Journal of Personality Assessment (2022) , the US case-law review; Yoxall, Bahr & O'Neill, Psychiatry, Psychology and Law (2017) , the Australian workers-compensation validity-index rates; Neal & Grisso, Criminal Justice and Behavior (2014, author-archived) , the forensic-practice usage figures; Cashel, Rogers, Sewell & Martin-Cannici, Assessment (1995) , the defensiveness discriminant function; Frazier, Naugle & Haggerty, Psychological Assessment (2006) and Ward, Arnold, Cunningham & Liljequist, Journal of Clinical Psychology (2018) , the short-form composition and comparability evidence; Maffly-Kipp & Morey, Assessment (2024) , temporal stability. Buros Center for Testing, tests reviewed in the Twelfth Mental Measurements Yearbook : the PAI's independent-review listing. MBS Online search (no item for personality-inventory administration, verified July 2026); distributors SIGMA Assessment Systems (Canada) and Psychological Assessments Australia ; and the APA Ethics Code (Standards 9.06, 9.09, and 9.11). ## Related templates MMPI-3 Report Write-Up Psychological Evaluation Report Neuropsychological Evaluation Report Psychiatric Diagnostic Evaluation MCMI-IV Report Write-Up PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/rbans-report All note templates # RBANS Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The RBANS (Repeatable Battery for the Assessment of Neuropsychological Status) is a roughly 30-minute neuropsychological screening battery: 12 subtests yield five index scores and a Total Scale score, with four parallel forms for retesting. Neuropsychologists, psychologists, and medical teams use it to screen and track cognition in dementia, stroke, and psychiatric care. This page covers how to write up RBANS results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Neuropsychologists and clinical psychologists, with trained examiners administering under supervision; publisher qualification level B Audience Referring physicians, neurologists, and geriatricians, memory clinics, rehabilitation teams, psychiatry services, courts and payers, families Typical length 250 to 600 words for the RBANS section · administration about 30 minutes Format family Norm-referenced neuropsychological screening battery When it's used Dementia and mild cognitive impairment workups, post-stroke and rehabilitation baselines, psychiatric cognition checks, serial tracking with alternate forms Standards context Published by NCS Pearson (1998; RBANS Update 2012); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the RBANS? The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) is a brief, individually administered neuropsychological battery published in 1998 by The Psychological Corporation, now NCS Pearson, and authored by Christopher Randolph. Twelve short subtests yield five index scores (Immediate Memory, Visuospatial/Constructional, Language, Attention, and Delayed Memory) plus a Total Scale score, each on a standard-score metric with a mean of 100 and a standard deviation of 15, standardized on a US census-matched adult sample. Administration takes about 25 to 30 minutes. Its defining feature is in the name: four parallel forms (A through D) exist so cognition can be retested without repeating identical items, with Form A carrying the primary norms and equating adjustments linking the other forms, including a Spanish adaptation. The current edition is the RBANS Update (2012), which extended the age range from the original 20 to 89 down to 12:0 through 89:11, added subtest-level norms, and updated the manual with adolescent data and a research review. Pearson states plainly that "The items presented to the client have not changed", so older stimulus materials remain usable, and no newer edition has been announced as of July 2026. The load-bearing distinction for any write-up is that the RBANS is a screening instrument, not a comprehensive evaluation. Pearson positions it as a stand-alone "core" battery for detecting and characterizing dementia and as a "screen" battery when longer assessment is impractical, and the National Academy of Neuropsychology's education paper draws the line that governs interpretation: "cognitive screening tests should not be used as a replacement for comprehensive neuropsychological testing". The battery also contains no executive function, letter-fluency, or motor subtests, so a defensible RBANS section names those absent domains rather than letting a normal Total Scale imply a clean cognitive bill of health. In practice a RBANS write-up either stands alone as a screening summary that ends in a referral decision or sits inside a full neuropsychological report alongside comprehensive instruments such as the WAIS and the D-KEFS . ## Who uses RBANS reports and when Neuropsychologists and clinical psychologists interpret the RBANS, and Pearson lists it at qualification level B in the US catalog, so appropriately trained clinicians in memory clinics, neurology and stroke rehabilitation services, geriatric medicine, psychiatry, and hospital consultation settings all produce RBANS write-ups. The publisher positions it two ways, and the write-up should say which one applied: a stand-alone "core" battery for detecting and characterizing dementia in older adults, or a "screen" battery when a comprehensive evaluation is impractical or premature. It earns its place in serial work: four parallel forms exist specifically so cognition can be rechecked without repeating identical items, which is why rehabilitation teams and memory clinics tracking change reach for it first. Upstream of it sit bedside tools documented in a mental status exam ; downstream sits the comprehensive evaluation written up as a full neuropsychological report , often built around instruments like the WAIS with executive measures such as the D-KEFS covering what the RBANS by design does not. A RBANS section usually lives inside that larger report or stands alone as a brief screening summary that ends with a referral decision. ## How to structure a RBANS results section No statute, payer, or publisher mandates a RBANS results-section format. The structure below is the convention experienced clinicians converge on because it survives review: it names the edition, form, norms, and mode of administration, reports the five indexes individually before any total, treats change scores statistically, and states out loud what a 30-minute screen cannot do. Each section carries the pitfall that most often undermines it. Measures and methods statement. Name the battery and edition (RBANS, 1998, or RBANS Update, 2012), the form administered (A, B, C, or D, or Spanish record form), the platform (paper, Q-interactive, or Q-global scoring), the mode (in person or telepractice), and the norm set used. One methods sentence carries all of it: which stimulus form, which norms, and why the RBANS fits the referral question. Pitfall: writing "the RBANS" with nothing else. Form, norms, and mode each change what a score means, and a reviewer who cannot reconstruct them cannot defend the numbers. Behavioral observations and validity statement. Two or three sentences on arousal, engagement, sensory or motor barriers, and effort, closing with an explicit statement of how performance validity was assessed. If an embedded RBANS effort indicator was computed, interpret it with published false-positive rates for the population in view; in genuine dementia, a stand-alone performance validity measure is the safer anchor. Pitfall: a validity sentence that quietly applies an embedded effort cutoff to a memory-clinic patient. The literature on false positives in dementia is exactly what a savvy reviewer will cite back. Index-by-index results. Report each of the five indexes by name (Immediate Memory, Visuospatial/Constructional, Language, Attention, Delayed Memory) with its standard score and percentile, in the same order every time. Subtest-level scores, normed since the 2012 Update, belong here only when they add interpretive value. Describe the profile shape in prose: memory-predominant, patchy and lesion-congruent, or diffusely low. Pitfall: burying a divergent index inside a reassuring average. If Delayed Memory sits far below the others, the write-up says so plainly. Total Scale handling. Report the Total Scale score once, then interpret the indexes. When indexes diverge substantially, say whether the spread exceeds base-rate expectations in healthy adults rather than treating any scatter as pathology, and keep diagnostic weight off the composite. Pitfall: leading with a single Total Scale number. A composite built from divergent indexes describes no ability the patient actually has. Serial comparison. For retest write-ups, name the interval, the alternate form used, and the change method: a reliable change index or standardized regression-based change estimate, not raw point differences. State whether observed change clears the reliable-change threshold, and remember that an absent practice effect can itself be informative in a memory clinic. Pitfall: calling a small raw gain "improvement" or a small raw drop "decline". Measurement error and practice effects both move scores; reliable-change statistics exist to keep those out of the conclusion. Domains not assessed and limits. One short paragraph stating that the battery samples five domains in about 30 minutes and contains no executive function, letter-fluency, or motor subtests, so conclusions in those domains come from other measures or remain open. Add the screen-versus-assessment boundary: findings inform whether comprehensive neuropsychological assessment is warranted. Pitfall: silence about scope. A reader who assumes the RBANS covered executive function will over-read a normal Total Scale as a clean bill of cognitive health. Impression and recommendations linkage. Tie the profile to the referral question and to collateral data: informant report, daily-function changes, medical history. Every recommendation traces to a documented finding, and the referral recommendation states what a comprehensive evaluation would add. Pitfall: a screening profile carrying a diagnosis on its own. The RBANS contributes data to a diagnostic process; it does not finish one. ### Blank template (copy and adapt) RBANS RESULTS SECTION SKELETON (adapt; delete guidance in parentheses before signing) MEASURES AND METHODS Edition (RBANS 1998 / RBANS Update 2012) and form (A / B / C / D / Spanish): ____ Platform (paper / Q-interactive / Q-global scoring) and mode (in person / telepractice): ____ Norm set used and why the RBANS fits the referral question: ____ BEHAVIORAL OBSERVATIONS AND VALIDITY Arousal, engagement, sensory or motor barriers: ____ How performance validity was assessed (stand-alone measure preferred in dementia): ____ INDEX-BY-INDEX RESULTS (standard scores; report each index, then the total) Immediate Memory: ____ Visuospatial/Constructional: ____ Language: ____ Attention: ____ Delayed Memory: ____ Total Scale (interpret indexes first when they diverge): ____ Profile shape in prose (memory-predominant / patchy / diffuse): ____ SERIAL COMPARISON (retest write-ups only) Interval, alternate form used, and prior scores: ____ Reliable change or SRB result, not raw point difference: ____ DOMAINS NOT ASSESSED AND LIMITS (No executive function, letter-fluency, or motor subtests; screen, not a comprehensive assessment; state what remains open): ____ INTEGRATION WITH COLLATERAL AND FUNCTION Informant report, daily-function changes, medical history: ____ IMPRESSION AND RECOMMENDATIONS Impression tied to the referral question: ____ Recommendations linked to findings, including whether comprehensive neuropsychological assessment is warranted: ____ Clinician signature, credentials, date: ____ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample RBANS write-up (fictional) Scenario: a 74-year-old retired teacher referred by her primary care physician after an annual wellness visit flagged memory concerns that her husband confirmed. The RBANS was administered as a screening battery in a hospital memory clinic; this is the RBANS section of the resulting consultation note, condensed but structurally complete. All details are fictional. Client: R.M., 74 · Referral: memory concerns raised at an annual wellness visit, referred by primary care · Evaluator: L. Chen, PhD, Licensed Psychologist · Testing date: 07/09/2026 · Report date: 07/14/2026 Measures and methods: Repeatable Battery for the Assessment of Neuropsychological Status (RBANS Update, 2012), Form A, administered on paper in person and scored in Q-global against the test's standardization norms. The RBANS was selected as a brief screening battery to characterize the reported memory change and inform whether comprehensive neuropsychological evaluation is warranted; administration took about 30 minutes. A stand-alone performance validity measure was also administered. Behavioral observations and validity: Mrs. M. arrived on time with her husband, wore her prescribed glasses and hearing aids, and engaged readily. She worked carefully, asked for one repetition of task instructions, and showed mild frustration when she could not recall list items, saying "they slip away". Performance on the stand-alone validity measure was acceptable, and observed engagement supported valid results. Embedded RBANS effort indicators were not used as validity anchors, given their documented false-positive rates in older adults with genuine memory impairment. Results: Index standard scores have a mean of 100 and a standard deviation of 15. Scores below are reported index by index; the profile, not the composite, carries the interpretation. Index Standard score Percentile Immediate Memory 78 7 Visuospatial/Constructional 98 45 Language 96 39 Attention 88 21 Delayed Memory 75 5 Total Scale 84 14 The profile is memory-predominant. Both memory indexes sit well below expectations for age (Immediate Memory at the 7th percentile, Delayed Memory at the 5th), while Visuospatial/Constructional and Language performance falls in the middle of the distribution (45th and 39th percentiles) and Attention is modestly reduced (21st percentile). Free recall of learned material after a delay was the weakest performance of the session, and recognition prompts recovered little of it, an observation recorded here because it sharpens the referral question for comprehensive testing. The Total Scale score of 84 (14th percentile) is reported for completeness but is not the finding: with a 23-point spread between the highest and lowest indexes, the memory-specific pattern is what the composite would otherwise hide. Obtained scores carry measurement error, and confidence intervals accompany these scores in the full report. Scope of this screen: The RBANS samples five cognitive domains in about 30 minutes and contains no executive function, letter-fluency, or motor subtests. Conclusions about those domains, and any diagnostic conclusion, are outside what this screening battery can establish on its own. Collateral and function: Mr. M., interviewed separately as an independent historian, described about eighteen months of gradually increasing repetition of questions, misplaced household items, and new reliance on written lists, with Mrs. M. recently handing over the couple's bill paying. He described her mood as good and her daily self-care as unchanged. Impression and plan: Screening results show a memory-predominant profile that is consistent with the concerns raised at the wellness visit and by the family, in the context of preserved visuospatial and language performance and intact daily self-care with emerging instrumental-task changes. These screening findings do not establish a diagnosis. Comprehensive neuropsychological evaluation is recommended to characterize the memory impairment, assess domains this screen does not cover, and support diagnostic and care planning with her physician; medical workup of contributing conditions will proceed in parallel per the referring practice's pathway. If serial screening is used to track change, an alternate RBANS form with reliable-change methods is recommended rather than repeating Form A and comparing raw scores. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works The methods statement is complete. Edition, form, platform, mode, and norm set are all named in one place, so any reviewer, auditor, or future clinician can reconstruct exactly what produced the numbers. Scores and narrative agree, and the profile leads. Every claim in the prose matches the table, the indexes are interpreted individually, and the Total Scale is reported without being asked to carry a diagnosis the scatter will not support. Validity is handled with the right tools. A stand-alone performance validity measure anchors the validity statement, and the write-up says out loud why embedded effort indicators were not used for that job in this population. The screen states its own limits. Domains the battery does not assess are named, the screen-versus-assessment boundary is explicit, and the diagnostic question is routed to comprehensive evaluation instead of being settled by a 30-minute instrument. Collateral data and recommendations connect. The independent historian's account and the functional changes sit next to the scores, and every recommendation, including the alternate-form retest plan, traces to a documented finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Know which rule you are writing under, because the RBANS sits at the junction of three US regimes. The Medicare Annual Wellness Visit is required by regulation (42 CFR 410.15) to include "Detection of any cognitive impairment", defined as assessment by direct observation with due consideration of patient and family report; no specific instrument is mandated, which is LAW leaving tool choice to the clinician. When impairment is detected, the cognitive assessment and care plan service (99483) is PAYER POLICY with teeth: CMS contractor guidance states that 99483 requires an independent historian, notes the service is permanently covered via telehealth, and expects it no more than about once every 180 days. Formal testing billed under 96132/96133 (with administration under 96136 through 96139) must be medically necessary, and the same contractor policy is blunt about the boundary: testing when no mental illness or disability is suspected "would be considered screening and would not be covered by Medicare", while brief screens used in isolation are not classifiable as psychological or neuropsychological testing at all. The same policy lists assessment of decision-making capacity as a covered indication, and here convention matters: RBANS results contribute data to capacity, driving, and independent-living questions, but those are clinical and legal judgments that no screening score decides. In Canada, outpatient neuropsychological assessment is typically accessed privately, through extended health benefits, or inside hospital programs, and specifics vary by province, which is CONVENTION rather than a published entitlement. In Australia, a search of the Medicare Benefits Schedule for "neuropsychological" returns no general item for standalone adult assessment (verified July 2026); public access runs through specialist consultations, Better Access arrangements, or the under-25 complex neurodevelopmental pathway, with most adult work hospital-based or private. The methods statement is where RBANS write-ups earn or lose their defensibility, because "the RBANS" is not one thing. Name the edition (1998 original or 2012 Update), the form (A through D, or Spanish), the platform (paper, Q-interactive, or Q-global scoring), the mode, and the norm set. The Update's published change list added subtest-level norms and the adolescent extension without changing the items, so as of July 2026 the core index norms trace to the original standardization program, and alternatives exist that change the numbers: age- and education-corrected norms built from 718 community-dwelling older adults recruited in primary care (Duff and colleagues, 2003), clinical norms for schizophrenia from 575 US patients (Wilk and colleagues, 2004) and 174 British Columbia inpatients (Iverson, Brooks, and Haley, 2009), and published Australian normative data (Green and colleagues, 2008), so the report must say which yardstick was applied. Telepractice deserves its own sentence: Pearson's guidance states that "Telepractice is a deviation from the standardized administration" and that the normative data were collected in person, and the supporting evidence base, while positive, remains small (an early videoconference study supporting feasibility and reliability enrolled 18 older adults), so remote administration is documented as a deviation with clinical judgment stated. Purchase requires Pearson qualification level B in the US catalog; qualification to buy is not competence for every interpretive use, which remains a licensing and professional-standards matter. Retention and access rules follow the parent record: a RBANS section inside a neuropsychological report or psychological evaluation report inherits that report's obligations, including the raw-data and test-security handling those pages cover. RBANS is a trademark of NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common RBANS write-up errors reviewers flag The published numbers behind these flags are striking. When researchers applied the embedded RBANS Effort Index to people with genuine impairment, 37% of nursing-home residents and 33% of patients with probable Alzheimer's disease scored past the suggested cutoffs, against 3% of cognitively intact older adults (Duff and colleagues, 2011), and a mixed-dementia study put overall Effort Index failure at 48%, with the newer Effort Scale failing 4% of Alzheimer's cases but 31% of non-Alzheimer's dementias (Burton and colleagues, 2015). One-year index stability in a community-dwelling elderly sample ran from .58 to .83 (Duff and colleagues, 2005). Numbers like these are why reviewers read RBANS sections skeptically. The BastionGPT Clinical Advisory Board sees the same errors most often in RBANS write-up reviews: The Total Scale carries the conclusion. When indexes diverge, a single composite describes no ability the patient has, and published base-rate tables show index discrepancies occur in healthy adults (Patton and colleagues, 2006). Reviewers expect index-by-index reporting with scatter judged against base rates, not a lone summary number. Raw retest differences presented as change. Test-retest reliability is respectable for a screen (intraclass correlations of .84 in schizophrenia and .77 in healthy comparisons, Wilk and colleagues, 2002) but nowhere near perfect, and practice effects are real even with alternate forms. Defensible serial write-ups use reliable change indices or the updated standardized regression-based equations, published after earlier equations failed to generalize (Hammers and colleagues, 2020; externally validated in amnestic MCI, 2022). Embedded effort indicators over-read in dementia. An Effort Index failure in a memory-clinic patient is often the impairment talking, not the effort, per the false-positive rates above. Reviewers flag reports that lean on embedded indicators alone where stakes are high; stand-alone performance validity measures and stated clinical judgment belong in the validity paragraph. No edition, form, or norm statement. A score without its yardstick cannot be audited: 1998 original versus 2012 Update, Form A through D or Spanish, and standardization versus age- and education-corrected or population-specific norms all change interpretation. Reviewers treat a missing methods sentence as a defect, not a style choice. A screen doing an assessment's job. Write-ups that diagnose dementia from a RBANS alone, or that stay silent about the absent executive, letter-fluency, and motor domains, overreach the instrument. The defensible move is explicit: state the scope, state what remains open, and route the diagnostic question to comprehensive evaluation. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste a RBANS score summary (edition, form, index scores, observations) and BastionGPT drafts the results narrative with the methods statement, percentile framing, and the screen-not-diagnosis boundary already in place for your review. It cross-checks a finished section for the reviewer flags on this page: a Total Scale doing diagnostic work, raw retest differences presented as change, a missing form or norms statement, or scores that disagree with the narrative. It turns finished findings into a plain-language summary for the patient, family, or referring provider. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do RBANS scores mean? The RBANS reports five index scores (Immediate Memory, Visuospatial/Constructional, Language, Attention, Delayed Memory) and a Total Scale score, each a standard score with a mean of 100 and a standard deviation of 15, so 100 sits at the middle of the distribution for the normative group, 85 falls about one standard deviation below it (roughly the 16th percentile), and 70 about two below (near the 2nd percentile). Subtest-level scores have their own norms since the 2012 Update. Keep the percentile next to the number, and if you use qualitative descriptor words, name the labeling convention and apply it consistently; a neuropsychology consensus on uniform score labels exists for exactly this reason (Guilmette and colleagues, 2020). No RBANS score is a diagnostic cutoff: cut-scores in the research literature are study- and population-specific, and no consensus diagnostic threshold exists. Is there a new edition of the RBANS? No. The current edition is the RBANS Update, released in 2012 as a revision of the 1998 original, and no newer edition has been announced as of July 2026. The Update extended the age range downward from the original 20 to 89 to 12:0 through 89:11, added subtest-level norms, added equating studies for Forms C, D, and Spanish Form A, and updated the manual with adolescent data and a review of RBANS research. Pearson's support materials state that "The items presented to the client have not changed", which is why older stimulus books remain usable and why your methods statement should name the edition and norms rather than assuming the reader knows. If a client was tested years apart, the same items may have been scored against different norm sets, which belongs in any serial comparison. How is the RBANS different from the MoCA or MMSE? Depth and repeatability. Bedside screens like the MoCA and MMSE take minutes and produce a single global result, usually recorded within a mental status exam . The RBANS takes about 30 minutes and produces five co-normed index scores plus a total, with four parallel forms built for retesting, so it can characterize a profile (memory-predominant versus diffuse, for example) that a single number cannot. Medicare contractor policy even draws the line administratively: brief screening measures used in isolation are not classified as psychological or neuropsychological testing. What the RBANS does not do is graduate out of the screening category: it omits executive function, letter fluency, and motor domains, and a diagnostic answer still requires comprehensive evaluation. Can the RBANS diagnose dementia by itself? No. The National Academy of Neuropsychology's education paper is explicit that screening results do not substitute for comprehensive neuropsychological testing, and Pearson itself positions the RBANS as a "core" battery for detecting and characterizing dementia, which is a contribution to a diagnostic process rather than the process. The instrument discriminates well at the group level (in one Alzheimer's study every index separated patients from controls, with the memory indexes strongest), but a diagnosis of a neurocognitive disorder integrates history, informant report, daily function, medical workup, and usually fuller testing. The defensible write-up reports the profile, states the screen's limits, and routes the diagnostic question to a comprehensive evaluation documented in a neuropsychological report . Is the RBANS required for the Medicare wellness visit or code 99483? No rule names it. The Annual Wellness Visit regulation ( 42 CFR 410.15 ) requires "Detection of any cognitive impairment", defined as assessment by direct observation with consideration of patient and family report; that is LAW, and it mandates no instrument. If impairment is detected, the cognitive assessment and care plan service (99483) is PAYER POLICY: CMS contractor guidance ( article A59036 ) states it requires an independent historian, is permanently covered via telehealth, and is expected no more than about every 180 days. Formal testing under 96132/96133 requires documented medical necessity, and testing done purely as screening is not covered. Choosing the RBANS to answer a detected concern is CONVENTION: defensible, common, and never mandated. How should I report change across repeat RBANS testing? Statistically, never as raw point differences. The alternate forms exist to reduce item practice, but scores still move with measurement error and practice: one-year index stability ran from .58 to .83 in a community-dwelling elderly sample (Duff and colleagues, 2005), and a large meta-analysis quantifies practice effects across repeated neuropsychological testing generally (Calamia and colleagues, 2012). The write-up names the interval and the form used, then applies a reliable change index or standardized regression-based (SRB) change equation. Use the updated SRB equations published in 2020, which replaced earlier equations that had failed to generalize, and which were externally validated in amnestic mild cognitive impairment in 2022 (Hammers and colleagues). State whether the observed change clears the reliable-change threshold; that sentence is the finding. Can the RBANS be administered by telehealth? It can be, with the deviation documented. Pearson's guidance document states that "Telepractice is a deviation from the standardized administration" and reminds examiners that the normative data were collected in person; its subtest-level equivalence table adds that more research is needed across ages and tasks. The peer-reviewed evidence is positive but small: an early videoconference study in adults over 55 supported feasibility and reliability of remote administration (Galusha-Glasscock and colleagues, 2016). The defensible write-up names the mode, describes any facilitation at the patient's site, and interprets scores with the deviation stated, exactly as the publisher's telepractice guidance recommends. Do the embedded RBANS effort indicators prove valid effort? No, and in dementia they mislead badly if used alone. Researchers have derived several embedded indicators from RBANS performance: the Effort Index (Silverberg and colleagues, 2007), the Effort Scale (Novitski and colleagues, 2012), and the Performance Validity Index and CRIER (Paulson and colleagues, 2015). The false-positive problem is well documented: 37% of nursing-home residents and 33% of probable Alzheimer's patients exceeded Effort Index cutoffs against 3% of cognitively intact older adults (Duff and colleagues, 2011), and a mixed-dementia sample failed the Effort Index at 48% overall, with the Effort Scale failing 31% of non-Alzheimer's dementias (Burton and colleagues, 2015). Professional consensus expects validity to be addressed proactively in every evaluation, so the working standard is a stand-alone performance validity measure plus stated clinical judgment, with embedded indicators treated as context calibrated to the population, never as a verdict. Can BastionGPT help with RBANS write-ups? Yes. BastionGPT is trained and clinically tested on psychological and neuropsychological evaluation reports, the parent documents RBANS sections live inside. Paste a score summary (edition, form, index scores, observations) and it drafts the results narrative with the methods statement, percentile framing, and screen-not-diagnosis boundary in place for your review; it can also check a finished section for a Total Scale doing diagnostic work, raw retest differences presented as change, or a missing form and norms statement, and produce a plain-language summary for the family or referring provider. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from a score summary you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: NCS Pearson, RBANS Update product page : 2012 publication, ages 12:0 through 89:11, 30-minute completion time, qualification level B, the four forms and Spanish record forms, Q-interactive and Q-global options, and the core-battery and screen-battery positioning quoted on this page. Pearson Support, RBANS Update: Features : the Update change list (age extension from 20 to 89 down to 12:0 through 89, equating studies for Forms C, D, and Spanish Form A, subtest-level scores, manual updates) and the items-unchanged statement quoted on this page. Pearson, Telepractice and the RBANS : the standardization-deviation and in-person-norms statements quoted on this page, and the subtest input/output equivalence table with its more-research caution. 42 CFR 410.15 : the Annual Wellness Visit element "Detection of any cognitive impairment" and its direct-observation definition. CMS Medicare Coverage Database, Article A59036 : the 99483 independent-historian requirement, permanent telehealth coverage, and frequency expectation. CMS Medicare Coverage Database, LCD L35101 : the screening exclusion, the brief-screens-in-isolation limitation, and decision-making capacity as a covered testing indication. MBS Online search : no general item for standalone adult neuropsychological assessment (verified July 2026); item 82000 defines the narrow under-25 pathway. Buros Center for Testing: the original RBANS is reviewed in the 14th Mental Measurements Yearbook and the RBANS Update is listed in the 19th . Roebuck-Spencer et al., Cognitive Screening Tests Versus Comprehensive Neuropsychological Test Batteries , National Academy of Neuropsychology education paper, Archives of Clinical Neuropsychology 32(4), 2017: the screening-boundary sentence quoted on this page. Gold et al., American Journal of Psychiatry, 1999 : the score metric (total and five indexes, mean 100, SD 15), the census-matched standardization, and the 25-minute administration figure. Peer-reviewed norms and change studies: Duff et al. 2003 (age- and education-corrected norms, 718 community-dwelling older adults), Wilk et al. 2004 and Iverson et al. 2009 (schizophrenia clinical norms), Green et al. 2008 (Australian normative data), Patton et al. 2006 (index-discrepancy base rates), Duff et al. 2005 (one-year stability), Hammers et al. 2020 and 2022 (SRB change equations and external validation), and Wilk et al. 2002 (test-retest reliability). Effort-indicator and labeling literature: Silverberg et al. 2007 , Novitski et al. 2012 , Paulson et al. 2015 (indicator development), Duff et al. 2011 and Burton et al. 2015 (false-positive rates in dementia), Duff et al. 2008 (index discrimination), and Guilmette et al. 2020 (uniform score labels). ## Related templates Neuropsychological Evaluation Report WAIS-5 / WAIS-IV Report Write-Up D-KEFS Report Write-Up Psychological Evaluation Report Mental Status Exam (MSE) Note MoCA Score Documentation MMSE Score Documentation Stroop Test Score Documentation Trail Making Test Score Documentation Boston Naming Test Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/srs-2-report All note templates # SRS-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The SRS-2 (Social Responsiveness Scale, Second Edition) is a 65-item rating scale quantifying autism-related social impairment from age 2 and a half through adulthood, completed by parents, teachers, and adults about themselves. Psychologists use it as a corroborating measure in autism evaluations and a progress measure in ABA programs. This page covers how to write up SRS-2 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and developmental pediatricians; publisher qualification level C Audience Autism evaluation teams, IEP teams, referring physicians, ABA programs and payers, parents and caregivers Typical length 150 to 400 words for the SRS-2 section · administration 15 to 20 minutes per form Format family Multi-informant autism trait rating scale When it's used Autism diagnostic evaluations, ABA outcome measurement, school evaluations, adult and late-identification assessments, treatment monitoring Standards context Published by WPS (2012); SRS-3 released in 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the SRS-2? The Social Responsiveness Scale, Second Edition (SRS-2) is a 65-item, norm-referenced rating scale by John N. Constantino, MD, with Christian P. Gruber, published by WPS in 2012 , that identifies autism-related social impairment and quantifies its severity across the lifespan. Four forms cover the age span: Preschool (ages 2.5 to 4.5) and School-Age (ages 4 to 18), each completed by a parent or teacher, plus an Adult form completed by a relative or other informant and an Adult Self-Report, both for ages 19 and up. Every form yields T scores (mean 50, standard deviation 10) for a Total score, two DSM-5 compatible subscales (Social Communication and Interaction; Restricted Interests and Repetitive Behavior), and five treatment subscales, against norms from 1,906 individuals and 4,709 ratings, separated by rater type and by the age and gender of the person rated. Scoring runs by hand on AutoScore forms or online through WPS's platform. Edition status matters as of July 2026: the SRS-3 published in 2026 with re-normed forms and structural changes, Pearson's Australian distributor page has carried an "SRS-3 Coming Soon in 2026" banner, and the SRS-2 remains on sale and in wide payer and program use during the transition. The load-bearing distinction for write-ups: the SRS-2 measures a trait dimension, it does not decide a diagnosis. Its scores quantify how much autism-related social difficulty an informant observes, and elevations are consistent with, but not specific to, autism: peer-reviewed comparisons show strong separation of autism from typical development but substantially weaker separation from disruptive behavior disorders , social anxiety and selective mutism , and other conditions, and unadjusted scores can read as "general levels of impairment" rather than autism-specific severity. That is why the SRS-2 enters a psychological evaluation report or a diagnostic evaluation as one converging data stream alongside direct observation such as the ADOS-2 , developmental history, and clinical judgment. A results section that reports the scores dimensionally, addresses what else could elevate them, and lets the whole evaluation carry the diagnostic conclusion is most of what separates a defensible SRS-2 write-up from a score printout with sentences around it. ## Who uses SRS-2 write-ups and when Psychologists and school psychologists write up SRS-2 results inside autism diagnostic evaluations and school evaluations, where it supplies the quantified, multi-informant view of everyday social behavior that direct observation cannot; developmental pediatricians and psychiatrists order it for diagnostic clarification inside a diagnostic evaluation ; ABA programs administer it on a schedule because TRICARE's Autism Care Demonstration requires the Parent Form at baseline and every year as an outcome measure; and clinicians assessing adults use the Adult and Adult Self-Report forms in late-identification workups, where the self-report has published validation in autistic adults . It is the right tool when the question is dimensional: how much autism-related social difficulty do the people around this person observe, in which settings, and how has it changed. The instruments around it then carry their own loads: the ADOS-2 for structured direct observation, the ADI-R for the developmental interview, a broadband scale such as the BASC-3 when the referral question spans attention, mood, and conduct rather than social reciprocity alone, and adaptive measures such as the Vineland-3 for daily-living skills. The SRS-2 write-up is where the informant evidence gets integrated: multiple raters, one dimensional account, clearly subordinated to the full evaluation. ## How to structure an SRS-2 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names what was administered and who the raters were, reports the Total score before the subscales, integrates informants instead of averaging them, and closes with the specificity statement that this instrument, more than most, cannot do without. Each section carries the pitfall that most often undermines it. Measures, forms, raters, and norms. Name the instrument and edition, the form completed by each rater (Preschool, School-Age, Adult, Adult Self-Report), each rater's relationship and setting, completion dates, and the norm reference, since SRS-2 norms are separated by rater type and by the age and gender of the person rated. Pitfall: writing "an SRS" with no edition or form. With the SRS-3 published in 2026 and re-normed, an unnamed edition leaves reviewers unable to tell which norms produced your T scores. Rating quality and completeness. The SRS-2 has no formal validity indexes, so the write-up carries that weight in prose: state that forms were complete, each rater's opportunity to observe, and anything that could color the ratings, such as rater stress or a recent crisis. Pitfall: treating informant ratings as objective measurement. They are structured observations from a vantage point, and the new Inconsistent Responding Index in the SRS-3 exists precisely because rating quality varies. Total score first. Report the Total T score with its qualitative severity descriptor in words (within normal limits, mild, moderate, severe) and its plain-language meaning. The Total score is the instrument's primary, continuously distributed index of severity. Pitfall: reporting a numeric threshold as if it were a diagnostic cutoff. The descriptor bands characterize normative elevation, and no T score establishes or excludes a diagnosis. DSM-5 compatible subscales. Report Social Communication and Interaction and Restricted Interests and Repetitive Behavior, which map onto the two DSM-5 criterion domains and have factor-analytic support in a sample of more than 9,000. Pitfall: building the interpretation on the five treatment subscales instead. The publisher states they are not used for screening or diagnosis; they are intervention-planning aids, and reviewers know the difference. Cross-informant integration. Report each informant's scores and interpret convergence and divergence rather than summarizing forms in sequence. Parent and teacher agreement is modest by design across instruments, averaging r = .28 in a meta-analysis of 341 studies , and the documented SRS-2 conventions are to treat an elevation from either informant as a flag for full evaluation and to seek additional rater input when scores differ by more than 10 T points ( Texas Statewide Leadership for Autism Training ). Pitfall: averaging raters into a single number, or crowning one rater correct. Divergence is usually setting information, and the write-up should say what it means. Differential and specificity statement. One sentence stating that elevations are consistent with, but not specific to, autism, followed by the contributors you considered: attention and behavior problems, anxiety, language and cognitive level, trauma. Published comparisons show weaker discrimination from disruptive behavior disorders and overlap with social anxiety and selective mutism . Pitfall: omitting the statement entirely. This is the single most common substantive gap reviewers flag in SRS-2 sections, because the instrument's documented weakness is specificity, not sensitivity. Integration with direct measures and history. State where the SRS-2 findings converge with or diverge from direct observation (such as the ADOS-2 ), the developmental interview, and records. Correlations between the SRS-2 and direct diagnostic measures are moderate at best, so some disagreement is expected. Pitfall: explaining divergence away. A rating-scale elevation the observation did not corroborate, or the reverse, is a finding to interpret, not an inconvenience to bury. Interpretive summary and recommendations linkage. Answer the referral question dimensionally, state what the scores do and do not establish, and tie each recommendation to a finding, including the re-rating plan: who completes which form, when, and why, matching any program cadence such as TRICARE's annual requirement. Pitfall: recommendations that could follow any profile. If nothing in the plan traces to the informant evidence, either the plan or the results section is incomplete. ### Blank template (copy and adapt) MEASURES, FORMS, AND RATERS Instrument/edition: SRS-2 Form(s): [Preschool / School-Age / Adult / Adult Self-Report] Raters: [name or role, relationship, setting, date completed] Norm reference: [rater type; age and gender of person rated] RATING QUALITY [Forms complete? Each rater's opportunity to observe; factors that could color ratings; how any concern was handled] TOTAL SCORE [Total T per rater + qualitative descriptor in words + plain-language meaning; no numeric cutoffs] DSM-5 COMPATIBLE SUBSCALES Social Communication and Interaction: [T scores, descriptor] Restricted Interests and Repetitive Behavior: [T scores] Treatment subscales (if reported): [intervention planning only] CROSS-INFORMANT INTEGRATION [Where raters converge; what divergence means (setting, perspective); >10 T-point difference: note added input] DIFFERENTIAL AND SPECIFICITY STATEMENT [Elevations consistent with, not specific to, autism; contributors considered: ADHD, anxiety, language, other] INTEGRATION WITH DIRECT MEASURES AND HISTORY [Convergence/divergence with observation, interview, records] INTERPRETIVE SUMMARY (answer the referral question) [Dimensional statement; what the scores do NOT establish] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-rating plan: form, rater, timing (payer cadence if applicable)] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample SRS-2 write-up (fictional) Scenario: an 8-year-old referred by his pediatrician for an autism diagnostic evaluation after two years of social difficulties at school. Parent and teacher School-Age forms diverge by more than 10 T points, the cross-informant shape where the write-up's job is to interpret the difference instead of averaging it. This is the SRS-2 section of the larger evaluation report only, condensed but structurally complete. All details are fictional. Client: L.T., 8 · Referral: autism diagnostic evaluation, referred by pediatrician · Evaluator: M. Okafor, PsyD, Licensed Psychologist · Ratings completed: 07/06/2026 (parent), 07/08/2026 (teacher) · Report date: 07/16/2026 Measures and administration: The Social Responsiveness Scale, Second Edition (SRS-2) School-Age Form was completed independently by L.T.'s mother and by his third-grade classroom teacher as part of an evaluation that also included the ADOS-2, a developmental history interview, cognitive screening, and record review, reported in their own sections. Scores are T scores (mean 50, SD 10) against the SRS-2 norms for each rater type and for L.T.'s age and gender. Severity descriptors below follow the manual's qualitative conventions and are stated in words; no score is treated as a diagnostic threshold. Rating quality: Both forms were fully completed. L.T.'s mother sees him daily across home and community settings; his teacher has observed him across seven months of structured classroom, lunchroom, and recess contexts. No recent family disruption or rater concern was identified that would be expected to color the ratings, and both raters described their responses as reflecting typical recent behavior. SRS-2 scale Parent T score Teacher T score Total 78 66 Social Communication and Interaction (DSM-5) 77 66 Restricted Interests and Repetitive Behavior (DSM-5) 74 63 Social Awareness (treatment) 68 58 Social Cognition (treatment) 74 64 Social Communication (treatment) 76 65 Social Motivation (treatment) 70 61 Total score: L.T.'s mother's ratings produced a Total T score of 78, in the range the manual describes as severe, indicating that she observes social difficulties of a kind and frequency strongly associated with clinically significant interference in everyday interaction. His teacher's ratings produced a Total T score of 66, in the moderate range, indicating clearly elevated but less pervasive difficulty in the classroom setting. DSM-5 compatible subscales: Both raters elevated both domains. Social Communication and Interaction (parent T = 77; teacher T = 66) and Restricted Interests and Repetitive Behavior (parent T = 74; teacher T = 63) were consistent with the Total score pattern for each rater, indicating that the reported difficulty spans both DSM-5 criterion domains rather than concentrating in one. Treatment subscale scores followed the same gradient for both raters, with relative strength in Social Awareness, which his teacher rated within normal limits; these subscale scores are used here for intervention planning only. Cross-informant integration: The 12-point Total score difference exceeds the 10-point convention for seeking additional perspective, and additional input was obtained from L.T.'s after-school program coordinator by interview, which described social behavior closer to the mother's account in unstructured settings. The divergence is interpreted as setting information: the classroom's predictable structure and adult scaffolding appear to reduce the visible load of L.T.'s social difficulty, while less structured home and community settings expose it. Both raters' elevations independently exceed the level treated as a flag for full evaluation, and neither is treated as more correct. Differential and specificity statement: Elevated SRS-2 scores are consistent with, but not specific to, autism; attention problems, anxiety, language weaknesses, and disruptive behavior can each raise scores. Rating-scale and interview data in this evaluation showed no attentional or oppositional pattern of that kind, language screening was age-typical, and the anxiety symptoms described were situational and secondary to social demands, so these contributors do not account for the elevations observed. Integration with direct measures and history: The SRS-2 pattern converges with the ADOS-2 observation, reported separately, and with a developmental history of social-communication differences evident before age 3. Cross-informant rating-scale evidence, direct observation, and history therefore point in the same direction, and the SRS-2 findings strengthen, but do not by themselves establish, the diagnostic formulation presented in the summary of this report. Recommendations linkage: The cross-setting social-communication findings support the social-communication intervention goals and the structured-peer-interaction recommendation in the plan; the relative strength in awareness supports building on motivation-side supports rather than insight-first approaches. Re-rating with the SRS-2 School-Age Form by the same two raters in 12 months is recommended to monitor change on the same instrument, edition, and norms, aligned with the program's annual outcome-measure cadence. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, form, each rater's relationship and setting, and the norm reference are named, which makes the T scores interpretable now and comparable at re-rating. Severity language stays qualitative and consistent: every T score appears with its descriptor in words, and no number is treated as a diagnostic threshold. Informant divergence is quantified, interpreted as setting information, and answered with additional input, instead of being averaged away or adjudicated. The specificity statement is present and specific: elevations are called consistent with, but not specific to, autism, and the considered contributors are named with the evidence that addressed them. The summary keeps the SRS-2 subordinate to the whole evaluation, and every recommendation, including the dated re-rating plan on the same edition and norms, traces to a finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the SRS-2 section knowing which program will read it, because the instrument's status changes sharply by payer and jurisdiction. Under TRICARE's Autism Care Demonstration it is PAYER POLICY at its most explicit: the TRICARE Operations Manual requires the SRS-2 Parent Form at baseline and every year thereafter as an outcome measure alongside the Vineland-3 and the PDDBI, treats ABA-provider-completed measures as indirect services under 97151, and sets a documentation rule worth taping to the wall: submissions must include the full publisher print report or hand-scored protocol with summary score sheets, because "Imbedding T-scores within the treatment plan or other clinical documents is insufficient." Georgia Medicaid's pathway, as implemented in Peach State's clinical policy GA.CP.BH.504 (revised October 2025), requires a minimum of two assessment tools, one clinician-administered and one caregiver tool, and lists the SRS-2 among the accepted caregiver tools; the evaluation report must summarize each instrument with scores, dates, and the evaluator's signature and credentials. Most commercial policies run the other way: Aetna's clinical policy CPB 0648 names the ADI-R, ADOS-2, CARS-2, and Asperger Syndrome Diagnostic Scale as diagnostic tools and does not name the SRS-2, so an SRS-2-forward report can read as thin to a reviewer working from that list. In US schools, the LAW frame is IDEA's instrument-neutral rule that evaluators may "not use any single measure or assessment as the sole criterion" ( 34 CFR 300.304(b)(2) ). Testing and rating-scale work is billed inside the psychological and developmental testing code families by bare code number (for example 96110, 96112, 96127 for screening instruments; 96130, 96131, 96136 through 96139 for testing services), which is payer policy and coding convention, not law. In Canada, British Columbia's Autism Funding pathway states that for a private diagnosis "both the ADOS-2 and ADI-R are required instruments" ( MCFD form CF0904 ), so the SRS-2 supports but never qualifies a BC funding application, and the Canadian Paediatric Society's diagnostic standards treat rating scales as adjuncts, noting that findings from a diagnostic tool "cannot be used alone to diagnose ASD" while observing that, uniquely on its list, training is not required to use the SRS-2. In Australia, the National Guideline (2018, second edition 2023, NHMRC-approved recommendations) describes a comprehensive assessment process without mandating any specific instrument, the NDIS asks for diagnosis plus functional-capacity evidence, and the MBS names no instruments at all (a July 2026 search of MBS Online for the scale returns no items), while record rules come from the Psychology Board's code of conduct : reports true and objective, records retained seven years, or until age 25 for child clients. Edition currency became a live question in 2026. The SRS-3 published in 2026 with a new normative sample of 1,180 individuals and 2,340 ratings, a new School-Age Self-Report Form, three norm options for every form (female, male, or all-sample), the Restricted Interests and Repetitive Behavior treatment subscale split into Cognitive Flexibility and Sensory-Motor Behaviors for six treatment subscales in all, a new Inconsistent Responding Index to flag random response patterns, and Spanish forms available online only. During the transition the defensible practices are specific: name the edition and norm reference in every report; finish an evaluation on the edition it started on; keep a monitoring sequence on one edition and norms, since a re-rating on new norms is not comparable to a baseline on old ones and any cross-edition comparison belongs in the text as a stated limitation; and check payer language before switching, because while TRICARE's manual entry already reads "SRS-2 (or Current Edition)", contractor submission systems and state tool lists such as Georgia's still name the SRS-2, and independent peer review of the SRS-3 will lag its launch. The qualification boundary is the publisher's: both editions are qualification level C, and WPS states that it "only qualifies individuals for test purchase, not for test use", which leaves use governed by licensing boards and scope of practice; forms are completed by laypeople, but interpretation and write-up are professional acts. One more tension belongs in the writer's head rather than on the page: the publisher's marketing describes the SRS-2 as offering "the convenience of a screener and the power of a diagnostic tool", while the peer-reviewed record documents weak discrimination from other clinical conditions, and a defensible report writes to the evidence, reporting elevations dimensionally and letting the full evaluation carry the diagnosis. SRS, SRS-2, and SRS-3 are trademarks of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common SRS-2 write-up errors reviewers flag There is no payer audit series for rating-scale write-ups; the accountability record here is psychometric, plus one payer paper trail that is unusually explicit. The peer-reviewed findings that matter: the SRS separates autism from typical development almost perfectly but discriminates far less well from disruptive behavior disorders , it overestimated autism in boys with selective mutism , the Adult Self-Report showed poor discriminant validity against anxiety , and unadjusted scores can index "general levels of impairment" rather than autism-specific severity. On the payer side, TRICARE's ACD submission rules reject write-ups whose scores exist only in the narrative. Those findings predict almost every failure below. The BastionGPT Clinical Advisory Board sees the same errors most often in SRS-2 write-up reviews: An elevation written as a diagnosis. "Total score in the severe range" becomes "results confirm autism spectrum disorder." The SRS-2 quantifies a trait dimension; the Canadian Paediatric Society's standards state that findings from a diagnostic assessment tool "cannot be used alone to diagnose ASD", and the diagnostic conclusion belongs to the whole evaluation. The specificity sentence missing. The section reports elevations without stating that they are consistent with, but not specific to, autism, and without naming the contributors considered. Given the published overlap with ADHD, anxiety, disruptive behavior, and language problems, a reviewer reads its absence as an interpretation gap, not a style choice. Informant disagreement averaged or adjudicated. The report splits the difference between a parent-severe and teacher-moderate profile, or declares one rater right. The documented conventions run the other way: an elevation from either informant flags the need for full evaluation, a difference of more than 10 T points calls for additional input, and divergence is setting information to interpret. Treatment subscales read as a diagnostic profile. The five treatment subscales drive the interpretation, or subscale scatter gets clinical meaning of its own. The publisher states they are not used for screening or diagnosis, and the factor-analytic support concentrates in the Total score and the two DSM-5 compatible subscales. Edition, form, and norms unnamed, or scores that live only in the narrative. "An SRS was administered" names no edition in the first year the SRS-3 and its new norms are on the market, no form, and no rater-referenced norm context, leaving the T scores uninterpretable; and for TRICARE ACD submissions the manual requires the full publisher print report or hand-scored protocol with summary score sheets, stating that "Imbedding T-scores within the treatment plan or other clinical documents is insufficient." How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (raters, forms, T scores, severity descriptors) and get a drafted SRS-2 results-section narrative with the cross-informant integration and specificity statement scaffolded in place for your review. Cross-check a finished draft for the gaps reviewers flag: scores that disagree with their descriptor language, a missing differential sentence, informant divergence averaged away, and an unnamed edition or norm reference. Translate the results section into a plain-language summary for parents, teachers, and care teams that keeps severity descriptors framed as observed elevation, not a diagnosis. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do SRS-2 scores actually mean? Every SRS-2 form yields T scores, standard scores with a mean of 50 and a standard deviation of 10, for a Total score, two DSM-5 compatible subscales, and five treatment subscales, referenced to norms separated by rater type and by the age and gender of the person rated . Higher scores mean the rater observes more autism-related social difficulty. The severity descriptors (within normal limits, mild, moderate, severe) are qualitative characterizations of normative elevation, subject to measurement error and rater perspective. Write them as descriptors in words, pair each with its plain-language meaning, and let any diagnosis come from the whole evaluation, never from a score. Should I report SRS-2 or SRS-3 results in 2026? Both are current as of July 2026. The SRS-3 published in 2026 with a new normative sample (1,180 individuals, 2,340 ratings), a new School-Age Self-Report Form, three norm options for every form (female, male, or all-sample), the Restricted Interests and Repetitive Behavior treatment subscale split into Cognitive Flexibility and Sensory-Motor Behaviors, a new Inconsistent Responding Index, and Spanish forms online only, while the SRS-2 remains on sale and named in payer documents. A report on either edition is defensible if it names the edition and norm reference, finishes on the edition it started on, and treats any cross-edition score comparison as a stated limitation. Before a program-wide switch, check payer language: TRICARE's manual entry reads "SRS-2 (or Current Edition)", but contractor systems and state tool lists still name the SRS-2, and independent peer review of the SRS-3 will lag its launch. Is the SRS-2 required for an autism diagnosis or for services? No law or diagnostic standard names it, and its status varies by program. Where it is required, it is payer policy: the TRICARE Autism Care Demonstration requires the SRS-2 Parent Form at baseline and every year as an outcome measure for ABA services. Where it is accepted, it is one option: Georgia Medicaid's pathway ( GA.CP.BH.504 ) lists it among caregiver tools in a required two-tool minimum. Where it is unnamed, other instruments are: Aetna's policy names the ADI-R, ADOS-2, and CARS-2, and British Columbia's Autism Funding pathway requires the ADOS-2 and ADI-R for a private diagnosis. In schools, IDEA's rule is instrument-neutral: no single measure may be the sole criterion ( 34 CFR 300.304(b)(2) ), and Australia's National Guideline mandates no instrument at all. The honest report sentence: the SRS-2 is never sufficient, sometimes required as an outcome measure, and often not named at all. Parent and teacher SRS-2 scores disagree. Which one is right? Neither, and the write-up should not choose. Agreement between different informant types is modest across instruments, averaging r = .28 in a meta-analysis of 341 studies , and parents often rate higher than teachers. The documented SRS-2 conventions, summarized by the Texas Statewide Leadership for Autism Training : treat an elevation from either informant as a flag for full evaluation, and obtain additional rater input when scores differ by more than 10 T points. Divergence usually carries setting information, structure and scaffolding in one environment, demands and exposure in another, and the defensible paragraph reports both scores, interprets the difference, and refuses to average it. Can a high SRS-2 score alone establish autism? No, and this is the instrument's best-documented limitation. The SRS separates autism from typical development almost perfectly, but in matched clinical comparisons discrimination fell substantially against disruptive behavior disorders , it overestimated autism in boys with selective mutism , and the Adult Self-Report showed poor discriminant validity against anxiety . Influential work concluded that without accounting for behavior problems, age, and language, scores can index "general levels of impairment" rather than autism-specific severity. The Canadian Paediatric Society's standards say it directly: findings from a diagnostic assessment tool "cannot be used alone to diagnose ASD". Write elevations as consistent with, but not specific to, autism, name the differential contributors you considered, and let observation, interview, and history carry the conclusion. How does the SRS-2 differ from the ADOS-2, CARS-2, ASRS, or SCQ? By method and job. The SRS-2 is an informant and self-report rating scale: graded-frequency items, lifespan forms, and a severity-quantifying T score, which makes it the dimensional, cross-setting layer. The ADOS-2 is a clinician-administered, semi-structured direct observation, the diagnostic anchor for most pathways. The CARS-2 is a clinician rating completed from observation and information, the ASRS is a DSM-aligned multi-informant rating system, and the SCQ is a brief yes-or-no screener. The battery convention in autism evaluations pairs a dimensional rating scale (SRS-2) with direct observation (ADOS-2), a developmental interview (ADI-R), and cognitive, language, and adaptive measures, and no instrument on that list substitutes for the comprehensive evaluation or for clinical judgment. Where can I find the SRS-2 questions or a free scoring key? Not on this page, and not legitimately on any public page. The 65 items, scoring keys, and norm tables are protected test materials: the SRS-2 is qualification level C, WPS states that it "only qualifies individuals for test purchase, not for test use", and psychologists carry an ethical duty to maintain test security under APA Ethics Standard 9.11. Consumer sites offering SRS-2 self-scoring undermine exactly the controls that make the norms meaningful. What a report can include: T scores, severity descriptors in words, and your own prose describing what the scales measure. What it cannot: item text, scoring rules, or reproduced norm tables. Can adults complete the SRS-2 about themselves? Yes. The Adult Self-Report form covers ages 19 and up and has published validation in a sample of 237 autistic adults , which matters in late-identification evaluations where childhood informants are scarce. Interpret it with two cautions. First, self-reported elevations overlap heavily with anxiety: the Adult Self-Report showed poor discriminant validity between autistic and highly anxious adults (the figure usually quoted from that study is a sensitivity value from that clinical comparison, not a specificity in anxious adults). Second, current measures and procedures may miss some autistic adults, particularly women , whose observable presentation can understate their internal experience. Pair the self-report with the Adult relative or other-informant form, developmental history, and direct assessment, and say in the write-up how the streams converged. Can BastionGPT help with SRS-2 write-ups? Yes. BastionGPT is trained and clinically tested on psychological and psychoeducational evaluation reports, the parent documents SRS-2 sections live inside. Paste a score summary (raters, forms, T scores, descriptors) and it drafts the results-section narrative with the cross-informant integration and the specificity statement in place for your review; it can also check a finished section for score-descriptor mismatches, a missing differential sentence, or an unnamed edition, and produce a plain-language summary for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: WPS, SRS-2 product page : authorship, 2012 publication, the four 65-item forms and age ranges, 15 to 20 minute completion, qualification level C, score architecture, norms (1,906 individuals, 4,709 ratings, by rater type and by age and gender of the person rated), scoring platforms, Spanish item-translation forms, and the purchaser-versus-user qualification statement. WPS, SRS-3 product page : 2026 publication, new norms (1,180 individuals, 2,340 ratings), the School-Age Self-Report Form, three norm options per form, the Cognitive Flexibility and Sensory-Motor Behaviors subscale split, the Inconsistent Responding Index, T-score reporting, and online-only Spanish forms. Pearson Clinical Assessment AU&NZ, SRS-2 distributor page : the "SRS-3 Coming Soon in 2026" banner and the distributes-but-does-not-publish note. TRICARE Operations Manual 6010.59-M, Chapter 18, Section 4 (Comprehensive Autism Care Demonstration) , as of July 1, 2026: the SRS-2 Parent Form required at baseline and annually, the "(or Current Edition)" entry, outcome measures as indirect services under 97151, and the full-score-report submission rule. Peach State Health Plan, clinical policy GA.CP.BH.504 (revised October 2025) , implementing Georgia Medicaid ASD services policy: the two-tool minimum and the caregiver-tool list naming the SRS-2. Aetna, clinical policy CPB 0648 : the named diagnostic tools (ADI-R, ADOS-2, CARS-2, Asperger Syndrome Diagnostic Scale). British Columbia Ministry of Children and Family Development, form CF0904, Autism Funding non-BCAAN (private) diagnosis : the ADOS-2 and ADI-R requirement for funding eligibility. Canadian Paediatric Society, Standards of diagnostic assessment for autism spectrum disorder (Brian, Zwaigenbaum, and Ip; Paediatr Child Health 2019;24(7):444-451): the adjunct framing, the tools table with the SRS-2 row, and the training-requirement footnote. Autism CRC, National Guideline for the assessment and diagnosis of autism in Australia (2018; second edition 2023; NHMRC-approved recommendations): a comprehensive assessment process with no mandated instrument; Medicare Benefits Schedule online search (July 2026) returning no instrument-named items. Psychology Board of Australia, Code of Conduct : record accuracy and retention rules for Australian psychologists. 34 CFR 300.304 : the no-single-measure rule for special-education evaluations. Buros Center for Testing, Tests reviewed in the Nineteenth Mental Measurements Yearbook , and Bruni, 2014, Journal of Psychoeducational Assessment, test review of the SRS-2 ( ERIC record ): independent review status and the Constantino and Gruber authorship attribution. Texas Statewide Leadership for Autism Training, SRS-2 evaluation summary : the T-score metric and the either-informant and 10-T-point informant-discrepancy conventions. Specificity literature: Cholemkery et al., 2014, European Child & Adolescent Psychiatry, ASD versus disruptive behavior disorders ; Cholemkery et al., 2014, Journal of Autism and Developmental Disorders, ASD versus social phobia and selective mutism ; South et al., 2017, Autism Research, the Adult Self-Report in anxious and autistic adults ; Hus et al., 2013, Journal of Child Psychology and Psychiatry, factors influencing SRS scores , with the Constantino and Frazier reply . Validation and structure literature: Frazier et al., 2014, Autism, DSM-5 factor structure (N = 9,635) ; Chan et al., 2017, Autism Research, Adult Self-Report validation (N = 237) ; Sturm et al., 2017, item response theory analysis and 16-item short form ; De Los Reyes et al., 2015, Psychological Bulletin, cross-informant meta-analysis (341 studies) ; Ratto et al., 2018, sex differences in adaptive and diagnostic outcomes . ## Related templates ADOS-2 Report Write-Up BASC-3 Report Write-Up Psychological Evaluation Report Psychiatric Diagnostic Evaluation NEPSY-II Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up MIGDAS-2 Report Write-Up Autism Spectrum Evaluation Report Sensory Profile 2 Report Write-Up GARS-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/vineland-3-report All note templates # Vineland-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The Vineland-3 (Vineland Adaptive Behavior Scales, Third Edition) is a norm-referenced measure of adaptive behavior: the communication, daily living, and social skills a person uses to function independently, from birth through adulthood, gathered by interview or by caregiver and teacher rating forms. Psychologists and developmental teams use it in autism, intellectual-disability, and disability evaluations. This page covers how to write up Vineland-3 results, with a fictional sample and a template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and developmental clinicians; publisher qualification level B Audience IEP and eligibility teams, physicians and therapists, disability programs (SSA, provincial services, NDIS), parents and caregivers Typical length 300 to 700 words for the results section · administration about 10 to 50 minutes by form Format family Norm-referenced adaptive behavior scale When it's used Autism and intellectual-disability evaluations, developmental assessments, special-education eligibility, SSA and NDIS disability determinations, progress monitoring Standards context Published by NCS Pearson (2016); current edition as of July 2026, no Vineland-4 announced; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Vineland-3? The Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) is a norm-referenced measure of adaptive behavior: the everyday communication, daily living, and social skills a person actually uses to get through life independently. Authored by Sara S. Sparrow, Domenic V. Cicchetti, and Celine A. Saulnier and published by NCS Pearson in 2016, it gathers information through a semi-structured Interview Form, a Parent/Caregiver Rating Form, or a Teacher Rating Form, each offered in a Comprehensive and a shorter Domain-Level version, and it spans birth through age 90 and beyond (the Teacher Form covers ages 3 to 21). Edition status matters here, and it runs opposite to most tests: as of July 2026 the Vineland-3 is the current edition, and Pearson lists no Vineland-4 as announced, released, or coming soon on its US , Canadian, or Australian sites, so a report can name the edition without the churn that shadows the Wechsler and BASC families. The load-bearing distinction for write-ups is not what the Vineland measures but how it scores. Two different metrics run through every report. The Adaptive Behavior Composite and the three domains (Communication, Daily Living Skills, Socialization) are standard scores with a mean of 100 and a standard deviation of 15, while the nine subdomains beneath them are v-scale scores with a mean of 15 and a standard deviation of 3. A subdomain v-scale score of 12 is low-average; a domain standard score of 12 would signal profound impairment; the two numbers look alike and mean nothing alike, and confusing them is the most common Vineland write-up error. A second boundary matters too: adaptive behavior is what a person typically does, which is different from cognitive ability, what a person can do on a test, and different again from the problem-behavior focus of a scale like the BASC-3 . When a decision turns on daily-living skills, the ABAS-3 is the nearest adaptive-behavior alternative, and the Vineland's semi-structured Interview Form is its distinguishing feature. Keeping the two score metrics and the adaptive-versus-cognitive boundary straight is most of what separates a defensible Vineland-3 write-up from a score printout with sentences around it. ## Who uses Vineland-3 write-ups and when Psychologists and school psychologists write up Vineland-3 results inside autism evaluations, where adaptive behavior supplies the everyday-functioning stream alongside direct observation such as the ADOS-2 and separate cognitive testing; developmental and disability teams use it in intellectual-disability determinations, where adaptive scores pair with an individually administered cognitive measure such as the WISC-V , since current diagnostic standards weigh adaptive functioning heavily. It also enters a psychological evaluation report or a neuropsychological report as the adaptive layer whenever a decision turns on real-world skills rather than test-room ability, and it supports disability determinations for programs such as SSA, provincial developmental services, and the NDIS. It is the right tool when the question is how well a person manages the demands of daily life, and it spans birth through adulthood. Where the question narrows, other instruments carry the load: the ABAS-3 is a rating-only adaptive measure some evaluators prefer for speed, the DABS is built specifically around the intellectual-disability adaptive cutoff, and the SIB-R pairs adaptive behavior with a structured problem-behavior measure. The Vineland-3 write-up is where the adaptive evidence gets anchored: named form, named informant, one integrated account. ## How to structure a Vineland-3 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names the form and informant before interpreting anything, keeps the two Vineland-3 score metrics visibly separate, and treats the one move that trips up most write-ups, reading a subdomain v-scale score as if it were a domain standard score, as an explicit labeling discipline rather than an accident. Each section carries the pitfall that most often undermines it. Measures, form, informant, and method. Name the instrument and edition, which form was used (Interview, Parent/Caregiver Rating, or Teacher Rating), who the informant was and their relationship to the person, the date, the administration method (in person, or telepractice through Q-global on-screen administration), and the norm set. Pitfall: not naming the form or informant. A Vineland score means nothing without the source behind it, and remote Interview administrations carry a documented caveat, because Pearson states the Interview Form was not standardized in a telepractice format. Informant context and validity of the ratings. Describe the informant's opportunity to observe, the setting the ratings reflect, any language or reliability considerations, and then state plainly whether you consider the results valid estimates of current everyday functioning. Pitfall: presenting a rating as objective fact. A Vineland-3 profile records what one informant reports seeing in one set of settings; a teacher's structured-classroom window and a parent's home-and-community view are different data, not competing truths. Score summary, with the metrics labeled. Report the Adaptive Behavior Composite and the three domain scores (Communication, Daily Living Skills, Socialization) as standard scores with a mean of 100 and a standard deviation of 15, each with its confidence interval and percentile; report the subdomains as v-scale scores with a mean of 15 and a standard deviation of 3, labeled as a separate metric. Pitfall: the two-metric trap. A subdomain v-scale of 12 and a domain standard score of 12 are worlds apart, the first is low-average, the second would be profound impairment, so a table that places the two numbers in the same column without labeling the metric is the single most common Vineland error. Domain-by-domain results in prose. Move through Communication, Daily Living Skills, and Socialization, and spend the prose on what each result means for this person's day-to-day functioning, not on re-reading the table aloud. Report the optional Motor Skills domain only if the person is age 9 or younger, since standard scores are not produced above that age, and report the optional Maladaptive Behavior index only if that section was administered. Pitfall: the score dump. A paragraph per subdomain in administration order documents that testing happened; it does not answer the referral question. Composite and interpretive summary. Lead with the Adaptive Behavior Composite as the best-supported summary score, and when you describe strengths and weaknesses across domains, lean on statistical significance and the base rate of a difference, the frequency of that gap in the norm sample, rather than eyeballing scatter. Pitfall: over-reading the domain profile. Published structural-validity work finds the composite interpretable but cautions that the three domain scores may not carry enough unique reliable variance to interpret in isolation, so a story built on small domain differences can outrun the evidence. Informant reconciliation, when there is more than one rater. Where parent and teacher, or two caregivers, disagree, attribute the difference to setting and observation opportunity rather than declaring one informant wrong, and do not average scores across raters. Pitfall: making the discrepancy disappear. Splitting the difference or crowning one rater correct throws away the setting information the disagreement actually contains; a multi-rater comparison is designed to hold both. Diagnostic or eligibility linkage. Where adaptive data feed an intellectual-disability question, pair the Vineland with an individually administered cognitive measure and align any severity language with DSM-5-TR, which sets intellectual-disability severity by adaptive functioning rather than IQ, then state what the adaptive data do and do not establish. Pitfall: an eligibility or severity verdict from one Vineland number. Federal special-education law requires that no single measure be the sole criterion ( 34 CFR 300.304(b)(2) ), and the results section supplies calibrated evidence, not the ruling. Recommendations linkage. Tie each recommended support to a specific adaptive finding, and where you expect to re-measure, say so and name the mechanism. Pitfall: recommendations that could follow any profile. If a daily-living-skills support does not trace back to the Daily Living Skills result, either the recommendation or the results section is incomplete; growth-scale values, not standard scores, are the metric built to track change over re-administrations. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: Vineland-3 Form: [Interview / Parent-Caregiver / Teacher] Informant: [name / relationship] Norms: [US census-based] Administration: [in person / telepractice via Q-global OSA] Date: INFORMANT CONTEXT AND VALIDITY [Opportunity to observe, setting, language, reliability; statement that results are valid estimates of current everyday functioning, or the stated limitation if not] SCORE SUMMARY (table, metrics labeled) [Adaptive Behavior Composite + Communication, Daily Living, Socialization = STANDARD scores (M 100, SD 15) with 95% CI and percentile; subdomains = V-SCALE scores (M 15, SD 3), labeled as a separate metric; optional Motor (birth-9 only) and Maladaptive index only if administered] DOMAIN-BY-DOMAIN RESULTS (prose, functional meaning) Communication / Daily Living Skills / Socialization: [what each means for this person's everyday functioning] COMPOSITE AND INTERPRETIVE SUMMARY [ABC in context; significance AND base rates for any scatter; domain-score interpretability caution acknowledged] INFORMANT RECONCILIATION (if more than one rater) [Discrepancies attributed to setting/observation, not averaged] DIAGNOSTIC / ELIGIBILITY LINKAGE [Adaptive paired with cognitive for ID; DSM-5-TR severity keyed to adaptive functioning; single-measure limit acknowledged] RECOMMENDATIONS LINKAGE [Each support tied to a finding; re-administration via GSVs] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample Vineland-3 write-up (fictional) Scenario: an 8-year-old is referred for an autism evaluation. Alongside direct observation and cognitive testing reported elsewhere, the Vineland-3 Parent/Caregiver Rating Form was completed by his mother to document adaptive functioning. The profile shows the shape often seen in autism: a relative Socialization weakness with Daily Living Skills as a relative strength, and overall adaptive scores below what average-range cognitive testing alone would predict. This is the adaptive-behavior results section only, condensed but structurally complete. All details are fictional. Client: M.K., 8 (grade 3) · Referral: autism evaluation, adaptive functioning · Informant: mother (Parent/Caregiver Rating Form) · Evaluator: R. Alvarez, PhD, Licensed Psychologist · Administration date: 07/10/2026 · Report date: 07/16/2026 Measures and administration: Adaptive functioning was assessed with the Vineland Adaptive Behavior Scales, Third Edition (Vineland-3), Parent/Caregiver Rating Form, completed by M.K.'s mother and scored against the United States normative sample. The Adaptive Behavior Composite and the three domains (Communication, Daily Living Skills, Socialization) are standard scores with a mean of 100 and a standard deviation of 15, each reported with its 95% confidence interval (CI) and percentile rank. The subdomains are reported as v-scale scores with a mean of 15 and a standard deviation of 3; these are a different metric from the domain standard scores and are not directly comparable to them. The optional Maladaptive Behavior section was also completed. Informant context and validity: M.K.'s mother is his primary caregiver and sees him across home, family, and community settings daily, giving her broad opportunity to observe the everyday behaviors the Vineland-3 samples. Her responses were consistent, and there were no indications of a response set that would undermine the ratings. The results below are considered valid estimates of M.K.'s current adaptive functioning as observed by his mother; ratings reflect one informant's view and are interpreted alongside the teacher input and direct observation reported elsewhere in this evaluation. Domains and composite (standard scores, mean 100, SD 15) Domain Standard score 95% CI Percentile Communication 84 79-89 14 Daily Living Skills 93 88-98 32 Socialization 72 67-78 3 Adaptive Behavior Composite 79 75-83 8 Subdomains (v-scale scores, mean 15, SD 3) Domain Subdomain v-scale score Communication Receptive 12 Communication Expressive 11 Communication Written 13 Daily Living Skills Personal 14 Daily Living Skills Domestic 15 Daily Living Skills Community 13 Socialization Interpersonal Relationships 9 Socialization Play and Leisure Time 9 Socialization Coping Skills 11 Domain results: M.K.'s Daily Living Skills are a relative strength (standard score 93, 32nd percentile, average range): his mother reports age-appropriate self-care and household participation, with community skills slightly behind personal and domestic skills. His Communication is lower (84, 14th percentile), with receptive, expressive, and written subdomains clustered in the low-average range and no dramatic split among them. His Socialization is the clear relative weakness (72, 3rd percentile): the Interpersonal Relationships and Play and Leisure Time subdomains, each a v-scale score of 9, fall two standard deviations below the subdomain mean of 15, while Coping Skills at 11 is somewhat stronger. Reading these v-scale scores against the subdomain mean of 15 is essential: a 9 here is a marked weakness, not the near-total impairment that a standard score of 9 would signal, and the two metrics are never placed side by side as if comparable. Composite and interpretation: M.K.'s Adaptive Behavior Composite of 79 (CI 75-83, 8th percentile) summarizes adaptive functioning that is below age expectations overall, driven by the Socialization weakness and pulled up by relatively intact daily living skills. Two points frame the interpretation. First, the domain differences here are large enough to describe, but domain-level profiles are interpreted cautiously: published structural-validity work supports the composite while questioning how much unique reliable information the separate domain scores carry, so the write-up leans on the composite and the sizable, base-rate-uncommon Socialization gap rather than on small subdomain movements. Second, cognitive testing reported elsewhere in this evaluation fell in the average range, which alone would predict stronger adaptive functioning than these ratings show. That gap between measured ability and everyday adaptive skill is well documented in autism and tends to be wider, not narrower, in children without intellectual disability; it does not indicate invalid ratings, and being cognitively able does not mean being adaptively able. Maladaptive behavior: On the optional Maladaptive Behavior section, the Internalizing subscale was elevated, consistent with the mother's report of anxiety in unfamiliar social settings; the Critical Items were reviewed at the item level, as the publisher intends, rather than summed into a scaled score, and none indicated an acute safety concern. Summary: The Vineland-3 documents below-average overall adaptive functioning (ABC 79) with a relative Socialization weakness, a relative Daily Living Skills strength, and an adaptive profile below what cognitive scores alone would predict, a pattern consistent with the referral question. These adaptive findings are one stream of evidence: an autism determination integrates them with direct observation, developmental history, and cognitive results, and no single measure serves as the sole basis for the diagnosis. All scores carry measurement error and describe current functioning as reported by one informant. Linkage to recommendations: The Socialization finding supports recommendations 1 and 2 (structured social-skills instruction and supported peer interaction, with the Interpersonal Relationships and Play and Leisure Time subdomains named as targets). The elevated Internalizing finding supports recommendation 3 (screening for anxiety and a graded approach to unfamiliar settings). Because growth-scale values, not standard scores, track change, recommendation 4 sets a re-administration at the next annual review to measure progress against this baseline. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The two metrics stay separate and labeled: domain and composite standard scores (mean 100, SD 15) and subdomain v-scale scores (mean 15, SD 3) are never placed side by side as if comparable, and the v-scale of 9 is read against its own mean of 15. Every number matches its narrative: each score appears with its interpretation, and no sentence upgrades or downgrades a result along the way. The edition, form, informant, and norm set are named, which makes the scores interpretable now and comparable at re-evaluation. Domain differences are interpreted with base rates and the published domain-interpretability caution in mind, so the story leans on the composite and a large, uncommon gap rather than small subdomain scatter. The adaptive-below-cognitive gap is explained rather than treated as an error, and the ratings are framed as one informant's view within a broader evaluation. The summary states what the scores do not establish (a diagnosis), and every recommendation traces to a specific finding, with a re-administration keyed to growth-scale values. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decisions it will feed, and label each requirement by its strength. In US special education, adaptive-behavior assessment is genuinely required for an intellectual-disability identification: the federal definition pairs subaverage intellectual functioning with concurrent deficits in adaptive behavior ( 34 CFR 300.8(c)(6) ), and no single measure may be the sole criterion ( 34 CFR 300.304(b)(2) ), so a Vineland is commonly part of that evidence. That is LAW; the specific cutoffs and the number of deficit areas required are set by each state, so the thresholds vary and the write-up should not assert a single national number. In clinical diagnosis, DSM-5-TR (American Psychiatric Association, 2022) sets intellectual-disability severity by adaptive functioning rather than IQ, which is the current CONVENTION and the reason the adaptive results carry real weight. For adult disability, the Social Security Administration's Listing 12.05 and the childhood Listing 112.05 require both subaverage intellectual functioning and significant adaptive deficits with onset before age 22, but here a widespread belief is simply wrong: SSA states it does not require a standardized adaptive-behavior test and will consider one only if the record already contains it, which is POLICY worth knowing before you tell a family a Vineland is mandatory. In capital cases, the Atkins line of decisions, most recently Moore v. Texas (2017), directs courts to follow current medical standards and focus the adaptive inquiry on deficits rather than offsetting them with unconnected strengths. Payer coverage is a separate layer: psychological and neuropsychological testing services are billed under the 96130, 96131, and 96136 through 96139 code families as payer policy, and a Vineland is usually one instrument inside a larger, medically necessary battery rather than a standalone service. Edition currency and access rules round out the defensibility picture. As of July 2026 the Vineland-3 is the current edition and no fourth edition is announced, but this is a fact with a shelf life, so re-verify it before a high-stakes report; publishers do not pre-announce far ahead. The norms are United States census-based, which is a stated limitation when the instrument is used in Canada or Australia, where it is widely accepted but not separately normed (a French-Canadian edition is distributed for francophone evaluators, and Canadian developmental-services and Australian NDIS processes accept adaptive evidence without naming the Vineland specifically, which is convention, not statute). Two mechanical limits belong in any write-up that touches them: the optional Motor Skills domain produces standard scores only through age 9, so it cannot be reported for an older examinee, and the Interview Form was not standardized in a telepractice format, so a remote Interview administration carries a documented interpretive caveat. Access is publisher-controlled: the Vineland-3 is a qualification level B instrument, purchase and administration are restricted accordingly, and Pearson's generated reports carry their own handling notices, so route any records request that touches raw protocols through your test-security obligations rather than reflexively printing the packet. Your integration and conclusions belong in the ordinary record and should be written to be read there. Vineland and Pearson are trademarks, in the United States and other countries, of Pearson Education, Inc., or its affiliates; the Vineland-3 is copyright NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Vineland-3 write-up errors reviewers flag There is no payer audit series for adaptive-behavior write-ups; the accountability literature here is psychometric, and it is sharper than most template pages admit. The Vineland-3 was reviewed in the Buros Center's 21st Mental Measurements Yearbook (a separate 2018 Canadian Journal of School Psychology review by Pepperdine and McCrimmon is sometimes miscited as the Buros entry). Independent structural-validity work by Pandolfi and Magyar supports interpreting the Adaptive Behavior Composite but finds the three domain scores add little the composite does not already capture, and Farmer, Floyd, and McNicholas found the Interview Form's dimensional status genuinely unresolved. Those findings predict most write-up failures. The BastionGPT Clinical Advisory Board sees the same errors most often in Vineland-3 write-up reviews: The two-metric trap. A subdomain v-scale score (mean 15, SD 3) is read against the domain and composite metric (mean 100, SD 15), or the two are placed in one column as if comparable. A v-scale of 12 is low-average, not the near-total impairment a standard score of 12 would signal; every score needs its metric named. Domain profiles over-read. A story is built on small differences among Communication, Daily Living Skills, and Socialization without checking base rates, the frequency of that gap in the norm sample. The published caution is that the domain scores may not carry enough unique reliable variance to interpret in isolation, so the composite and large, uncommon gaps carry the interpretation, not subdomain scatter. Informant disagreement averaged away. A parent-high and teacher-average profile gets split down the middle, or one rater is crowned correct. Discrepancies are expected and usually carry setting information: a teacher's structured-classroom window and a parent's home-and-community view are different data, and the strong move is a sentence that says what the difference means rather than arithmetic that erases it. Form and cutoff myths. The write-up implies the semi-structured Interview Form is required, or that a single Vineland composite number establishes intellectual disability. No statute, SSA rule, or IDEA regulation names a specific Vineland form or a single composite cutoff; the Parent/Caregiver and Teacher forms yield valid norm-referenced scores, and severity is a clinical judgment anchored to adaptive functioning, not one number. Edition, norms, or scope unnamed. "The Vineland" with no edition, no form, and no norm set, a Motor Skills score reported for a child older than 9 (where standard scores are not produced), or an adaptive profile that is never reconciled with the cognitive results it sits beside in an autism evaluation, where adaptive scores commonly fall below what ability alone would predict. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (composite, domains, subdomains, informant) and get a drafted results-section narrative organized domain by domain, with the standard-score and v-scale metrics kept distinct and validity and informant context in place for your review. Cross-check a finished draft for the gaps reviewers flag: a v-scale score read against the wrong mean, numbers that disagree with their interpretation, domain scatter over-read past its base rate, and an unnamed form or norm set. Translate the results section into a plain-language summary for parents and caregivers that explains adaptive functioning in everyday terms without confusing the two score metrics. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Vineland-3 scores actually mean? Two score metrics do the work, and they are not comparable. The Adaptive Behavior Composite and the three domains (Communication, Daily Living Skills, Socialization) are standard scores with a mean of 100 and a standard deviation of 15, each reported with a confidence interval and a percentile. The nine subdomains beneath them are v-scale scores with a mean of 15 and a standard deviation of 3. Because the metrics differ, a subdomain score of 12 is low-average while a domain score of 12 would be profound impairment, so every number should carry its metric label, and a v-scale score should never sit in the same column as a standard score. Growth-scale values and age equivalents are also produced, but they are built for tracking change over time, not for diagnostic interpretation. Is the Vineland-3 still current, or is a Vineland-4 coming? Keep testing on the Vineland-3. As of July 2026 it is the current edition, and no Vineland-4 is announced, released, or listed coming soon on Pearson's US , Canadian, or Australian sites, the reverse of the WISC, WAIS, and BASC families, which all carry next-edition signals. Because publishers do not pre-announce far ahead, treat this as a fact to re-verify before a high-stakes report rather than a permanent one. The practical rule is unchanged: name the edition and the norm set in the report, and if a Vineland-4 ever enters the picture, treat any cross-edition comparison as a stated limitation. Do I have to use the Interview Form? No. A common belief holds that the semi-structured Interview Form is required for diagnosis or eligibility, but no statute, SSA rule, IDEA regulation, or diagnostic manual names a specific Vineland form. The Parent/Caregiver Rating Form and the Teacher Rating Form yield valid, norm-referenced scores. The Interview Form adds clinician-guided probing and scoring control that some evaluators prefer for forensic or disputed cases, and published factor analyses show the administration modes are not perfectly interchangeable, so the defensible practice is to choose the form based on informant availability and the referral question, and to disclose which form you used and why. Is a Vineland required for an intellectual-disability diagnosis, SSA, or an IEP? It depends on the decision, and the answer is more nuanced than most pages say. For US special-education intellectual-disability eligibility, adaptive-behavior assessment is federally required ( 34 CFR 300.8(c)(6) ), but the specific instrument and cutoffs are set by states, and no single measure may be the sole criterion. For SSA disability, the agency requires documented adaptive deficits but states that it does not require a standardized adaptive test and will consider one only if the record already contains it ( Listing 12.05 ). For a clinical diagnosis, DSM-5-TR requires clinically significant adaptive deficits and sets severity by adaptive functioning, without fixing a Vineland number. So a Vineland is often the right evidence, but it is rarely, strictly speaking, mandatory. Vineland-3 or ABAS-3? Both are norm-referenced adaptive-behavior measures and they correlate, but they are not interchangeable. The ABAS-3 is rating-only and organized around the conceptual, social, and practical skill areas, which some evaluators find faster; the Vineland-3 adds the semi-structured Interview Form and is often preferred where floor effects matter in more severely affected people. Studies find the two produce different absolute scores in the same individuals, so they should not be swapped within one person's longitudinal file. Choose one instrument and stay with it across re-evaluations, and if a prior report used the other, note the change as a limitation rather than comparing the numbers directly. How low does a score have to be to count as a deficit? There is no single mandated cutoff. DSM-5-TR requires clinically significant deficits in adaptive functioning and sets intellectual-disability severity by that functioning rather than by an IQ number; the deficit threshold is conventionally interpreted as roughly two standard deviations below the mean in at least one domain or the composite, applied with clinical judgment rather than as a bright line. For special-education eligibility the operative thresholds are set by each state, so they vary. The write-up should report the scores with their confidence intervals, describe the functional deficits concretely, and let the diagnostic or eligibility team apply its own criteria rather than announcing the determination from one number. Can I see the actual Vineland-3 questions, or include them in a report? No. Test items, record forms, and scoring materials are protected: access is restricted to qualified professionals (the Vineland-3 is Pearson qualification level B), and psychologists are ethically obligated to protect test security under their ethics code and publisher agreements. Item exposure also damages the instrument, because the norms assume the tasks have not been rehearsed. Reports should carry scores, descriptors, and interpretation, never item content, and records requests that touch raw protocols route through your test-security obligations rather than reflexive printing. What a family can get is more useful anyway: the scores with a plain-language explanation of what each domain means for daily life and what happens next. How often should I re-administer the Vineland-3 for a review? There is no single mandated interval; payers and programs set their own, and applied-behavior-analysis payers in the US commonly expect adaptive re-measurement roughly every six months, while disability-program reviews vary. When you do re-administer, compare growth-scale values across administrations rather than standard scores, because growth-scale values are the metric built to track change, while standard scores move with age norms and can mask or exaggerate real progress. One scope limit belongs in any re-evaluation: the optional Motor Skills domain yields standard scores only through age 9, so it drops out for older examinees. Can BastionGPT draft my Vineland-3 results section? Yes. Paste your score summary (composite, domains, subdomains, and the form and informant used) and it drafts the results-section narrative for your review: organized domain by domain, with the standard-score and v-scale metrics kept distinct, functional meaning in prose, and informant context in place. It can also cross-check a draft you wrote for the errors reviewers flag, a v-scale read against the wrong mean, numbers that disagree with their interpretation, or an unnamed form, and produce a plain-language summary of the findings for parents and caregivers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from the scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, Vineland-3 product page and brochure: the 2016 publication, qualification level B, the three forms in Comprehensive and Domain-Level versions, the standard-score and v-scale metrics, the Motor Skills age-9 limit, and the absence of any Vineland-4 as of July 2026. 34 CFR 300.8(c)(6) and 34 CFR 300.304(b)(2) : the IDEA intellectual-disability definition requiring adaptive-behavior deficits, and the rule that no single measure may be the sole eligibility criterion. Social Security Administration, Listing 12.00/12.05 (adult) and childhood Listing 112.05: the intellectual-disorder criteria, onset before age 22, and the provision that a standardized adaptive test is not required but will be considered if it already exists. US Supreme Court, Moore v. Texas , 581 U.S. 1 (2017), with Atkins v. Virginia (2002) and Hall v. Florida (2014): the requirement that adaptive-functioning inquiry follow current medical standards and focus on deficits. Pandolfi & Magyar, 2021, American Journal on Intellectual and Developmental Disabilities : the finding that the Adaptive Behavior Composite is interpretable while the domain scores add little beyond it. Farmer, Floyd & McNicholas, 2021, Assessment : the unresolved dimensional status of the Comprehensive Interview Form. Buros Center for Testing, Tests reviewed in the Twenty-First Mental Measurements Yearbook (2020): the independent review of the Vineland-3, distinct from the 2018 Canadian Journal of School Psychology review. ## Related templates ADOS-2 Report Write-Up Psychological Evaluation Report Neuropsychological Evaluation Report WISC-V Report Write-Up BASC-3 Report Write-Up ABAS-3 Report Write-Up WPPSI-IV Report Write-Up NEPSY-II Report Write-Up Bayley-4 Report Write-Up Child & Adolescent Developmental Assessment NDIS Functional Capacity Assessment Report DAYC-2 Report Write-Up Battelle Developmental Inventory (BDI-3) Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/wais-report All note templates # WAIS Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The Wechsler Adult Intelligence Scale (WAIS) is the standard adult intelligence battery for ages 16 to 90: the WAIS-5 (2024) is the current edition, and the WAIS-IV (2008) remains in use during the transition. Psychologists and neuropsychologists use it in diagnostic, disability, accommodation, and forensic evaluations. This page covers how to write up WAIS results, with a fictional sample and the errors reviewers flag. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and neuropsychologists; publisher qualification level C Audience Referring physicians and psychiatrists, disability adjudicators, courts and attorneys, testing-accommodation reviewers, schools and employers Typical length 500 to 900 words for the results section · administration 45 to 90 minutes by edition and battery Format family Norm-referenced cognitive battery When it's used Adult cognitive and neuropsychological referrals: ADHD and learning-disability documentation, intellectual-disability determinations, capacity and forensic questions, cognitive-decline workups Standards context Published by NCS Pearson (WAIS-5, 2024; WAIS-IV, 2008); described here for write-up purposes; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the WAIS? The Wechsler Adult Intelligence Scale (WAIS) is an individually administered, norm-referenced intelligence battery for ages 16:0 to 90:11, published by NCS Pearson in a lineage that runs from the original WAIS of 1955 through the WAIS-IV (2008) to the current WAIS-5 (2024). The WAIS-5 yields five primary index scores, Verbal Comprehension, Visual Spatial, Fluid Reasoning, Working Memory, and Processing Speed, plus a Full Scale IQ derived from seven subtests and 15 ancillary indexes including the General Ability Index (GAI), Cognitive Proficiency Index (CPI), and new Nonverbal, Nonmotor, and Motor-Reduced Processing Speed accessibility indexes. Composites have a mean of 100 and a standard deviation of 15, subtest scaled scores a mean of 10, and the battery was normed in 2023 and 2024 on 2,020 examinees, co-normed with the Wechsler Memory Scale, Fifth Edition (WMS-5, 2025) for ability-memory comparison. Administration is paper or digital via Q-interactive , with scoring by hand, on Q-global, or on Q-interactive, at publisher qualification level C. The 7-subtest FSIQ takes about 45 minutes; the 10 primary index subtests take about 60. The load-bearing fact for write-ups, as of July 2026, is that two editions are legitimately in circulation and they are not interchangeable. The WAIS-5 restructured the battery: the WAIS-IV's Perceptual Reasoning Index was split into separate Visual Spatial and Fluid Reasoning indexes, the FSIQ went from ten subtests to seven, five subtests are new, and Picture Completion and Cancellation are gone. The score reports changed too: WAIS-IV confidence intervals are built on the standard error of measurement, while the WAIS-5 report states "Confidence intervals are calculated using the Standard Error of Estimation," and the descriptor vocabulary was rewritten, with Borderline, Superior, and Very Superior retired in favor of terms like Above average and Very high. Pearson still sells the WAIS-IV, whose product page now carries a "WAIS-5 - is now available!" banner, and no discontinuation date has been published. A defensible modern WAIS write-up therefore names the edition, the norm set, and, during the transition, the reason for the choice, the same discipline that governs the WISC-V as its own next edition approaches. A WAIS score is evidence, not a verdict: eligibility and legal determinations belong to the agencies and courts that read the whole evaluation , and the results section's job is to feed them calibrated, honestly bounded numbers. ## Who uses WAIS write-ups and when Clinical and counseling psychologists write up WAIS results inside adult psychological evaluation reports for diagnostic clarification, ADHD and learning-disability documentation, and testing-accommodation requests; neuropsychologists fold it into a neuropsychological report as the cognitive core alongside memory (often the co-normed WMS-5), executive, and performance-validity measures; and forensic psychologists carry it into capacity, disability, and Atkins intellectual-disability evaluations, where the write-up faces cross-examination. It also supplies the IQ evidence in Social Security adult disability claims under Listing 12.05. Age drives instrument choice at one edge: at the 16-year overlap, convention favors the WISC-V when lower ability is suspected (better floor) and the WAIS-5 for high-ability adolescents (better ceiling). When only an estimate is needed, the WASI-2 screens general ability but does not substitute for a full battery in high-stakes determinations, and brief repeatable batteries such as the RBANS serve serial neurocognitive tracking rather than full ability measurement. In Canada, administration runs through provincially regulated psychologists with a live norms question this page covers below; in Australia and New Zealand, the released A&NZ standardised edition carries local norms. ## How to structure a WAIS results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names the measurement conditions before the numbers, presents composites in a table and meaning in prose, and makes the two genuinely contested moves in adult cognitive reporting, the edition choice and the FSIQ-versus-GAI decision, explicit stated decisions instead of silent ones. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and format. Name the instrument and edition (WAIS-5 or WAIS-IV), the norm set (US, Canadian for the WAIS-IV, or Australian and New Zealand standardised), which subtests were administered, the administration format (paper or Q-interactive), and the scoring platform. During the transition, add the one-sentence rationale for the edition used. Pitfall: writing "the Wechsler" or "the WAIS" with no edition. Two editions are legitimately in circulation as of July 2026, Canadian administrations on US norms carry Pearson's own "clinically appropriate" caveat that belongs in the report, and an edition-less score set is ambiguous at re-evaluation. Behavioral observations and validity. Describe effort, engagement, language and sensory considerations, breaks, and anything that departed from standardized procedure, then state plainly whether the results are valid estimates of current functioning. In medico-legal, disability, and accommodation contexts, document that performance validity was formally considered. Pitfall: boilerplate that contradicts the interpretation. "Effort was excellent" followed by scores you then attribute to disengagement hands a reviewer two incompatible reports in one document. Composite summary table. Report each composite with its standard score, confidence interval, and percentile rank, and give the qualitative descriptor exactly as your scoring platform prints it, from one edition-correct vocabulary. State the interval convention once: the WAIS-5 report calculates confidence intervals "using the Standard Error of Estimation" and prints a 95% column, while WAIS-IV intervals are SEM-based. Pitfall: bare standard scores. A score without its interval and percentile invites over-precision, and threshold readers (SSA, courts, accommodation reviewers) are exactly the audiences that need the error band; NCBE's guidelines specifically require age-based standard scores, with percentiles alone insufficient. Index-by-index results in prose. Move through Verbal Comprehension, Visual Spatial, Fluid Reasoning, Working Memory, and Processing Speed (or the WAIS-IV's four indexes), and spend the prose on what each result means functionally for this adult, in work, study, or daily life, not on re-reading the table aloud. Pitfall: the score dump. A paragraph per subtest in administration order documents that testing occurred; it does not answer the referral question, and it drifts toward the subtest-level interpretation the measurement literature warns against. FSIQ, GAI, and the interpretive decision. The WAIS-5 FSIQ comes from seven subtests, and the independent factor-analytic literature finds general ability the best-measured thing on the test. When working memory or processing speed depress the composite, convention offers the General Ability Index, with the Cognitive Proficiency Index as its paired contrast. Whatever you foreground, report both scores and state the choice and the reason. Pitfall: silent substitution. Which score you use changes outcomes at cutoffs (159 versus 196 children meeting an intellectual-disability threshold in one published sample), so an unexplained GAI, or an FSIQ quietly dropped for scatter, reads as score-shopping. Premorbid estimate and change, where the question is decline. In neurocognitive referrals, frame current scores against expected prior ability using a word-reading estimate or demographic prediction, and treat cross-edition comparisons to old records as different measurements, not the same ruler. Pitfall: reading an Average score as "no decline" for a person whose premorbid estimate was well above average, or comparing a WAIS-5 score to a 2015 WAIS-IV score point-for-point when the editions differ in structure, norms vintage, and interval convention. Interpretive summary and referral linkage. Answer the referral question directly, state what the cognitive data do and do not establish, and place the WAIS explicitly as one source among several: SSA's intellectual-disorder listing pairs the IQ criterion with adaptive-functioning deficits, and no accommodation or eligibility framework reads an IQ score alone. Pitfall: a verdict from a score. Determinations belong to the agencies, teams, and courts applying their criteria; the results section supplies calibrated evidence, including its error bands. Recommendations linkage. Tie each recommendation to a specific finding, and for accommodation requests give the functional-impact rationale reviewers ask for: what the limitation is, how it shows under standard conditions, and why each requested accommodation addresses it. Pitfall: recommendations that could follow any profile. USMLE's guidance asks for a clear rationale for why each recommended accommodation is necessary; extended time that does not trace back to the processing-speed and working-memory findings is an assertion, not an argument. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: [WAIS-5 / WAIS-IV] Norms: [US / CDN (IV) / A&NZ] Edition rationale (transition): [current norms / longitudinal comparison / local-norms availability] Subtests given: [7 FSIQ / 10 primary / list ancillary] Format: [paper / Q-interactive] Scoring: [Q-global / manual] BEHAVIORAL OBSERVATIONS AND VALIDITY [Effort, engagement, language/sensory factors, breaks, departures from standard procedure; performance validity addressed; statement that results are valid estimates, or the stated limitation if not] COMPOSITE SUMMARY (table) [Composite | standard score | CI (state SEE or SEM basis) | percentile | platform descriptor, one edition-correct vocabulary throughout] INDEX-BY-INDEX RESULTS (prose, functional meaning) Verbal Comprehension: [score context + what it means here] Visual Spatial / Fluid Reasoning: [same] Working Memory / Processing Speed: [same] FSIQ / GAI INTERPRETIVE DECISION [FSIQ with CI; if indexes diverge, the ability estimate you are using, WITH the stated rationale; GAI-CPI contrast described qualitatively, no reproduced base-rate tables] PREMORBID ESTIMATE AND CHANGE (decline referrals) [Estimate basis; current-vs-expected framing; cross-edition comparisons treated as different measurements] INTERPRETIVE SUMMARY (answer the referral question) [What the data establish and what they do not; single-measure limits; measurement error acknowledged] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; functional-impact rationale for any requested accommodation] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample WAIS write-up (fictional) Scenario: a 27-year-old law graduate seeks updated documentation of longstanding ADHD to support a bar-examination testing-accommodation request. The profile splits, average-or-better reasoning over notably lower working memory and processing speed, the shape that forces the FSIQ-versus-GAI decision and the functional-impact argument accommodation reviewers read for. This is the WAIS-5 results section only, condensed but structurally complete. All details are fictional. Client: R.T., 27 · Referral: updated ADHD documentation for a bar-examination accommodation request, coordinated with treating psychiatrist · Evaluator: L. Okafor, PsyD, Licensed Psychologist · Testing date: 07/08/2026 · Report date: 07/16/2026 Measures and administration: The Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5) was administered via Q-interactive as part of a comprehensive evaluation that also included achievement testing, self- and observer-report ADHD ratings, and academic record review, reported in their own sections. All ten primary subtests were given and scored against the United States normative sample (2023 to 2024 norms), yielding the Full Scale IQ (FSIQ), the five primary index scores, and, given the pattern below, the General Ability Index (GAI) and Cognitive Proficiency Index (CPI). The WAIS-5 is used as the current edition for this new evaluation. Composites have a mean of 100 and a standard deviation of 15 and are reported with the WAIS-5 report's 95% confidence intervals, which are calculated on the Standard Error of Estimation; qualitative descriptors follow the WAIS-5 report vocabulary in the narrative below. Behavioral observations and validity: R.T. arrived on time, was cooperative and candid, and worked with visible investment across the session. She engaged quickly with reasoning tasks, often elaborating on her answers; on timed clerical tasks her pace was slow and effortful, with frequent re-checking, a pattern consistent across tasks rather than tied to any single lapse of attention. Performance validity was formally assessed and was intact, and embedded indicators were consistent with adequate effort. Administration followed standardized procedures, and the results below are considered valid estimates of current cognitive functioning. Composite Standard score 95% CI Percentile Verbal Comprehension (VCI) 111 105-116 77 Visual Spatial (VSI) 104 97-110 61 Fluid Reasoning (FRI) 108 101-114 70 Working Memory (WMI) 85 80-92 16 Processing Speed (PSI) 83 76-92 13 Full Scale IQ (FSIQ) 96 92-101 39 General Ability (GAI) 109 103-114 73 Cognitive Proficiency (CPI) 82 77-90 12 Index results: R.T.'s verbal reasoning is a relative strength. Her Verbal Comprehension score of 111 (77th percentile, described as Above average on the WAIS-5) indicates well-developed word knowledge and verbal concept formation, consistent with her academic record. Visual Spatial (104) and Fluid Reasoning (108) scores fall in the Average range: she analyzes visual material and reasons through novel patterns at an age-typical level, and she solved untimed reasoning items methodically and well. In contrast, her Working Memory score of 85 (16th percentile) falls below the Average range, reflecting difficulty holding and mentally reworking spoken information such as number sequences, and her Processing Speed score of 83 (13th percentile) reflects slow performance on simple timed visual-clerical tasks even with accurate work. Functionally, this is the profile of an adult who reasons at or above the level of her peers but loses long spoken instructions, re-reads under time pressure, and does not finish timed sections despite mastery of the content, which matches her educational history and current report. Overall ability and the interpretive decision: The FSIQ of 96 (CI 92 to 101, 39th percentile, Average range) is reported for completeness, but it averages across abilities that differ sharply and so describes no single ability R.T. actually shows. The contrast between her General Ability Index of 109 (73rd percentile, reasoning-focused) and her Cognitive Proficiency Index of 82 (12th percentile, working memory and processing speed) is large and uncommon relative to her overall level. For the functional questions in this evaluation, the GAI is used as the estimate of reasoning ability, because the referral question is whether cognitive efficiency limits her demonstration of that ability under standard conditions, and the FSIQ would fold the areas of difficulty into the expectation itself. This is a stated interpretive choice, not a substitution: both scores are reported above, and the FSIQ remains the more broadly validated summary score. Functional impact: The cognitive findings converge with the history documented elsewhere in this evaluation: childhood diagnosis with school accommodations from age 10, extended time throughout law school, timed practice-examination scores well below untimed performance on the same material, and current observer ratings. The specific accommodation of extended testing time addresses the identified limitations directly: processing speed at the 13th percentile and working memory at the 16th percentile constrain timed retrieval and response recording, not knowledge or reasoning, which are Average to Above average. Summary: Cognitive testing shows Above average verbal reasoning and Average visual-spatial and fluid reasoning over notably lower working memory and processing speed, with a large gap between reasoning (GAI 109) and cognitive efficiency (CPI 82). This pattern is consistent with, though not by itself diagnostic of, the ADHD documented in the diagnostic section of this evaluation; no single measure carries that conclusion alone. All scores are estimates that carry measurement error, describe current performance rather than fixed capacity, and are reported with the edition, norms, and format stated so they remain interpretable at any future re-evaluation. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works The edition, norm set, administration format, and scoring platform are all named, with the confidence-interval convention stated, so the scores stay interpretable at re-evaluation and defensible during the edition transition. Every composite appears with its 95% confidence interval and percentile, every number in the prose matches the table, and descriptor language follows the WAIS-5 vocabulary with no WAIS-IV terms leaking in. The FSIQ-versus-GAI decision is made in the open: both scores are reported, and the GAI anchor comes with its stated rationale rather than a silent substitution. Functional impact carries the argument: the score pattern is tied to timed real-world performance and history, which is what accommodation reviewers say approval turns on, and performance validity is addressed. The summary states what the data do not establish, keeps the single-measure limits explicit, and every recommendation traces to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label each requirement's strength honestly. In capital litigation the framework is constitutional law: Atkins v. Virginia (2002) bars executing persons with intellectual disability, Hall v. Florida (2014) holds that "intellectual disability is a condition, not a number" and that a defendant whose IQ score falls within the test's margin of error must be allowed to present adaptive-deficit evidence, and Moore v. Texas (2017) requires courts to be informed by current medical standards, reading a score of 74 as the range 69 to 79 once measurement error is applied. How courts should weigh multiple IQ scores remains genuinely unsettled: the Supreme Court took the question up twice in Hamm v. Smith, then dismissed the case as improvidently granted on May 21, 2026, so the write-up convention of reporting every score with its error band, dates, and edition is what fills the gap. Capital and other forensic write-ups are specialist work; follow your board's forensic guidelines and coordinate with counsel. For Social Security, adult Listing 12.05B keys on a "full scale (or comparable) IQ score of 70 or below," or 71 to 75 with a part score of 70 or below, on an individually administered standardized test, paired with significant adaptive-functioning deficits and evidence the disorder began before age 22; SSA requires a qualified specialist licensed at the independent level to administer, and if an assistant administered the test, a supervisory specialist must interpret and co-sign ( 20 CFR 404, Subpart P, Appendix 1, 12.00H ). For testing accommodations, agency policy is the framework: NCBE's documentation guidelines require "Age-based standard scores" for all normed measures, state that "Percentiles alone are not sufficient," ask for a comprehensive battery that relies on no single test, and in most cases want evaluation within the past five years, while USMLE asks for comprehensive documentation with a functional-limitations rationale tied to each requested accommodation. Payer billing for testing time runs under the psychological and neuropsychological testing code families (96130 through 96139) with the evaluation services personally performed rather than delegated, and at least one Medicare contractor describes technician administration as requiring direct supervision ( A57481 ); the billing mechanics belong to the parent report pages, the psychological evaluation report and neuropsychological report . Edition and norms currency is the WAIS's live defensibility question, and it has a quantified core. Default to the WAIS-5 for new US evaluations: the norms are current (2023 to 2024), APA Ethics Standard 9.08 tells psychologists not to base decisions on obsolete measures or outdated results, and Pearson reports that most practitioners transition within the first year after a release. No statute or board sets a hard switch deadline, so the WAIS-IV remains defensible with a stated reason: longitudinal comparison to prior WAIS-IV data, or a jurisdiction whose local norms have not released. Canada is that jurisdiction as of July 2026: the WAIS-5 Canadian edition is still in standardization, and Pearson's own Canadian materials state the US version is available "when the examiner determines that U.S. norms are clinically appropriate," a determination that belongs in the report, not just the file. Australia and New Zealand have a released A&NZ standardised edition with 2023 to 2024 norms; note that Australian public funding is narrow, with no general MBS item for standalone adult cognitive assessment on a July 2026 search of the schedule. Norms aging is measurable: the classic Flynn-effect estimate is about 3 points per decade, but the WAIS-5 counterbalanced standardization studies found "a dramatic, unexpected Flynn effect of 1.2 IQ points per decade" (Winter, Trudel & Kaufman, Journal of Intelligence, 2024), whose analysis also concludes WAIS-IV scores now run about five points high and that retest practice gains shrank from roughly 4.5 to 5 points on the WAIS-IV to 3.7 on the WAIS-5 for most adult ages. Where a Flynn or practice-effect adjustment matters, name the edition-specific evidence rather than applying the historical rule reflexively, and treat scores from different editions as different measurements: structure, norms vintage, descriptor vocabulary, and even the confidence-interval basis (SEM on the WAIS-IV, SEE on the WAIS-5) all changed. Access boundaries are the publisher's: qualification level C restricts purchase and administration to doctorally trained or equivalently credentialed professionals, and Pearson's own equivalence study (Q-interactive Technical Report 16, 2024) supports treating digital and paper formats as comparable, with Spatial Addition's digital norms equated, while the write-up convention remains to name the format used. WAIS is a trademark, in the US and other countries, of Pearson plc; the WAIS-5 and WAIS-IV are published by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common WAIS write-up errors reviewers flag There is no payer audit series for cognitive write-ups; the accountability literature here is psychometric and forensic, and right now it is unusually active. The WAIS-IV was reviewed in the Buros Center's 18th Mental Measurements Yearbook ; the WAIS-5 has no Buros review yet, and the first independent factor analysis of its standardization sample, published in Assessment in February 2026, reported that its results "did not support the purported WAIS-5 structure," finding four factors and a dominant general factor instead of the five marketed indexes. Meanwhile the edition transition is producing a new error class entirely, and even agencies lag it: NCBE's learning-disability documentation guidelines still name the WAIS-IV among their example measures. The BastionGPT Clinical Advisory Board sees the same errors most often in WAIS report reviews: WAIS-IV vocabulary on WAIS-5 scores. Borderline, Superior, and Very Superior are WAIS-IV descriptor terms; the WAIS-5 report prints a different vocabulary (Average, Above average, Very high). Carrying the old labels forward, or describing a SEE-based WAIS-5 confidence interval as if it were the WAIS-IV's SEM-based interval, tells a careful reader the report was written from an old template rather than the current score report. The historical Flynn adjustment applied reflexively. Correcting a WAIS-5 score by the classic 3 points per decade ignores the edition's own counterbalanced data showing about 1.2, and the difference is potentially dispositive: in Winter, Trudel and Kaufman's worked example, a 5-point WAIS-IV-era correction yields an IQ of 75 and intellectual-disability eligibility, while the 1.2-per-decade correction yields 78 and does not. Cite edition-specific evidence for any adjustment, and remember courts must account for measurement error either way. The FSIQ-GAI move made silently. Which summary score you foreground changes real outcomes: in a pediatric sample, using the GAI instead of the FSIQ changed how many children met an intellectual-disability cutoff (159 versus 196, Koriakin et al., 2013). Report both scores, state which one anchors your conclusions, and give the reason; a GAI that appears without rationale reads as score-shopping. Five indexes treated as five validated abilities. The Visual Spatial and Fluid Reasoning split is the WAIS-5's headline change, but independent analyses found the two merge back into one factor with measurement dominated by general ability, echoing a decade of WAIS-IV findings that composites, not subtest patterns or index scatter, carry the interpretive weight. Keep index interpretation cautious, keep subtest commentary qualitative and bounded, and never let scatter carry a diagnosis. Edition and norms unnamed during a live transition. "The WAIS" with no edition, no norm set, and no format is now ambiguous in a way it was not in 2020: two editions circulate, Canadian testing on US norms requires a documented rationale in Pearson's own words, and cross-edition score comparisons are measurements on different rulers. Name edition, norms, format, and platform every time; it costs one sentence. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your composite summary (scores, confidence intervals, percentiles) and get a drafted results-section narrative organized index by index, with the FSIQ-GAI decision framed for your review and rationale. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with the table, WAIS-IV descriptor terms in a WAIS-5 report, and ability estimates substituted without a stated reason. Translate the results section into a plain-language summary for the client, referring providers, or disability reviewers that keeps scores as estimates, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do WAIS scores actually mean? Three score types do the work. Composites (the FSIQ and the index scores) are standard scores with a mean of 100 and a standard deviation of 15; subtest scaled scores have a mean of 10, with roughly 8 to 12 typical; and every composite carries a percentile rank and a confidence interval. Read the interval by edition: the WAIS-5 report states its intervals are "calculated using the Standard Error of Estimation" and prints a 95% column, while WAIS-IV intervals are SEM-based. Descriptors are edition-specific too: the WAIS-5 vocabulary uses terms like Average, Above average, and Very high, and the WAIS-IV terms Borderline, Superior, and Very Superior are retired. Write scores as estimates: interval and percentile belong next to every composite, and a descriptor should never appear without the number it describes. Should I test on the WAIS-5 or the WAIS-IV now? Default to the WAIS-5 for new US evaluations: it is the current edition (2024) with 2023 to 2024 norms, APA Ethics Standard 9.08 tells psychologists not to base decisions on obsolete measures or outdated results, and Pearson reports most practitioners transition within the first year after a release. No statute or board sets a hard deadline, so the WAIS-IV stays defensible with a stated reason: longitudinal comparison to prior WAIS-IV data, or a jurisdiction without released local norms. Canada is the live example, with the WAIS 5-CDN still in standardization and Pearson's Canadian materials offering the US version "when the examiner determines that U.S. norms are clinically appropriate"; Australia and New Zealand already have a released A&NZ standardised edition. Whichever edition you use, name it, name the norms, and give the transition rationale in one sentence. Are WAIS-IV and WAIS-5 scores interchangeable? No. The editions differ in structure (Perceptual Reasoning split into Visual Spatial and Fluid Reasoning, a 7-subtest FSIQ, five new subtests), norms vintage (2007 to 2008 versus 2023 to 2024), descriptor vocabulary, and even the confidence-interval basis. The quantitative gap is now measured: the WAIS-5 counterbalanced standardization studies found "a dramatic, unexpected Flynn effect of 1.2 IQ points per decade," far below the roughly 3 points per decade of the classic literature, and the same analysis concludes the WAIS-IV now yields scores about five points high ( Winter, Trudel & Kaufman, 2024 ). Retest practice gains also shrank, from roughly 4.5 to 5 FSIQ points on the WAIS-IV to about 3.7 on the WAIS-5 for most adult ages. Treat cross-edition comparisons as different measurements, describe change qualitatively with the edition difference named, and cite edition-specific evidence for any score adjustment. Do I have to report the FSIQ, and when does the GAI belong? No authority mandates the FSIQ by name; SSA's listing keys on a "full scale (or comparable) IQ score," and the or-comparable wording admits an equivalent composite. Convention still favors reporting it, and the psychometric case is strong: the first independent factor analysis of the WAIS-5 standardization sample found measurement dominated by general ability, the quantity the FSIQ estimates. The GAI belongs when working memory or processing speed depress the composite and the referral question concerns reasoning ability: report both scores, state which one anchors your conclusions, and give the reason. The choice has teeth at cutoffs: in a published pediatric sample, using the GAI instead of the FSIQ changed intellectual-disability classification from 196 children to 159 . Silent substitution in either direction is the error; a stated, reasoned choice is the convention. Is the WAIS required for SSA disability or intellectual-disability decisions? No instrument is named anywhere in the framework. SSA's adult intellectual-disorder listing requires a "full scale (or comparable) IQ score of 70 or below," or 71 to 75 with a part score of 70 or below, on an individually administered standardized test, together with significant adaptive-functioning deficits and evidence of onset before age 22 ( Listing 12.05B ); SSA also requires a qualified specialist licensed at the independent level, with a supervisory specialist interpreting and co-signing if an assistant administered. In court, the constitutional line runs through Atkins, Hall v. Florida ("intellectual disability is a condition, not a number," with scores read through their margin of error), and Moore v. Texas, and the method for weighing multiple IQ scores is genuinely unsettled: the Supreme Court dismissed Hamm v. Smith without deciding it in May 2026. The practical rule: report every score with its interval, date, edition, and norms, and let the adaptive-functioning evidence do its half of the work. What do accommodation reviewers (bar exam, USMLE) want with WAIS scores? Full standard scores, a comprehensive battery, recent testing, and a functional argument. NCBE's documentation guidelines require age-based standard scores for all normed measures and state flatly that "Percentiles alone are not sufficient," ask that the assessment "not rely on any single test or subtest," and in most cases want evaluation within the past five years ; their example measures even still name the WAIS-IV, a reminder that agency lists lag editions and your report should name the edition used. USMLE asks for comprehensive documentation with a description of current functional limitations and "a clear rationale" for each requested accommodation. What carries approvals is the linkage: scores, history, and real-world timed-versus-untimed evidence converging on the specific accommodation requested, which is exactly how the sample on this page argues it. WAIS-5 or WISC-V for a 16-year-old? Both cover the overlap (the WISC-V runs to 16:11, the WAIS-5 starts at 16:0), so the choice is about measurement room. Convention follows floor-and-ceiling logic: choose the WISC-V when lower ability is suspected, because a test built for children has more items at the low end to differentiate with, and choose the WAIS-5 for a high-ability 16-year-old who would hit the WISC-V's ceiling. Practical factors count too: if re-evaluation will happen in adulthood, starting the adult battery avoids an instrument switch, and if ability-memory analysis is coming, the WAIS-5's co-norming with the WMS-5 argues for the adult pair. Whichever you choose, say why in one sentence; reviewers read instrument selection at boundary ages as a marker of care. Can I see the actual WAIS questions, or include them in a report? No. Test items, stimuli, and record forms are protected test materials: access is restricted to qualified professionals (the WAIS-5 is Pearson qualification level C), psychologists are ethically obligated to protect test security under APA Ethics Standard 9.11 and their publisher agreements, and Pearson's own score reports carry trade-secret notices limiting what may be excerpted. Item exposure also damages the test itself, because norms assume examinees have not rehearsed the tasks. Reports should carry scores, descriptors, and interpretation, never item content, and records requests that touch raw protocols route through your test-security obligations rather than reflexive printing. What clients and reviewers can get is better anyway: the scores with a plain-language explanation of what each index means and what follows from it. Can BastionGPT draft my WAIS results section? Yes. Paste your composite summary (scores, confidence intervals, percentiles) and it drafts the results-section narrative for your review: organized index by index, functional meaning in prose, the FSIQ-GAI decision framed with a stated rationale, and descriptor language kept to one edition-correct vocabulary. It can also cross-check a draft you wrote for score-versus-narrative mismatches, WAIS-IV terms in a WAIS-5 report, and unstated interpretive choices, and produce a plain-language summary of the findings for the client or referral source. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, WAIS-5 product page : 2024 publication, ages 16:0 to 90:11, qualification level C, 45-minute 7-subtest FSIQ and 60-minute 10-subtest administration, paper and Q-interactive formats. Pearson, WAIS-IV vs. WAIS-5 comparison flyer and WAIS-5 overview brochure : the edition-by-edition table (administration times, norms vintages, subtest rosters, 10- versus 7-subtest FSIQ, 4-plus-optional versus 5-plus-15 index structure), the five new subtests, the accessibility indexes, the 2,020-examinee normative sample, and WMS co-norming. Pearson, WAIS-5 sample score report and WAIS-IV score report : the Standard Error of Estimation versus SEM confidence-interval bases, the 95% interval column, and the two editions' descriptor vocabularies. Pearson, WAIS-IV product page : the 2008 publication and the "WAIS-5 - is now available!" banner on a still-active product line. Pearson Clinical Canada, WAIS-5 Canadian brochure and WAIS 5-CDN standardization study : the US-version availability caveat and the in-development Canadian edition. Pearson Clinical AU&NZ, WAIS-5 A&NZ sample score report : the released Australian and New Zealand standardised edition (adaptation copyright 2025). Pearson, Q-interactive Technical Report 16 (2024): digital-paper equivalence findings and the Spatial Addition equating; Pearson, WMS-5 sample score report (2025): the co-normed memory battery. Winter, Trudel & Kaufman, 2024, Journal of Intelligence : the 1.2-points-per-decade Flynn finding, the five-points-high WAIS-IV conclusion, and the edition practice-effect comparison; Trahan, Stuebing, Fletcher & Hiscock, 2014, Psychological Bulletin : the classic Flynn meta-analysis (2.31 points per decade; 2.93 for modern tests). Canivez & Watkins, 2010, Psychological Assessment (WAIS-IV factor structure); Canivez, Watkins, McGill & Dombrowski, 2026, Assessment (first independent WAIS-5 factor analysis); Koriakin et al., 2013, Developmental Medicine & Child Neurology (FSIQ-versus-GAI classification); Estevis, Basso & Combs, 2012, The Clinical Neuropsychologist (WAIS-IV practice effects at 3 and 6 months). 20 CFR Part 404, Subpart P, Appendix 1 : adult Listing 12.05 criteria and the 12.00H testing and qualified-specialist requirements. Atkins v. Virginia (2002); Hall v. Florida , 572 U.S. 701 (2014); Moore v. Texas (2017); Hamm v. Smith, 604 U.S. 1 (2024) and the May 21, 2026 dismissal : the capital intellectual-disability line. NCBE, learning-disability and ADHD documentation guidelines; USMLE, test-accommodation documentation : the standard-scores, currency, and functional-rationale requirements. Buros Center for Testing, Tests reviewed in the Eighteenth Mental Measurements Yearbook : independent review status of the WAIS-IV; APA, Ethical Principles of Psychologists and Code of Conduct : Standards 9.08 (obsolete tests) and 9.11 (test security); CMS Medicare Coverage Database, Article A57481 : testing-code billing context. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report WISC-V Report Write-Up Psychiatric Diagnostic Evaluation MMPI-3 Report Write-Up D-KEFS Report Write-Up RBANS Report Write-Up Psychological Testing Administration Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/wiat-4-report All note templates # WIAT-4 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The WIAT-4 (Wechsler Individual Achievement Test, Fourth Edition) is an individually administered academic achievement battery for ages 4 through 50, measuring reading, mathematics, written expression, and oral language with standard scores set to a mean of 100. Psychologists, school psychologists, and specific-learning-disability evaluation teams use it to quantify academic skills against age or grade norms. This page covers how to write up WIAT-4 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and other qualified assessors; publisher qualification level B Audience IEP and Section 504 teams, parents, teachers, accommodations reviewers, physicians Typical length 500 to 900 words for the results section · administration time varies by grade and subtests given Format family Norm-referenced academic achievement battery When it's used Specific learning disability and psychoeducational evaluations, dyslexia and dyscalculia referrals, academic accommodation requests, ages 4 to 50 Standards context Published by NCS Pearson (2020); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the WIAT-4? The Wechsler Individual Achievement Test, Fourth Edition (WIAT-4) is an individually administered, norm-referenced academic achievement battery for ages 4:0 to 50:11, published by NCS Pearson in 2020 as the fourth edition of a line that began with the original WIAT in 1992. It measures learned academic skills, word reading, decoding, reading comprehension and fluency, spelling and written expression, math computation, problem solving and fluency, and oral language, and reports them as composites: Reading, Mathematics, Written Expression, and Oral Language, an overall Total Achievement score (which summarizes the reading, math, and writing subtests), and supplemental composites covering basic reading, decoding, fluency, phonological and orthographic processing, and a Dyslexia Index. The current battery carries 20 subtests , five of them new to this edition and one of the five (Orthographic Choice) available only on Q-interactive, and it yields 32 scores in all, so a report should name which subtests were actually given rather than a generic count. One metric detail matters more than it looks: WIAT-4 subtests and composites are both standard scores with a mean of 100 and a standard deviation of 15, unlike Wechsler cognitive batteries, where subtests use scaled scores with a mean of 10. Scoring runs through Q-global or Q-interactive against age-based or grade-based norms (grade norms in fall, winter, and spring windows), with percentile ranks, confidence intervals, and Growth Scale Values for progress monitoring. Edition status as of July 2026: Pearson's product page calls the WIAT-4 "the most current version of the WIAT" and no fifth edition has been announced; Canada gained a Canadian-normed WIAT-4-CDN in 2024 with a French-Canadian adaptation; Australia and New Zealand have no WIAT-4 adaptation, so clinicians there use the US-normed edition or the older WIAT-III A&NZ . The load-bearing distinction for write-ups is that achievement is not ability. The WIAT-4 quantifies what a person has learned to do academically; it says nothing by itself about reasoning, and it earns its place in an evaluation next to a cognitive measure such as the WISC-V , whose linkage studies support ability-achievement and pattern-of-strengths-and-weaknesses analyses. The second distinction is interpretive altitude. The technical manual, unusually, contains no factor analyses, a gap both published reviews flag: Dombrowski and Casey (2022) judge the test "well-conceptualized, and generally psychometrically sound" while faulting the missing structural evidence, and Beaujean and Parkin (2022) , after running the analyses the manual omits, found support for equal-interval score properties insufficient and advised psychologists to "completely refrain" from composite-score uses that require them. The defensible posture, and the one this page teaches, leads with normative composite interpretation, confidence intervals attached, treats ipsative and discrepancy analytics as stated choices with stated limits, and feeds the eligibility decision, which belongs to a team applying whole-evaluation criteria , rather than announcing it. ## Who uses WIAT-4 write-ups and when School psychologists write up WIAT-4 results inside special-education eligibility evaluations, where its domain structure tracks the eight achievement areas named in the federal specific-learning-disability regulation; clinical and pediatric psychologists use it for dyslexia, dyscalculia, and written-language referrals; and neuropsychologists fold achievement testing into a neuropsychological report when learning questions ride along with attention or memory ones. It also supplies the academic evidence in college-admissions and professional-exam accommodation requests, which for adults is why the norms running to age 50 matter. Instrument choice is mostly a pairing decision: per Pearson's product page the WIAT-4 "links directly to the WISC-V and KABC-2 NU," the KTEA-3 (also Pearson, normed for ages 4:0 to 25:11) carries its own dyslexia index, and the Woodcock-Johnson achievement battery pairs with its own cognitive counterpart for CHC-style cross-battery work; head-to-head validity studies among the three are sparse, so the honest defense of your choice is the referral question and the cognitive test you are linking to. For frequent progress monitoring, curriculum-based measures do that job better; the WIAT-4 is the norm-referenced diagnostic snapshot, and its Growth Scale Values add only limited monitoring capacity between full administrations. Achievement results rarely stand alone: they sit beside cognitive results, behavior ratings such as the BASC-3 , intervention-response data, and classroom evidence in the full evaluation report . ## How to structure a WIAT-4 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names the data before interpreting it, leads with composites and their confidence intervals, and makes the two genuinely contested moves in achievement reporting, which norm set anchors the comparison and how far below the composite you interpret, explicit stated decisions instead of silent ones. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and format. Name the instrument and edition, the norm set with its basis (US or Canadian; age-based or grade-based, and for grade norms the season window), the subtests administered, the format (paper or Q-interactive, which differ by four subtests), and the scoring platform. State the age-versus-grade choice as a decision: age norms when the comparison is same-age peers or an age-normed cognitive test, grade norms when the question is grade-level attainment, both for retained or young-for-grade students. Pitfall: scores with no stated norm basis. An 88 against age norms and an 88 against grade norms answer different questions, and ability-achievement analyses require the achievement and cognitive scores to share a reference frame. Behavioral observations and validity. Describe effort, attention, language and sensory considerations, breaks, and any departure from standardized procedure, then state plainly whether the results are valid estimates of current academic functioning. Pitfall: observations that contradict the interpretation. Noting fatigue and rushed responding, then interpreting timed fluency scores without a caveat, hands a reviewer two incompatible accounts in one report. Composite summary table. Report each composite with its standard score, confidence interval, percentile rank, and descriptor from one labeled system, and state the metric once: WIAT-4 subtests and composites are standard scores with a mean of 100 and a standard deviation of 15. That sentence prevents the most common cross-battery misreading, because Wechsler cognitive subtests use a mean-10 scaled metric and mixed reports invite readers to apply the wrong ruler. Pitfall: bare standard scores. A score without its confidence interval and percentile presents a measured estimate as an exact point, and achievement decisions downstream (eligibility, accommodations) are threshold decisions where error bands matter. Domain-by-domain results with error analysis. Move through reading, mathematics, written expression, and oral language, spending the prose on functional meaning for this student, and add what the numbers cannot show: how the student erred, decoding attempts, self-corrections, strategy use, because achievement tests support qualitative skills analysis in a way cognitive batteries rarely do. Pitfall: the score dump. A paragraph per subtest in administration order documents that testing occurred; it does not answer the referral question, and it buries the error-pattern observations that make achievement write-ups useful to teachers. Composite integrity and the interpretive decision. When subtests inside a composite diverge, say so and interpret at the level the data support; when composites are intact, resist drilling down. The published critiques give this section its spine: the technical manual contains no factor analyses, and the independent evaluation found support for equal-interval score properties insufficient, advising psychologists to refrain from composite-score uses that require them, which is a direct caution about ipsative strength-and-weakness analytics and mechanical discrepancy arithmetic. Lead with normative interpretation, and where you run planned comparisons, present them as supplementary with their base-rate context. Pitfall: an intact-looking composite hiding a split. Reporting Reading as one number when word reading and comprehension sit twenty points apart writes the referral question out of the report. Ability-achievement integration and the SLD model. Name the framework the operative state uses (discrepancy, intervention response, or pattern of strengths and weaknesses), name the cognitive estimate anchoring any comparison and its norm frame, and remember what federal law actually says: states must not require a severe discrepancy, must permit an intervention-response process, and may permit other research-based procedures (34 CFR 300.307). Pitfall: writing "a severe discrepancy is required" in the report. The regulation prohibits states from requiring it; asserting it as federal law dates the report and misstates the rule reviewers know. Dyslexia Index and screening boundaries. Report it like any composite, score, confidence interval, percentile, then frame it as what the publisher says it is: a risk screener interpreted inside a comprehensive evaluation. Note the grade-band design where relevant, because the index draws on different subtests before and after grade 4, so index values from different bands describe different skill mixes. Pitfall: "the Dyslexia Index confirms a diagnosis of dyslexia." It flags risk; the diagnosis, where one is made, rests on the whole evaluation, and the word choice is the difference between a defensible report and an exhibit. Interpretive summary and referral linkage. Answer the referral question directly, state what the achievement data establish and what they do not, and place the WIAT-4 as one source among several: federal law requires that no single measure or assessment serve as the sole criterion for eligibility (34 CFR 300.304(b)(2)), and the eligibility determination carries its own documentation requirements. Pitfall: an eligibility verdict from an achievement score. Identification belongs to the team applying state criteria; the results section supplies calibrated evidence, not the ruling. Recommendations linkage. Tie each recommendation to a specific finding, keep instructional suggestions at the skill level the data support, and handle grade equivalents deliberately: report standard scores and percentiles, and if a family asks for a grade equivalent, provide it with the explanation that it is an interpolated score, not an instructional placement. Pitfall: recommendations written from grade equivalents. "Reads at a third-grade level, so provide third-grade texts" converts the test's least interpretable score into an instructional decision it cannot support. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: WIAT-4 Norms: [US / Canadian] [age / grade (season)] Subtests given: [list; paper battery has 16, Q-interactive 20] Format: [paper / Q-interactive] Scoring: [Q-global / Q-interactive] Norm-choice rationale: [age norms to match cognitive test / grade norms for attainment question / both, with reason] BEHAVIORAL OBSERVATIONS AND VALIDITY [Effort, attention, breaks, departures from standard procedure; statement that results are valid estimates, or the limitation] COMPOSITE SUMMARY (table) [Composite | standard score | 95% CI | percentile | descriptor; one descriptor system; note subtest metric is mean 100, SD 15] DOMAIN-BY-DOMAIN RESULTS (prose + error analysis) Reading / Basic Reading / Fluency: [scores in context + error patterns observed, e.g., decoding attempts, self-corrections] Mathematics: [same] Written Expression: [same] Oral Language: [same] COMPOSITE INTEGRITY NOTES [Within-composite splits named; interpretation kept at the level the data support; planned comparisons labeled supplementary] ABILITY-ACHIEVEMENT INTEGRATION [State SLD model named; cognitive anchor + shared norm frame; no discrepancy-formula-is-required language] DYSLEXIA INDEX (if given) [Score + CI + percentile; framed as risk screening within the comprehensive evaluation, never as a diagnosis] INTERPRETIVE SUMMARY (answer the referral question) [What the data establish and do not; single-measure limits; measurement error acknowledged] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; grade equivalents only with context, never as placement decisions] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample WIAT-4 write-up (fictional) Scenario: a school team refers a 9-year-old in grade 4 after two years of reading intervention with limited response. The achievement profile shows the word-level reading pattern that most often prompts a dyslexia question: low decoding, fluency, and phonological processing against intact oral language and mathematics. This is the achievement results section only, condensed but structurally complete. All details are fictional. Student: J.T., 9 (grade 4) · Referral: school student-support team, reading concerns after two intervention cycles · Evaluator: L. Okafor, PhD, Licensed Psychologist · Testing date: 07/10/2026 · Report date: 07/16/2026 Measures and administration: The Wechsler Individual Achievement Test, Fourth Edition (WIAT-4) was administered in paper format as part of a broader psychoeducational evaluation and scored on Q-global against the United States age-based norms. Age norms were selected so that achievement and cognitive results in this evaluation share the same reference frame. Subtests were selected for the referral question and sampled reading, mathematics, written expression, and oral language; the Dyslexia Index was also derived. WIAT-4 subtest and composite scores are standard scores with a mean of 100 and a standard deviation of 15, and each composite below is reported with its 90% confidence interval (CI), matching the intervals printed on Q-global score reports, and its percentile rank. Behavioral observations and validity: J.T. worked cooperatively across two sessions and took one scheduled break. He engaged readily with math and listening tasks. On reading tasks he slowed noticeably, subvocalized while working through unfamiliar words, and self-corrected several times using sentence context. Effort was consistent, administration followed standardized procedures, and the results are considered valid estimates of his current academic skills. Composite Standard score 90% CI Percentile Oral Language 104 99-109 61 Mathematics 101 96-106 53 Written Expression 88 83-93 21 Reading 82 78-86 12 Phonological Processing 81 77-86 10 Reading Fluency 79 74-84 8 Basic Reading 78 74-82 7 Decoding 76 72-80 5 Dyslexia Index 80 76-85 9 Total Achievement 88 85-91 21 Reading: J.T.'s word-level reading skills are well below age expectations. His Decoding composite of 76 (CI 72-80, 5th percentile) and Basic Reading composite of 78 (7th percentile) reflect consistent difficulty reading real words and pronounceable nonwords, and his Phonological Processing score of 81 (10th percentile) indicates the underlying sound-manipulation skills are similarly weak. The Reading composite of 82 (12th percentile) masks a split worth naming: word reading (subtest standard score 76, 5th percentile) sits well below reading comprehension (90, 25th percentile), which his strong oral language appears to support when passages are short. Reading Fluency of 79 (8th percentile) shows the downstream cost in rate. In error terms, J.T. attempted unfamiliar words by naming the first sound and predicting from word length, confused vowel teams, and read more accurately without time pressure; his self-corrections relied on sentence meaning rather than decoding. Mathematics and written expression: Mathematics of 101 (CI 96-106, 53rd percentile) is age-typical across computation and problem solving, and no error pattern emerged beyond occasional misread word problems that he solved correctly when reread aloud. Written Expression of 88 (21st percentile) is a relative weakness driven by spelling (subtest standard score 82), while his sentence-level composition (95) carried age-appropriate ideas; his spelling errors mirrored his decoding errors, phonetically plausible but orthographically incorrect. Oral language: Oral Language of 104 (CI 99-109, 61st percentile) is solidly age-typical. J.T. understood spoken passages and expressed ideas verbally at a level that contrasts sharply with his printed-word skills, the contrast his teachers describe in class. Dyslexia Index and integration: The Dyslexia Index of 80 (CI 76-85, 9th percentile) is a screening indicator, and this result flags elevated risk consistent with the pattern above; it is not by itself a diagnosis. The Total Achievement score of 88, which summarizes the reading, mathematics, and writing subtests (oral language is not part of it), is reported for completeness but averages across domains that differ sharply, so the domain composites carry the interpretation. Within this evaluation's pattern-of-strengths-and-weaknesses framework, these achievement results supply the academic-weakness evidence; they are read against the cognitive results reported separately, on the same age-norm frame, and against intervention-response and classroom data. Baseline Growth Scale Values were recorded for the reading subtests; per the publisher's guidance, change will be interpreted only after at least three months and across at least three data points. Summary: Achievement testing shows an age-typical oral-language and mathematics foundation against consistently low word-level reading skills: decoding at the 5th percentile, basic reading at the 7th, reading fluency at the 8th, with spelling weakness on the same phonological line and a Dyslexia Index in the elevated-risk range. This pattern is consistent with the referral concern about word-level reading. It does not by itself establish a specific learning disability: that determination belongs to the eligibility team integrating cognitive results, intervention response, observation, and classroom data, and no single measure may serve as the sole criterion. All scores are estimates that carry measurement error and describe current performance. Linkage to recommendations: The decoding and phonological findings support recommendation 1 (explicit, systematic instruction in word-level decoding and spelling, delivered with cumulative practice). The fluency finding supports recommendation 2 (a trial of extended time and audio access for content-area text, with the team monitoring whether comprehension holds as passages lengthen). The intact oral-language finding supports recommendation 3 (preserve access to grade-level ideas through discussion and read-aloud while word-level skills are remediated). Progress should be monitored with curriculum-based measures between administrations, with WIAT-4 re-administration reserved for the re-evaluation cycle. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, norm set and its basis (US age norms, with the rationale stated), subtests given, and administration format are all named, so the scores are interpretable now and comparable at re-evaluation. Every composite appears with its confidence interval and percentile, the prose matches the table, and the score metric is stated once, preventing the mean-10 scaled-score misreading that mixed cognitive-achievement reports invite. Interpretation leads with normative composites, the within-composite split (word reading versus comprehension) is surfaced rather than averaged away, and error-pattern observations give teachers something scores cannot. The Dyslexia Index is reported as screening evidence inside a comprehensive evaluation, no grade equivalents are interpreted, and the state's identification framework is named rather than assumed. The summary states what the data do not establish, honors the no-single-measure rule, and every recommendation traces to a specific finding, with progress monitoring assigned to the right tool. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label the strength of each requirement honestly. In US special education, federal law is instrument-neutral: no regulation names the WIAT-4, and evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ). For specific learning disability the state "must not require the use of a severe discrepancy" between ability and achievement, must permit an intervention-response process, and may permit other research-based procedures ( 34 CFR 300.307 ), so which model your write-up serves is state policy, and the same WIAT-4 data serve any of them. The eight achievement areas the regulation lists ( 34 CFR 300.309(a)(1) ) read like a WIAT-4 table of contents, and the technical manual itself says the Basic Reading composite "closely aligns with the definition of basic reading skills" in IDEA. State practice genuinely differs: a comprehensive review of state criteria found only 67% of states allowed the discrepancy approach and 20% prohibited it outright as of the 2013 policy year ( Maki, Floyd, and Roberson ), and the strengths-and-weaknesses model, adopted by at least 14 states, was found by a systematic review of the diagnostic-accuracy evidence to identify SLD "at the level of chance" ( School Psychology Review ), which is exactly why the report should name the framework it feeds instead of implying one is federally required. Two federal clarifications belong in every school evaluator's head: intervention response "cannot be used to delay or deny" a full evaluation ( OSEP Memo 11-07 ), and nothing in IDEA prohibits the words dyslexia, dyscalculia, and dysgraphia in evaluations, eligibility determinations, or IEPs ( OSERS letter, October 2015 ); the SLD definition names dyslexia itself ( 34 CFR 300.8(c)(10) ). Accommodations bodies read achievement testing on their own clocks: College Board policy asks that for learning disorders "the educational evaluation and testing should be no more than five years old" while cognitive testing may be older, an IEP or 504 plan alone does not transfer (students "will still need to request accommodations"), and full documentation review "can take up to seven weeks"; ACT's criteria for learning disorders expect a comprehensive battery including "Results of a complete achievement battery" from a credentialed evaluator. Funding is the quiet boundary: school SLD evaluations are education-funded under IDEA, while clinically indicated testing bills payers under the psychological-testing code family (96130 and 96131 for evaluation services, 96136 through 96139 for administration and scoring) as payer policy, and Medicare's statute "does not extend coverage to screening procedures" ( Billing Article A57481 ), so achievement testing run purely for educational eligibility is generally not a medical claim. Edition and norms currency is quieter here than on the cognitive side, and worth one honest paragraph. As of July 2026 Pearson's own product page states "The WIAT-4 is the most current version of the WIAT," no fifth edition has been announced anywhere, and any WIAT-5 date you encounter is speculation. The norms have a specific property reports should carry quietly in mind: the standardization sample of 1,832 was collected between October 2018 and February 2020, entirely before pandemic school closures ( Beaujean and Parkin, 2022 ), so a score today compares a student to pre-pandemic peers, which matters when instruction was disrupted. Administration time is not a fixed number: per Pearson it "varies by grade level and number of subtests administered," one more reason the report lists what was given. Jurisdiction determines the norms available. Canada has a Canadian-normed WIAT-4-CDN (normative copyright 2024) plus a French-Canadian adaptation, so Canadian reports should say which norm set scored the protocol, especially for records that span the 2024 transition. Australia and New Zealand have no WIAT-4 adaptation: Pearson's local catalogue lists the WIAT-III A&NZ as its standardised achievement edition (checked July 2026), so Australian clinicians either run the US-normed WIAT-4 and say so as a stated limitation, or use the older local standardisation, and Medicare's Better Access items, which fund treatment of a clinically diagnosed mental disorder, are not a funding pathway for educational assessment by policy. Access is one tier broader than the cognitive scales: the WIAT-4 is publisher qualification level B, which is why school psychologists, specialist teachers with assessment credentials, and allied assessors administer it in settings where a level-C instrument would be out of reach; eligibility to purchase is set by Pearson's local affiliate. For re-testing, use the scores built for it: Growth Scale Values track a student against their own past performance (anchored so the average grade-3 GSV is 500 on every subtest), they compare only within the same subtest, and Pearson's GSV guidance cautions against interpreting change when "fewer than 3 months have passed" between sessions and against reading a trend from fewer than three data points; standard scores answer the different question of standing relative to peers. Wechsler Individual Achievement Test, WIAT, Q-global, and Q-interactive are trademarks, in the US and other countries, of Pearson plc or its affiliates; the WIAT-4 is published by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common WIAT-4 write-up errors reviewers flag There is no payer audit series for achievement write-ups; the accountability literature here is psychometric, and it is current. The WIAT-4 is reviewed in the Buros Center's 22nd Mental Measurements Yearbook (published December 2025), the Journal of Psychoeducational Assessment review judged it "well-conceptualized, and generally psychometrically sound" while faulting the manual's missing factor analyses and item-analysis results, and the independent Journal of Intelligence evaluation ran the analyses the manual omits, recommended limiting composite interpretation to a few reading and mathematics scores, and advised psychologists to "completely refrain" from composite-score uses that require equal-interval values. Those findings converge on the same theme: the danger in WIAT-4 reporting is arithmetic that outruns the measurement. The BastionGPT Clinical Advisory Board sees the same errors most often in WIAT-4 report reviews: Grade equivalents written as instructional placements. "Reads at a 3rd-grade level" converts an interpolated, non-equal-interval score into a curriculum decision it cannot support, and it is the single most common misreading parents and teams take away. Report standard scores and percentiles; if a family asks for the grade equivalent, give it with its explanation. The wrong ruler. WIAT-4 subtests are standard scores with a mean of 100, unlike the mean-10 scaled scores on Wechsler cognitive batteries, and mixed psychoeducational reports invite readers to apply one test's metric, or one test's descriptor vocabulary, to the other. State each instrument's metric once and keep descriptor systems separate and labeled. "A severe discrepancy is required." Federal law says the opposite: states must not require it (34 CFR 300.307), and models genuinely vary by state. The paired error is running an ability-achievement or strengths-and-weaknesses analysis without naming the state framework, the cognitive estimate that anchors it, or the shared norm frame (age norms on both sides). The Dyslexia Index reported as a diagnosis. Pearson positions it as risk screening with "risk assessment, strength of risk" language, and its subtest composition changes at grade 4, so index values from different grade bands describe different skill mixes. Write it as screening evidence inside the comprehensive evaluation, never as the conclusion. Equal-interval and ipsative overreach. Strength-and-weakness tables promoted to findings, mechanical score-difference arithmetic, Growth Scale Values compared across different subtests, growth claims from two data points, or re-tests inside three months read as precision but contradict the published measurement evidence and the publisher's own GSV guidance. Interpret normatively first, and label supplementary analytics as supplementary. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your composite summary (scores, confidence intervals, percentiles) and get a drafted results-section narrative organized by academic domain, with the norm-set statement, error-pattern observations, and the state-model framing set up for your review. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with the table, metric or descriptor drift across a mixed battery, grade-equivalent statements written as placements, and a Dyslexia Index phrased as a diagnosis. Translate the results section into a plain-language summary for parents and teachers that keeps scores as estimates and explains what the eligibility team does next. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do WIAT-4 scores actually mean? One metric almost everywhere: WIAT-4 subtests and composites are both standard scores with a mean of 100 and a standard deviation of 15, which is worth stating in the report because Wechsler cognitive subtests use a mean-10 scaled metric and mixed batteries invite misreading. Each score carries a percentile rank, a confidence interval ( Pearson's sample report prints 90% intervals), and a descriptive category. Grade and age equivalents, stanines, and normal-curve equivalents exist for education audiences but carry the least interpretive weight, and Growth Scale Values sit on a separate absolute scale used only for progress comparison against the same student. Write every score as an estimate: number, interval, percentile together. Is the WIAT-4 the current edition, or is a WIAT-5 coming? The WIAT-4 is current everywhere as of July 2026, and no fifth edition has been announced: Pearson's own product page states "The WIAT-4 is the most current version of the WIAT," so treat any WIAT-5 date you read as speculation. What has changed recently is jurisdictional: Canada gained the Canadian-normed WIAT-4-CDN (normative copyright 2024) plus a French-Canadian adaptation, while Australia and New Zealand still have only the WIAT-III A&NZ as a local standardisation. Name the edition and the norm set in every report and the WIAT-5 question takes care of itself when it eventually arrives. Is the WIAT-4, or a discrepancy formula, required for SLD eligibility? Neither. Federal law names no instrument, forbids any single measure from being the sole criterion ( 34 CFR 300.304(b)(2) ), and prohibits states from requiring a severe ability-achievement discrepancy for SLD identification ( 34 CFR 300.307 ): states must permit an intervention-response process and may permit other research-based methods, and intervention response in turn "cannot be used to delay or deny" a full evaluation ( OSEP Memo 11-07 ). Achievement testing stays standard evidence in nearly every framework because the regulation defines SLD through eight achievement areas the WIAT-4 samples directly. Know your state's model and say which one the report feeds; note that the strengths-and-weaknesses model, though adopted by at least 14 states, fared poorly in a recent diagnostic-accuracy review , one more reason to present scores as evidence rather than verdict. How many subtests does the WIAT-4 have, and how many do I give? Twenty subtests, per Pearson's own comparison flyer (the WIAT-III had 16; five subtests are new, and one of the five, Orthographic Choice, runs only on Q-interactive). Counting scores rather than subtests, the instrument yields 32, which is why "how many subtests" answers around the web disagree with each other. How many you give is referral-driven: no authority sets a minimum or a required battery, and a dyslexia referral, a math referral, and a full psychoeducational evaluation each justify different selections. The write-up rule that follows is simple: list exactly which subtests were administered and which composites they produced, because "the WIAT-4 was administered" describes twenty different possible batteries. Does the Dyslexia Index diagnose dyslexia? No. Pearson positions it as a brief performance-based screener that provides "risk assessment, strength of risk, and interpretive information," taking about 5 minutes at the PK to grade 3 level and under 20 minutes for grades 4 through 12 and adults. Two details matter for write-ups. First, composition changes by grade band ( per Pearson , Phonemic Proficiency and Word Reading through grade 3; Word Reading, Orthographic Fluency, and Pseudoword Decoding from grade 4), so index values from different bands describe different skill mixes. Second, a low index flags risk to be examined inside a comprehensive evaluation, and the word "diagnosis" belongs to the whole evaluation, not the screener. If dyslexia is confirmed, federal guidance is explicit that nothing in IDEA prohibits using the term in evaluations, eligibility determinations, or IEPs. Should I report grade equivalents? Report them only deliberately, and interpret them almost never. A grade equivalent says a student's raw score matched the median raw score of students at some grade point; it is interpolated, it is not an equal-interval scale, and it says nothing about whether the student can do that grade's curriculum. "Reads at a 3rd-grade level" is the most repeated misreading of achievement testing, and instructional placement decisions built on it inherit the error. The defensible pattern: carry the meaning with standard scores, confidence intervals, and percentiles; when a parent or team asks for the grade equivalent, provide it alongside a one-sentence explanation of what it is and is not. The same logic applies to age equivalents, and it is why this page's sample report and templates lead with composites and percentiles throughout. WIAT-4, KTEA-3, or Woodcock-Johnson: which achievement battery? Choose by the cognitive test you are pairing with and the referral, because head-to-head validity studies are sparse. Pearson's product page states the WIAT-4 "links directly to the WISC-V and KABC-2 NU," which is the practical draw when a WISC-V anchors the evaluation and you want supported ability-achievement or strengths-and-weaknesses analyses on a shared reference frame. The KTEA-3 (also Pearson, norms for ages 4:0 to 25:11) is the sibling battery with its own Dyslexia Index, and the Woodcock-Johnson achievement tests pair with their own cognitive battery for CHC-style cross-battery work. The WIAT-4's distinctive additions are the phonological, orthographic, and fluency processing measures and automated essay scoring. Whichever you choose, the report should say why in one sentence, and for repeat testing, staying within one battery family keeps Growth Scale Values usable. Can I see the actual WIAT-4 items, or include them in a report? No. Test items, stimuli, and record forms are protected test materials: purchase and use are restricted by publisher qualification (the WIAT-4 is level B), and evaluators are ethically obligated to protect test security under APA Ethics Standard 9.11 and their publisher agreements. Item exposure also damages the instrument itself, because norms assume examinees have not rehearsed the tasks, and achievement items leak especially easily into test-prep circulation. Reports should carry scores, error-pattern description, and interpretation, never item content: "he confused vowel teams on unfamiliar words" documents the skill; quoting the words he misread would document the test. Records requests that touch protocols route through your test-security obligations, and what families actually need, the scores explained in plain language with next steps, belongs in the report anyway. What is the WIAT-4 age range? The WIAT-4 covers ages 4 years 0 months through 50 years 11 months, one of the widest ranges in academic assessment, and can be scored against either age-based norms or grade-based norms (Pre-K through 12+, with fall, winter, and spring windows). The choice matters and belongs in the report: age norms compare the examinee with same-age peers regardless of schooling, while grade norms compare with same-grade peers, and the two can classify the same performance differently for students who are old or young for their grade, retained, or accelerated. Canadian evaluations can draw on the Canadian-normed WIAT-4-CDN. Past 50:11 the WIAT-4 simply has no norms, so adult evaluations beyond that age move to an achievement measure normed for the range, and the report says so rather than stretching the closest table. Can BastionGPT draft my WIAT-4 results section? Yes. Paste your composite summary (scores, confidence intervals, percentiles) and it drafts the results-section narrative for your review: organized by academic domain, functional meaning in prose, the norm-set and state-model statements in place, and grade equivalents kept out of the interpretation. It can also cross-check a draft you wrote for score-versus-narrative mismatches, metric confusion, and a Dyslexia Index phrased as a diagnosis, and it can produce a plain-language summary of findings for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, WIAT-4 product page : "The WIAT-4 is the most current version of the WIAT," ages 4:0 to 50:11, prekindergarten through grade 12, administration time that "varies by grade level and number of subtests administered," Q-global and Q-interactive delivery, and the WISC-V and KABC-2 NU linkage. Pearson, Get the 411 on WIAT-4 and What's New on the WIAT-4 flyers: 2020 publication, qualification level B, 20 subtests versus the WIAT-III's 16, the five new subtests with Orthographic Choice on Q-interactive only, the new processing composites, and automated Essay Composition scoring. Pearson, WIAT-4 sample score report : score-report structure and the 90% confidence-interval convention. Pearson, WIAT-4 Dyslexia Index product page and Dyslexia Index scores flyer : screener positioning ("risk assessment, strength of risk"), administration times, and the grade-band subtest composition. Pearson, Growth Scale Values overview : same-subtest comparison rules, the three-month practice-effect caution, and the three-data-point trend rule. Pearson Canada, WIAT-4-CDN product page : the 2024 Canadian normative edition; Pearson Clinical AU&NZ, WIAT-III A&NZ product page : the Australian and New Zealand edition status. Beaujean & Parkin, 2022, Journal of Intelligence : the 32-score architecture, the 1,832-person norming sample collected October 2018 to February 2020, the reliability values cited from the technical manual, the independent factor analyses, and the equal-interval recommendation. Dombrowski & Casey, 2022, Journal of Psychoeducational Assessment : the "well-conceptualized, and generally psychometrically sound" verdict and the noted omissions; Buros Center for Testing, tests reviewed in the 22nd Mental Measurements Yearbook : independent review status. 34 CFR 300.304 , 300.307 , 300.309 , and 300.8 : the no-single-measure rule, the severe-discrepancy prohibition, the eight SLD achievement areas, and the SLD definition. OSEP Memo 11-07 (2011) : intervention response cannot delay or deny an evaluation; OSERS Dear Colleague Letter (October 23, 2015) : dyslexia terminology in IDEA documents. College Board, learning-disorder documentation guidelines : the five-year educational-testing currency line and request requirements; ACT, criteria for diagnostic documentation : the complete achievement battery requirement. Maki, Floyd & Roberson, 2015, School Psychology Quarterly : state SLD-criteria variability; Dombrowski, Benson & Maki, School Psychology Review : the PSW diagnostic-accuracy systematic review; CMS, Billing Article A57481 : the psychological-testing code family and the Medicare screening exclusion. ## Related templates WISC-V Report Write-Up BASC-3 Report Write-Up Psychological Evaluation Report Neuropsychological Evaluation Report Psychiatric Diagnostic Evaluation Woodcock-Johnson (WJ IV / WJ V) Report Write-Up Psychoeducational Evaluation Report Beery VMI Report Write-Up CELF-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/wisc-v-report All note templates # WISC-V Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The WISC-V (Wechsler Intelligence Scale for Children, Fifth Edition) is an individually administered intelligence test for ages 6 to 16, yielding a Full Scale IQ and five index scores covering verbal comprehension, visual spatial, fluid reasoning, working memory, and processing speed. Psychologists use it in special-education, learning-disability, and disability-benefit evaluations. This page covers how to write up WISC-V results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and neuropsychologists; publisher qualification level C Audience IEP and Section 504 teams, parents, physicians, SSA and NDIS reviewers, courts Typical length 500 to 900 words for the results section · administration about 60 minutes (10 primary subtests) Format family Norm-referenced cognitive battery When it's used Special-education and SLD evaluations, intellectual disability and gifted assessment, disability determinations, ages 6 to 16 Standards context Published by NCS Pearson (2014); WISC-6 in standardization; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the WISC-V? The Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V) is an individually administered, norm-referenced intelligence test for ages 6:0 to 16:11, published by NCS Pearson in 2014 as the fifth edition of a battery that began with the original WISC in 1949. Ten primary subtests yield five primary index scores, Verbal Comprehension (VCI), Visual Spatial (VSI), Fluid Reasoning (FRI), Working Memory (WMI), and Processing Speed (PSI), and seven of them yield the Full Scale IQ; ancillary indexes including the General Ability Index (GAI) and Cognitive Proficiency Index (CPI) serve specific interpretive questions. Composites have a mean of 100 and a standard deviation of 15, subtest scaled scores a mean of 10, and results carry percentile ranks and 95% confidence intervals. Administration is paper or digital via Q-interactive , with scoring on Q-global or by hand, and separate Canadian and Australian and New Zealand standardised editions carry local norms. Edition status matters as of July 2026: the WISC-V is current, and a WISC-6 is in active development, with Pearson's US product page announcing "WISC-6 is coming," its November 2025 field-research Q&A stating the standardization "will continue for the next 10 months," and a Canadian WISC-6 standardization expected to run Spring 2026 to early 2027. No release date has been announced in any country. The load-bearing question for write-ups is interpretation altitude. Pearson markets five interpretable primary indexes, but independent factor-analytic studies of the US norms ( exploratory and confirmatory ) and of the UK and German versions repeatedly find that general ability dominates what the subtests measure, that the index scores carry limited unique information beyond it, and, in the UK analysis, that the factor-level scores were "of questionable interpretive value independent of g." Subtest-level and profile-scatter interpretation has failed psychometric scrutiny for decades. The defensible posture, and the one this page teaches, interprets the FSIQ and the index pattern cautiously, treats subtest commentary as bounded qualitative description, and states explicitly which ability estimate anchors any ability-achievement comparison. A WISC-V score is also evidence, not a verdict: eligibility categories belong to teams and agencies applying whole-evaluation criteria , and the results section's job is to feed that decision with calibrated, honestly bounded numbers. ## Who uses WISC-V write-ups and when School psychologists write up WISC-V results inside special-education eligibility evaluations; clinical child psychologists use it for learning-disability, intellectual-disability, and gifted referrals; and pediatric neuropsychologists fold it into a neuropsychological report as the cognitive core alongside memory, executive, and achievement measures. It also supplies the cognitive evidence in Social Security childhood disability claims and, in Australia, in NDIS access requests. Age drives instrument choice at the edges: below age 6 the WPPSI-IV covers early childhood, and at the 16-year overlap convention favors the WISC-V when lower ability is suspected (better floor) and the WAIS-5 , published in 2024, for high-ability adolescents (better ceiling). When language or hearing complicates verbal testing, nonverbal batteries or the WISC-V's own Nonverbal Index carry the load. Cognitive results rarely stand alone: they anchor a psychological evaluation report next to achievement testing , behavior ratings such as the BASC-3 , and, where intellectual disability is the question, adaptive-behavior measures that the law makes mandatory, not optional. ## How to structure a WISC-V results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it names the data before interpreting it, presents composites in a table and meaning in prose, and makes the one genuinely contested move in WISC-V reporting, what to do when the index scores diverge, an explicit stated decision instead of a silent one. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and format. Name the instrument and edition, the norm set (US, Canadian, or Australian and New Zealand standardisation), which subtests were administered, the administration format (paper or Q-interactive), and the scoring platform. Pitfall: writing "the Wechsler" or "the WISC" with no edition or norms. With Pearson's own product page announcing " WISC-6 is coming " as of July 2026, an undated, edition-less report will be ambiguous within a few years. Behavioral observations and validity of the session. Describe effort, engagement, language considerations, breaks, and anything that departed from standardized procedure, then state plainly whether you consider the results valid estimates of current functioning. Pitfall: boilerplate observations that contradict the interpretation. "Rapport was easily established" followed by index scores you then attribute to disengagement gives a reviewer two incompatible reports in one document. Composite summary table. Report each composite with its standard score, 95% confidence interval, percentile rank, and qualitative descriptor, and keep descriptors from one labeled system. Pearson's own sample reports model the convention: scores from 90 to 109 are typically described as Average, and the WISC-V descriptor ladder runs from Extremely Low through Extremely High. Pitfall: bare standard scores. A score without its confidence interval and percentile invites over-precision, and agency readers expect both; SSA's childhood listings, for example, specify individually administered, standardized measures reported by a qualified specialist. Index-by-index results in prose. Move through Verbal Comprehension, Visual Spatial, Fluid Reasoning, Working Memory, and Processing Speed, and spend the prose on what each result means functionally for this child, not on re-reading the table aloud. Pitfall: the score dump. A paragraph per subtest in administration order documents that testing occurred; it does not answer the referral question, and it drifts toward the subtest-level interpretation the literature warns against. FSIQ, GAI, and the interpretive decision. Pearson's reports describe the FSIQ as derived from seven subtests, and the independent literature finds it the best-measured score on the test. When the primary indexes diverge widely, convention offers the General Ability Index (verbal, visual spatial, and fluid reasoning subtests) with the Cognitive Proficiency Index (working memory and processing speed) as a paired supplement. Whatever you choose, state the choice and the reason in the report. Pitfall: silent substitution. A GAI that appears without a stated rationale, or an FSIQ quietly dropped because of scatter, reads as score-shopping; note that the practice of discarding the FSIQ for variability has itself been seriously questioned . Subtest-level commentary, tightly bounded. If you discuss subtests at all, keep it descriptive and qualitative, and say why: individual subtest scores are less reliable than composites, and ipsative strength-and-weakness profiling has failed to hold up psychometrically for decades. Pitfall: diagnostic conclusions from scatter. "The Coding weakness indicates ADHD" is the single move most likely to draw fire from a reviewing psychologist, because the evidence says subtest patterns cannot carry that weight. Interpretive summary and referral linkage. Answer the referral question directly, state what the cognitive data do and do not establish, and place the WISC-V explicitly as one source among several: federal special-education law requires that no single measure be the sole criterion for eligibility ( 34 CFR 300.304(b)(2) ). Pitfall: an eligibility verdict from an IQ score. Identification decisions belong to teams and agencies applying their criteria; the results section supplies calibrated evidence, not the ruling. Recommendations linkage. Tie each recommendation to a specific finding, and where cognitive results drove it, say which ones. Pitfall: recommendations that could follow any profile. If extended time, chunked instructions, or assistive supports do not trace back to the working memory and processing speed findings, either the recommendations or the results section is incomplete. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: WISC-V Norms: [US / Canadian / A&NZ] Subtests given: [10 primary / 7 FSIQ / list ancillary] Format: [paper / Q-interactive] Scoring: [Q-global / manual] BEHAVIORAL OBSERVATIONS AND VALIDITY [Effort, engagement, language, breaks, departures from standard procedure; statement that results are valid estimates, or the stated limitation if not] COMPOSITE SUMMARY (table) [Composite | standard score | 95% CI | percentile | descriptor, one labeled descriptor system used throughout] INDEX-BY-INDEX RESULTS (prose, functional meaning) Verbal Comprehension: [score context + what it means here] Visual Spatial / Fluid Reasoning: [same] Working Memory / Processing Speed: [same] FSIQ / GAI INTERPRETIVE DECISION [FSIQ with CI; if indexes diverge, the ability estimate you are using, WITH the stated rationale; GAI-CPI contrast described qualitatively, no reproduced base-rate tables] SUBTEST COMMENTARY (only if needed, qualitative, bounded) INTERPRETIVE SUMMARY (answer the referral question) [What the data establish and what they do not; single-measure limits; measurement error acknowledged] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-evaluation plan] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample WISC-V write-up (fictional) Scenario: a school team refers a 9-year-old for evaluation after two years of reading interventions with limited progress. The cognitive profile splits, average-or-better reasoning over notably lower working memory and processing speed, the shape that most often forces the FSIQ-versus-GAI decision. This is the cognitive results section only, condensed but structurally complete. All details are fictional. Student: M.R., 9 (grade 4) · Referral: school student-support team, reading and work-completion concerns · Evaluator: D. Whitfield, PhD, Licensed Psychologist · Testing date: 07/09/2026 · Report date: 07/15/2026 Measures and administration: The Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V) was administered in paper format as part of a broader psychoeducational evaluation. All ten primary subtests were given and scored against the United States normative sample, yielding the Full Scale IQ (FSIQ), the five primary index scores, and, given the pattern described below, the General Ability Index (GAI) and Cognitive Proficiency Index (CPI). Composite scores have a mean of 100 and a standard deviation of 15; each is reported with its 95% confidence interval (CI), percentile rank, and qualitative descriptor. Behavioral observations and validity: M.R. arrived rested, transitioned easily, and worked persistently through both sessions of testing, asking for one scheduled break. He engaged readily with reasoning tasks, and his effort was consistent. On timed paper-and-pencil tasks he worked slowly and carefully, checking his work; this pace was consistent across tasks and does not appear to reflect momentary inattention. Administration followed standardized procedures, and the results below are considered valid estimates of his current cognitive functioning. Composite Standard score 95% CI Percentile Descriptor Verbal Comprehension (VCI) 110 103-115 75 High Average Visual Spatial (VSI) 102 95-109 55 Average Fluid Reasoning (FRI) 105 98-111 63 Average Working Memory (WMI) 82 76-90 12 Low Average Processing Speed (PSI) 80 74-91 9 Low Average Full Scale IQ (FSIQ) 96 91-101 39 Average General Ability (GAI) 106 100-112 66 Average Cognitive Proficiency (CPI) 79 74-88 8 Very Low Index results: M.R.'s verbal reasoning is a relative strength. His Verbal Comprehension score (110, High Average) indicates he expresses word knowledge and verbal concepts as well as or better than about 75 percent of children his age, consistent with his teachers' description of strong class discussion. His Visual Spatial (102) and Fluid Reasoning (105) scores fall solidly in the Average range: he analyzes visual detail, constructs designs, and identifies underlying conceptual rules at an age-typical level. In contrast, his Working Memory score (82, Low Average, 12th percentile) indicates difficulty holding and mentally reworking information such as number sequences, and his Processing Speed score (80, Low Average, 9th percentile) indicates slow, effortful performance on simple timed visual-clerical tasks even with accurate work. Functionally, this is the profile of a student who can reason through grade-level ideas but loses instructions delivered in long strings, and who produces written work slowly enough that assignments go unfinished. Overall ability and the interpretive decision: M.R.'s FSIQ of 96 (CI 91-101, 39th percentile, Average) is reported for completeness, but it averages across abilities that differ sharply and so describes no single ability he actually shows. The contrast between his General Ability Index of 106 (reasoning-focused, with reduced working memory and processing speed contribution) and his Cognitive Proficiency Index of 79 is large, and a difference of this size occurs in a small minority of children with similar overall scores. For the ability-achievement comparisons in this evaluation, the GAI is used as the estimate of reasoning ability, because the question those comparisons ask is whether achievement is below expectation given his reasoning, and the FSIQ would fold the areas of difficulty into the expectation itself. This is a stated interpretive choice, not a replacement of one score by another: both are reported above, and the FSIQ remains the more broadly validated summary score. Summary: Cognitive testing shows High Average verbal reasoning and Average visual-spatial and fluid reasoning over Low Average working memory and processing speed, with a large, uncommon gap between reasoning (GAI 106) and cognitive proficiency (CPI 79). This pattern is consistent with the referral concerns about reading fluency and work completion. It does not by itself establish or rule out a specific learning disability: that determination integrates achievement testing, intervention response, and classroom data, and no single measure may serve as the sole criterion. All scores carry measurement error and describe current performance, not fixed capacity. Linkage to recommendations: The working memory finding supports recommendations 1 and 2 (instructions chunked and paired with visual supports; key directions repeated or written). The processing speed finding supports recommendation 3 (a trial of extended time and reduced copying demands, with the team monitoring whether completion improves). The GAI-CPI contrast supports recommendation 4: the eligibility team should read achievement results against the GAI-based expectation, alongside the intervention-response data, in making its determination. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, norm set, subtests given, and administration format are all named, so the scores are interpretable now and comparable at re-evaluation, and the report stays unambiguous once the WISC-6 arrives. Every composite appears with its confidence interval, percentile, and descriptor, every number in the prose matches the table, and one descriptor system is used throughout. The FSIQ-versus-GAI decision is made in the open: both scores are reported, and the GAI anchor for ability-achievement comparison comes with its rationale in the report itself. Index results are written as functional meaning for this student, and subtest-level claims are absent, matching what the measurement evidence can actually support. The summary states what the data do not establish, honors the no-single-measure rule, and every recommendation traces to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label the strength of each requirement honestly. In US special education, federal law is instrument-neutral: no regulation names the WISC-V, evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ), and for specific learning disability the state "must not require the use of a severe discrepancy" between ability and achievement ( 34 CFR 300.307 ), so whether your state runs on discrepancy, intervention response, or a strengths-and-weaknesses pattern is state policy, and the same WISC-V data serve any of them. The federal intellectual-disability definition pairs "significantly subaverage general intellectual functioning" with concurrent "deficits in adaptive behavior" ( 34 CFR 300.8(c)(6) ), which is law, not preference: an IQ score alone cannot make that identification. Social Security's childhood intellectual-disorder listing keys on a "full scale (or comparable) IQ score of 70 or below," or 71 to 75 with a part score of 70 or below, administered by a qualified specialist ( Listing 112.05 ); the "or comparable" wording means the FSIQ specifically is not the only acceptable composite, and courts have separately rejected treating rigid IQ cutoffs as decisive without accounting for measurement error (Hall v. Florida, 2014). Gifted identification has no federal mandate at all; score thresholds are district and state convention. Testing time is billed to payers under the psychological-testing code families (96130 and 96131 for evaluation services; 96136 through 96139 for administration and scoring) as payer policy with plan-specific units and authorization rules. For accommodations bodies, College Board policy asks that the educational evaluation be no more than five years old while "cognitive testing may be older than five years," with the caveat that testing before third grade may not reflect current ability, and evaluators must be licensed. Edition and norms currency is the WISC-V's live defensibility question. As of July 2026 the WISC-V remains the current edition, and the WISC-6 is in the field: Pearson's product page carries a "WISC-6 is coming" recruitment banner, its November 2025 field-research Q&A says the standardization "will continue for the next 10 months" and confirms a real administration change (verbal query procedures removed, because "querying appropriately is actually the number one error" examiners make), and the Canadian standardization is slated for Spring 2026 to early 2027. Pearson has announced no release date, and its timelines can slip, as the WAIS-5's did before its 2024 publication. Until a WISC-6 publishes, testing on the WISC-V is fully defensible; what ages a report is not the edition but the omissions, so name the edition and norm set every time, treat any future cross-edition comparison as a stated limitation, and expect referral sources to ask about the transition. Norms currency has a quantified side: standardization samples age, and the largest meta-analysis of the Flynn effect estimates scores drift about 2.31 points per decade (2.93 for modern Wechsler and Stanford-Binet tests), which is one more reason re-evaluations should compare like with like. Administration format is worth one line in the report: Pearson's own equivalence study of Q-interactive and paper found "none of the effect sizes exceeded the criterion of 0.20," about half a scaled-score point ( Q-interactive Technical Report 8 ), publisher-conducted evidence that supports treating formats as comparable while still naming the format used. Access and use boundaries are the publisher's: the WISC-V is qualification level C, restricted to doctorally trained or equivalently credentialed professionals, and in Canada and Australia administration runs through provincially regulated psychologists and the A&NZ standardised edition respectively, with Australian school-adjustment and NDIS processes treating cognitive results as supporting evidence alongside functional and adaptive documentation by convention. Retest timing is convention too: practice effects make re-administration within about a year hard to interpret, and school re-evaluation cycles usually make the question moot. Wechsler Intelligence Scale for Children, WISC, and Pearson are trademarks, in the US and other countries, of Pearson Education, Inc. or its affiliates; the WISC-V is published by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common WISC-V write-up errors reviewers flag There is no payer audit series for cognitive write-ups; the accountability literature here is psychometric, and it is unusually pointed. The WISC-V was reviewed in the Buros Center's 20th Mental Measurements Yearbook , independent factor analyses across the US , UK , and German norms found the five marketed indexes hard to replicate and of limited value independent of general ability, and even Pearson's WISC-6 materials concede that "querying appropriately is actually the number one error" in current administrations. Administration errors are the publisher's problem to engineer away; the errors below are write-up errors, and they are yours. The BastionGPT Clinical Advisory Board sees the same ones most often in WISC-V report reviews: Subtest scatter written as diagnosis. "The low Coding score indicates ADHD" or a strengths-and-weaknesses table promoted to clinical findings. Individual subtests are the least reliable scores on the test, and ipsative profile analysis has failed validation for thirty years; keep subtest commentary qualitative and bounded, and let composites carry the interpretive weight. The FSIQ-GAI move made silently. A GAI appears without rationale, or the FSIQ vanishes because the indexes diverged. Both directions draw fire: unexplained substitution reads as score-shopping, and reflexive invalidation of the FSIQ for scatter has itself been challenged in the literature . Report both, state the choice, give the reason. Descriptor drift and mixed vocabularies. A 110 is High Average in the table and "superior" in the prose, or WISC-IV-era labels (Borderline, Very Superior) leak into a WISC-V report. Reviewers read for one labeled descriptor system applied consistently; drift makes a sound report look careless and makes scores sound better or worse than they are. Bare standard scores. Composites reported without confidence intervals or percentiles, presenting measured estimates as exact points. Agency readers expect the full triplet, and the interpretive summary should say plainly that scores carry error, especially anywhere a threshold decision (SSA, eligibility, gifted cutoffs) is downstream. Edition, norms, and format unnamed. "The WISC" with no edition, no norm set (US, Canadian, A&NZ), and no administration format. With the WISC-6 in standardization, norms aging measurably, and paper and Q-interactive both in circulation, an unanchored score set is uninterpretable at re-evaluation and indefensible under cross-examination. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your composite summary (scores, CIs, percentiles, descriptors) and get a drafted results-section narrative organized index by index, with the FSIQ-GAI interpretive decision framed for your review and rationale. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with the table, descriptor drift or mixed edition vocabularies, and ability estimates substituted without a stated reason. Translate the results section into a plain-language summary for parents and teachers that keeps descriptors as comparison language and scores as estimates, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do WISC-V scores actually mean? Three score types do the work. Composites (the FSIQ and the index scores) are standard scores with a mean of 100 and a standard deviation of 15; subtest scaled scores have a mean of 10, with roughly 8 to 12 typical; and every composite carries a percentile rank and a 95% confidence interval. Qualitative descriptors translate ranges into words, and Pearson's own sample reports model the convention: scores from 90 to 109 are typically described as Average, with the ladder running from Extremely Low through Extremely High. Write scores as estimates: the confidence interval and percentile belong next to every composite, and the descriptor should never appear without the number it describes. Is the WISC-V still current, or should I wait for the WISC-6? Keep testing on the WISC-V. As of July 2026 it is the current edition everywhere, and the WISC-6 has no announced release date in any country. What is public: Pearson's US product page carries a "WISC-6 is coming" field-research banner, the November 2025 examiner Q&A says the US standardization "will continue for the next 10 months" and that the research edition removes verbal query procedures, and the Canadian standardization is expected to run Spring 2026 to early 2027. Publisher timelines also slip, as the WAIS-5's did before its 2024 release. The practical rule: test on the current edition, name it in the report, and treat any future cross-edition comparison as a stated limitation. Do I have to report the FSIQ? No authority mandates it by name, and the strongest-sounding requirement actually says otherwise: SSA's childhood intellectual-disorder listing keys on a "full scale (or comparable) IQ score" ( Listing 112.05 ), and the "or comparable" wording admits an equivalent composite. Convention still favors reporting the FSIQ, because it is the best-measured score on the test. When the indexes diverge widely, the defensible pattern is to report the FSIQ, note the divergence, and state which estimate anchors your ability-achievement comparisons and why, often the GAI. Be aware the underlying practice is contested: the argument that scatter invalidates the FSIQ has been seriously challenged , which is exactly why the report should show its reasoning instead of silently swapping scores. Is the WISC-V required for an IEP or SLD eligibility? No. Federal law names no instrument, requires that no "single measure or assessment" be the sole criterion for eligibility ( 34 CFR 300.304(b)(2) ), and for SLD it forbids states from requiring an ability-achievement discrepancy at all ( 34 CFR 300.307 ); states choose among discrepancy, intervention response, and pattern-of-strengths-and-weaknesses frameworks. Cognitive testing remains standard evidence in most of them, which is convention and state policy, not federal mandate. For intellectual disability the law is stricter in the other direction: the definition itself pairs subaverage intellectual functioning with adaptive-behavior deficits ( 34 CFR 300.8(c)(6) ), so a WISC-V without an adaptive measure cannot carry that identification. WISC-V or WAIS-5 for a 16-year-old? Both cover the overlap (the WISC-V runs to 16:11), so the choice is about measurement room. Convention follows the floor-and-ceiling logic: choose the WISC-V when lower ability is suspected, because a test built for children has more items at the low end to differentiate with, and choose the WAIS-5 (published 2024) for a high-ability 16-year-old who would hit the WISC-V's ceiling. Practical factors count too: if a re-evaluation will happen in adulthood, starting the adult battery avoids an instrument switch. At the other edge, the WPPSI-IV covers early childhood with overlap at age 6. Whichever you choose, say why in one sentence; reviewers read instrument selection at boundary ages as a marker of care. Should I interpret individual subtest scores or scatter? Sparingly, qualitatively, and never diagnostically. Individual subtests are the least reliable scores on the test, and the case against ipsative profile analysis is one of the most replicated findings in the assessment literature, argued since McDermott, Fantuzzo, and Glutting's 1990 "Just Say No to Subtest Analysis." The modern factor-analytic work points the same direction at the index level: independent analyses of the US , UK , and German versions find general ability dominates and the five marketed indexes carry limited unique information. Composites first, indexes cautiously, subtests as bounded qualitative observation. A sentence like "performance was slower on paper-and-pencil tasks, consistent with teacher reports" is defensible; "the Coding weakness indicates ADHD" is not. Can I see the actual WISC-V questions, or include them in a report? No. Test items, stimuli, and record forms are protected test materials: access is restricted to qualified professionals (the WISC-V is Pearson qualification level C), and psychologists are ethically obligated to protect test security under APA Ethics Standard 9.11 and their publisher agreements. Item exposure also damages the test itself, because norms assume examinees have not rehearsed the tasks. Reports should carry scores, descriptors, and interpretation, never item content; records requests that touch raw protocols route through your test-security obligations rather than reflexive printing. What parents can get is better anyway: the scores with plain-language explanation of what each index means and what happens next, which is what the parent-summary section of a good report is for. Is Q-interactive administration equivalent to paper? The published evidence says effectively yes, with a caveat about its source. Pearson's equivalence study ( Q-interactive Technical Report 8 , 2014) reports that "none of the effect sizes exceeded the criterion of 0.20," with the two largest at +0.20 (Block Design) and -0.20 (Comprehension), and the report itself translates 0.2 as slightly more than half a scaled-score point. The caveat: this is publisher-conducted research. Practice treats the formats as interchangeable, and the norms apply to both; the write-up convention is simply to name the format used, so that any future examiner reading your scores knows exactly how they were collected. Can BastionGPT draft my WISC-V results section? Yes. Paste your composite summary (scores, confidence intervals, percentiles, descriptors) and it drafts the results-section narrative for your review: organized index by index, functional meaning in prose, the FSIQ-GAI decision framed with a stated rationale, and descriptors kept to one consistent system. It can also cross-check a draft you wrote for score-versus-narrative mismatches and descriptor drift, and produce a plain-language summary of the findings for parents and teachers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, WISC-V product page : 2014 publication, ages 6:0 to 16:11, qualification level C, core-subtest time of about 60 minutes, paper and Q-interactive administration, Q-global and manual scoring, and the "WISC-6 is coming" field-research banner. Pearson, WISC-6 Q&A session recap (November 5, 2025) : the 10-month standardization statement and the removal of verbal query procedures; Pearson Clinical Canada, WISC-6 Canadian Edition standardization study : Spring 2026 to early 2027 timeline. Pearson, WISC-V interpretive considerations sample report : the seven-subtest FSIQ, ten primary subtests, descriptor conventions including the 90-to-109 Average range, and GAI-CPI comparison practice. Pearson, Q-interactive Technical Report 8 (2014) : the paper-digital equivalence findings; Pearson, WISC-V A&NZ product page : the 2016 Australian and New Zealand standardised edition. 34 CFR 300.304 , 34 CFR 300.307 , and 34 CFR 300.8 : the no-single-measure rule, the severe-discrepancy prohibition, and the intellectual-disability definition. Social Security Administration, childhood mental-disorder listings (112.00) : the 112.05 full-scale-or-comparable IQ criteria and qualified-specialist requirement. College Board, learning-disorder documentation guidelines : the five-year currency policy and its cognitive-testing exception. Buros Center for Testing, Tests reviewed in the Twentieth Mental Measurements Yearbook : independent review status of the WISC-V. Canivez, Watkins & Dombrowski, 2016 ( exploratory ) and 2017 ( confirmatory ) factor analyses, Psychological Assessment; Canivez, Watkins & McGill, 2019, British Journal of Educational Psychology (UK edition); Canivez et al., 2021, Assessment (German edition): the five-factor replication findings. McGill, 2016, Archives of Assessment Psychology : the counter-argument on invalidating the FSIQ for factor-score variability. Trahan, Stuebing, Fletcher & Hiscock, 2014, Psychological Bulletin : the Flynn-effect meta-analysis (2.31 points per decade; 2.93 for modern Wechsler and Stanford-Binet tests). Pearson, WISC-V Technical and Interpretive Manual Supplement : trademark and copyright attribution; Pearson, WAIS-5 overview brochure : the 2024 adult-scale publication. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report BASC-3 Report Write-Up Psychiatric Diagnostic Evaluation WAIS-5 / WAIS-IV Report Write-Up WIAT-4 Report Write-Up ADOS-2 Report Write-Up Vineland-3 Report Write-Up Woodcock-Johnson (WJ IV / WJ V) Report Write-Up Psychoeducational Evaluation Report WPPSI-IV Report Write-Up NEPSY-II Report Write-Up IEP Psychoeducational Input Beery VMI Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/woodcock-johnson-report All note templates # Woodcock-Johnson Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The Woodcock-Johnson is an individually administered, norm-referenced battery of cognitive and achievement tests built on Cattell-Horn-Carroll theory, currently sold in two editions: the WJ IV (2014) and the digital WJ V (2025). Psychologists, school psychologists, and educational diagnosticians use it in special-education, learning-disability, gifted, and accommodation evaluations. This page covers how to write up Woodcock-Johnson results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and educational diagnosticians; graduate-level qualification required by publisher policy Audience IEP and Section 504 teams, parents, physicians, gifted programs, postsecondary disability offices Typical length 500 to 900 words for the results section · administration about 5 to 10 minutes per test administered Format family Norm-referenced cognitive and achievement battery When it's used SLD and psychoeducational evaluations, gifted identification, adult accommodation documentation, three-year re-evaluations Standards context Published by Riverside Insights; WJ V (2025) and WJ IV (2014) both current; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Woodcock-Johnson? The Woodcock-Johnson is a family of individually administered, norm-referenced tests of cognitive abilities and academic achievement, first published in 1977 and now sold by Riverside Insights in two editions at once: the WJ IV (2014, paper-based, ages 2 to 90+) and the WJ V (released February 2025, fully digital, ages 4 through 90+, normed on a 2020-Census-aligned sample collected from 2022 to 2024). It is the most explicit operationalization of Cattell-Horn-Carroll (CHC) theory among the major batteries, and its distinctive design decision is co-norming: the cognitive and achievement batteries share one normative sample, so ability-achievement comparisons run inside a single reference group instead of across linked but separate tests. Alongside the familiar standard scores (mean 100, SD 15) and percentile ranks, WJ reports can carry two metrics most other batteries lack, the Rasch-based W score that underlies all WJ scoring and the criterion-referenced Relative Proficiency Index (RPI), which is written against 90 and predicts the examinee's success rate on tasks typical age or grade peers perform with 90% success. The load-bearing fact for write-ups, as of July 2026, is that the WJ V is not a reskinned WJ IV, and both editions are in active use with no US retirement announcement for the older one. Per the publisher's own technical documentation , the WJ V recomputed the General Intellectual Ability composite as eight equally weighted tests, removed Auditory Processing (Ga) from the composite and the cognitive battery entirely, split the WJ IV's Long-Term Retrieval cluster into Long-Term Storage (Gl) and Retrieval Fluency (Gr), rebuilt Fluid Reasoning on new tests (Matrices and Verbal Analogies), and eliminated the standalone Oral Language battery, moving those measures into the achievement battery and a new 15-test Virtual Test Library. Scores from the two editions are therefore not interchangeable: the normative samples sit more than a decade apart and the composite and several clusters are built from different tests, which makes the edition-and-norms statement the single highest-value line in a Woodcock-Johnson report. The battery also sits differently in evaluations than a Wechsler scale does: where a WISC-V is typically paired with a separately normed achievement test such as the WIAT-4 , the WJ offers both sides co-normed, and many evaluators mix the two traditions in cross-battery work. A WJ score set is evidence, not a verdict: eligibility decisions belong to teams and agencies applying whole-evaluation criteria , and the results section's job is to feed that decision with calibrated, honestly bounded numbers. ## Who uses Woodcock-Johnson write-ups and when School psychologists and educational diagnosticians write up Woodcock-Johnson results inside special-education eligibility evaluations, where the co-normed cognitive and achievement batteries feed every state's specific-learning-disability method, discrepancy, intervention response, or pattern of strengths and weaknesses. Clinical and pediatric psychologists use it in learning-disability and gifted referrals, and neuropsychologists fold WJ achievement clusters into a neuropsychological report beside their cognitive and memory measures. Because the norms extend past age 90, it also carries adult work that child-normed batteries cannot: adult basic education, vocational rehabilitation, and the documentation behind postsecondary and licensing-exam accommodation requests. Instrument choice follows the referral question. Many evaluators pair a WISC-V with the WJ achievement battery, or run WJ COG and ACH together to keep ability-achievement comparisons inside one norm group; the WIAT-4 and KTEA-3 compete on the achievement side with direct statistical links to the Wechsler scales. For Spanish-English bilingual evaluations, the co-normed Batería IV is the WJ IV's Spanish parallel, and no Batería V exists as of July 2026. Oral-language and phonological measures in the WJ overlap with, but do not replace, the comprehensive language batteries a speech-language pathologist administers for a language-disorder diagnosis; in a psychoeducational workup the WJ versions serve as converging evidence, not the diagnostic instrument. ## How to structure a Woodcock-Johnson results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced evaluators converge on because it survives review: it anchors the scores to an edition and norm set before interpreting them, presents clusters in a table and meaning in prose, and makes the two genuinely WJ-specific moves, explaining the proficiency metrics and handling the WJ IV to WJ V transition, explicit stated decisions instead of silent ones. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and format. Name the instrument and edition (WJ IV or WJ V), the norm set, the batteries and clusters administered, and the administration format and scoring platform. Pitfall: writing "the Woodcock-Johnson" with no edition. Both editions are on the market at once, and cluster names that look interchangeable are not: the WJ V recomputed the General Intellectual Ability composite, removed Auditory Processing from it, and split Long-Term Retrieval into two clusters, so an edition-less score set is uninterpretable at re-evaluation. Behavioral observations and validity of the session. Describe effort, engagement, language considerations, breaks, and anything that departed from standardized procedure, then state plainly whether you consider the results valid estimates of current functioning. For WJ V administrations, note the digital format and any connectivity or device accommodations. Pitfall: boilerplate observations that contradict the interpretation. "Worked quickly and confidently" followed by fluency scores you then attribute to slow, effortful responding gives a reviewer two incompatible reports in one document. Score metrics statement. Say which metrics you are reporting, standard scores (mean 100, SD 15), percentile ranks, and, if you include them, the W score and Relative Proficiency Index, and give lay readers one plain sentence for each WJ-specific metric. The RPI is criterion-referenced and always written against 90, and Riverside's interpretive bulletin fixes the reading: an RPI of 60/90 means about 60% success on tasks that typical age or grade peers perform with 90% success. Pitfall: pasting RPI columns with no explanation. IEP teams read 55/90 as a fraction, a percentile, or a typo; either explain the metric in one sentence where it first appears or deliberately omit it, and apply that choice consistently. Cognitive results, composite first, then cluster by cluster. Lead with the general ability composite you are using (GIA, or the Brief Intellectual Ability when a screener was given), then move through the broad CHC clusters, spending the prose on what each result means functionally for this examinee rather than re-reading the table aloud. Pitfall: the score dump. A paragraph per test in administration order documents that testing occurred; it does not answer the referral question, and narrow-cluster interpretation is exactly where the independent factor-analytic literature urges the most restraint. Achievement results by academic domain. Organize by reading, mathematics, written language, and oral language rather than test by test, and tie each domain to the referral concern. State whether scores were computed with age or grade norms if both are in play in your setting. Pitfall: grade equivalents promoted to instructional levels. A grade equivalent of 3.2 means the examinee's raw score matched the median for that grade on this test's items; it is not a placement recommendation, and reviewers flag reports that treat it as one. Variation and comparison procedures. If you report intra-ability variations or an ability-achievement comparison, name the procedure and the decision framework it feeds (the state's SLD method, or the specific pattern-of-strengths-and-weaknesses model), and describe base rates in words. Pitfall: the unnamed model. A "significant discrepancy" conclusion with no named method, or score-report comparison tables pasted wholesale, reads as software output rather than clinical reasoning, and PSW methods themselves remain scientifically contested, which makes naming your framework more important, not less. Prior testing and the edition note. When earlier Woodcock-Johnson results exist, name the edition and norms of both administrations. If the editions differ, say directly that WJ IV and WJ V scores are not directly comparable and describe change qualitatively. Pitfall: cross-edition arithmetic. "Broad math fell six points since 2023" across a WJ IV to WJ V gap treats two different instruments as one ruler, the exact move the test's own authors caution against. Interpretive summary and referral linkage. Answer the referral question directly, state what the data do and do not establish, and place the WJ explicitly as one source among several: federal special-education law requires that no single measure be the sole criterion for eligibility ( 34 CFR 300.304(b)(2) ). Pitfall: an eligibility verdict from a score table. Identification decisions belong to teams and agencies applying their criteria; the results section supplies calibrated evidence, not the ruling. Recommendations linkage. Tie each recommendation to a specific finding, and where the WJ's proficiency metrics drove it, say so; the instructional-zone language (what this student finds easy versus difficult) converts directly into intervention planning. Pitfall: recommendations that could follow any profile. If fluency supports, chunked instruction, or extended time do not trace back to named cluster findings, either the recommendations or the results section is incomplete. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: [WJ V / WJ IV] Norm set: [named] Batteries and clusters given: [COG / ACH / VTL selections] Format: [WJ V digital / WJ IV paper] Scoring: [platform] BEHAVIORAL OBSERVATIONS AND VALIDITY [Effort, engagement, language, breaks, departures from standard procedure; statement that results are valid estimates, or the stated limitation if not] SCORE METRICS STATEMENT [Metrics reported (standard score, percentile, RPI, W); one plain-language sentence for RPI if included] COGNITIVE RESULTS (composite first, cluster by cluster) GIA or BIA: [score, band, percentile + what it means here] Broad clusters: [functional meaning per cluster, not a test-by-test dump] ACHIEVEMENT RESULTS (by academic domain) Reading / Math / Written Language / Oral Language: [domain results tied to the referral question] VARIATION AND COMPARISON PROCEDURES [Named procedure and framework (state method or PSW model); base rates described in words] PRIOR TESTING AND EDITION NOTE [Prior edition and norms named; cross-edition caution stated when editions differ] INTERPRETIVE SUMMARY (answer the referral question) [What the data establish and what they do not; single- measure limits; measurement error acknowledged] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-evaluation plan] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample Woodcock-Johnson write-up (fictional) Scenario: a three-year re-evaluation for a student found eligible for special education in mathematics in second grade. The prior evaluation used the WJ IV; this one uses the WJ V, so the report has to handle a cross-edition comparison without treating two different instruments as one ruler. This is the Woodcock-Johnson results section only, condensed but structurally complete. All details are fictional. Student: J.T., 11 (grade 5) · Referral: three-year re-evaluation, continuing eligibility and math concerns · Evaluator: R. Calloway, EdS, NCSP, School Psychologist · Testing dates: 07/08/2026 and 07/10/2026 · Report date: 07/16/2026 Measures and administration: The Woodcock-Johnson V Tests of Cognitive Abilities (WJ V COG) and selected clusters from the Woodcock-Johnson V Tests of Achievement (WJ V ACH) were administered digitally on the publisher's platform, examiner device and student tablet, and scored against the WJ V United States age-based norms. Standard scores have a mean of 100 and a standard deviation of 15 and are reported with a 95% confidence band and percentile rank. The Relative Proficiency Index (RPI) is also reported: it is written against 90, so an RPI of 55/90 predicts about 55% success on tasks that typical age peers perform with 90% success. J.T.'s prior evaluation (March 2023, grade 2) used the WJ IV; the two editions are normed and structured differently, and scores are not directly comparable across them (see Prior testing, below). Behavioral observations and validity: J.T. engaged readily with the tablet format, followed the recorded instructions without repetition, and sustained effort across both sessions with one scheduled break each. On untimed reasoning tasks he worked deliberately and self-corrected. On timed and math tasks he counted on his fingers, skipped items he judged hard, and twice said "math is not my thing," while continuing to work when encouraged. Administration followed standardized procedures, and the results are considered valid estimates of current functioning. Cluster Standard score (95% band) Percentile RPI General Intellectual Ability (GIA) 96 (92-100) 39 88/90 Comprehension-Knowledge (Gc) 108 (103-113) 70 95/90 Fluid Reasoning (Gf) 104 (99-109) 61 93/90 Visual Processing (Gv) 102 (96-108) 55 92/90 Auditory Working Memory Capacity (Gwm) 82 (77-87) 12 61/90 Cognitive Processing Speed (Gs) 84 (78-90) 14 66/90 Brief Reading 102 (97-107) 55 93/90 Brief Writing 94 (88-100) 34 84/90 Brief Math 78 (73-83) 7 47/90 Number Concepts 80 (74-86) 9 52/90 Cognitive results: J.T.'s General Intellectual Ability score of 96 (39th percentile) is in the average range, and the clusters beneath it split in an interpretable way. Comprehension-Knowledge (108, 70th percentile) is a relative strength: his vocabulary and general knowledge are developing ahead of most age peers, consistent with his strong classroom participation in science discussions. Fluid Reasoning (104) and Visual Processing (102) are age-typical: he identifies conceptual rules and works with visual patterns as well as most 11-year-olds. In contrast, Auditory Working Memory Capacity (82, 12th percentile, RPI 61/90) and Cognitive Processing Speed (84, 14th percentile, RPI 66/90) are limited relative to peers: holding and reworking information he hears, and completing simple tasks quickly, are effortful for him. Functionally, this is a student who understands grade-level ideas but loses the middle steps of multi-step directions and works slowly enough that timed tasks go unfinished. Achievement results: Reading (Brief Reading 102) and writing (Brief Writing 94) are age-appropriate. Mathematics is the area of concern, and the two math clusters agree: Brief Math of 78 (7th percentile) and Number Concepts of 80 (9th percentile). The proficiency metrics say the same thing in instructional terms: with an RPI of 47/90, grade-level math tasks that typical age peers perform with 90% success are predicted to be at about 47% success for J.T., placing them in his difficult range. Error patterns on scored responses showed accurate single-step computation with breakdowns on regrouping and multi-step problems, consistent with the working memory finding; no test items are reproduced here. Variations and comparisons: Within cognitive abilities, J.T.'s working memory and processing speed clusters are weaknesses relative to his own average, and his math achievement is below the level his overall ability score would predict; differences of this size occur infrequently among students with similar ability scores. These results feed the eligibility team's specific-learning-disability analysis under the state's identification method, alongside classroom data and intervention response; the pattern here, average reasoning and knowledge with concordant weaknesses in working memory, processing speed, and mathematics, is the shape that analysis looks for. The comparison procedures and their base rates come from the publisher's scoring platform and are described here rather than reproduced. Prior testing and edition note: J.T.'s March 2023 evaluation reported WJ IV mathematics cluster scores in the low 80s. Those scores were produced by a different edition, normed on a different sample collected more than a decade earlier, and several WJ V clusters are built from different tests than their WJ IV counterparts. This report therefore does not read the current Brief Math of 78 as a decline. Stated qualitatively: mathematics was below age expectation in 2023 on the WJ IV, and it remains below age expectation in 2026 on the WJ V, despite documented intervention. Summary: Cognitive results show average general ability with above-average knowledge and age-typical reasoning over limited working memory and processing speed. Achievement is age-appropriate in reading and writing and well below age expectation in mathematics, with proficiency metrics placing grade-level math in J.T.'s difficult range. This pattern is consistent with his existing identification in mathematics. It does not by itself decide continuing eligibility: that determination belongs to the team, integrating these results with intervention response and classroom evidence, and no single measure may serve as the sole criterion. All scores are estimates that carry measurement error. Linkage to recommendations: The working memory finding supports recommendations 1 and 2 (multi-step directions chunked and paired with written steps; worked examples visible during practice). The math fluency and Number Concepts findings support recommendation 3 (explicit instruction and daily brief fluency practice at his instructional level, with progress monitoring). The processing speed finding supports recommendation 4 (a trial of extended time on timed math work, with the team monitoring completion). The edition note supports recommendation 5: at the next re-evaluation, compare within the WJ V rather than across editions. This sample is fictional and for educational purposes. It does not describe a real student or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, norm set, batteries given, and administration format are all named, so the scores are interpretable now and stay unambiguous while WJ IV and WJ V circulate side by side. The cross-edition question is handled the way the test authors direct: prior WJ IV scores are named but never subtracted from current WJ V scores, and change is described qualitatively. The RPI is explained in one plain sentence where it first appears, then used to translate scores into instructional terms, instead of appearing as unexplained notation. Every number in the prose matches the table, cluster results are written as functional meaning for this student, and the comparison procedures are named and described rather than pasted from the scoring platform. The summary states what the data do not establish, honors the no-single-measure rule, and every recommendation traces to a specific cluster finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label the strength of each requirement honestly. In US special education, federal law is instrument-neutral: no regulation names the Woodcock-Johnson, evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ), and for specific learning disability the state "must not require the use of a severe discrepancy" between ability and achievement ( 34 CFR 300.307 ). Which method your state runs, discrepancy where still permitted, intervention response, or a pattern-of-strengths-and-weaknesses model, is state policy, and the WJ's co-normed comparison procedures serve any of them; what stays constant is the convention of naming the method and model in the report. Gifted identification has no federal mandate at all, so WJ-based thresholds are district and state policy. Testing time is billed to payers under the psychological-testing code families (96130 and 96131 for evaluation services; 96136 through 96139 for administration and scoring) as payer policy with plan-specific units, while school-based IDEA evaluations are education-funded rather than billed to health insurance, a boundary that routinely sends families seeking accommodation documentation to private evaluators. For those accommodation bodies, currency rules are the gatekeeper: ETS's guidelines , for example, describe learning-disability documentation "completed within the past 5 years and/or when the test taker was at least 16" as potentially helpful, treat re-administration of an adult cognitive measure as unnecessary in an update, and ETS's current disability guidance names the WJ V cognitive battery among the adult measures it recognizes. Verify the receiving agency's documentation policy before you choose the battery and norms. Edition and norms currency is the Woodcock-Johnson's live defensibility question, and as of July 2026 it has an unusual answer: both editions are current products. Riverside released the WJ V in February 2025 as a digital-only system and still sells the WJ IV with no US retirement statement (the publisher posts one when a product ends; the WJ III has one, the WJ IV does not), telling university trainers that short-term WJ IV use is expected while recommending against new WJ IV kit purchases; its Canadian distributor is more concrete, estimating transition support into 2027 to 2028. No IDEA rule forces an edition switch mid-cycle; re-evaluation timing follows the ordinary three-year cycle and district policy. That makes the choice yours, and the documentation non-negotiable: name the edition, the norm basis, and the platform every time, because the WJ V's norms are 2020-Census-aligned and post-pandemic while the WJ IV's date from more than a decade earlier, and standardization samples age measurably, about 2.31 points per decade in the largest Flynn-effect meta-analysis. Scores from the two editions are separately normed and differently structured, so re-evaluations that span the transition should describe change qualitatively and restart trend lines inside the WJ V. Digital administration deserves one line in the report: the WJ V was standardized on a tablet-based format with an offline mode, and remote administration runs through Presence, named its exclusive remote provider in August 2025 , with equivalence evidence that is publisher- and vendor-sponsored, worth labeling as such in high-stakes contexts. Access boundaries are the publisher's: Woodcock-Johnson batteries are restricted to qualified professionals under Riverside's user-qualification policy, and outside the US the picture is jurisdiction-specific. Canadian distribution of the WJ V runs digitally through Nelson with data stored on Canadian servers and no separate Canadian norming announced, and Australia's distributor lists the WJ V as the US adaptation while the Australasian adaptation with Australian norms, as of July 2026, remains a WJ IV product, so name the norm set explicitly when writing outside the US. Woodcock-Johnson is a registered trademark, and WJ IV and WJ V are trademarks, of Riverside Assessments, LLC; the tests are published by Riverside Insights. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Woodcock-Johnson write-up errors reviewers flag There is no payer audit series for cognitive and achievement write-ups; the accountability literature here is psychometric, and the Woodcock-Johnson has drawn pointed independent scrutiny. The WJ IV was reviewed in the Buros Center's 20th Mental Measurements Yearbook , where the reviewer wrote that "abundant caution should be exercised in interpretation of lower-order scores." Independent factor-analytic studies ( exploratory and confirmatory ) concluded that the WJ IV cognitive battery supports about four distinct factors rather than the seven it markets, with general ability dominating, and noted that the technical manual reported no separate factor analyses of the cognitive battery on its own. The WJ V is new enough that its independent evidence base is thin: the first independent review appeared in 2026 and, while judging the psychometrics favorably, found the evidence "less convincing for children under six." The errors below are write-up errors, and they are yours. The BastionGPT Clinical Advisory Board sees the same ones most often in Woodcock-Johnson report reviews: Cross-edition scores compared as change. "Math dropped six points since the last evaluation" across a WJ IV to WJ V gap treats two differently normed, differently structured instruments as one ruler: the composite is computed differently, several clusters are built from different tests, and the normative samples sit more than a decade apart. Name both editions and describe change qualitatively. Edition and norms unnamed mid-transition. "The Woodcock-Johnson" with no edition, norm basis, or platform. With both editions on the market, the GIA computed differently in each, and familiar-looking cluster names covering different tests, an unanchored score set cannot be interpreted at re-evaluation or defended under review. Proficiency metrics pasted without translation. An RPI column appears (55/90, 82/90) with no sentence explaining the metric, and readers parse it as a fraction or a percentile. Either explain it in plain language where it first appears, or deliberately omit it; no authority requires the RPI, so the defensible move is a stated house rule, applied consistently. Grade equivalents promoted to placement. "A grade equivalent of 3.2 means fourth-grade math is inappropriate" misreads a developmental score as an instructional prescription. Grade equivalents describe where a raw score falls on this test's growth curve; standard scores, bands, and proficiency metrics carry the interpretive weight. Narrow clusters over-read. Diagnostic conclusions built on a single narrow cluster or test-level contrast, when the independent factor-structure literature finds most of the reliable variance sits with general ability. Interpret composite-first, use broad clusters cautiously, and keep test-level commentary qualitative and bounded. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your cluster summary (standard scores, bands, percentiles, RPIs) and get a drafted results-section narrative organized composite-first and cluster by cluster, with the edition-and-norms statement and the RPI explained in plain language for your review. Cross-check a finished draft for the gaps reviewers flag: numbers that disagree with the table, unexplained proficiency notation, and cross-edition WJ IV to WJ V comparisons written as change. Translate the results section into a plain-language summary for parents and teachers that keeps scores as estimates and converts proficiency metrics into what feels easy and what feels difficult. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Woodcock-Johnson scores actually mean? Four score types do the work. Standard scores compare the examinee to age or grade peers, with a mean of 100 and a standard deviation of 15, and each carries a percentile rank and a confidence band. Underneath them sits the W score, the Rasch-based equal-interval metric all WJ scoring is built on; because it tracks absolute ability rather than standing among peers, it is the metric that shows real growth over time even when the standard score holds still. The Relative Proficiency Index (RPI) is criterion-referenced and always written against 90: Riverside's own interpretive bulletin gives the reading, an RPI of 60/90 means about 60% success on tasks that typical peers perform with 90% success. Age and grade equivalents describe where a raw score falls on the test's growth curve and deserve the least interpretive weight. Write standard scores as estimates, and translate the WJ-specific metrics in one plain sentence each. Should I test on the WJ IV or the WJ V? Either is defensible as of July 2026, and that is the unusual part. Riverside released the WJ V in February 2025 as a fully digital system and still sells the WJ IV with no US retirement statement, while its university training page acknowledges some programs will stay on the WJ IV in the short term and recommends against buying new WJ IV kits; the Canadian distributor estimates transition support into 2027 to 2028. No IDEA rule requires the newest edition the day it ships, and re-evaluation timing follows the ordinary three-year cycle. The honest trade: the WJ V carries current, post-pandemic, 2020-Census-aligned norms and the redesigned battery, while the WJ IV carries a decade of familiarity, the paper format, and norms that are now aging. Whichever you choose, name the edition and norms in the report and expect the choice to get asked about mid-transition. Can I compare WJ V scores with a student's earlier WJ IV scores? Not as arithmetic. The reasons are structural, not cosmetic: the WJ V's General Intellectual Ability composite is eight equally weighted tests where the WJ IV's was differentially weighted, Auditory Processing no longer contributes to it, the old Long-Term Retrieval cluster is now two clusters, and Fluid Reasoning is measured by new tests, all against a normative sample collected more than a decade after the WJ IV's. Even the vocabulary moved: the WJ IV reports a Short-Term Working Memory cluster, the WJ V an Auditory Working Memory Capacity cluster. A same-named-looking score is not the same measure, and norms drift on their own, about 2.31 points per decade in the largest Flynn-effect meta-analysis. The defensible pattern is the one the sample on this page models: report the prior edition's results as historical findings, describe change qualitatively (still below age expectation, no longer below age expectation), and restart numeric trend lines inside the WJ V. What is the RPI, and do I have to report it? The Relative Proficiency Index is the WJ's criterion-referenced metric, always written against 90. It answers a different question than a standard score: not "how does this student rank," but "when peers succeed 90% of the time on a task, how often will this student?" An RPI of 47/90 predicts 47% success on material typical peers handle at 90%, which is why the RPI converts so directly into instructional planning, and why it can reveal a proficiency problem a mid-80s standard score understates. No authority requires you to report it, and plenty of experienced evaluators omit it from the body precisely because unexplained 55/90 notation confuses IEP teams. Both practices are defensible; a silent mix of the two is not. Set a house rule, and when the RPI does appear, spend one sentence translating it in plain language. Is the Woodcock-Johnson required for SLD or IEP eligibility? No. Federal law names no instrument, requires that no "single measure or assessment" be the sole criterion for eligibility ( 34 CFR 300.304(b)(2) ), and forbids states from requiring an ability-achievement discrepancy for specific learning disability ( 34 CFR 300.307 ); states choose among discrepancy, intervention response, and pattern-of-strengths-and-weaknesses frameworks. The Woodcock-Johnson's place in those evaluations is convention and district practice, earned by its co-normed cognitive-achievement design, which lets every comparison run inside one norm group. If your district or agency names a specific battery, that is local policy, and the practical answer is the one evaluators have always given: document what the policy requires, and choose the instrument that answers the referral question. Woodcock-Johnson, or WISC-V plus WIAT-4? Both stacks are standard; they solve the linking problem differently. The WISC-V with the WIAT-4 gives you the Wechsler clinical tradition and a publisher-linked ability-achievement pair normed on separate but statistically connected samples. The Woodcock-Johnson gives you one co-normed system for both sides, the most explicit CHC-theory alignment of any major battery, and adult norms that extend past age 90. Convention splits by setting: Wechsler scales dominate clinical and neuropsychological work, while the WJ is a school-evaluation workhorse and the backbone of cross-battery assessment, where evaluators combine WJ clusters with Wechsler indexes deliberately. The referral question, your state's SLD method, and what the prior evaluation used (staying inside one battery family preserves comparability) decide it case by case. Can I see the actual Woodcock-Johnson test items, or include them in a report? No. Test items, stimuli, and record forms are protected test materials: purchase and use are restricted under Riverside's qualification policy, and psychologists are ethically obligated to maintain test security under APA Ethics Standard 9.11 and their publisher agreements. Item exposure damages the instrument itself, because norms assume examinees have not rehearsed the tasks. Reports should carry scores, bands, percentiles, proficiency metrics, and interpretation, never item content; the sample on this page describes error patterns in category terms for exactly that reason. What parents and teams can get is better anyway: the scores with a plain-language explanation of what each cluster means, what the proficiency metrics predict for classroom work, and what happens next. Is the digital, tablet-based WJ V administration valid? The WJ V is digital by design rather than a paper test ported to a screen: it was standardized in the tablet-based format, with the examiner running the session from one device and the examinee responding on a tablet, an offline mode for settings without reliable connectivity, and printed response booklets where written responses are required. That means format equivalence with paper is not the right question inside the WJ V; the norms are the digital administration. Two caveats belong in high-stakes write-ups. First, use the standardized device class the publisher specifies, an examinee tablet with a 10-inch-plus screen, "as that is how the test was standardized" in Riverside's words, rather than improvising hardware. Second, remote administration runs through Presence, named the exclusive remote provider in 2025, and the supporting equivalence research is publisher- and vendor-sponsored; naming the administration arrangement in the report is the convention that keeps the scores defensible. Can BastionGPT draft my Woodcock-Johnson results section? Yes. Paste your cluster summary (standard scores, bands, percentiles, RPIs) and it drafts the results-section narrative for your review: organized composite-first and cluster by cluster, the edition-and-norms statement up front, proficiency metrics translated into plain language, and prior-edition results handled qualitatively instead of as arithmetic. It can also cross-check a draft you wrote for score-versus-narrative mismatches and cross-edition comparisons, and produce a parent-friendly summary of the findings. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and drafting from scores you paste means no protocol or item content ever needs to leave your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Riverside Insights, WJ V product page : the digital-only platform, the 15-test Virtual Test Library, 2022 to 2024 norming against 2020 US Census demographics, the November 1, 2026 Nonverbal launch, and the publisher's own time-saving and post-pandemic-norms claims. Riverside Insights, WJ V launch release (February 2025) : the WJ V COG and ACH batteries and Virtual Test Library available February 1, 2025, hosted on Riverside Score. Riverside Insights, WJ V Technical Abstract : ages 4 through 90+, the 5,837-person norming sample, the eight-test equally weighted GIA, the removal of Ga from the GIA and COG battery, the Long-Term Retrieval split into Gl and Gr, the new Matrices and Verbal Analogies tests, oral language measures moved into the ACH battery, and the trademark attribution. Riverside Insights, WJ V brochure : the cluster list, including Auditory Working Memory Capacity (Gwm) and Cognitive Processing Speed (Gs), and the "Coming November 1st, 2026" Nonverbal line. Riverside Insights, Assessment Service Bulletin on the W score and RPI (Jaffe): the Rasch-based, equal-interval W scale and the RPI's 90% peer-success reading. Riverside Insights, User Qualifications Guide and university training page : the graduate-level qualification tiers, the short-term WJ IV acknowledgment, and the recommendation against new WJ IV kit purchases. Riverside Insights, WJ IV Technical Abstract and WJ IV store page : the 2014 publication, ages 2 to 90+, the 1977 first edition, continued availability, and the five-to-ten-minutes-per-test administration estimate. Nelson, WJ V Canada page : Canadian-server hosting, the transition-support estimate, and the absence of a separate Canadian norming; PAA, WJ IV Australasian Adaptation and WJ V US Adaptation listings: the Australian norms status. 34 CFR 300.304 and 34 CFR 300.307 : the no-single-measure rule and the severe-discrepancy prohibition. ETS, guidelines for writing diagnostic reports and disability documentation guidance : the five-year currency language, the adult cognitive re-administration position, and the WJ V named among current adult measures. Buros Center for Testing, Tests reviewed in the Twentieth Mental Measurements Yearbook : independent review status of the WJ IV, with the reviewer's lower-order-scores caution. Dombrowski, McGill & Canivez, exploratory factor analysis, Psychological Assessment , and confirmatory follow-up, Archives of Scientific Psychology : the four-factor, g-dominant findings; Foster, Perazzo & Decker, Journal of Psychoeducational Assessment (2026) : the first independent WJ V review; Trahan et al., Psychological Bulletin (2014) : the Flynn-effect meta-analysis. PR Newswire, Presence named exclusive remote provider of the WJ V (August 2025) : the remote-administration arrangement. ## Related templates WISC-V Report Write-Up WIAT-4 Report Write-Up Psychological Evaluation Report Neuropsychological Evaluation Report Psychiatric Diagnostic Evaluation Psychoeducational Evaluation Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/wppsi-iv-report All note templates # WPPSI-IV Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The WPPSI-IV (Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition) is an individually administered intelligence test for ages 2 years 6 months to 7 years 7 months, published in 2012. Psychologists use it for preschool special-education eligibility, kindergarten and gifted decisions, and the cognitive core of early-childhood evaluations. This page covers how to write up WPPSI-IV results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and school psychologists; publisher qualification level C Audience IEP and preschool special-education teams, early-intervention programs, pediatricians, gifted and admissions committees, parents Typical length 400 to 800 words for the results section · administration 30 to 60 minutes by age band Format family Norm-referenced preschool cognitive battery When it's used Preschool special-education and developmental-delay eligibility, early-intervention to preschool transitions, kindergarten-readiness and gifted referrals, autism and developmental evaluations, re-evaluations Standards context Published by Pearson (2012, current edition); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the WPPSI-IV? The Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) is an individually administered, norm-referenced intelligence test for children ages 2 years 6 months through 7 years 7 months, published by Pearson in 2012 as the successor to the WPPSI-III (2002) and standardized on 1,700 children. Its defining design fact is that it is effectively two batteries divided at age 4. The younger band (2:6 to 3:11) yields a Full Scale IQ (FSIQ) and three primary index scores, Verbal Comprehension (VCI), Visual Spatial (VSI), and Working Memory (WMI), in the publisher's stated 30 to 45 minutes of core-subtest time. The older band (4:0 to 7:7) adds Fluid Reasoning (FRI) and Processing Speed (PSI) for five primary indexes, at 45 to 60 minutes. Ancillary indexes serve specific questions: the Vocabulary Acquisition Index (VAI) and Nonverbal Index (NVI) at both bands, the General Ability Index (GAI) at both bands with band-dependent composition, and the Cognitive Proficiency Index (CPI) at 4:0 to 7:7 only. Subtest scaled scores run 1 to 19 with 7 to 12 usually considered average; index scores run 40 to 160 with 90 to 109 considered average, each carrying percentile ranks and confidence intervals. Tasks are deliberately preschool-shaped, "engaging tasks for children as young as 2.5 years old" in Pearson's words, including picture-based working-memory subtests and an ink dauber in place of a pencil on processing-speed tasks. Administration is paper or digital via Q-interactive with scoring on Q-global or by hand, and separate Canadian and Australian and New Zealand standardised editions carry local norms. Edition status as of July 2026: the WPPSI-IV remains current, and no WPPSI-5 has been announced in any market. The contrast is visible on Pearson's own pages, where the WISC-V product page banners "WISC-6 is coming" while the WPPSI-IV page carries no equivalent notice. The load-bearing question for write-ups is architecture and altitude, because most of what ranks in search gets both wrong. Consumer and test-prep pages present a clean five-index battery for all ages, repeat the WPPSI-III's 7:3 upper age bound instead of 7:7, and describe ancillary indexes without noting that the CPI does not exist below age 4 or that the GAI's subtest composition changes at the band boundary, a detail even a published clinical-trial protocol has garbled. The interpretive altitude question is sharper. In the dual Buros review of the WPPSI-IV in the 19th Mental Measurements Yearbook, one reviewer judged it psychometrically strong while Canivez's review documented that exploratory factor analyses "were not reported" in the technical manual despite four subtests being deleted and five added, that incremental validity of the index scores was not reported, and that several index factors rest on only two subtests. His conclusion for practice: clinicians who follow the manual's interpretation schemes "risk overinterpretation and misinterpretation of WPPSI-IV scores in clinical application", and subtest and profile interpretation methods "should not be used in clinical decision-making until psychometric support for them is provided". The defensible posture, and the one this page teaches, names the age band before any number, interprets the FSIQ and reported indexes cautiously, keeps subtest commentary qualitative, and treats the score set as evidence feeding a team or clinical decision , never the decision itself. ## Who uses WPPSI-IV write-ups and when School psychologists write up WPPSI-IV results inside preschool special-education eligibility evaluations, most often around the transition from early intervention to preschool services at age 3, where federal law puts an IEP or IFSP in place no later than the third birthday. Clinical child psychologists use it for developmental-delay questions, kindergarten-readiness and gifted referrals, and as the cognitive component of autism and developmental evaluations, where it sits alongside observation measures and adaptive scales such as the Vineland-3 rather than standing alone. It also appears, by convention rather than clinical necessity, in private-school admissions testing in some markets. Age drives instrument choice at both edges: below 2:6, and for very low-functioning children in the overlap years, developmental measures such as the Bayley-4 carry the referral, while at the 6:0 to 7:7 overlap the choice between the WPPSI-IV and the WISC-V follows the publisher's floor-and-ceiling logic, the WPPSI-IV for suspected below-average ability and the school-age scale for high-ability children. A WPPSI-IV section rarely travels alone: it anchors the cognitive portion of a psychoeducational report or broader psychological evaluation report next to speech-language findings, adaptive data, and the behavioral observations that preschool validity rests on. ## How to structure a WPPSI-IV results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced preschool evaluators converge on because it survives review: it names the age band before any score, gives behavioral observation the weight preschool validity actually rests on, keeps composite interpretation at the altitude the psychometric evidence supports, and ends with the caveats that make an early-childhood score usable. Each section carries the pitfall that most often undermines it. Measures, edition, age band, and format. Name the instrument and edition, the age band administered (2:6 to 3:11 or 4:0 to 7:7), the norm set (US, Canadian, or Australian and New Zealand), the administration format (paper or Q-interactive), the subtests given, and any substitution or proration. Pitfall: an unnamed age band. The two bands yield different index sets, three primary indexes for the younger band and five for the older, so an unlabeled score table is uninterpretable at re-evaluation, and a five-index template forced onto a 3-year-old announces template reuse to any reviewer who knows the test. Behavioral observations and the validity statement. Document separation from the caregiver, rapport, attention, activity level, frustration tolerance, response to the play-based materials, and breaks, then close with an explicit validity statement tied to what you observed. Preschool cognitive results are only as good as the engagement that produced them. Pitfall: a boilerplate validity sentence. "Results are considered valid" with no observational basis reads as template text; reviewers look for the linkage between observed behavior and the validity claim. Composite summary. Report the FSIQ and the band's primary indexes, each with confidence interval, percentile rank, and one named descriptor set applied consistently. Pitfall: unlabeled or mixed descriptor vocabularies. The publisher itself calls the qualitative descriptors suggestions rather than evidence-based categories, so the write-up must say which set it uses, and a 92 labeled Average in the table cannot become "solidly average ability" in the summary. Index-by-index results. Translate each reported index into functional meaning for this child: what the score says about how the child handled that family of tasks, in language a parent and a team can use. Pitfall: subtest-level storytelling. Individual subtest scores are the least reliable numbers on the test, and published reviews of the WPPSI-IV specifically caution against profile and pattern interpretation; keep subtest commentary qualitative and let composites carry the weight. Ancillary indexes and the reporting decision. When a Vocabulary Acquisition, Nonverbal, General Ability, or (at 4:0 to 7:7) Cognitive Proficiency Index earns a place, report it alongside the FSIQ with the reason stated. Pitfall: a silent composite swap. An ancillary index appearing without rationale reads as score shopping, and at the younger band the GAI still contains an expressive naming task, so it is not the language workaround there that it is at school age. Stability and norm-currency caveats. State that scores at this age are estimates of current functioning, less stable and less predictive than school-age scores, name the re-evaluation plan, and note the edition and norm vintage. Pitfall: prediction language. "Will remain delayed" and trajectory claims outrun what preschool cognitive scores can support, and reviewers flag them. Integration with adaptive and developmental data. Read the cognitive results against adaptive behavior, developmental history, speech-language findings, and observation across settings. Pitfall: cognitive-only reasoning. Eligibility and diagnostic frameworks alike require multiple sources; a results section that lets one composite carry the conclusion invites reversal. Interpretive summary. Answer the referral question, state what the scores do and do not establish, and place the determination where it belongs. Pitfall: eligibility conclusions written as the examiner's verdict. Category decisions belong to teams applying their jurisdiction's criteria; the results section supplies calibrated evidence, not the ruling. Recommendations linkage. Tie each recommendation to a specific finding, including the re-evaluation timeline and, near age 6 or 7, the instrument-selection note for the next evaluation. Pitfall: recommendations unmoored from findings. A recommendation list that could sit under any profile tells the reader the data did not drive it. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: WPPSI-IV (2012) Age band: [2:6-3:11 / 4:0-7:7] Norms: [US / Canadian / A&NZ] Format: [paper / Q-interactive] Subtests given: [core set; any substitution or proration stated] BEHAVIORAL OBSERVATIONS AND VALIDITY [Separation, rapport, attention, activity level, breaks; validity statement tied to observed behavior] COMPOSITE SUMMARY [FSIQ + the band's primary indexes: score, CI, percentile, one named descriptor set applied consistently] INDEX-BY-INDEX RESULTS [Functional meaning per index; no subtest-level claims] ANCILLARY INDEXES (IF REPORTED) [VAI / NVI / GAI, CPI at 4:0-7:7 only; rationale stated; younger-band GAI includes an expressive naming task] STABILITY AND NORM-CURRENCY CAVEATS [Current estimate, not a fixed trait; re-evaluation plan; edition and norm vintage named] INTEGRATION WITH ADAPTIVE AND DEVELOPMENTAL DATA [Adaptive scores, speech-language findings, history, observation; team decision under jurisdiction criteria] INTERPRETIVE SUMMARY (answer the referral question) [What converges; what the scores do NOT establish] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-evaluation timing; instrument-selection note near the band boundary] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample WPPSI-IV write-up (fictional) Scenario: a 3-year-old approaching the transition from early intervention to preschool special education, referred with expressive-language concerns. Testing uses the younger age band (2:6 to 3:11), which yields three primary indexes rather than five, the architecture detail most sample reports miss. This is the cognitive results section only, condensed but structurally complete. All details are fictional. Child: L.T., age 3 years 4 months · Referral: preschool special-education team, early-intervention transition, expressive-language concerns · Evaluator: R. Okafor, PsyD, Licensed Psychologist · Testing date: 07/08/2026 · Report date: 07/14/2026 Measures and administration: The Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) was administered in paper format using the 2:6 to 3:11 age band and United States norms. At this band the battery yields the Full Scale IQ (FSIQ) and three primary index scores, Verbal Comprehension (VCI), Visual Spatial (VSI), and Working Memory (WMI); given the referral question, the Vocabulary Acquisition Index (VAI), Nonverbal Index (NVI), and General Ability Index (GAI) were also derived. All subtests were administered per standard procedures with no substitution or proration. Composites are reported with 95% confidence intervals (CI), percentile ranks, and the qualitative descriptor set printed in the publisher's score reports, named here so the labels are interpretable. Behavioral observations and validity: L.T. separated from her mother after a brief warm-up period and settled quickly with the play-based materials. She responded readily to pointing-format tasks, used the ink dauber on the visual scanning task with evident enjoyment, and communicated mostly in single words and gestures, consistent with the referral concern. Attention wandered late in the session; one break restored engagement, and standardized procedures were maintained throughout. Because engagement, not effort in the adult sense, is what preschool validity rests on, the validity statement is tied to these observations: the results below are considered a valid estimate of current cognitive functioning, interpreted in the context of her expressive-language presentation. Composite Standard score 95% CI Percentile Descriptor Verbal Comprehension (VCI) 78 73-86 7 Borderline Visual Spatial (VSI) 92 86-100 30 Average Working Memory (WMI) 85 79-93 16 Low Average Full Scale IQ (FSIQ) 81 77-87 10 Low Average Vocabulary Acquisition (VAI) 71 66-81 3 Borderline Nonverbal (NVI) 90 85-96 25 Average General Ability (GAI) 80 75-87 9 Low Average Index results: L.T.'s Visual Spatial score (92, Average, 30th percentile) shows age-typical block construction and puzzle assembly, and her Working Memory score (85, Low Average, 16th percentile) sits at the lower edge of the average range on the picture-based memory tasks. Her Verbal Comprehension score (78, Borderline, 7th percentile) is notably lower and should be read alongside a design fact the publisher's own materials state: no subtest contributing to the FSIQ or the primary indexes at either band requires an expressive response on its floor items. Her limited spoken output therefore did not mechanically invalidate these composites, which she could answer by pointing, but the VCI still samples verbal comprehension and is plausibly constrained by her language development rather than by reasoning alone. The Vocabulary Acquisition Index (71, Borderline, 3rd percentile), which does include an expressive naming task, is the profile's lowest score and is reported as the language-acquisition indicator, converging with the speech-language evaluation rather than substituting for it. Overall ability and the reporting decision: The FSIQ of 81 (CI 77-87, 10th percentile, Low Average) is reported for completeness, but it blends verbal and nonverbal abilities that differ here by nearly a standard deviation. The Nonverbal Index of 90 (CI 85-96, 25th percentile, Average), derived entirely from the four subtests with no expressive-response demand, is used as the anchor for statements about her nonverbal cognitive ability. The General Ability Index (80) is reported but not used as that anchor, for a compositional reason write-ups often miss: at the 2:6 to 3:11 band the GAI includes the expressive naming task, so it does not sidestep the language concern the way the school-age GAI sidesteps working memory and processing speed. Each choice is stated, both scores remain in the table, and no composite has been swapped in silently. Stability and integration: Cognitive scores obtained at age 3 are current estimates, not fixed traits: they are less stable and less predictive than school-age scores, and re-evaluation is planned before kindergarten entry. On the adaptive side, the Vineland-3 (reported separately) shows Communication well below age expectations with stronger Daily Living and Socialization skills, a pattern consistent with the cognitive profile above. Eligibility for preschool special education is a team determination under this state's developmental-delay criteria; no single measure, including this one, may serve as the sole basis for that decision. Summary and linkage to recommendations: Testing shows Average nonverbal ability (NVI 90) alongside Borderline verbal comprehension (VCI 78) and a Borderline vocabulary-acquisition indicator (VAI 71), with the FSIQ (81) blending the two domains. The nonverbal-verbal contrast supports recommendation 1 (continuation and intensification of speech-language services through the transition) and recommendation 2 (language-rich preschool programming that lets her demonstrate learning through visual and hands-on formats). The stability caveat supports recommendation 3 (cognitive re-evaluation before kindergarten, noting she will cross into the 4:0 to 7:7 band, where the index architecture changes). The full profile, with the adaptive and speech-language findings, feeds recommendation 4: the eligibility team should apply the state's developmental-delay criteria to the whole evaluation, not to any single score. This sample is fictional and for educational purposes. It does not describe a real child or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, age band, norm set, and administration format are all named before any score appears, so the three-index architecture reads as a design fact rather than missing data, and the results stay interpretable at re-evaluation. The validity statement is tied to specific observed behavior, separation, engagement with the play materials, response formats, and breaks, which is the linkage preschool reviewers look for in place of a boilerplate sentence. Every composite appears with its confidence interval, percentile, and one named descriptor set, every number in the prose matches the table, and no subtest-level claims appear anywhere. The reporting decision is made in the open: the FSIQ stays in the table while the Nonverbal Index anchors nonverbal-ability statements, with the compositional reason stated, including why the younger-band GAI does not solve an expressive-language problem. The stability caveat, the adaptive-data integration, and the no-single-measure rule place the eligibility decision with the team, and every recommendation traces to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label each requirement's strength honestly. US federal special-education law is instrument-neutral: no regulation names the WPPSI-IV, and evaluators may "not use any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ). The eligibility category preschool evaluations most often feed, developmental delay, is a state option, not a federal mandate: states may apply it to ages 3 through 9 or any subset ( 34 CFR 300.8(b) ), a district may not use the category unless its state adopted it ( 300.111(b) ), and the ECTA Center's state-by-state table shows 47 of the 50 states use it with age windows that vary widely, while "CA, IA, PR, and TX do not use the developmental delay category" and "BIE applies the developmental delay category to ages 4-9". So the same profile that qualifies a child in one state routes through a different category next door, and eligibility language in the write-up should track your state's definition, not a generic one. Timing is law: the obligation to make FAPE available begins "no later than the child's third birthday" with an IEP or IFSP in effect by that date ( 300.101(b) , 300.124 ), which is why so many WPPSI-IV administrations cluster just before age 3 transitions, and re-evaluation "must occur at least once every 3 years" unless parent and agency agree otherwise (300.303(b)(2)). Gifted identification has no federal mandate at all: where thresholds exist they are state or district policy, and Pennsylvania's regulation is the instructive exemplar, naming "an IQ of 130 or higher" while directing that "Determination of gifted ability will not be based on IQ score alone" (22 Pa. Code 16.21). Testing time bills to payers under the psychological-testing code families (96130 and 96131 for evaluation services, 96136 and 96137 for administration and scoring, with 96112 and 96113 covering developmental testing) as payer policy with plan-specific rules. Edition and norms currency is the WPPSI-IV's quiet defensibility question. As of July 2026 the WPPSI-IV (2012) is the current edition and no WPPSI-5 has been announced anywhere, while Pearson's own WISC-V page banners the coming WISC-6, so a report that names the edition is future-proofing itself for the transition question referral sources will eventually ask; when a new edition does publish, Pearson's own transition guidance is that most practitioners move to the new edition within 8 to 12 months of its release. The norms themselves date to the test's 2012 development cycle, and aging norms have a quantified cost: the largest Flynn-effect meta-analysis estimates scores drift about 2.31 points per decade (2.93 for modern Wechsler-family tests), renorming visibly moves individual children, students near eligibility thresholds "lost an average of 5.6 points when retested on a renormed test" in the classic Kanaya, Scullin, and Ceci analysis , and the drift is not uniform across domains, with Wechsler-family analyses noting the particular sensitivity of nonverbal scales and cautioning against mechanical point adjustments. The write-up move is simple: name the edition and norm vintage, never hand-adjust scores for norm age, and compare like with like at re-evaluation. Format deserves one line too: Pearson's equivalence study for digital administration, Q-interactive Technical Report 14 , found all subtests within its 0.2 effect-size criterion, concluding "the WPPSI-IV produces consistent scores regardless of format", and the digital kit still uses physical manipulatives, paper response booklets for processing-speed subtests, and a paper stimulus book for Picture Memory, so "Q-interactive" and "paper" are not shorthand for fully digital versus fully analog; as of July 2026 Pearson also lists no WPPSI-IV-specific telepractice equivalence report alongside its WISC-V and WAIS-family guidance, so any remote administration deserves an explicit documented rationale. Access boundaries are the publisher's: qualification level C in the US, with the Canadian edition (French materials available) and the Australian and New Zealand standardised edition, whose normative sample was stratified for "age, gender, parental education, geographic location, and indigenous status", administered through registered psychologists, and Australian school-adjustment and NDIS processes treating cognitive results as supporting evidence alongside functional documentation by convention. Retest timing is convention as well: practice effects make short-interval re-administration hard to interpret, and the publisher's stability data cover only short retest windows, so state the plan rather than implying a rule. Wechsler, WPPSI, and Pearson are trademarks, in the US and other countries, of Pearson Education, Inc. or its affiliates; WPPSI-IV materials are copyrighted by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common WPPSI-IV write-up errors reviewers flag There is no payer audit series for preschool cognitive write-ups; the accountability literature here is psychometric, and it cuts close to everyday practice. The WPPSI-IV drew a split verdict in the Buros Center's 19th Mental Measurements Yearbook : one reviewer found it psychometrically strong, while Canivez documented that the manual reported no exploratory factor analyses despite wholesale subtest turnover, no incremental-validity evidence for the index scores, and interpretation schemes resting on what he called "a shared professional myth of subtest and profile utility". Administration mechanics are the publisher's problem; the errors below are write-up errors, and they are yours. The BastionGPT Clinical Advisory Board sees the same ones most often in WPPSI-IV report reviews: The age-band architecture botched. A five-index template forced onto a 2:6 to 3:11 administration, Fluid Reasoning and Processing Speed reported as "not obtained" as if they were missing data, a Cognitive Proficiency Index claimed for a 3-year-old, or the WPPSI-III's 7:3 upper bound repeated in a WPPSI-IV report. The two bands are different batteries by design, and even a published clinical-trial protocol has described the GAI as a fixed four-subtest composite when its composition is band-dependent. Name the band first; everything else reads through it. Subtest scatter and profile analysis written as findings. Strengths-and-weaknesses tables, pairwise subtest comparisons, and pattern claims promoted to clinical conclusions. The critical Buros review is blunt: subtest and profile interpretation methods "should not be used in clinical decision-making until psychometric support for them is provided", and clinicians following the manual's schemes "risk overinterpretation and misinterpretation of WPPSI-IV scores in clinical application". Keep composites in charge and subtest commentary qualitative. Descriptor drift and unlabeled vocabularies. A 92 labeled Average in the table but "solidly average ability" in the prose, classic and neutral descriptor sets mixed in one report, or no statement of which set is in use. The publisher's own FAQ says qualitative descriptors are "only suggestions and are not evidence-based; alternate terms may be used as appropriate", which makes naming your set mandatory, not optional. Prediction language a preschool score cannot carry. "Will remain significantly delayed", trajectory claims, or a gifted determination treated as settled from one administration. Stability climbs with age and shrinks with retest interval in the temporal-stability literature, and the manual's own short-term retest figures (FSIQ about .88 over a mean 23-day interval) are a best-case ceiling, not long-range prediction. Write current estimates, state the re-evaluation plan. Silent composite swaps and quiet score repairs. An ancillary index replacing the FSIQ without stated rationale, a substitution slipped into a primary index (the publisher allows none there, and at most one substitution for the FSIQ and ancillary composites, with proration available for the FSIQ only), or the younger-band GAI offered as the fix for an expressive-language problem when its composition still includes an expressive naming task. Report both scores, state the choice, give the reason. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your composite summary (band, scores, CIs, percentiles, descriptors) and get a drafted results-section narrative organized index by index, with the behavioral-observation linkage and any FSIQ-versus-ancillary reporting decision framed for your review. Cross-check a finished draft for the gaps reviewers flag: an unnamed age band, numbers that disagree with the table, descriptor drift, prediction language a preschool score cannot carry, and composites swapped in without a stated reason. Translate the results section into a plain-language summary for parents and preschool teams that keeps scores as current estimates and eligibility as the team's decision. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do WPPSI-IV scores actually mean? Two metrics carry everything. Subtest scaled scores run 1 to 19, with 7 to 12 usually considered average; composite scores, the FSIQ and the index scores, run 40 to 160, with 90 to 109 considered average, and each composite carries a percentile rank and a confidence interval. The manual supplies estimated-true-score confidence intervals, and its own text notes "there may be a preference for using the obtained score confidence interval" when the question is the child's functioning at the time of testing, a precision nuance the Buros review walks through. Qualitative labels are looser than most readers assume: the publisher's own FAQ says descriptors are "only suggestions and are not evidence-based; alternate terms may be used as appropriate", and Pearson's printed reports use classic Wechsler-era terms such as Borderline and Extremely Low. The defensible write-up reports score, interval, and percentile together, names the descriptor set it uses, and applies it consistently. Is there a WPPSI-5, and is the WPPSI-IV outdated? As of July 2026 there is no WPPSI-5: no announcement, pre-order page, or standardization recruitment in any market. The absence is informative because Pearson visibly advertises Wechsler revisions that do exist, the WAIS-5 published in 2024 and the WISC-V product page currently banners "WISC-6 is coming", while the WPPSI-IV page carries no such notice. The WPPSI-IV (2012) is therefore fully current and fully defensible; what ages a report is not the edition but the omissions. Norms from the 2012 development cycle do drift, about 2.31 points per decade in the largest Flynn-effect meta-analysis, which is a reason to name the edition and norm vintage in every report, not to hand-adjust scores. And when a fifth edition eventually arrives, the publisher's own transition guidance for the current edition was that most practitioners move to a new edition within 8 to 12 months of its release. Why does my report show only three index scores? Because the WPPSI-IV is two batteries divided at age 4, and a child tested at 2:6 to 3:11 gets the younger one by design. That band yields the FSIQ plus Verbal Comprehension, Visual Spatial, and Working Memory; Fluid Reasoning and Processing Speed exist only in the 4:0 to 7:7 band, which yields all five primary indexes. The ancillary architecture follows the same split: the Vocabulary Acquisition and Nonverbal Indexes exist at both bands, the General Ability Index exists at both but with band-dependent subtest composition, and the Cognitive Proficiency Index exists only from 4:0. Two related corrections worth knowing: the younger band's FSIQ is built from five core subtests including the Picture Memory working-memory task, which the publisher notes makes it broader than the WPPSI-III's four-subtest version, and the test's upper bound is 7 years 7 months, not the 7:3 figure many secondary pages copied forward from the WPPSI-III. A three-index preschool report is correct, not incomplete. Is the WPPSI-IV required for preschool special education or developmental delay eligibility? No law names it. IDEA is instrument-neutral and bars using "any single measure or assessment as the sole criterion" for eligibility ( 34 CFR 300.304(b)(2) ). The developmental-delay category itself is a state option: states may apply it to ages 3 through 9 or any subset ( 300.8(b) ), and a district may not use it unless the state adopted it (300.111(b)). The ECTA Center's current table shows 47 of the 50 states use a developmental-delay category with widely varying age windows, while "CA, IA, PR, and TX do not use the developmental delay category" and "BIE applies the developmental delay category to ages 4-9". What is fixed federally is timing: FAPE must be available "no later than the child's third birthday" with an IEP or IFSP in effect by that date (300.101(b)). So the WPPSI-IV earns its place by convention, as one appropriate cognitive measure inside a multi-source evaluation, and your write-up's eligibility language should track your state's definition. Should I use the WPPSI-IV or the WISC-V for a 6- or 7-year-old? The 6:0 to 7:7 overlap is deliberate, and the publisher's own selection logic is floor versus ceiling: for children with suspected below-average ability, the WPPSI-IV's lower floor differentiates better, while high-ability children need the school-age scale's higher ceiling. Pearson's overlap FAQ, written when the school-age edition was the WISC-IV, also noted a practical difference, the WPPSI-IV reaches an FSIQ in six subtests versus ten, and the trade-offs carry forward to the WISC-V , whose FSIQ uses seven. Developmental fit matters as much as psychometrics at this age: the WPPSI-IV's game-like, dauber-and-pictures format suits developmentally younger or shy children, while a verbally advanced 7-year-old is usually better served, and better challenged, on the WISC-V. Whichever you choose, say why in the report, name the norm set, and keep the choice consistent with the referral question, because at re-evaluation the two tests' different index architectures make cross-battery score comparison a stated limitation, not a trend line. When is the Bayley-4 a better choice than the WPPSI-IV at age 2 or 3? Choose by referral question. A developmental measure like the Bayley-4 answers "is development proceeding as expected across domains", covers motor, language, and social-emotional ground the WPPSI-IV never touches, and serves low-functioning children through its extended floor; the WPPSI-IV answers "what is this child's cognitive ability relative to peers" with an extended ceiling and the Wechsler common metric. That floor-versus-ceiling logic is the publisher's own, stated in its instrument-selection FAQ for the previous generation (Bayley-III versus WPPSI-III), and worth knowing in that dated form: the Bayley for suspected delay and trajectory questions, the Wechsler when an ability score is required. Publisher concordance data for the current editions show the two land close on average, a mean WPPSI-IV FSIQ of 103.3 against a Bayley-4 cognitive standard-score equivalent of 98.5 in the comparison sample, so crossing between them at the age-3 transition is expected, not alarming. Either way, cognitive scores at this age belong next to adaptive data such as the Vineland-3 and a speech-language evaluation, not in place of them. Can you get valid WPPSI-IV results from a minimally verbal or hard-to-test preschooler? Often yes, and the test's design helps more than most write-ups acknowledge. Pearson's own FAQ states that only Vocabulary and Picture Naming require expressive responses on the floor items, and "neither of these subtests are core to the primary index scores or FSIQ", so a child who points reliably can produce interpretable composites. The Nonverbal Index goes further, drawing entirely on subtests with no expressive demand, while the Vocabulary Acquisition Index serves as the language-acquisition indicator rather than a general-ability score; note that at the younger band the General Ability Index still includes an expressive naming task, so it is not the language workaround it becomes at school age. The non-negotiable part is behavioral documentation: separation, rapport, attention, breaks, and response to the play materials, ending in a validity statement tied to what you observed. When standardized administration genuinely breaks down, the defensible moves are to qualify the results as a likely underestimate, or to shift to a developmental measure and say why. Where can I find WPPSI-IV practice questions or sample items? Not on this page, and not legitimately on any public page. WPPSI-IV items, stimuli, and norms are protected test materials sold at qualification level C , and psychologists carry an ethical duty to maintain test security under APA Ethics Standard 9.11 . The pressure here is real: a test-prep market exists around preschool admissions and gifted screening, and that is exactly why coached exposure matters clinically, because familiarity with item types distorts the norm-referenced comparison every downstream decision relies on, at an age when scores are already their least stable. Even where gifted thresholds are codified, as in Pennsylvania's regulation naming "an IQ of 130 or higher", the same rule directs that the determination "will not be based on IQ score alone". A report may include scores, percentiles, descriptors, and your own prose about what the tasks measure; it may not include item content, stimuli, scoring rules, or reproduced norm tables. Can BastionGPT help with WPPSI-IV write-ups? Yes. BastionGPT is trained and clinically tested on psychological and psychoeducational evaluation reports, the parent documents a WPPSI-IV section lives inside. Paste a score summary (age band, composites, CIs, percentiles, descriptors, and your observation notes) and it drafts the results-section narrative with the band architecture right, the validity statement tied to observed behavior, and any FSIQ-versus-ancillary reporting decision framed with its rationale for your review. It can also check a finished section for the errors reviewers flag, an unnamed age band, numbers that disagree with the table, descriptor drift, prediction language, or a silent composite swap, and produce a plain-language summary for parents and preschool teams. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, WPPSI-IV product page : publication year, age range 2:6 to 7:7, qualification level C, completion times by band, scoring platforms, and the FAQ text quoted on this page (descriptor wording, expressive floor-item design, the five-subtest younger-band FSIQ, the WISC overlap and edition-transition guidance, and substitution and proration rules). Pearson, WPPSI-IV brochure : the per-band index architecture and ancillary-index compositions, the 1,700-child normative sample and its stratification, new subtests, and the ink-dauber processing-speed design. Pearson, Q-interactive Technical Report 14: WPPSI-IV Equivalence of Q-interactive and Paper Formats (Drozdick, Getz, Raiford, and Zhang, June 2016): the 0.2 effect-size criterion and format-equivalence conclusion quoted on this page. Pearson Q-global sample reports for the WPPSI-IV (score reports for ages 2:6 and 6, and the interpretive report, linked from the product page): printed descriptor vocabulary, confidence-interval columns, and the per-band ancillary index rows reflected in the fictional sample. Pearson Clinical Canada, WPPSI-IV CDN product page , and Pearson Clinical Australia and New Zealand, WPPSI-IV A&NZ product page : local editions, French Canadian materials, and the A&NZ normative stratification wording. Pearson, Bayley-III product page FAQ (the prior-generation Bayley-versus-WPPSI selection guidance) and the Bayley-4 Technical Report 1 (WPPSI-IV concordance figures cited in the FAQ above). Canivez, G. L., and Thorndike, T., reviews of the WPPSI-IV in the 19th Mental Measurements Yearbook (Buros Center for Testing, 2014; author preprint ; Buros 19th MMY test list ): reliability and stability figures, the CFA structure, the EFA and incremental-validity critique, the interpretation-methods conclusion, and the confidence-interval discussion quoted on this page. US Department of Education regulations, 34 CFR: 300.8(b) , 300.101(b) , 300.111(b) , 300.124 , 300.303(b)(2), and 300.304(b)(2) . ECTA Center, States and jurisdictions that use developmental delay for IDEA Part B eligibility (sourced to OSEP 2021 child-count documentation): the 47-state count, age-window table, the CA, IA, PR, TX exclusion line, and the BIE age range. Trahan LH, Stuebing KK, Fletcher JM, Hiscock M, The Flynn effect: a meta-analysis , Psychological Bulletin, 2014;140(5):1332-1360: the 2.31 and 2.93 points-per-decade estimates. Kanaya T, Scullin MH, Ceci SJ, The Flynn effect and U.S. policies , American Psychologist, 2003;58(10):778-790: the renorming retest loss quoted on this page. Zhou X, Zhu J, Weiss LG, Peeking Inside the "Black Box" of the Flynn Effect: Evidence From Three Wechsler Instruments, Journal of Psychoeducational Assessment, 2010;28(5):399-411 ( doi:10.1177/0734282910373340 ): domain-uneven Flynn evidence and the caution against routine score adjustment. Schuerger JM, Witt AC, The temporal stability of individually tested intelligence, Journal of Clinical Psychology, 1989;45(2):294-302 ( doi link ): age and retest interval as predictors of IQ stability. Pennsylvania Code, 22 Pa. Code 16.21 : the codified gifted threshold and the bar on IQ-only determinations. CMS Medicare Coverage Database, billing and coding articles A57481 and A57780 : the psychological, neuropsychological, and developmental testing code families. ClinicalTrials.gov, NCT05545540 : the published protocol whose fixed four-subtest GAI description illustrates the band-composition confusion discussed in the common-errors section. American Psychological Association, Ethical Principles of Psychologists and Code of Conduct , Standard 9.11 (maintaining test security). ## Related templates Psychoeducational Evaluation Report WISC-V Report Write-Up Psychological Evaluation Report Vineland-3 Report Write-Up Bayley-4 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/care-coordination-note All note templates # Care Coordination Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A care coordination note documents the work of organizing a client's care across providers: contacts made, referrals and their status, consent to share information, handoffs, and outcomes. Therapists, case managers, and integrated care teams write one whenever they arrange, share, or follow up on care with someone outside the session. Most run 75 to 250 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, case managers, care coordinators, integrated behavioral health teams Audience Care team, next provider, supervisors, payers, auditors Typical length 75 to 250 words · 5 to 10 minutes by hand (clinical team estimate) Format family Operational continuity note (compare: case management note, transfer-of-care summary) When it's used When arranging, sharing, or following up on care across providers Standards context A documentation convention; the billable care-management services it supports carry specific payer rules On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a care coordination note? A care coordination note records one kind of clinical work: organizing a client's care across the people involved in it, and sharing the information each of them needs. The framing has a clear source. The Agency for Healthcare Research and Quality (AHRQ) defines care coordination as "deliberately organizing patient care activities and sharing information" among everyone concerned with a patient's care. Clinicians also call these contact notes, coordination-of-care notes, or collateral-contact notes, and the activity often sits inside a broader case management note. One distinction prevents most of the confusion. A care coordination note is documentation of coordination activity, not a billable service, and it has no CPT code of its own. Medicare defines separate service families for coordinating care over time, including Chronic Care Management, Transitional Care Management, and Behavioral Health Integration, and each carries its own consent, care-plan, and time requirements. Writing a care coordination note does not by itself make the time billable. When you are documenting to support one of those services, the note has to meet that service's rules; when you are simply recording that you called a psychiatrist or sent a referral, it is an ordinary entry in the client's record. ## Who uses care coordination notes and when Care coordination notes appear wherever a client's care crosses a boundary: a referral to a psychiatrist or a primary care provider, a discharge from a higher level of care, a school or agency involved in a young person's treatment, or a benefits system that needs a report. Case managers and care coordinators write them constantly, and solo therapists write them whenever they pick up the phone about a shared client. The note earns its place when coordination is frequent or high stakes, such as a treatment plan that depends on other providers doing their part, or a care transition where a dropped handoff can undo months of work. For the clinical reasoning those handoffs carry forward, see the case formulation . ## Care coordination note structure: what goes in each section A care coordination note is short and operational. It answers four questions a reviewer will ask: why did coordination happen, who did you contact, what were you allowed to share, and what happens next. These sections cover all four, and each carries the pitfall that most often weakens it. Reason for coordination. The trigger: a new referral, a care transition, a risk event, or a scheduled review. State it in one line so the note has a purpose a reader can see. Pitfall: a coordination note with no stated reason, which reads as time spent for no documented purpose. Participants and contacts. Who you contacted: name, role, organization, the method (phone, secure message, portal, fax), the direction (you reached out or they reached you), and the date. Pitfall: "coordinated with the PCP" with no who, when, or how, which a reviewer cannot verify. Consent to share. The basis for the disclosure: the treatment exception, a signed release of information on file, or a substance use disorder consent under 42 CFR Part 2. Pitfall: sharing protected information without naming the authority that permits it. Referrals and status. Each referral, to whom and for what, plus where it sits in the loop: sent, accepted, scheduled, completed, results returned. Pitfall: documenting the referral you sent and never closing the loop, so no one knows whether the client arrived. Information exchanged and actions. What you shared or requested, the decisions reached, and the tasks assigned with an owner. Pitfall: "discussed the case" with none of the substance or the decisions. Outcome and follow-up. The result of the coordination, the next step, who owns it, and the date or trigger for the next check. Pitfall: a follow-up with no owner and no date, so the task falls through the transition. ### Blank template (copy and adapt) Client: [initials] Date: Clinician: Reason for coordination: [new referral / care transition / risk event / review] Participants and contacts: [name, role, org, phone/portal/fax, outreach or received, date] Consent to share: [treatment exception / signed ROI on file (date) / 42 CFR Part 2 consent (date) / N/A] Referral(s) and status: [to whom, for what, urgency; sent / accepted / scheduled / completed / results returned] Information exchanged and actions: [what was shared or requested; decisions; tasks and owners] Outcome and follow-up: [result; next step; responsible party; next check date or trigger] Time spent (if billing a care-management service): [total minutes; who performed] Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample care coordination note Scenario: an adult client re-entering outpatient care after an inpatient psychiatric stay, with the therapist coordinating the handoff to primary care and psychiatry. All details are fictional. Client: R.K., 41 · Date: 07/09/2026 · Clinician: J. Rivera, LCSW Reason for coordination: Care transition after inpatient psychiatric discharge on 07/07/2026; re-establishing outpatient care and confirming the handoff. Participants and contacts: Phone call to County General discharge social worker (A. Alvarez, LCSW) 07/09 at 10:15, outbound. Secure message to PCP (Dr. Nguyen) 07/09. Portal message left for client 07/09. Consent to share: Signed release of information on file dated 06/02/2026 authorizing exchange with the PCP and the inpatient team. No substance use disorder records involved. Referrals and status: Psychiatry referral (Dr. Patel) for medication management sent 07/09, accepted, first appointment scheduled 07/16. PCP follow-up scheduled 07/14. Discharge summary requested from County General, received 07/09. Information exchanged and actions: Shared the current medication list and updated safety plan with the PCP. Confirmed the client has a two-week medication supply. Reconciled discharge medications against the outpatient record; no discrepancies. Outcome and follow-up: Handoff complete; psychiatry and PCP appointments booked and therapy resumes 07/13. Writer to confirm the client attended psychiatry and reconcile any medication changes by 07/17. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every contact carries a name, a role, a method, and a date, so the coordination is verifiable rather than a vague "coordinated care." The consent basis is named and dated, a signed release on file, so each disclosure is defensible. Referrals show loop status, not just that they were sent, which is what actually closes the referral loop. The follow-up names an owner and a date, so nothing falls through the transition. Time and the person who spent it can be recorded, so the entry can support a care-management service if one is billed. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A care coordination note is part of the designated record set: the client can request it, other providers may rely on it, and a payer can audit it. Most coordination is a disclosure for treatment, and under HIPAA a provider may share protected health information with another provider for treatment without the client's authorization ( 45 CFR 164.506 ); the rule defines treatment to include the coordination of care, and the minimum-necessary standard does not apply to disclosures for treatment. Two limits matter in behavioral health. Substance use disorder records from a Part 2 program generally need the client's written consent to share ( 42 CFR Part 2 ), and true psychotherapy notes stay segregated with their own authorization. Naming the consent basis in the note is what shows you applied the right rule. The confusion worth resolving is billing. Recording that you coordinated care does not make the time reimbursable, and a care coordination note has no billing code of its own. The care-management services that pay for coordination, including Chronic Care Management and Behavioral Health Integration , each require specific elements: the patient's consent to the service, a documented care plan, and a running total of coordination time performed by identified staff. HIPAA permitting the disclosure is a separate question from the payer requiring that consent. Clinical staff may document coordination under the billing provider's general supervision, and the provider authenticates the entry. The note is a convention; those service requirements are payer policy, and an auditor checks the note against them. If you are handing a client off rather than coordinating over time, a treatment plan or a transfer summary may fit the moment better. ↑ Back to the template and downloads ## Common care coordination note errors auditors flag The errors that draw audit attention cluster around coordination that gets billed as a care-management service. In an August 2021 audit ( A-07-19-05122 ), the HHS Office of Inspector General found about $1.9 million in overpayments across 50,192 Chronic Care Management claims from 2017 and 2018, driven mostly by duplicate and overlapping billing rather than by the coordination work itself. The BastionGPT Clinical Advisory Board sees the same documentation gaps most often in care coordination note reviews: Open referral loops. The referral is documented as sent, then never reconciled, so the record cannot show the client reached the next provider. Coordination billed without its scaffolding. Time is claimed for a care-management service with no documented consent, care plan, or time total, which is the fastest way to lose the claim. Unverifiable contacts. Entries such as "spoke with the team" with no name, role, date, or method. Consent basis missing. Protected or Part 2 information shared with no note of the release or the exception that permitted it. Copy-forward coordination. The same coordination text repeated across entries, which reads to a reviewer as a record that does not reflect distinct work. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on care coordination notes. Draft a care coordination note from your contact log, referral details, and a few bullets. Combine handoff details, a discharge summary, and referral updates into one clear coordination summary. Check a note for open referral loops, a missing consent basis, or a follow-up with no owner before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a care coordination note be? Most care coordination notes run 75 to 250 words and take 5 to 10 minutes to write. It is an operational note: long enough to show who you contacted, what you were allowed to share, and what happens next, short enough to write right after the call. Length beyond that rarely adds anything a reviewer needs. Is a care coordination note the same as a case management note? They overlap, but a care coordination note is narrower. It documents a specific coordination episode: the contacts, referrals, consent status, and follow-up. A case management note is broader, tracking a client's needs, goals, barriers, and progress against a care plan over time. Coordination activity often lives inside case management work, and many teams keep both. Can I bill for the time I spend coordinating care? Coordination by itself is not separately billable, and a care coordination note has no billing code. Time can count only inside an approved care-management service such as Chronic Care Management, Transitional Care Management, or Behavioral Health Integration, each with its own rules. Chronic Care Management, for example, expects at least 20 minutes of clinical-staff coordination time in a month, the patient's consent to the service, and a care plan. See the treatment plan those services build on. Do I need the client's consent to coordinate care or share information? Under HIPAA you may share protected health information with another provider for treatment without a separate authorization, and coordinating a client's care counts as treatment (45 CFR 164.506). Two caveats: substance use disorder records under 42 CFR Part 2 need the client's written consent, and billing a care-management service requires documenting the patient's consent to that service. Record which basis applies. The psychotherapy notes authorization page covers the separate rules for process notes. How is a care coordination note different from a referral letter or a transfer-of-care summary? A referral letter asks another provider to take on a piece of care. A transfer-of-care summary hands the whole case to a new provider. A care coordination note documents the ongoing work of keeping providers aligned while care continues: the calls, the referral tracking, the consent status, and the follow-up. You often write a coordination note about a referral you sent. Who can write a care coordination note? Any qualified member of the care team can document coordination. When the work supports a billable care-management service, Medicare lets clinical staff perform and document it under the billing provider's general supervision, and the billing provider authenticates the entry. Note who did the work and who is billing, and sign with your credentials. How long do I need to keep care coordination notes? There is no single HIPAA retention rule for patient records; retention comes from state, payer, and program rules. Medicare commonly points to a seven-year window for records that support claims, and states set their own minimums, often longer for records of minors. Because a coordination note can support a billed care-management service, keep it at least as long as the claim rules require and check your jurisdiction. Can BastionGPT write a care coordination note? Yes. Give it your contact log, referral details, and a few bullets, and it drafts a structured coordination note for your review. It can also pull a coordination summary together from a discharge summary and referral updates, and flag open referral loops or a missing consent basis before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: Agency for Healthcare Research and Quality, Care Coordination : the definition and the specific coordination activities. 45 CFR 164.506 : HIPAA permits uses and disclosures for treatment without authorization. 45 CFR 164.501 : HIPAA defines treatment to include the coordination of care among providers. 42 CFR Part 2 : consent requirements for substance use disorder records. CMS, Chronic Care Management Services (MLN909188) : consent, care plan, and time requirements. CMS, Behavioral Health Integration Services (MLN909432) : Collaborative Care and BHI documentation. HHS Office of Inspector General, report A-07-19-05122 (2021) : $1.9 million in Chronic Care Management overpayments, 2017 to 2018. 42 CFR 410.26 : documentation of services by clinical staff under general supervision. ## Related templates Treatment Plan Psychotherapy Progress Note SOAP Note Case Formulation Note DAP Note Crisis Psychotherapy Note Discharge Summary Release of Information (ROI) Authorization Collaborative Care (BHI/CoCM) Note Multidisciplinary Team Meeting Note Case Conference Note Hospital Consult & Liaison Note Consultation Response & Report Chronic Care Management (CCM) Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Australia Telehealth & MBS Case Conference Note Communication Log & Portal Message Note Appointment No-Show & Outreach Note 42 CFR Part 2 Consent Case Management Note Transfer-of-Care Summary CANS Assessment Documentation Braden Scale Documentation Apgar Score Documentation Morse Fall Scale Documentation Pressure Injury Staging Documentation OASIS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/discharge-summary All note templates # Mental Health Discharge Summary: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A mental health discharge summary is the closing document of an episode of care: why treatment started, the course and response, final diagnoses, the client's condition and risk status at discharge, and the aftercare plan. Hospitals and community mental health centers must produce one under federal rules; in outpatient therapy it is professional convention, with some state exceptions. A typical summary runs 300 to 1,200 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Discharging clinician or treatment team: therapists, psychologists, counselors, psychiatrists, program staff Audience Next treating provider, primary care, payers, auditors, the client Typical length 300 to 1,200 words · 20 to 45 minutes by hand (clinical team estimate) Format family Episode summary and transition document (compare: transfer-of-care summary, care coordination note) When it's used At the end of an episode of care: program or hospital discharge, therapy termination, or transfer Standards context Required record content in hospitals and community mental health centers; a professional convention in most outpatient practice On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a mental health discharge summary? A mental health discharge summary is the closing document of an episode of care. It condenses why treatment started, what happened, and what should happen next into a page or two the next reader can act on: a receiving provider, a primary care physician, a payer reviewer, or the client. Clinicians and payers also call it a discharge note, a termination summary (the common name in outpatient therapy), a closing summary, or a transition record. No one invented the format: it descends from the hospital-standardization movement, whose 1919 Minimum Standard first pushed hospitals toward complete, reviewable records, and today's wording has sat in Medicare's hospital rules since 1986. Its regulatory status depends on the setting, and that distinction organizes everything on this page. In hospitals the document is required by law: the Medicare Conditions of Participation list a "discharge summary with outcome of hospitalization, disposition of case, and provisions for follow-up care" among required record content, psychiatric hospitals carry their own three-part standard (a recapitulation of the hospitalization, aftercare recommendations, and the condition on discharge), and community mental health centers must forward theirs within 2 working days of a transfer. In private outpatient practice no federal rule names the document: writing one at termination is professional convention, though some regulators write it into law, Ohio's discharge summary rule and Washington's discharge-statement requirement among them. A discharge summary looks backward at a whole episode; a care coordination note documents contacts while care is still active. ## Who uses mental health discharge summaries and when Every level of care produces one. Inpatient psychiatric units write them at discharge, residential and PHP or IOP programs at completion or step-down, community mental health centers at transfer, and outpatient therapists at termination, planned or not. The audiences differ more than the document does: a hospital summary feeds the follow-up appointment in the first week after discharge, when risk is highest, while an outpatient termination summary mostly serves the record itself, the next treating clinician, and any payer that later asks how the episode ended. Use a discharge summary when an episode is ending. For contacts during active treatment, use a care coordination note ; a referral letter asks another provider to begin care rather than closing your own. ## Mental health discharge summary structure: what goes in each section A discharge summary condenses an entire episode of care into a document the next reader can use. These are the sections receiving providers, accreditors, and auditors expect, with the pitfall that most often undermines each. Client and episode identification. Client identifiers, the program or level of care, the intake or admission date, the discharge date, and the treating clinician or team. Pitfall: no episode dates. Without them a reviewer cannot bound the episode or match the summary to billed services. Reason for treatment. Two or three sentences on why care started: the presenting problem, the diagnosis at intake, and any precipitating events. Pitfall: restating the whole intake note. Orient the reader; the intake note already holds the detail. Course of treatment. Modalities, session frequency, attendance, response to treatment, and significant clinical events such as level-of-care changes, crises, or medication changes. Pitfall: a session-by-session replay. Summarize the arc; the progress notes hold the sessions. Final diagnoses. The discharge diagnoses, with any changes from intake stated rather than left for the reader to discover. Pitfall: an intake diagnosis carried forward unexamined when the course of treatment plainly revised it. Condition at discharge. Clinical status at the final contact, with outcome measures at baseline and discharge where you have them. Pitfall: writing "improved" with no measure, example, or functional detail to support it. Reason for discharge. State it plainly: goals met, mutual decision, client-initiated, lost to contact after documented outreach, transfer to a higher level of care, or administrative discharge. Pitfall: euphemism. "Completed treatment" written over a lost-to-contact discharge misstates the record. Medications at discharge. If a prescriber is involved: the final medication list, changes made during the episode, and who manages each prescription going forward. Pitfall: a list that changed during treatment with no reconciliation, so the next prescriber inherits a contradiction. Risk status at discharge. The risk picture at the final contact, and any safety plan completed or updated during the episode. Pitfall: silence. The closing document of record is where a later treater, or a later reviewer, will look first. Aftercare plan. Referrals with names and contact details, scheduled or recommended appointments with timeframes, crisis resources provided, and the criteria that should bring the client back. Pitfall: "follow up as needed." Name the provider, the interval, and the return criteria. Pending items. Records to forward, outstanding results, unfinished authorizations, and who owns each one. Pitfall: pending items with no owner. Once the episode closes, unowned tasks die quietly. ### Blank template (copy and adapt) Client: [initials] Program/Level of care: Episode: [intake or admission date] to [discharge date] Clinician/team: [name, credentials] Referral source: Reason for treatment (presenting problem and diagnosis at intake): Course of treatment (modalities, frequency, attendance, response, key events): Final diagnoses (note any changes from intake): Condition at discharge (status; outcome measures baseline vs final): Reason for discharge: [goals met / mutual / client-initiated / lost to contact after outreach / transfer / administrative] Medications at discharge (final list; who manages each going forward): Risk status at discharge (assessment; safety plan status): Aftercare plan (referrals with contacts, appointments, crisis resources, return criteria): Pending items (item and owner): Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample mental health discharge summary Scenario: outpatient individual therapy, planned termination after treatment goals were met. All details are fictional. Client: J.R., 41 · Program: Outpatient individual psychotherapy · Episode: 02/10/2026 to 07/10/2026 · Clinician: A. Morales, LPC · Summary written: 07/10/2026 Reason for treatment: J.R. began weekly therapy on 02/10/2026 for a depressive episode following a divorce, with low mood, loss of interest, early-morning waking, and withdrawal from friends. Intake diagnosis: major depressive disorder, single episode, moderate (F32.1). Baseline PHQ-9: 16. Course of treatment: 19 sessions attended of 21 scheduled, weekly through April and biweekly from May. Treatment used cognitive behavioral therapy with behavioral activation, then relapse-prevention work in the final month. Client's primary care physician started sertraline 50 mg in March and has managed it throughout. Mood, sleep, and social activity improved steadily from session 8 onward; no crises or level-of-care changes during the episode. Final diagnoses: Major depressive disorder, single episode, in partial remission (F32.4). Changed from F32.1 at intake to reflect treatment response. Condition at discharge: PHQ-9 at final session: 5 (baseline 16). Client reports stable mood, restored sleep of about 7 hours, regular attendance at work, and weekly social contact. Functioning consistent with pre-episode baseline. Reason for discharge: Treatment goals met. Termination planned collaboratively across the final three sessions. Medications at discharge: Sertraline 50 mg daily, prescribed and managed by the client's primary care physician, Dr. K. Patel; unchanged for the final two months. No medications managed by this practice. Risk status at discharge: Client denied suicidal ideation throughout treatment and at the final session; no risk indicators observed at discharge. Crisis resources reviewed and provided in writing. Aftercare plan: Relapse-prevention plan completed in session 18: early warning signs, skills list, and support contacts, copy given to client. Client declined a scheduled booster session; agreed return criteria: PHQ-9 above 9 for two weeks or any return of hopelessness. With a signed release, a copy of this summary was sent to Dr. Patel on 07/10/2026. Crisis line provided: call or text 988. Pending items: None. Chart closed 07/10/2026. Signed: A. Morales, LPC, 07/10/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The episode is bounded: dates, session count, and level of care let any reader place the summary against the chart and the billing record. PHQ-9 at baseline and discharge (16 to 5) turns "improved" into evidence a reviewer can verify. The reason for discharge is stated plainly, and the aftercare plan matches it, including the criteria that would bring the client back. Risk status is documented at the final contact, with crisis resources given in writing. Diagnoses and medications are reconciled: the final diagnosis reflects treatment response, and one named prescriber owns the ongoing medication. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A discharge summary travels further than any other document in the chart, and the federal rules are unusually specific about timing. Hospital records may take up to 30 days after discharge to complete, but the discharge-planning rules require the necessary medical information to move at the time of discharge, and a community mental health center must forward its discharge summary within 2 working days of a transfer. The clinical logic matches the regulatory one: Joint Commission sentinel-event data for 2024 classified 75% of suicide sentinel events as occurring within 7 days of discharge or while in intensive behavioral health care, and among 139,694 hospitalized youths, follow-up within 7 days was associated with less than half the suicide risk of later contact. An aftercare plan that books follow-up inside that first week, and a summary that reaches the follow-up provider before the visit, are the two levers this document controls. In outpatient practice the format is a convention and the content is the requirement, and a few regulators have made the content explicit: Ohio lists eight required discharge summary fields for licensed agencies, Washington requires a discharge statement when a client leaves without notice, and Ontario's CRPO expects a conclusion-or-termination entry with reasons and referrals. Whatever headings you use, show how the episode ended: the reason for discharge stated plainly, condition at discharge with a measure where you have one, risk status at the final contact (a VA Office of Inspector General review found safety plans missing for 12% of discharged patients), and an aftercare plan with named providers and return criteria. When a client cannot be reached, write what you know: the date of the last direct assessment, what it found, and that current status could not be reassessed. Absence of new information is not a finding of "no risk." Sending the summary outside your organization still requires a signed release, and retention follows the record it belongs to (see the table below). For the risk documentation itself, see the suicide risk assessment and safety plan templates. ↑ Back to the template and downloads ## Common mental health discharge summary errors auditors flag The research and survey record on discharge summaries is unusually specific about failure. A JAMA systematic review of hospital discharge communication found a summary in hand at the first post-discharge visit only 12% to 34% of the time, tests still pending at discharge missing from 65% of summaries, and follow-up plans missing from up to 43%. In the Joint Commission's latest published survey cycle , "complete discharge summary" accounted for 138 record-of-care findings, among the most frequent record-of-care issues. The requirement itself also moved in the January 2026 manuals: general hospitals now find it at RC.12.01.01, psychiatric hospitals at RC.11.01.01, not the retired RC.02.04.01 EP 3 that most template pages still cite. The BastionGPT Clinical Advisory Board sees the same errors most often in discharge summary reviews: The summary never reaches the next provider. Writing it is half the job. Hospital rules require the necessary information to move at the time of discharge, CMS surveyors probe whether the material actually reached the patient, family, and next source of care, and in a practice setting nothing moves without a signed release and a named recipient. "Improved" with no measure. Surveyor guidance for psychiatric hospitals expects a baseline of psychiatric, physical, and social functioning at the time of discharge. A condition section with no score, example, or functional detail cannot show it. A euphemized reason for discharge. "Completed treatment" written over a lost-to-contact or against-advice ending misstates the record, and "no current risk" written about a client nobody could reach converts silence into a finding. State what happened, and what could not be assessed. Risk silence at the final contact. The closing document of the episode says nothing about risk either way, in the window where sentinel-event data concentrates. Document status even when it is unremarkable. Pending items with no owner. Results, records, and authorizations still open at discharge go unmentioned, and unowned items die when the episode closes. ### Mental health discharge summaries in the US, Canada, and Australia Aspect United States Canada Australia Status Law in hospitals and community mental health centers (42 CFR 482.24, 482.61, 485.914); convention in private outpatient practice, with state exceptions such as Ohio's field list and Washington's discharge statement No national rule; Ontario hospitals must keep discharge summaries by regulation, CPSO expects physician completion within 48 hours, and CRPO requires a conclusion-or-termination record No named national mandate; NSQHS accreditation requires defined handover content, NSW public mental health forwards the summary within 12 hours, and it is a defined My Health Record document Terminology Discharge summary; termination summary (outpatient); discharge statement (Washington no-notice) Discharge summary; record of conclusion or termination (CRPO) Discharge/transfer summary; written end-of-course report (Better Access) What changes Setting-specific content lists and clocks; surveyor and payer chart review Provincial hospital regulation plus college standards; supervisor co-signature for supervised psychology services Handover content is organisation-defined; My Health Record upload conventions; report to the referrer at course end Retention Hospital records at least 5 years federally; state rules govern practices (California 7 years, New York 6) Ontario 10+ years from last interaction or 18th birthday; BC hospitals keep discharge summaries 10 years as primary documents, BC clinicians 16 years 7 years from last entry, or to age 25 for minors; federal privacy principles can require destruction once no retention duty remains The episode-closing structure travels well across all three countries. What changes is the administrative layer around it: who must produce it, how fast it must move, and how long it must be kept. Adapt those details to your setting and jurisdiction. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on mental health discharge summaries. Draft a complete discharge summary from the chart you already have: it condenses the intake note, progress notes, and treatment plan reviews into one episode narrative. Check a finished draft for the elements reviewers look for: episode dates, final diagnoses, outcome measures, reason for discharge, risk status, and a specific aftercare plan. Turn the clinical summary into a plain-language aftercare letter the client can actually use, with the same facts at an accessible reading level. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to write a discharge summary when therapy ends? It depends on the setting. Hospitals must have one in the record by federal rule, psychiatric hospitals have their own three-part requirement, and community mental health centers must forward one when a client transfers. In private outpatient practice, no federal regulation names the document: a closing record is professional convention, though some regulators require one. Ohio lists eight required fields , and Ontario’s CRPO expects a conclusion-or-termination entry, which can be a closing note rather than a separate form. The safest reading: the title is optional, the closure is not. An episode that ends with no closing documentation is a gap every reviewer notices. What do I write when a client just stops coming? No national rule in the US, Canada, or Australia requires a separately titled summary when a client stops attending, and none sets a deadline for writing one. The record must still show an adequate closing: the last service date, the last-known clinical and risk status with its date, your documented outreach and any response, why the chart is closing, and the referrals or crisis resources you offered. Where rules exist they cut both ways: Washington requires a discharge statement when a client leaves without notice, and Ohio’s field list drops the referral requirement when the client discontinued without notice. One wording rule matters most: never write "no current risk" for someone you could not reach. Record the date of the last real assessment and that reassessment was not possible. How soon after discharge does the summary have to be written? It depends on who you are. Hospitals get 30 days to complete the record, but the necessary transition information must move at the time of discharge. A community mental health center forwards its summary within 2 working days of a transfer. Ontario’s college of physicians expects the most responsible physician to complete it within 48 hours. New South Wales public mental health services must forward it within 12 hours and hand the client a copy at discharge. Washington gives licensed agencies seven working days for a planned discharge. Private outpatient practice has no regulatory deadline: write it close to the final contact, while the episode is fresh. Can I bill for writing a discharge summary? The document itself is not a billable service. In facilities, the discharging practitioner’s service on the day of discharge is billed as discharge day management, which pays for the visit, not the paperwork. On the receiving side, the practitioner who picks the client up can bill transitional care management , which requires interactive contact within 2 business days of discharge and a face-to-face visit within 7 or 14 calendar days. For an outpatient therapist, no psychotherapy code pays separately for writing a termination summary: documentation is part of the service. Australia ties its Better Access end-of-course report to the covered course rather than paying for it separately. What is the difference between a discharge summary, a transfer summary, and a referral letter? A discharge summary closes an episode and looks backward: course, response, final status, aftercare. A transfer-of-care summary moves an active episode to a new provider and leads with current status, active risks, and pending items; it may need to move before the chart is complete. A referral letter asks another provider to begin care and leads with the question you want answered. Coordination during treatment belongs in a care coordination note . In outpatient work, a final progress note can serve as the closing record when it carries the closure content; where a regulation names a discharge summary, a progress note cannot replace it. What do I document when a client leaves against medical advice? Record the fact and time of the departure, the capacity and mental-status findings you actually made, the risks you discussed, what you offered (medications, prescriptions, alternatives, follow-up, crisis resources), who was notified, and the client’s response. British Columbia’s hospital regulation makes the point explicit: an against-advice departure must be stated on the discharge order itself. Leaving against advice ends the stay, not your record duties; the summary should show what could not be completed because the client left. Who may write a discharge summary, and does a supervisor have to co-sign? Federal hospital rules require entries to be authenticated by the person responsible for the service and leave authorship to state law and hospital policy; the psychiatric-hospital summary standard names no profession. Ontario shows both patterns at once: the college of physicians makes the most responsible physician accountable for completing the hospital summary within 48 hours, and the college of psychologists requires a supervisor to co-sign any document others will rely on for decisions affecting care. For pre-licensed clinicians in the US, the co-signature answer lives in your state’s supervision rules, your payer’s policies, and your agency’s requirements, not in a federal discharge-summary rule. How long do I keep a discharge summary? It follows the record it belongs to. HIPAA sets no retention period for clinical records ; Medicare requires hospital records be kept at least 5 years, and state rules govern practices: California expects 7 years after termination, New York at least 6. Ontario expects 10 or more years from the last interaction or from the client’s 18th birthday. British Columbia’s hospital regulation classes discharge summaries as primary documents kept 10 years, and its clinician college expects 16. Australian boards expect 7 years from the last entry, or until a former minor turns 25, and the federal privacy principles can require destruction once no retention duty remains. Keep it retrievable, not just retained: it is usually the first document a later requester wants. Can BastionGPT draft a discharge summary from my chart? Yes. Give it the intake note, progress notes, and treatment plan reviews, or just bullets, and it drafts the episode narrative with the elements in this template: course, final diagnoses, condition with measures, reason for discharge, risk status, and aftercare. It can also produce a plain-language version for the client. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 42 CFR § 482.24 : hospital medical records, including the discharge summary content line, the 30-day completion clock, and the 5-year federal retention minimum. 42 CFR § 482.61(e) : the psychiatric-hospital discharge summary standard (recapitulation, aftercare recommendations, condition on discharge). 42 CFR § 482.43(b) : necessary medical information must accompany the patient at the time of discharge. 42 CFR § 485.914(e) : community mental health center discharge summary content and the 2-working-day forwarding rule on transfer. CMS, State Operations Manual, Appendix AA : surveyor guidance on what the psychiatric discharge summary must actually demonstrate (Tags B133 and B134). Joint Commission, updated accreditation manual, record of care : the January 2026 relocation of the discharge summary requirement. Joint Commission, Sentinel Event Data 2024 Annual Review : 75% of suicide sentinel events occurred within 7 days of discharge or during intensive behavioral health care. Ohio Administrative Code 5122-27-05 : the discharge summary field list for licensed behavioral health providers. Washington WAC 246-341-0640 : discharge statement for no-notice departures; seven working days for planned discharge information. Ontario Reg. 965 (Hospital Management), s. 19 : discharge summaries as required hospital-record content. College of Physicians and Surgeons of Ontario, Transitions in Care : the 48-hour completion expectation for the most responsible physician. CRPO, Clinical Records standard : the record of conclusion or termination, reasons, and referrals. NSW Health, PD2019_045 : consumer copy at discharge and the 12-hour forwarding requirement. Australian Commission on Safety and Quality in Health Care, Communicating for Safety Standard : organisation-defined minimum handover content. Australian Digital Health Agency, discharge summaries in My Health Record : the national document type and its content. Psychology Board of Australia, Code of Conduct : continuity duties when services end and record retention (7 years, or age 25 for minors). ## Related templates Care Coordination Note Intake Note Treatment Plan Safety Plan Psychotherapy Progress Note Release of Information (ROI) Authorization Treatment Plan Review & Update Collaborative Care (BHI/CoCM) Note Multidisciplinary Team Meeting Note Case Conference Note Hospital Consult & Liaison Note Consultation Response & Report Chronic Care Management (CCM) Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Utilization Review Summary Patient Instructions & After-Visit Summary Appointment No-Show & Outreach Note Records Request & Chart Copy Packet Transfer-of-Care Summary Referral Letter CIWA-Ar Protocol Documentation Glasgow Coma Scale (GCS) Documentation Braden Scale Documentation Morse Fall Scale Documentation MEWS & NEWS2 Early Warning Score Documentation Pressure Injury Staging Documentation OASIS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychotherapy-notes-authorization All note templates # Psychotherapy Notes Authorization: HIPAA Rules, Template & Sample Form Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychotherapy notes authorization is the standalone HIPAA authorization required before a practice may use or disclose psychotherapy notes, the private process notes a clinician keeps separate from the medical record. Under 45 CFR 164.508(a)(2) it cannot be combined with any other release. Records staff and clinicians use it whenever a third party requests these notes. Most forms run one page. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Records staff or the treating clinician; signed by the client Audience Client, requesting third party, privacy officer, auditors Typical length One page, 300 to 500 words · 5 to 10 minutes by hand (clinical team estimate) Format family Consent and privacy form (compare: release of information, Part 2 consent) When it's used Whenever anyone asks a practice to use or disclose psychotherapy notes Standards context Required by law when psychotherapy notes are released (45 CFR 164.508); the elements are mandated, the layout is not On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychotherapy notes authorization? A psychotherapy notes authorization is the written permission HIPAA requires before a practice may use or disclose psychotherapy notes. The underlying category comes from the Privacy Rule, published December 28, 2000: 45 CFR 164.501 defines psychotherapy notes as a mental health professional's notes documenting or analyzing conversation during a session, kept separate from the rest of the medical record. Clinicians usually call them process notes. HHS created the category to shield a therapist's private working impressions, material other providers do not need for treatment, and the shield only applies while the notes stay separated. You may also see the form called a psychotherapy notes release or an authorization for release of psychotherapy notes. The distinction that matters: your regular session documentation is a progress note. It lives in the medical record, payers can request it, and a standard release of information covers it. Psychotherapy notes are different by definition and by handling. They exist only if you choose to keep them, and 45 CFR 164.508(a)(2) requires this dedicated authorization before they leave the practice, even for requests that would otherwise count as routine treatment, payment, or operations. The regulation states this form "may only be combined with another authorization for a use or disclosure of psychotherapy notes" (45 CFR 164.508(b)(3)(ii)), so it always stands alone. If a request covers anything else, use two forms. The psychotherapy progress note page covers the record-set side of the boundary. ## Who uses psychotherapy notes authorizations and when Solo therapists, group practices, and health-system records departments all reach for this form in the same situation: someone wants psychotherapy notes to leave the practice. Common triggers include a client asking you to share process notes with a new therapist, an attorney requesting a complete record, or a personal representative gathering records. Use it only when actual psychotherapy notes exist and are truly what is being requested. Most requests that say "send the therapy notes" are satisfied by progress notes under a standard release of information , and the most compliant response is often to release the record-set documents and leave the psychotherapy notes where they are. The progress note and SOAP note pages cover what belongs in the shareable record. ## What a valid psychotherapy notes authorization includes No official government form exists. HIPAA mandates the content, not the layout: the core elements of 45 CFR 164.508(c), three required statements, and a signature. A one-page form covers all of it: Client and practice identification: who the notes are about and which clinician holds them. Description of the information: named as psychotherapy notes, with a date range. Who may disclose: the clinician or practice releasing the notes. Who may receive: the specific person or organization. Purpose: why, or "at the request of the individual." Expiration: a date or an event. Required statements: revocation, conditioning, redisclosure. Signature and date: the client or their personal representative. Each element in more detail, with the pitfall that most often invalidates it: Client and practice identification. Full client name and date of birth, plus the practice or clinician holding the notes. Pitfall: listing multiple providers or an entire health system as the source; the form should match the originator who actually keeps the notes. Description of the information. Name psychotherapy notes explicitly and bound them, for example "psychotherapy notes from sessions between January and June 2026." Pitfall: writing "all records" or "mental health records"; a description that does not specifically identify psychotherapy notes does not authorize their release. Who may disclose. The person or class authorized to release the notes: usually the treating clinician or the records custodian. Pitfall: leaving it blank because it feels obvious; it is a core element, and a missing core element makes the authorization invalid. Who may receive. The specific recipient: a named clinician, practice, or attorney. Pitfall: catch-alls like "any treating provider"; recipients that vague invite over-disclosure and get forms bounced by privacy officers. Purpose. State why the notes are being released. When the client initiates the request, HIPAA accepts "at the request of the individual" as sufficient. Pitfall: over-explaining; the purpose line is an administrative statement, not a place for clinical detail. Expiration. A date or an event, such as "90 days from signing" or "upon completion of the records transfer." Pitfall: writing "none" or leaving it blank; an authorization without an expiration date or event is defective on its face. Required statements. Three notices the regulation requires on the form: the client may revoke in writing (with the procedure for doing it), treatment and payment cannot be conditioned on signing, and disclosed notes may be redisclosed by the recipient and lose federal protection. Pitfall: reusing a general release's statement block that omits the revocation procedure; a missing statement invalidates the whole form. Signature and date. The client signs and dates the form; a personal representative also records their authority, such as parent of a minor or healthcare power of attorney. Pitfall: accepting a representative's signature without documenting the authority; note it on the form or attach it. ### Blank template (copy and adapt) AUTHORIZATION FOR RELEASE OF PSYCHOTHERAPY NOTES (Standalone form. May not be combined with any other authorization.) Client name: DOB: Practice / clinician holding the notes: Information to be released: Psychotherapy notes kept by the clinician named above, from sessions between ____ and ____ May be disclosed by: May be released to (name, organization, contact): Purpose: [ ] At the request of the individual [ ] Other: This authorization expires (date or event): I understand that: (1) I may revoke this authorization in writing at any time, except where the practice has already acted on it; the revocation procedure is: ____________ (2) Treatment, payment, enrollment, or eligibility for benefits may not be conditioned on signing this form. (3) Notes disclosed under this authorization may be redisclosed by the recipient and may no longer be protected by federal privacy law. Client signature: Date: Personal representative and authority (if applicable): Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-release checklist; the DOCX is the blank form, ready to adapt. ## Sample psychotherapy notes authorization Scenario: an adult client relocating to another city asks their psychologist to share psychotherapy notes with the new treating psychologist. All details are fictional. AUTHORIZATION FOR RELEASE OF PSYCHOTHERAPY NOTES · Standalone form; not combined with any other authorization Client: R.L. · DOB: 04/12/1991 Practice holding the notes: M. Okafor, PsyD, Lakeside Psychology, Austin, TX Information to be released: Psychotherapy notes kept by Dr. Okafor from individual sessions between 01/14/2026 and 06/30/2026. No other records are covered by this authorization. May be disclosed by: M. Okafor, PsyD May be released to: S. Patel, PhD, Cedar Counseling Group, Denver, CO Purpose: Continuity of care, at the request of the individual Expires: Upon completion of the records transfer or 90 days from signing, whichever comes first I understand that: (1) I may revoke this authorization in writing at any time, except where Lakeside Psychology has already acted on it, using the practice's written revocation procedure. (2) My treatment, payment, enrollment, or eligibility for benefits cannot be conditioned on signing this form. (3) Notes disclosed under this authorization may be redisclosed by the recipient and may no longer be protected by federal privacy law. Client signature: R.L. · Date: 07/09/2026 Personal representative and authority: Not applicable This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works The description names psychotherapy notes specifically and bounds them to a date range, so the release cannot be read to cover the wider record. The form stands alone: nothing else is authorized on the same page, which keeps it valid under the no-combination rule. The expiration is an event with a hard 90-day backstop, so the authorization cannot linger indefinitely. All three required statements appear, including how to revoke, so the form survives a privacy officer's element check. The purpose is stated in the client's terms, and no clinical content leaks into an administrative form. The signature block resolves who signed and confirms no representative authority was needed. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Psychotherapy notes sit outside the designated record set. Clients have no HIPAA right of access to them ( 45 CFR 164.524 ), and payment reviews run on the record set, not on process notes. That protection depends on real separation: keep them physically or electronically apart from the chart, and keep diagnosis, medications, session times, test results, and treatment-plan content out of them, because material in those categories is record-set content no matter where you write it. Mixed notes lose the shield. If what you keep is really a working conceptualization that informs treatment, the case formulation page covers the record-set way to hold it. The authorization itself is paperwork with teeth. HIPAA prohibits conditioning treatment, payment, enrollment, or eligibility on signing one (45 CFR 164.508(b)(4)). Honor written revocations, and keep every signed form: HIPAA's documentation rule ( 45 CFR 164.530(j) ) requires retaining authorizations for six years. Requests that arrive as subpoenas deserve special care: a records subpoena is not a court order, and psychotherapist-patient privilege may apply. Consult your attorney or licensing board before responding; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ## Common psychotherapy notes authorization errors auditors flag OCR has published no enforcement action aimed specifically at psychotherapy notes authorizations, and no national statistics exist on how often these requests fail. That absence is itself useful: the failures that surface in privacy-office reviews and records disputes are element failures, defects in the form rather than exotic legal questions. The BastionGPT Clinical Advisory Board sees the same errors most often in psychotherapy notes authorization reviews: Combined with a general release. One form covering the chart and the psychotherapy notes. The regulation allows this authorization to be combined only with another psychotherapy notes authorization, so the combined form fails for the notes even when it is valid for everything else. The description never says "psychotherapy notes." Wording like "all mental health records" does not reach notes the Privacy Rule defines by their separateness, and a privacy officer reading it strictly should refuse to release them. Core elements missing. No expiration date or event, or a statement block borrowed from an older release that drops the right to revoke and the procedure for it. Either defect invalidates the authorization on its face. Releasing notes because a payer asked for "therapy notes." Payment and audit reviews run on progress notes, and HIPAA bars conditioning payment on a psychotherapy notes authorization. Sending the process notes over-discloses. Treating a subpoena as an authorization. A records subpoena without the client's authorization or a qualifying court order routes to your attorney, not to the copier. ### Psychotherapy notes authorizations in the US, Canada, and Australia Aspect United States Canada Australia Status Required by law: 45 CFR 164.508(a)(2) mandates a standalone authorization before psychotherapy notes are used or disclosed No equivalent category; consent for any health record follows PIPEDA and provincial acts such as Ontario's PHIPA No equivalent category; the Privacy Act 1988 treats all health information as sensitive and the APPs govern release Terminology Psychotherapy notes, process notes Process notes, part of the clinical record Process notes, part of the health record What changes Notes sit outside the record set; clients have no HIPAA right of access Clients generally have a right of access to the full record, including therapist notes, subject to limited exceptions APP 12 gives clients access to their health information, with harm-based exceptions Retention Keep the signed authorization 6 years (45 CFR 164.530(j)); the notes themselves follow state record rules Provincial rules; Ontario's CRPO expects 10+ years from the last interaction State rules; NSW requires 7 years for adults and until age 25 for minors The US concept does not travel. In Canada and Australia the same private notes are usually just part of the health record: clients can typically access them, and the paperwork is the standard consent. The separate-authorization workflow on this page is a US legal requirement first, and a useful discipline anywhere. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychotherapy notes authorizations. Check a signed authorization against the required elements (specific description, expiration, revocation and conditioning statements) before any notes leave your practice. Extract what an incoming records request actually covers and flag when progress notes under a standard release satisfy it without touching psychotherapy notes. Draft the response: the compliant release letter, the request for a corrected form, or the explanation of why the notes stay put. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a psychotherapy notes authorization the same as a release of information? No. A standard release covers the designated record set: progress notes, treatment plans, test results, billing records. A psychotherapy notes authorization covers only the private process notes a clinician keeps separate from that record, and HIPAA requires it to stand alone. If a request covers both, use two forms. The psychotherapy progress note page covers the record-set side. Can a client get a copy of their own psychotherapy notes? HIPAA gives clients no right of access to psychotherapy notes: 45 CFR 164.524 excludes them from the access right that covers the rest of the record. A clinician may still choose to share them, and some state laws grant more access than HIPAA does, so check your state before refusing. Most requests for "my records" are satisfied by the record set. Can I combine it with a standard release form? No. The regulation allows a psychotherapy notes authorization to be combined only with another psychotherapy notes authorization (45 CFR 164.508(b)(3)(ii)). A combined form is invalid for the notes, even when it works for everything else it covers. Does an insurance company have a right to psychotherapy notes? No. Payment and audit reviews run on the record set, and HIPAA prohibits conditioning treatment, payment, enrollment, or eligibility on signing a psychotherapy notes authorization (45 CFR 164.508(b)(4)). When a payer asks for "therapy notes," send progress notes under the appropriate release. The progress note page covers what payers actually review. When can psychotherapy notes be disclosed without an authorization? Only in the narrow cases listed in 45 CFR 164.508(a)(2): use by the clinician who wrote them for treatment, supervised training inside the practice, the practice defending itself in an action brought by the client, HHS enforcement, disclosures required by law such as mandatory abuse reporting, health oversight of the clinician who wrote them, coroners and medical examiners, and a serious and imminent threat. Everything else needs the signed form. Does a subpoena alone compel release of psychotherapy notes? Usually not. A records subpoena is not a court order, and psychotherapist-patient privilege, recognized in federal courts since Jaffee v. Redmond (1996), may apply on top of HIPAA. Route any subpoena for psychotherapy notes to your attorney before responding; state procedures vary. Am I required to keep psychotherapy notes at all? No. No law requires creating or keeping them. HIPAA only defines the protection they receive if you choose to keep them. Many clinicians document fully in progress notes and keep no separate process notes, which also means there is nothing extra to fight over in a records dispute. How long do I keep the signed authorization? Six years. HIPAA's documentation rule (45 CFR 164.530(j)) covers authorizations along with other required paperwork. Retention for the psychotherapy notes themselves is a separate question governed by your state's record rules and your board's standards. Can BastionGPT check an authorization before we release notes? Yes. Paste the form and the request, and it checks the required elements, flags a combined or expired form, and drafts the response letter. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.501 : the psychotherapy-notes definition, its exclusions, and the separation requirement. 45 CFR § 164.508 : the authorization requirement (a)(2), the no-combination rule (b)(3)(ii), the conditioning prohibition (b)(4), and the core elements and required statements (c). 45 CFR § 164.524 : the right of access and its psychotherapy-notes exclusion. 45 CFR § 164.530 : the six-year documentation retention rule. HHS Office for Civil Rights, HIPAA and mental health information : OCR's guidance hub on mental health disclosures. Ontario, Personal Health Information Protection Act, 2004 : the provincial access and consent regime, with no separate psychotherapy-notes category. OAIC, Australian Privacy Principles : health information as sensitive information under the Privacy Act 1988. HHS and SAMHSA, 42 CFR Part 2 final rule (2024) : the substance-use counseling-notes analogue to HIPAA's psychotherapy notes. ## Related templates Psychotherapy Progress Note SOAP Note Case Formulation Note DAP Note Treatment Plan Release of Information (ROI) Authorization Good Faith Estimate (No Surprises Act) Patient Instructions & After-Visit Summary Financial Policy & Fee Agreement Records Request & Chart Copy Packet Consent for Treatment of a Minor Record Amendment & Correction Response 42 CFR Part 2 Consent Subpoena Response & Records Affidavit Informed Consent for Treatment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/release-of-information All note templates # Release of Information (ROI) Authorization: HIPAA Rules, Template & Sample Form Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A release of information (ROI) is the written authorization a client signs before a mental health practice may disclose their records to a third party. HIPAA does not publish an official form; it mandates the content: six core elements and three required statements (45 CFR 164.508). Records staff use one whenever a disclosure is not already permitted by law. Most forms run one to two pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Records staff or the treating clinician; signed by the client Audience Client, requesting third party, privacy officer, auditors Typical length One to two pages, 400 to 700 words · 5 to 10 minutes by hand (clinical team estimate) Format family Consent and privacy form (compare: psychotherapy notes authorization, Part 2 consent) When it's used Before records leave the practice for any purpose HIPAA does not already permit Standards context The elements are law (45 CFR 164.508); no official government form exists, the layout is yours On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a release of information (ROI) authorization? A release of information is the written permission a client gives before a practice may disclose their health records to a third party. In HIPAA terms it is an authorization: the instrument required for any use or disclosure the Privacy Rule does not already permit. The rule behind it, 45 CFR 164.508 , comes from the Privacy Rule published in December 2000 and enforceable since April 2003, and it standardizes content, not paperwork: HHS lets every practice draft its own form as long as the core elements and required statements are present, and the regulation adds that the authorization "must be written in plain language." You will also see the form called an authorization to disclose protected health information, a consent to release records, or a medical records release. "HIPAA release" is common shorthand, and imprecise, because many disclosures need no release at all. Two boundaries do most of the work. First, an ROI is not the client's own records request: when a client asks for their chart, the access right under 45 CFR 164.524 applies, with its own 30-day clock and fee limits, and no authorization is involved. Second, an ROI never reaches psychotherapy notes: those need their own standalone psychotherapy notes authorization , which cannot be combined with any other release. A general form that quietly claims to cover them fails for the notes even where it works for everything else. ## Who uses release of information forms and when Everyone who lets records leave the practice: solo therapists, group practices, and records departments. Typical triggers are an attorney assembling a case file, a disability or life insurer, a school accommodation request, a family member the client wants involved, or coordination with someone outside the treatment relationship. The counterintuitive part is when an ROI is not legally required: HIPAA itself permits disclosures for treatment, payment, and health care operations without one ( 45 CFR 164.506 ), so sending progress notes to the psychiatrist treating your client needs no signature under federal law. Substance use disorder records under 42 CFR Part 2, psychotherapy notes, and stricter state statutes are the layers that put a form back in the workflow, and many practices require one for every disclosure as policy, which is caution, not law. The care coordination note and discharge summary pages cover the records that most often travel. ## What a valid release of information includes No official government form exists. HIPAA mandates the content: six core elements, three required statements, plain language, and a signature (45 CFR 164.508(b) and (c)). The identity fields most forms carry, date of birth, address, phone, record number, help match the right chart, but they are not legal elements, and HHS states plainly that no notarization or witness is required. One to two pages covers all of it: Client and practice identification: who the records are about, and which practice or clinician may disclose them. Description of the information: record types and a date range, in a "specific and meaningful fashion." Who may receive: the person or class, with contact details. Purpose: why, or "at the request of the individual." Expiration: a date or an event. Required statements: revocation, conditioning, redisclosure. Signature and date: the client or their personal representative. What stays off the form: psychotherapy notes and Part 2 records. Each element in more detail, with the pitfall that most often invalidates it: Client and practice identification. Full client name, plus the practice or clinician authorized to make the disclosure; the discloser is a core element in its own right. Pitfall: a form dense with identifiers that never names who may disclose. Date of birth and record number are matching fields, not legal elements, and a missing element voids the form while a missing identifier does not. Description of the information. Name record types and bound them with dates, for example "progress notes and treatment summary, January through June 2026." OCR accepts "entire medical record" as sufficiently specific when that is the intent; an undefined "all protected health information" may fail, because PHI sweeps far wider than the chart. Pitfall: the catch-all phrase. Illinois goes further by statute: blanket consent to unspecified mental health information is simply invalid. Who may receive. A named person or organization, or a class of recipients, with enough contact detail to route the records. Pitfall: recipients so vague the client cannot tell who will end up holding their information; classes are legal, ambiguity is the thing privacy officers bounce. Purpose. State why the disclosure is happening. When the client initiates the request, HIPAA accepts "at the request of the individual" as the complete answer. Pitfall: leaving it blank; purpose is a core element, not a courtesy. Expiration. A date, or an event that relates to the client or the purpose, such as "upon completion of this records transfer." HIPAA sets no maximum, but state law can: California defaults to one year or less unless the signer asks for longer. Pitfall: writing "none," which the rule reserves for research authorizations; and in Illinois a form with no calendar date authorizes release only on the single day it arrives. Required statements. Three notices the regulation requires: the client may revoke in writing (with the procedure, or a pointer to your notice of privacy practices), treatment and payment cannot be conditioned on signing, and disclosed information may be redisclosed by the recipient and lose federal protection. Pitfall: statement blocks borrowed from an older form that drop the revocation procedure; a missing statement is a defect on its face. Signature and date. The client signs and dates the form; a personal representative also records the authority they sign under, such as parent of a minor or healthcare power of attorney. Electronic signatures are fine, and California's statute says so explicitly. Pitfall: accepting a representative's signature without documenting the authority on or with the form. What stays off the form. Psychotherapy notes always need their own standalone authorization , and records from a Part 2 substance use program follow their own consent rules with different expiration mechanics. Pitfall: a single checkbox quietly absorbing a protected category; the release fails for that category even where it is valid for the rest. ### Blank template (copy and adapt) RELEASE OF INFORMATION (ROI) AUTHORIZATION (Psychotherapy notes need their own standalone authorization; this form does not cover them.) Client name: DOB: Practice / clinician authorized to disclose: Information to be released (types + date range): [ ] Progress notes [ ] Diagnosis / treatment summary [ ] Treatment plan [ ] Assessment or testing results [ ] Billing records [ ] Other: __________________________ Date range: ______________ Exclusions: ______________ May be released to (name, organization, contact): Purpose: [ ] At the request of the individual [ ] Other: This authorization expires (date or event): I understand that: (1) I may revoke this authorization in writing at any time, except where the practice has already acted on it; the revocation procedure is: ________________ (2) Treatment, payment, enrollment, or eligibility for benefits may not be conditioned on signing this form. (3) Information disclosed under this authorization may be redisclosed by the recipient and may no longer be protected by federal privacy law. Client signature: Date: Personal representative and authority (if applicable): Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-release checklist; the DOCX is the blank form, ready to adapt. ## Sample release of information Scenario: an adult client asks their therapy practice to send treatment records to the psychiatric nurse practitioner who is starting medication management. All details are fictional. RELEASE OF INFORMATION (ROI) AUTHORIZATION · Psychotherapy notes are not covered by this form Client: J.M. · DOB: 09/23/1989 Practice authorized to disclose: A. Reyes, LPC, Harbor Light Counseling, Portland, OR Information to be released: Progress notes, diagnosis and treatment summary, and the current treatment plan from sessions between 01/12/2026 and 07/10/2026. Psychotherapy notes and billing records are not included. May be released to: K. Osei, PMHNP-BC, Riverline Psychiatry, Portland, OR · (503) 555-0142 Purpose: Coordination of care, at the request of the individual Expires: Upon completion of this records transfer or 90 days from signing, whichever comes first I understand that: (1) I may revoke this authorization in writing at any time, except where Harbor Light Counseling has already acted on it, using the practice's written revocation procedure. (2) My treatment, payment, enrollment, or eligibility for benefits cannot be conditioned on signing this form. (3) Information disclosed under this authorization may be redisclosed by the recipient and may no longer be protected by federal privacy law. Client signature: J.M. · Date: 07/10/2026 Personal representative and authority: Not applicable This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works The description names record types and a date range: specific and meaningful under the rule, and readable by the staff member who pulls the chart. Psychotherapy notes are excluded by name, so the release cannot be read to reach them, and the exclusion doubles as client reassurance. The recipient is one named provider with contact details, not a class the client never met. The purpose is stated in the client's terms, and "at the request of the individual" would have been enough on its own. The expiration is an event with a hard 90-day backstop, so the authorization cannot linger past its usefulness. All three required statements appear, including how to revoke, so the form survives a privacy officer's element check. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The ROI sits at the intersection of three rule sets, and knowing which one is speaking prevents most mistakes. HIPAA permits treatment, payment, and operations disclosures with no form at all, so an ROI for a routine provider-to-provider exchange is policy, not federal law. What makes the form legally load-bearing is everything else: disclosures to attorneys, insurers beyond payment, schools, and family members need a valid authorization; substance use disorder records from Part 2 programs follow 42 CFR 2.31 , whose 2024 overhaul became mandatory on February 16, 2026; and state statutes can be stricter than HIPAA on both content and duration, the way California requires 14-point type with a one-year default limit and Illinois voids blanket mental health consents outright. One nuance worth knowing: HIPAA's minimum-necessary rule does not apply to disclosures made under a valid authorization ( 45 CFR 164.502(b)(2) ). The form's scope is the whole limit, which is exactly why the description element deserves care. Handling is as auditable as content. Validity is assessed at the moment of disclosure, not when the form was scanned: an expired date, a known revocation, or a missing element makes the authorization defective that day. Honor written revocations prospectively; anything already released in reliance stays released. Keep every signed form six years under HIPAA's documentation rule ( 45 CFR 164.530(j) ), and when your practice initiates the authorization, give the client a copy. When the client asks for their own records, route the request to the access workflow instead: the 30-day deadline and cost-based fee limits belong to 164.524 access requests, and after Ciox Health v. Azar those fee limits do not automatically govern third-party authorization requests. Requests that arrive as subpoenas deserve independent analysis: an attached ROI does not settle privilege, and a subpoena does not satisfy every confidentiality statute. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common release of information errors auditors flag Impermissible uses and disclosures were the single most common issue in the HIPAA complaints OCR resolved in 2024, ahead of right-of-access failures, per its 2024 report to Congress (660 of the top-five issue counts across 28,228 resolved complaints). The enforcement record shows what that looks like in practice: Cadia Healthcare paid $182,000 in 2025 after publishing 150 patients' information in website success stories without valid written authorizations, and Manasa Health Center , a New Jersey psychiatric practice, paid $30,000 after replying to negative online reviews with patient information. In Canada, Alberta's privacy commissioner ordered that an insurer's own authorization form did not satisfy the Health Information Act, so the clinic that relied on it made an unauthorized disclosure. The BastionGPT Clinical Advisory Board sees the same errors most often in release of information reviews: Releasing on the requester's form without checking it. Insurers and attorneys draft forms around their needs, not your statutory duties. The discloser is responsible for validity, so check every incoming form against the elements before honoring it. Catch-all descriptions. "Any and all information" reads broad and performs badly: undefined all-PHI wording can fail the specific-and-meaningful test, and Illinois-style statutes void blanket mental health consents entirely. Name the record types and dates. Honoring a defective form. Expiration passed, revocation received, an element blank: each makes the authorization invalid at the moment of disclosure, no matter how long it sat quietly in the chart. Protected categories riding along. Psychotherapy notes and Part 2 substance use records swept into a general release under one checkbox; each needs its own instrument, and the general form fails for them. Running a client's own request through the authorization workflow. Access requests carry a 30-day deadline and fee limits that authorizations do not, so the misrouting creates deadline violations in one direction and imaginary paperwork requirements in the other. ### Release of information forms in the US, Canada, and Australia Aspect United States Canada Australia Status Content mandated by law: six core elements, three statements, plain language (45 CFR 164.508); no official form Consent validity standards, not form checklists: PHIPA requires knowledgeable consent, express for disclosure to non-custodians; Alberta lists what the custodian must explain first Privacy Act and APP 6 govern disclosure; OAIC guidance expects consent to be informed, voluntary, current and specific, with capacity Terminology Authorization, release of information, ROI Consent to disclose personal health information Consent to disclose health information What changes A written, signed form is the rule for authorization disclosures; defects void it Express consent may be oral: Ontario's regulator says verbal or written, and Alberta accepts documented oral consent since June 2026 Express consent may be oral, written, or electronic, and implied consent is recognized in some contexts; the form is evidence, not a statutory box Retention Keep the signed authorization 6 years (45 CFR 164.530(j)); the records themselves follow state rules Provincial record rules govern the file; Ontario's CRPO expects 10+ years from last interaction APP 11.2 requires destroying or de-identifying information no longer needed; state record rules add minimums such as NSW's 7 years or age 25 for minors The pattern: the US form is a legal instrument whose content is prescribed, while Canadian and Australian forms are evidence that valid consent existed. The same document serves both jobs well, so draft to the strictest standard that touches your practice and document consent conversations even where a signature is not the legal trigger. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on release of information authorizations. Check a signed ROI against the required elements (specific description, expiration, the three statements) and flag anything that would make it defective before records go out. Extract what an incoming records request actually covers and whether the release on file matches its scope and dates. Prepare the release: assemble only what the form covers, draft the cover letter, and flag content that needs a different instrument, from psychotherapy notes to Part 2 records. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a release of information the same as a psychotherapy notes authorization? No. A release of information covers the designated record set: progress notes, treatment plans, assessments, billing records. Psychotherapy notes sit outside that record set, and HIPAA requires a separate standalone authorization before they leave the practice; the two cannot be combined on one form. If a request covers both, use two forms. The psychotherapy notes authorization page covers the notes side. Do I need a signed ROI to send records to another treating provider? Under HIPAA alone, no: disclosures for treatment are permitted without an authorization (45 CFR 164.506), so coordinating with your client's psychiatrist or primary care clinician is lawful with no signature. What changes the answer: Part 2 substance use records, psychotherapy notes, a stricter state mental health statute, or your own practice policy. Many practices still obtain one for trust and clarity, which is sound practice; just call it policy rather than law. Does a client's request for their own records need an ROI? No. A client asking for their own records exercises the HIPAA right of access (45 CFR 164.524), which is not an authorization: the practice must act within 30 days, fees are limited to cost-based charges, and psychotherapy notes are excluded. Routing an access request through the ROI workflow adds paperwork the law does not require and puts the 30-day deadline at risk. How long does a release of information last? Until its expiration date or event arrives, unless the client revokes it in writing first. HIPAA sets no maximum, so a one-year limit is convention rather than federal law, but stricter state law controls where it exists: California limits most authorizations to one year unless the signer requests longer, and in Illinois a mental health consent without a calendar date is good only on the day it arrives. "None" as an expiration is reserved for research authorizations. Does the form need to be notarized or witnessed? No. HHS states the Privacy Rule "does not require that a document be notarized or witnessed." A receiving organization may still ask for notarization as identity assurance; that is their policy, not a HIPAA requirement, and your form is not defective without it. Can a client give verbal consent instead of signing? Not in the United States: a HIPAA authorization must be a signed written document, and a chart note that the client agreed verbally does not substitute. Canada and Australia regulate consent validity rather than form content: Ontario's regulator confirms express consent may be given verbally or in writing, Alberta accepts documented oral consent under its amended Health Information Act, and Australian guidance recognizes oral express consent, with health information usually warranting the express kind. What changes for substance use disorder (Part 2) records? Records from a Part 2 substance use disorder program follow 42 CFR 2.31, and the 2024 alignment rule became fully mandatory on February 16, 2026. The consent list looks like HIPAA's with real differences: the patient's name is an express element, a single consent can cover all future treatment, payment, and operations uses, and for those TPO consents "end of treatment" or even "none" is a valid expiration, which a general HIPAA authorization does not allow. Redisclosure now follows HIPAA rules after a TPO consent, with one bright line left: the records cannot be used in proceedings against the client without separate consent or a court order. Can we charge a fee for records sent under an ROI? It depends on the pathway. When the client exercises their own access right, fees are capped at reasonable, cost-based charges. When a third party requests records under an authorization, the federal court in Ciox Health v. Azar (2020) vacated the attempt to extend that patient-rate cap, so it does not automatically apply; state fee schedules and contracts still can. Identify the legal pathway first, then the fee. Can BastionGPT check a release before records go out? Yes. Paste the incoming request and the signed form, and it checks the elements, flags defects such as a passed expiration or a missing revocation statement, and drafts the response letter or the request for a corrected form. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: 45 CFR § 164.508 : core elements (c)(1), required statements (c)(2), plain language (c)(3), defective authorizations (b)(2), compound and conditioning rules, revocation. 45 CFR § 164.506 : treatment, payment, and health care operations disclosures permitted without authorization. 45 CFR § 164.502 : the minimum-necessary exception for authorization disclosures, and personal-representative rules. 45 CFR § 164.524 : the right of access, its 30-day clock, and its exclusions. 45 CFR § 164.530 : the six-year documentation retention rule. 42 CFR § 2.31 and the 2024 Part 2 final rule : substance use disorder consent content and the February 16, 2026 compliance date. HHS OCR FAQs 471 (entire-record descriptions), 476 (expiration), and 478 (no notarization or witness). HHS, notice on Ciox Health v. Azar : the access fee limit does not extend to third-party directives. California Civil Code § 56.11 : 14-point type, single-purpose signature, one-year default limit, copy rights. Illinois 740 ILCS 110/5 : consent content, the day-of-receipt rule, and the invalidity of blanket consents. Ontario, Personal Health Information Protection Act, 2004 , s. 18 and 19, and the IPC's consent guidance : knowledgeable consent, express consent for non-custodians, oral express consent. Alberta, Health Information Act , s. 34: the pre-consent information list and documented oral consent. OAIC, Australian Privacy Principles : APP 6 disclosure grounds and the consent framework for health information. HHS OCR, 2024 Annual Report to Congress : complaint volumes and top alleged issues. ## Related templates Psychotherapy Notes Authorization Intake Note Discharge Summary Care Coordination Note Psychotherapy Progress Note Mandated Reporting & Duty-to-Protect Documentation Good Faith Estimate (No Surprises Act) Financial Policy & Fee Agreement Records Request & Chart Copy Packet Consent for Treatment of a Minor Record Amendment & Correction Response 42 CFR Part 2 Consent Capacity & Competency Evaluation FMLA & Leave Certification Support Note Court Report & Treatment Progress Update Subpoena Response & Records Affidavit Informed Consent for Treatment Telehealth Consent HIPAA Notice of Privacy Practices Acknowledgment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/billing-note All note templates # Claim-Support Billing Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A claim-support billing note is the documentation that ties a billed CPT or HCPCS code to what happened in session: the service, the time, the diagnosis it addressed, medical necessity, and an authenticated author. United States payers require these elements in the clinical record rather than a separate document. Clinicians use the template as a pre-claim check, a records-request response, or a compliant addendum. Most run 100 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The treating clinician (therapist, psychologist, psychiatrist, or a pre-licensed clinician with required co-signature); billers and coders work from it Audience Payers and claim reviewers, auditors, your own biller or coder, and your pre-sign self-check Typical length 100 to 300 words · 3 to 6 minutes by hand (clinical team estimate) Format family Billing & Insurance documentation; checklist plus structured addendum When it's used Before a claim goes out, when a payer requests records for review, and when a missing element needs a compliant late entry or addendum Standards context No regulation requires a separate billing note; Medicare law and payer policy require the record itself to substantiate every element of the claim. On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a claim-support billing note? A claim-support billing note is the documentation that ties a billed service code to what actually happened in session: the service delivered, the time it took, the diagnosis it addressed, why it was medically necessary, and who provided and authenticated it. The name comes from revenue-cycle practice rather than from any clinical tradition or standards body. Billers and payers also call it billing documentation, a charge-support note, or a billing addendum. It is a different artifact from the superbill (the itemized statement a client submits for reimbursement) and from the psychotherapy progress note , which is the clinical record of the session itself. Here is the distinction that matters: in the United States, payers do not ask for a document called a billing note. They hold the clinical record itself responsible for supporting the claim. Medicare's Program Integrity Manual defines progress notes broadly, stating they "may be in any form or format," and its reviewers judge whether that record substantiates the code billed. The same manual warns against the shortcut this page's title might suggest: check-box forms built mainly for reimbursement are, in CMS's words, "often insufficient to demonstrate that all coverage and coding requirements are met." A claim-support billing note therefore works as a pre-claim checklist, a payer-review response, or a compliant addendum that strengthens the record. It never replaces the clinical note. ## Who uses claim-support billing notes and when Solo practitioners who do their own billing, group practices with a biller or billing service, and agency clinicians whose claims pass through a revenue-cycle team all use some version of this document, whether or not they call it by this name. It earns its place at three moments: before a claim goes out, as a fast check that the note supports the code; when a payer requests records for prepayment or postpayment review, as the organized summary that accompanies the notes; and when a required element is missing from the original entry, as the vehicle for a compliant addendum. Pre-licensed clinicians and their supervisors get a fourth use: a place where supervision and co-signature requirements are visibly satisfied. For the session record itself, use a progress note or SOAP note . To argue for coverage before a service, use a prior authorization support letter ; to argue after a denial, a medical necessity letter or insurance appeal letter . A start-stop time attestation and a diagnosis list update are narrower cousins that handle one element each; this note handles the whole claim. ## Claim-support billing note structure: what goes in each section The note has two jobs: a pre-sign check that every claim element is present, and a payer-facing summary or addendum when the record needs one. Each section below maps to something a reviewer actually looks for. Complete it from the clinical note, never instead of it. Client and service identification. Client identifiers and date of birth, date of service, place of service, and the rendering clinician with credential and NPI. Pitfall: identification that stops at page one; Medicare contractor guidance expects every page of the record to be legible and carry patient identification and dates of service. Code billed and service description. The CPT or HCPCS code and units, plus a plain-words description of what you did. The two must agree: contractor guidance states the submitted code "must describe the service performed." Pitfall: a code-note mismatch, such as an intake code on a routine session or an add-on billed without its base code. Time. For the timed psychotherapy codes (90832, 90834, 90837), Medicare contractor policy is explicit: "Start and stop times or total times must be documented." Either format works: "time may be documented with start and stop times or with total time." Count face-to-face time only, and pick the code whose range contains your minutes: 16 to 37 for 90832, 38 to 52 for 90834, 53 or more for 90837. Sessions under 16 minutes are not reportable, and when psychotherapy is billed with an E/M service the two times stay separate. Pitfall: "approximately an hour" supports nothing; a reviewer needs countable minutes inside the billed code's range. Diagnosis linkage. The ICD-10 code on the claim and a line showing how today's service addressed it. The record "must support the use of the selected ICD-10-CM code(s)." Pitfall: a diagnosis carried forward for years with no current symptoms in the note reads as unsupported. Medical necessity and plan linkage. Current symptoms or functional impact, and the numbered treatment plan goal this session worked toward. Pitfall: relying on a check-box form; CMS warns that templates built mainly for reimbursement are "often insufficient" in claim review. Supervision and co-signature (when applicable). For pre-licensed clinicians, the supervisor's name, credential, and co-signature where the payer or state requires one. Pitfall: assuming one rule; incident-to billing, Medicaid supervision rules, and state board requirements each set their own conditions. Authentication. The rendering clinician's signature, credential, and date. Medicare requires "the legible signature of the physician or non-physician practitioner responsible" for the care. Pitfall: signing later without noting it, or letting anyone other than the author attest to a missing signature; an attestation can identify an author, never backdate one. Addendum or late entry (when needed). Label it as an addendum, date it, identify the author, state the reason, and leave the original entry intact. Medicare's manual requires the date and author of any amendment or delayed entry to be identifiable, with changes "clearly and permanently denoted." Pitfall: program-integrity reviewers list "excessive late entries" among the signs of possible falsification; add what is missing promptly and honestly, and let the addendum say why. ### Blank template (copy and adapt) CLAIM-SUPPORT BILLING NOTE (supplements the clinical note) Client: [initials] DOB: Date of service: Place of service: Clinician / credential: NPI: Code(s) billed: Units: Modifier(s): Service description (what was done, in plain words): Time: start stop total face-to-face minutes: Diagnosis (ICD-10) on claim: How today's service addressed it: Medical necessity (current symptoms / functional impact): Treatment plan goal or objective addressed: Supervision (if pre-licensed): supervisor / credential / co-signature: PRE-CLAIM CHECK [ ] Code matches the service described [ ] Time inside the code's range [ ] Diagnosis supported in today's note [ ] Plan goal linked [ ] Signed, credentialed, and dated [ ] No cloned text from prior notes ADDENDUM (only if needed) Date: Author: Reason for late entry: Content (original entry left unchanged): Clinician signature / credential: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample claim-support billing note Scenario: a payer requests records for one 90834 claim in postpayment review. The clinician completes a claim-support summary and adds one compliant addendum for a stop time omitted from the original note. All details are fictional. Client: A.M., 34 · DOB: 02/1992 · Date of service: 07/09/2026 · Place of service: 11 (office) · Clinician: R. Patel, LMFT · Code: 90834 x1, no modifiers Service description: Individual psychotherapy. Cognitive behavioral therapy for generalized anxiety: reviewed worry-log homework, practiced cognitive restructuring on two work-related predictions, and introduced a scheduled worry-time protocol with a written plan for the week. Time: 3:04 pm to 3:49 pm, 45 face-to-face minutes (within the 38 to 52 minute range for 90834). No E/M service performed. Diagnosis linkage: F41.1, generalized anxiety disorder, as on the claim. Session targeted the persistent worry and sleep-onset difficulty documented today; GAD-7 completed in session: 11 (moderate), down from 14 at intake. Medical necessity: Client reports worry interfering with concentration at work most days and delayed sleep onset three to four nights per week. Symptoms persist at a moderate level; skills acquisition is incomplete. Continued weekly psychotherapy remains indicated. Plan linkage: Treatment plan Goal 1 (reduce worry interference with work and sleep, target GAD-7 below 8). Today's restructuring and worry-time work advance Objective 1a. Pre-claim check: Code matches service; time in range; diagnosis supported in today's note; plan goal linked; signed and dated below; narrative individualized to this session. Signed: R. Patel, LMFT, 07/09/2026. ADDENDUM, 07/16/2026, R. Patel, LMFT: Addendum to the entry of 07/09/2026, added after the payer's records request identified a missing stop time. Session ran 3:04 pm to 3:49 pm per the appointment record. The original entry is unchanged. This sample is fictional and for educational purposes. It does not describe a real patient. The addendum is shown as its own dated, authored entry that leaves the original note intact, which is the format Medicare's Program Integrity Manual expects. ↑ Back to the template and downloads ### Why this sample works Time is documented with start and stop times and a total, and 45 minutes sits inside 90834's 38 to 52 minute range, so the billed code is supported on its face. The service description and the code agree: what the clinician says happened is individual psychotherapy of the billed length, with named interventions. The diagnosis on the claim is supported by today's symptoms and a scored measure, not carried forward from intake on faith. Medical necessity chains to a numbered treatment plan goal with a measurable target, which is the linkage reviewers read for. The addendum is dated, authored, labeled, and explains itself, and it leaves the original entry unchanged, so the correction strengthens the record instead of raising a falsification question. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Billing-support material lives in the ordinary clinical record, and that is by design. HIPAA's psychotherapy-notes definition at 45 CFR 164.501 excludes session start and stop times, treatment modalities and frequencies, test results, and summaries of diagnosis, symptoms, and progress: exactly the material a claim runs on. Medicare's Program Integrity Manual tells its reviewers they may never demand true psychotherapy notes, and makes the provider "responsible for extracting information" that supports the claim from the record they can share. Keep private reflections in separated psychotherapy notes and the claim-support elements in the record, and a payer request never forces a hard choice. One more line worth knowing from the same manual: when a scribe or AI technology captures the documentation, CMS does not require the scribe to sign; the treating clinician's signature is what "affirms the note adequately documents the care provided." The format is a convention; the content is the requirement, and the requirement starts in statute: Medicare pays only when the claim is backed by "such information as may be necessary" to determine the amount due (Social Security Act 1833(e)). Payer rules attach to elements (time, diagnosis linkage, necessity, authentication), and Medicare contractors publish them in billing articles rather than in a required form. The same content-first logic runs through the cloning prohibitions ("Documentation cloning is strictly prohibited," Arizona Medicaid's June 2026 claims guidance), through amendment rules that keep originals intact, and through audits that read a note asking whether it substantiates the code, in the words of the manual, in "any form or format." Retention is layered the same way: HIPAA sets no retention period for the clinical record itself, CMS tells Medicare-enrolled providers to keep medical records for 7 years from the date of service, and licensing boards set their own periods on top. Australia goes further than the US here: every Medicare-billed attendance needs its own record entry with "sufficient clinical information to explain the service," completed "at the time, or as soon as practicable after" it (Professional Services Review Scheme Regulations 2019, section 6), and the Professional Services Review can require repayment when records fail that test. In Canada, provincial plans generally cover psychologist and psychotherapist services only in public settings; in private practice the claim-support document is usually the insurance receipt behind an extended-health claim, and college standards still require financial records that show dates, services, and fees. Payer contracts and provincial rules vary; check the ones that bind you. ## Common claim-support billing note errors auditors flag The numbers say documentation, not fraud, is where most therapy claims fail. CMS's compliance tips for outpatient psychiatric care report a 16.1 percent improper payment rate, projected at $254.5 million: 78.3 percent of it from insufficient documentation and another 17 percent from claims with no documentation submitted at all (2024 reporting period). A 2023 HHS OIG audit estimated $580 million in improper Medicare psychotherapy payments after 128 of 216 sampled claims failed review, with missing time documentation and missing signatures among the leading defects. The newest data continue the pattern: in FY2025 CERT results, 86.2 percent of the improper-payment errors attributed to clinical psychologists came from insufficient documentation (a 65-claim sample, a signal rather than a rate). Enforcement follows the same line: a California behavioral-medicine practice paid $2.75 million in 2025 to resolve False Claims Act allegations that psychotherapy add-on services were not furnished as billed or "failed to sufficiently document that such services had been provided." The BastionGPT Clinical Advisory Board sees the same errors most often in claim-support reviews: Time that cannot support the code. No time at all, "about an hour," or minutes outside the billed range. The timed psychotherapy codes need countable face-to-face minutes: 16 to 37, 38 to 52, or 53 and up, with start and stop times or a total documented. Authentication gaps. Unsigned notes, missing credentials, undated signatures, and attestation misuse: only the author may attest, and an attestation can never backdate a signature that a rule required by a given date. Cloned documentation. Identical narratives across dates or clients. Payers treat cloning as a program-integrity issue: one Medicare contractor warns that cloned documentation "does not meet medical necessity requirements" and that identifying it "will lead to denial of services" and recoupment. A note that matches last week's tells a reviewer one service was documented twice rather than two services delivered. Diagnosis-claim mismatch. The ICD-10 code on the claim never appears in the note, or appears with no current symptoms behind it. The record must support the code selected, visit by visit. A billing form substituted for a clinical note. CMS discourages limited-space, check-box templates and warns that forms built mainly for reimbursement are "often insufficient" in claim review. The billing note supplements the narrative; when it becomes the whole record, it becomes the finding. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on claim-support billing notes. Draft a claim-support summary from your session note, a few bullets, or dictation, with time, diagnosis linkage, and plan linkage stated the way reviewers look for them. Check a finished note against the code billed before you sign: time inside the code's range, diagnosis supported, plan goal linked, signature elements present, no cloned text. Turn a payer records request into an organized response: extract dates, codes, times, and medical-necessity language across a set of notes, and draft the compliant addendum when an element is missing. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a separate billing note required for therapy claims? No United States payer we cite requires a document called a billing note. The requirement runs the other way: the clinical record must contain the elements that substantiate the claim, and Medicare's Program Integrity Manual accepts progress notes "in any form or format." In practice the progress note is the claim support. This template earns its keep as a pre-claim checklist, an organized response to a records request, and a compliant way to add what the original entry missed. Do I need start and stop times, or is total time enough? For Medicare, either works. The contractor billing article for psychotherapy states that "time may be documented with start and stop times or with total time," and applies the requirement to the timed codes 90832, 90834, and 90837. Count face-to-face time only. Some Medicaid programs and commercial payers write stricter rules into their manuals or contracts, so check the ones that bind you; when in doubt, start and stop times satisfy everyone. Can I bill 90837 if the session ran exactly 53 minutes? Yes. Medicare contractor guidance sets the reporting ranges at 16 to 37 minutes for 90832, 38 to 52 for 90834, and 53 or more for 90837, with the instruction to "choose the code closest to the actual time." A documented 53-minute session sits inside 90837's range. Two related rules matter: psychotherapy under 16 minutes is not reportable at all, and when psychotherapy is billed alongside an E/M service the psychotherapy minutes exclude the E/M work. I forgot to sign a note and the claim already went out. What now? Sign it now as a properly dated late authentication, and know the review-side fix: Medicare's signature rules let the author of an entry submit a signature attestation, a signed and dated statement identifying the patient and date of service. Only the author may attest; nobody signs for a colleague. An attestation resolves authorship, and it cannot backdate a signature that a rule required by a specific date. When a contractor requests one, the standard window to respond is 20 calendar days. How do I add a late entry without it looking like fraud? Label it, date it, sign it, and leave the original alone. No national Medicare rule sets a 24, 48, or 72 hour completion deadline for outpatient psychotherapy notes; prompt completion is professional practice, not a federal clock. Medicare's Program Integrity Manual expects services documented "at the time they are rendered"; when an entry comes later, the date and author must be identifiable and the change "clearly and permanently denoted." The same manual lists "excessive late entries" among the patterns integrity reviewers read as possible falsification, alongside white-out and inserted pages. One honest, labeled addendum that explains its reason strengthens a record; a pattern of quiet after-the-fact edits does the opposite. Can my biller, a scribe, or AI write the note? Drafting help is allowed; authorship is not transferable. CMS does not require a scribe to sign the entry, because the treating clinician's signature is what "affirms the note adequately documents the care provided," and its manual states the same concurrence requirement applies when AI technology captures the documentation (Program Integrity Manual, chapter 3, effective January 2025). Your biller can assemble the claim and flag gaps; the clinical record itself must be authored, reviewed, and signed by the clinician who rendered the service. Are billing notes psychotherapy notes under HIPAA? No. HIPAA's psychotherapy-notes definition at 45 CFR 164.501 excludes session start and stop times, treatment modalities and frequencies, test results, and summaries of diagnosis, symptoms, and progress. Everything a claim runs on is ordinary record material a payer can request, and Medicare reviewers are barred from demanding true psychotherapy notes; the provider extracts the claim-support information instead. If a payer asks for the separated private notes themselves, that requires the client's specific psychotherapy notes authorization . How does billing documentation work in Canada and Australia? Australia writes the duty into law: every Medicare-billed attendance needs its own dated record entry with enough clinical information to explain the service, completed at the time or as soon as practicable after it, and the Professional Services Review enforces the standard. In one December 2025 outcome, a psychologist repaid $165,000 after records "were often brief, incomplete, duplicative or out of sequence." Ontario makes time documentation a payment condition for physician time-based codes: without recorded start and stop times, "the service is not eligible for payment." Provincial plans generally cover psychologists and psychotherapists only in public settings, so private-practice claim support is usually the insurance receipt behind an extended-health claim, with college standards requiring financial records that show dates, services, and fees, kept ten years in Ontario. Can BastionGPT handle my claim-support documentation? Yes. Give it your session note, bullets, or dictation and it drafts the claim-support summary with time, diagnosis linkage, and plan linkage stated plainly. Before you sign, it checks the note against the code billed: time in range, diagnosis supported, goal linked, signature elements present, no cloned text. When a records request arrives, it assembles dates, codes, times, and necessity language across notes and drafts a compliant addendum where something is missing. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: CMS Medicare Coverage Database, Billing and Coding Article A57520 : start and stop times or total times for 90832, 90834, and 90837; the 16 to 37, 38 to 52, and 53-plus minute ranges; legible records with patient identification on every page; and the requirement that the record support the ICD-10 code selected. CMS, Medicare Program Integrity Manual, Chapter 3 : progress notes in any form or format; the warning that reimbursement-focused templates are often insufficient; signature attestation rules; amendments and delayed entries; the scribe and AI concurrence note; and "excessive late entries" as a falsification signal. Social Security Act section 1833(e), 42 U.S.C. 1395l(e) : the statutory root; no payment unless the information necessary to determine the amount due has been furnished. 45 CFR § 164.501 : the psychotherapy-notes definition and its exclusions, including session start and stop times and treatment modalities and frequencies. HHS Office of Inspector General, audit of Medicare psychotherapy services (2023) : an estimated $580 million in improper payments, with 128 of 216 sampled claims failing review. CMS Medicare Learning Network, provider compliance tips for outpatient psychiatric care : the 16.1 percent improper payment rate, $254.5 million projection, and the insufficient-documentation share (2024 reporting period). CMS, FY2025 Medicare fee-for-service supplemental improper payment data : provider-type error distributions, including the clinical-psychologist insufficient-documentation share. AHCCCS, Claims Clues, June 2026 : "Documentation cloning is strictly prohibited." CGS Administrators, Electronic Medical Record Tips : cloned documentation does not meet medical necessity requirements, and its identification leads to denial and recoupment. US Department of Justice, settlement announcement, August 2025 : a $2.75 million False Claims Act resolution over psychotherapy add-on services billed without sufficient documentation. Health Insurance (Professional Services Review Scheme) Regulations 2019 , section 6: Australia's record standards; a separate entry for each attendance, sufficient clinical information to explain the service, completed at the time or as soon as practicable after. Professional Services Review, Director's update for December 2025 : the psychologist outcome; $165,000 repaid with records found not adequate and contemporaneous. Ontario Ministry of Health, Schedule of Benefits for Physician Services (effective April 1, 2026), General Preamble GP6 to GP7: start and stop times must be recorded for time-based services, or the service is not eligible for payment. ## Related templates Prior Authorization Support Letter Psychotherapy Progress Note SOAP Note Treatment Plan Telehealth Psychotherapy Note Collaborative Care (BHI/CoCM) Note Chronic Care Management (CCM) Note Transitional Care Management (TCM) Note Good Faith Estimate (No Surprises Act) Communication Log & Portal Message Note Utilization Review Summary Patient Instructions & After-Visit Summary Appointment No-Show & Outreach Note Financial Policy & Fee Agreement Incident & Adverse Event Report Records Request & Chart Copy Packet Record Amendment & Correction Response Psychological Testing Administration Note Psychological Testing Scoring Note Peer-to-Peer Review Note Advance Beneficiary Notice (ABN) Diagnosis & Problem List Update (ICD) Start-Stop Time Attestation Medical Necessity Letter Insurance Appeal Letter Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/prior-authorization-letter All note templates # Prior Authorization Support Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A prior authorization support letter is a narrative letter a treating clinician sends to a health plan to justify why a mental health service should be authorized before it is delivered, or continued. It states the diagnosis, symptoms, functional impairment, treatment history, and the exact service requested, mapped to the plan's medical-necessity criteria. Many plans do not require it for routine outpatient therapy. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating therapists, psychologists, counselors, social workers, prescribers Audience Health plan utilization reviewer, behavioral health managed care Typical length 400 to 700 words · 15 to 30 minutes by hand (clinical team estimate) Format family Payer-facing medical-necessity letter (compare: medical necessity letter, appeal letter) When it's used Before a service that needs prior authorization, or to extend one (concurrent review) Standards context A payer-driven letter, not a mandated form; content maps to the plan's medical-necessity criteria On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a prior authorization support letter? A prior authorization support letter is the clinical narrative a treating clinician sends to a health plan to justify a mental health service before the plan will approve it. It belongs to the family of utilization-management and medical-necessity documentation. Clinicians and payers also call it a prior auth letter, a pre-authorization or precertification letter, or a letter supporting a prior authorization request. Prospective review happens before treatment starts; concurrent review happens while treatment is ongoing, to extend an existing authorization. Two distinctions carry the whole document. First, the letter is not the request. The request is the administrative transaction a plan processes through its portal, its own form, or the standard electronic transaction that federal rules recognize; the support letter is the clinical evidence attached to that request, and a letter on its own may not count as a valid request. Second, prior authorization is not universal. Many plans do not require it for routine outpatient psychotherapy: in a 2025 review, none of nine Medicare Advantage organizations required prior authorization for routine in-network counseling or psychotherapy. Plans reserve it for psychological and neuropsychological testing, higher levels of care such as intensive outpatient or residential treatment, and plans that authorize therapy in blocks of sessions. No statute prescribes a letter format. What the plan requires is medical-necessity content mapped to its own criteria, which makes this a letter of medical necessity written for a utilization reviewer, distinct from an appeal letter that a clinician writes after a denial. ## Who uses prior authorization support letters and when Therapists, psychologists, counselors, clinical social workers, and prescribers write these letters when a plan requires authorization before it will cover a service. That is most common for psychological and neuropsychological testing, applied behavior analysis, transcranial magnetic stimulation, and admissions to higher levels of care, and for Medicaid managed care and behavioral-health carve-out plans that authorize outpatient therapy in blocks and require a clinical update to extend. A support letter beats a bare form when the request needs a clinical story: a partial treatment response, a level-of-care decision, or an out-of-network single-case agreement. When the plan has already denied the service, the task shifts to an appeal letter instead. ## Prior authorization support letter structure: what goes in each part Header and recipient. Your practice identifiers and the plan's utilization-management or behavioral-health prior-authorization unit, with the fax or portal reference. Pitfall: sending it to general claims instead of the utilization-management address, so it never reaches a reviewer. Member and request identifiers. Member name, date of birth, member or policy ID, group number, the exact code and units or dates requested, and the prior authorization number for a continuation. Pitfall: omitting the member ID or the exact code and units, which stalls the request at intake. Diagnosis. The ICD-10 code or codes that support the requested service. Pitfall: treating the diagnosis code as the justification; the code anchors the request, the impairment justifies it. Clinical presentation and severity. Current symptoms with a dated, validated measure such as the PHQ-9, GAD-7, or PCL-5, and the course over time. Pitfall: adjectives without measures; a dated score carries more weight with a reviewer than the word "severe". Functional impairment. The concrete effect on work, school, relationships, self-care, and safety. Pitfall: leaving impairment out; medical necessity turns on function, not symptoms alone. Treatment history and response. What has been tried, the response so far, and why a lower level of care is insufficient. Pitfall: no treatment history, so the reviewer cannot see why the requested service is the right next step. Requested service and medical-necessity justification. The exact service, frequency, and duration or level of care, mapped to the plan's criteria, with the clinical risk if it is not authorized. Pitfall: an open-ended request; name the units and the dates so the reviewer can approve a defined span. Closing, peer-to-peer offer, and signature. Offer a peer-to-peer discussion, then sign with credentials, license number, and NPI, and add a supervisor co-signature where required. Pitfall: no signature block, or a missing supervisor co-signature for a pre-licensed author. ### Blank template (copy and adapt) [Practice letterhead: name, credentials, license #, NPI, address, phone, fax] [Date] [Health plan / behavioral health managed care organization] [Utilization Management / Prior Authorization Department] RE: Prior authorization request ( ) initial ( ) continued Member: [name] Date of birth: [ / / ] Member/policy ID: [ ] Group #: [ ] Requested service/code: [CPT/HCPCS] Units/frequency/dates: [ ] Prior authorization # (continuation): [ ] Dear Utilization Reviewer, I am the treating [discipline] for the member above and am requesting prior authorization for the service described below. Diagnosis: [ICD-10 code(s) and description] Clinical presentation and severity: [current symptoms; validated measure scores with dates, such as PHQ-9, GAD-7, PCL-5; onset and course] Functional impairment: [effect on work, school, relationships, self-care, safety] Treatment history and response: [prior and current treatment, medications, response, why a lower level of care is insufficient] Requested service and medical necessity: [exact service, frequency, and duration or level of care, mapped to the plan's criteria (MCG, InterQual, ASAM, LOCUS), and the clinical risk if it is not authorized] I am available for a peer-to-peer discussion. Please contact me at [phone]. Sincerely, [Signature] [Name, credentials, license #, NPI] [Supervisor name and credentials, if required] Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-send checklist; the DOCX is the blank template, ready to adapt. ## Sample prior authorization support letter Scenario: an adult member in outpatient therapy under a Medicaid managed care plan that authorizes psychotherapy in blocks and requires a clinical update to extend. Recurrent major depression with generalized anxiety. All details are fictional. From: J. Okafor, LCSW, NPI 1000000000, Riverside Counseling Associates, Fort Collins, CO · Date: July 15, 2026 · To: Mountain State Health Plan, Behavioral Health Utilization Management RE: Prior authorization, continued outpatient psychotherapy. Member: A.R. · DOB: 03/1988 · Member ID: MSHP000000000 · Requested: 12 sessions of 90837, weekly, 08/01/2026 to 10/31/2026 · Prior authorization #2026-000123 (10 sessions authorized 05/01 to 07/31/2026) Dear Utilization Reviewer, I am the treating clinician for the member above and am requesting authorization to continue weekly individual psychotherapy for a further 12 sessions. This letter sets out the clinical basis for continued treatment. Diagnosis: Major depressive disorder, recurrent, moderate (F33.1); generalized anxiety disorder (F41.1). Clinical presentation and severity: Treatment began 05/06/2026 with moderate to severe symptoms (PHQ-9 18, GAD-7 15). After 10 sessions of cognitive behavioral therapy, symptoms have improved but remain moderate (PHQ-9 12, GAD-7 11 on 07/08/2026). The member reports low mood most days, initial insomnia, and anticipatory worry about returning to full-time work. Functional impairment: The member is on a graduated return-to-work plan and cannot yet sustain full hours because of concentration difficulty and fatigue. Anxiety continues to limit driving and contact outside the home. Treatment history and response: The primary care physician started sertraline in April 2026, continued at a stable dose. Psychotherapy has produced a measurable but partial response; the member now uses cognitive restructuring with less coaching. The gains are consistent with continued treatment, and ending now carries a substantial relapse risk given the recurrent course. Requested service and medical necessity: I am requesting 12 additional weekly sessions of individual psychotherapy (90837) over the next authorization period, to consolidate partial gains, complete the return to work, and prevent relapse. The requested frequency and duration are consistent with the plan's outpatient behavioral-health medical-necessity criteria. I am available for a peer-to-peer review. Sincerely, J. Okafor, LCSW · License 000000 · NPI 1000000000 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The requested service, units, and dates are stated exactly, so the reviewer can authorize a defined span rather than guess. Severity is documented with dated, validated measures at two time points, which shows both need and response. Functional impairment is concrete and tied to the requested service, which is the core of medical necessity. The treatment history shows why continued care at this level is the right next step, not an open-ended request. The letter offers a peer-to-peer review and is signed with credentials, license number, and NPI. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The letter itself is a convention; what a plan enforces is medical-necessity content mapped to its own criteria. Behavioral-health plans commonly measure requests against a named framework: MCG or InterQual for general medical necessity, the ASAM Criteria for substance-use levels of care, and LOCUS or CALOCUS for psychiatric level-of-care decisions. Name the framework the plan actually uses rather than a generic reference, and show, point by point, how the member meets it. Mental health parity sits behind all of this. Since 2021, federal law has required health plans to perform and document a comparative analysis showing that any nonquantitative treatment limitation, and prior authorization is a classic one, is applied no more stringently to mental health and substance-use benefits than to comparable medical and surgical benefits. A clinician does not include that analysis in a routine letter, but it is worth requesting when a denial looks like a more stringent behavioral-health standard. Two practical points close the loop. First, plans owe a decision on a clock. For employer plans governed by ERISA, the federal claims-procedure rule sets the outer limits: 72 hours for an urgent request, 15 days for a standard pre-service request, and 30 days after service (LAW, 29 CFR 2560.503-1). Federal reform shortens standard decisions to 7 calendar days and urgent decisions to 72 hours for Medicare Advantage, Medicaid, and CHIP starting in 2026, though it does not cover drugs (LAW, CMS-0057-F), and states set their own rules: Minnesota now bars prior authorization for outpatient mental health and substance-use treatment other than medication. One caveat worth knowing: many state prior-authorization reforms do not reach self-funded employer plans, which ERISA governs. Second, an authorization is a coverage decision, not a promise of payment; it remains subject to eligibility, benefit limits, and correct billing. In Medicare Advantage, an approved authorization is binding on the plan, which may not later deny the service for lack of medical necessity except for good cause (LAW, 42 CFR 422.138). Keep the letter in the record; it becomes the backbone of an appeal if the request is denied. See also the letter of medical necessity for coverage requests that are not tied to a prior authorization. ## Common prior authorization letter errors auditors and reviewers flag The pattern behind most denials is a request that does not connect the service to documented medical necessity. Federal oversight shows how often the process itself errs: an HHS Office of Inspector General review found that 13 percent of the prior-authorization requests Medicare Advantage plans denied actually met Medicare coverage rules ( OIG, 2022 ). The denials are rarely tested: an analysis of 2024 data found that only 11.5 percent of denied Medicare Advantage prior-authorization requests were appealed, yet 80.7 percent of those appeals were overturned ( KFF, 2024 ). A clear letter is what prevents the denial in the first place. The BastionGPT Clinical Advisory Board sees the same errors most often in prior authorization letter reviews: Symptoms without function. The letter lists symptoms but never shows how they impair work, school, relationships, or safety, so the reviewer cannot map it to medical-necessity criteria. No measure, no baseline. Severity is asserted in adjectives with no dated, validated score, so there is nothing to show need at the start or response over time. An open-ended ask. The letter requests "continued treatment" without a specific code, number of units, frequency, and date span, so the plan cannot authorize a defined period. No link to the plan's criteria. The request is not framed against the plan's own framework, so it reads as a preference rather than a necessity. Cite the page, date, score, or episode instead of saying the information is in the chart. Missing identifiers or signature. A missing member ID, a missing prior authorization number on a continuation, or an unsigned letter stalls the request before a clinician ever reviews it. ### Prior authorization for mental health in the US, Canada, and Australia Aspect United States Canada Australia Status Set by each plan (PAYER POLICY); federal law regulates the process, not a letter format. Parity law, ERISA timeframes, and the 2026 CMS reform apply; some states, such as Minnesota, now bar prior authorization for outpatient mental health Prior authorization generally does not apply to publicly funded psychology; provincial drug plans use special-authority approval, and a federal benefit (NIHB) requires prior approval for mental-health counselling beyond an initial allowance MBS Better Access sessions are not prior-authorized; a mental health treatment plan is a care-planning record, not insurer approval. PBS Authority Required medicines and NDIS supports do need prior approval Terminology Prior authorization, pre-authorization, precertification, pre-cert, prospective and concurrent review Predetermination, special authority, exceptional access, exception drug status Authority required, prior approval, NDIS plan approval What changes Whether a service needs authorization at all, the plan's criteria framework, and whether the plan is fully insured or self-funded The public system rarely covers private therapy, so the letter usually supports a private insurer or a drug plan The referral and mental health treatment plan drive access; there is no session-level pre-authorization for therapy Retention No universal federal chart-retention rule; state law and payer contracts govern Provincial college standards; Ontario expects about 10 years from the last interaction State rules; New South Wales requires 7 years for adults and until age 25 for minors The letter travels across all three countries. What changes is whether authorization applies at all, and to which service, so adapt the request to the plan or program you are writing to. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on prior authorization support letters. Draft a complete support letter from the diagnosis, a few measure scores, and the service you are requesting. Map the clinical picture to a plan's medical-necessity criteria and flag what a reviewer will look for. Check a finished letter for a missing measure, functional impairment, exact units, or signature before you send it. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does routine outpatient therapy always need prior authorization? No, not inherently. Many plans cover routine outpatient psychotherapy without prior authorization: in a 2025 review, none of nine Medicare Advantage organizations required it for routine in-network counseling or psychotherapy, and some states, including Minnesota, now bar it for outpatient mental health. Prior authorization is more common for testing, higher levels of care, and plans that authorize therapy in blocks. Always check the member's plan before writing. What is the difference between a prior authorization letter and a letter of medical necessity? A prior authorization support letter backs a specific authorization request against a plan's criteria, before or during the service. A letter of medical necessity is the broader genre: it explains why a service, drug, device, or exception is medically necessary and can be used inside a prior authorization, a benefit exception, or an appeal. The title alone does not start a prior authorization. Is a prior authorization letter the same as an appeal letter? No. A support letter is proactive: it justifies a service before it is delivered, or to extend one. An appeal letter comes after an adverse determination, invokes the plan's review rights, and answers the specific reason the plan gave for the denial. Keep your support letter; it becomes the foundation of the appeal if the request is denied. Who writes the prior authorization letter, the therapist or the biller? The treating clinician supplies or authenticates the clinical rationale, because that is the part a reviewer relies on. Billing staff often transmit the request, but they should not attest to clinical facts. A prescriber handles medication-specific requests, and a pre-licensed clinician's supervisor co-signs where the payer or the professional rules require it. How fast does a plan have to decide? It depends on the plan type. For employer plans governed by ERISA, the outer limits are 72 hours for an urgent request, 15 days for a standard pre-service request, and 30 days after service. Starting in 2026, Medicare Advantage, Medicaid, and CHIP must decide standard requests within 7 calendar days and urgent requests within 72 hours, though that reform does not cover drugs. Many states set shorter deadlines of their own. Does an approved authorization guarantee payment? Not by itself. An authorization confirms the plan approved the service on the information it had; payment still depends on eligibility on the service date, benefit limits, network status, correct coding, and timely filing. Medicare Advantage is stronger: once a service is approved, the plan generally may not later deny it for lack of medical necessity except for good cause. What if the authorization lapses while the patient is still in treatment? Submit a concurrent-review letter before the current authorization ends, documenting remaining symptoms, progress, why step-down is premature, current risk, and the requested dates and units. Do not assume the plan will backdate coverage. Under ERISA, ending an approved course early is itself an adverse determination, and in Medicare Advantage an approved course stays valid as long as it is medically necessary. Does mental health parity affect prior authorization? Yes. Federal parity law treats prior authorization as a nonquantitative treatment limitation, and since 2021 plans must document that they do not apply it more stringently to mental health and substance-use benefits than to comparable medical and surgical benefits. If a behavioral-health denial looks more restrictive than the medical side, you can request the plan's comparative analysis and the exact criteria it used. Can BastionGPT draft a prior authorization support letter? Yes. Give it the diagnosis, a few measure scores, the functional impairment, the treatment history, and the service you are requesting, and it drafts a letter you can review and adapt to the plan's criteria. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: HHS Office of Inspector General, Medicare Advantage prior authorization denials (2022) : 13 percent of denied requests met Medicare coverage rules. KFF, Medicare Advantage prior authorization in 2024 : 11.5 percent of denials appealed, 80.7 percent overturned. US GAO, Behavioral health and Medicare Advantage prior authorization (2025) : none of nine reviewed organizations required prior authorization for routine outpatient psychotherapy. 29 CFR 2560.503-1 : ERISA claims-procedure timeframes and full and fair review. CMS, Interoperability and Prior Authorization Final Rule (CMS-0057-F) : 7-day and 72-hour decision timeframes from 2026. 42 CFR 422.138 : a Medicare Advantage approval is generally binding on the plan. US Department of Labor, Mental Health Parity and Addiction Equity Act rules : prior authorization as a nonquantitative treatment limitation. Minnesota Statutes, section 62M.07 : prior authorization barred for outpatient mental health and substance-use treatment other than medication. ## Related templates Treatment Plan Psychotherapy Progress Note SOAP Note Intake Note Psychiatric Diagnostic Evaluation Claim-Support Billing Note Treatment Plan Review & Update Good Faith Estimate (No Surprises Act) Communication Log & Portal Message Note Utilization Review Summary Patient Instructions & After-Visit Summary Appointment No-Show & Outreach Note Financial Policy & Fee Agreement Prior Authorization Request Medical Necessity Letter Insurance Appeal Letter Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/adi-r-report All note templates # ADI-R Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The ADI-R (Autism Diagnostic Interview-Revised) is a standardized, semi-structured caregiver interview that gathers developmental history and current behavior when autism is part of the question. Psychologists and other trained clinicians use it as one input to an autism evaluation, and it yields a categorical classification, not a diagnosis. This page covers how to write up ADI-R results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, developmental pediatricians, and clinicians trained on the ADI-R; publisher qualification level C Audience Autism evaluation teams, physicians, IEP and early-intervention teams, BC Autism Funding and NDIS reviewers, parents Typical length 300 to 600 words for the ADI-R section · administration 90 to 150 minutes including scoring Format family Semi-structured caregiver developmental-history interview When it's used Autism diagnostic evaluations across the lifespan, mental age above two years; paired with direct observation for ABA authorization, school, BC Autism Funding, and NDIS support Standards context Published by Western Psychological Services (2003); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the ADI-R? The Autism Diagnostic Interview-Revised (ADI-R) is a standardized, semi-structured interview conducted with a parent or caregiver who knows the person's early development well. Western Psychological Services published the revised edition in 2003, and it descends from the original ADI (Le Couteur and colleagues, 1994); its authors are Michael Rutter, Ann Le Couteur, and Catherine Lord. It is composed of 93 items across three functional domains, Language and Communication, Reciprocal Social Interactions, and Restricted, Repetitive, and Stereotyped Behaviors and Interests, and the interviewer codes the caregiver's descriptions across eight content areas covering background, early development, language, social development and play, interests, and behavior. It can be used with children and adults whose mental age is above two years, and administration with scoring runs about 90 to 150 minutes. As of July 2026 the 2003 edition remains current, with no successor edition released. One point is load-bearing for write-ups: the ADI-R yields a categorical result, not a score. WPS states that because it focuses on behaviors rare in unaffected individuals, it "provides categorical results rather than scales or norms," so there are no standard scores, percentiles, or severity metrics to report, only whether a developmental-history pattern met the instrument's threshold. Its Comprehensive Algorithm Form supports two kinds of algorithm: Diagnostic Algorithms, based on developmental history and used to support a formal diagnosis, and Current Behavior Algorithms, which describe present functioning for treatment and educational planning and are not diagnostic. The original diagnostic algorithm was built around DSM-IV Autistic Disorder, which creates a genuine alignment gap with the DSM-5-TR two-domain model, so the classification supports rather than settles a diagnosis. Convention pairs the caregiver interview with direct observation such as the ADOS-2 and a rating scale such as the SRS-2 , and the whole set feeds a psychological evaluation report , not a verdict on its own. ## Who uses ADI-R write-ups and when Autism diagnostic clinics, developmental-behavioral pediatricians, child and adult autism services, and psychologists use the ADI-R as the developmental-history component of an evaluation, drawing on a caregiver who can describe early childhood in detail. It rarely stands alone: convention pairs it with direct observation, usually the ADOS-2 , a social-communication rating scale such as the SRS-2 , and cognitive and adaptive measures such as an adaptive interview , and the whole set anchors a psychological evaluation report . The caregiver-history piece is where the ADI-R does work the others cannot: it carries weight when direct observation understates difficulty, as with an adolescent who has learned to mask socially or an adult whose current presentation is subtle but whose early history is clear. Two boundaries belong in the report itself. First, the diagnostic algorithm depends on early developmental history, so it needs a caregiver who was present and can recall that period; without one, for many adults, the algorithm cannot be validly completed. Second, WPS states that remote administration by phone or videoconference "has not been specifically validated," so a remote interview belongs in a limitations line, not a silent footnote. ## How to structure an ADI-R results section No statute, payer, or publisher mandates a format for the ADI-R section of an autism evaluation. The sequence below is the convention experienced evaluators converge on because it survives review, and because it is built around one rule the top search results routinely break: report the classification and describe the developmental history in your own words, and leave the interview's questions, coding conventions, algorithm totals, and cutoff numbers out. That omission is a test-security obligation under APA Ethics Standard 9.11 and WPS policy, not an oversight. Each section below carries the pitfall that most often undermines it. Instrument, informant, and administration. Name the ADI-R and its 2003 edition, identify the informant and their relationship to the person and how well placed they are to recall early childhood, note who administered and coded the interview and their ADI-R training, and state that it was one component of the evaluation. Pitfall: not identifying the informant or their recall quality, or writing the interview up as though it produced the diagnosis on its own. Developmental-history findings, in your own words. Describe what the caregiver reported across the three domains, early social development and reciprocity, language acquisition and any loss of skills, and restricted or repetitive behavior, with attention to the early-childhood window the diagnostic algorithm focuses on. Pitfall: reproducing the interview questions, prompts, or coding conventions. Transcribing the instrument exposes protected content and erodes its validity. Algorithm used and classification, reported not decoded. State which algorithm you applied, the developmental-history Diagnostic Algorithm for diagnostic support or a Current Behavior Algorithm for planning, and whether the diagnostic classification was consistent with autism. Pitfall: printing domain totals or cutoff values, presenting the categorical result as a score or percentile, or mixing the two algorithm families so a current-behavior result reads as diagnostic. Integration with observation, history, and other measures. Place the caregiver history next to direct observation such as the ADOS-2, prior records, and any rating, cognitive, or adaptive data, and say where the sources agree and where they diverge. Pitfall: an ADI-R section that never connects to the observation, or that leaves an interview-versus-observation disagreement unexplained. Diagnostic formulation under DSM-5-TR. State the diagnostic determination and the clinical judgment behind it, with the ADI-R named as one input, and note that its algorithm reflects DSM-IV criteria while the diagnosis is made against DSM-5-TR. Pitfall: a diagnosis that rests on the classification alone. An ADI-R classification is not a DSM-5-TR diagnosis, and a result not consistent with autism does not rule autism out. Limitations and informant recall. Note anything that bears on interpretation: the informant's recall confidence, retrospective bias, especially for adults reporting distant childhood, any masking or camouflaging that shaped the observation, and in-person versus remote administration. Pitfall: silence on informant recall quality, the single most common gap for adults and older adolescents. Recommendations linkage. Tie each recommendation, whether supports, services, or further assessment, to a specific finding. Pitfall: recommendations that could follow any evaluation and trace back to nothing in the interview. ### Blank template (copy and adapt) ADI-R RESULTS SECTION Instrument/edition: ADI-R (2003) Informant: [relationship, recall quality] Administered/coded by: [name, credentials, ADI-R training] Date: [ ] DEVELOPMENTAL-HISTORY FINDINGS (narrative, in your own words) [Early social development and reciprocity, language acquisition and any loss of skills, restricted and repetitive behavior; describe what the caregiver reported, not the interview questions] ALGORITHM USED AND CLASSIFICATION [Diagnostic Algorithm (developmental history) or Current Behavior Algorithm] [Whether the diagnostic classification was consistent with autism] [Categorical result only; no domain totals, cutoffs, or item scores] INTEGRATION WITH OBSERVATION, HISTORY, AND OTHER MEASURES [Direct observation (e.g., ADOS-2), records, rating scales, cognitive and adaptive data; state where they converge or diverge and reconcile it] DSM-5-TR FORMULATION [The diagnostic determination and the clinical judgment behind it; the ADI-R is one input, and its algorithm reflects DSM-IV criteria] LIMITATIONS AND INFORMANT RECALL [Recall confidence and retrospective bias, masking or camouflaging, in-person vs remote administration] RECOMMENDATIONS LINKAGE [Each recommendation tied to a specific finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample ADI-R write-up (fictional) Scenario: a 15-year-old is referred by the school counselor and family doctor for social difficulties and long-standing anxiety. Language was on time, and the adolescent is verbally fluent and socially motivated, but the mother describes years of one-sided friendships and rigid routines. The ADI-R was completed with the mother as informant, and an ADOS-2 was administered separately; the direct observation came in nearer the threshold than the developmental history. This is the ADI-R section only, and it deliberately reports the classification and narrative without item content, coding, or cutoff numbers, which are protected. All details are fictional. Client: M.R., 15 · Informant: mother, lived with client since birth, strong early-childhood recall · Evaluator: L. Okafor, PhD, Licensed Psychologist (ADI-R trained) · Interview date: 07/09/2026 · Report date: 07/15/2026 Instrument, informant, and administration: The Autism Diagnostic Interview-Revised (ADI-R, 2003) was completed with M.R.'s mother, who has lived with him since birth and recalled the early-childhood period in detail. The interview was administered and coded in person by the evaluator, who is trained on the ADI-R, and it served as the developmental-history component of a broader autism evaluation. Informant recall was rated as good, and the account was internally consistent with school records from the early grades. Developmental-history findings: Across the reciprocal social domain, the mother described early and persistent difficulty forming and keeping friendships, a tendency to relate to adults rather than peers, and limited sharing of interest and enjoyment in early childhood. Language milestones were reported on time, but early communication featured repetitive questioning and a formal, one-sided conversational style. In the restricted and repetitive domain, she reported intense circumscribed interests that dominated play, strong insistence on routine, and marked distress at small changes, together with early sensitivity to certain sounds and textures. The pattern was most pronounced in the early-childhood window the diagnostic algorithm focuses on. Algorithm used and classification: The developmental-history Diagnostic Algorithm was applied. The pattern the mother reported was consistent with autism on that algorithm. Consistent with test-security convention, domain totals and cutoff values are not reproduced here, and the result is reported as a categorical classification rather than a score. Integration with observation, history, and other measures: The caregiver history and the direct observation diverged in degree. On the ADOS-2, M.R. presented as verbally fluent and socially motivated, and the current-observation profile fell nearer the threshold than the developmental history did. This pattern is expected for an adolescent who has learned to compensate socially, and the divergence is reconciled rather than averaged: the detailed early history carries weight where present observation understates difficulty because of camouflaging. A completed social-communication rating scale fell in the elevated range, and school records from the early grades corroborated the caregiver account. DSM-5-TR formulation: Integrating the ADI-R developmental-history classification, the ADOS-2 observation, the rating-scale data, and clinical judgment, M.R. meets DSM-5-TR criteria for autism spectrum disorder, requiring support (Level 1), with co-occurring anxiety documented as an associated condition rather than the primary explanation. The ADI-R is one input to this determination, not the determination itself, and its algorithm reflects DSM-IV criteria while the diagnosis is made against DSM-5-TR. Limitations and informant recall: The findings rest on a single informant's retrospective account; recall was judged good and was corroborated, but it remains retrospective. The lower current-observation profile reflects likely camouflaging and does not negate the developmental history. Administration was in person; no remote interview was involved. Recommendations linkage: The reciprocal-communication history supports recommendation 1 (social-communication support at school with structured peer opportunities). The routine and sensory findings support recommendation 2 (predictability and sensory accommodations in the classroom). The co-occurring anxiety supports recommendation 3 (anxiety-focused therapy that accounts for an autistic profile). The diagnosis supports recommendation 4 (eligibility and service discussions by the school and payer teams applying their own criteria). This sample is fictional and for educational purposes. It does not describe a real person or record, and the classification and details are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The informant is named along with their relationship and recall quality, so a later reader can judge how much weight the developmental history can bear. The developmental history is described in the evaluator's own words, and no interview questions, coding, algorithm totals, or cutoff numbers appear, so the write-up respects test security and the instrument's validity. The classification is reported as categorical and the algorithm family is named, so the reader knows the result supports a developmental-history pattern rather than a score. The interview-versus-observation divergence is stated and reconciled through camouflaging and corroborating records, which is what turns a classification into evidence rather than a contradiction. The diagnosis is attributed to DSM-5-TR criteria and clinical judgment, not to the ADI-R alone, the DSM-IV-to-DSM-5 gap is acknowledged, and every recommendation traces back to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the ADI-R section knowing which decision it will feed, and label the strength of each requirement honestly. No United States or Australian authority legally requires the ADI-R for an autism diagnosis. DSM-5-TR governs the diagnosis, and the American Academy of Pediatrics clinical report ( Hyman, Levy, and Myers, 2020 , reaffirmed October 2025) affirms that clinicians comfortable with DSM-5 criteria can diagnose without mandating any single instrument. What names the ADI-R is payer or program policy, always as an example rather than a rule: Aetna's clinical policy lists it among diagnostic tools used with clinical assessment, and TRICARE's Autism Care Demonstration has encouraged standardized tools "such as the ADOS-2 and the ADI-R" while recognizing others (PAYER POLICY, not a clinical rule). The clear exception is jurisdictional: in British Columbia, for a private diagnosis to qualify for Autism Funding, both the ADOS-2 and the ADI-R are required instruments , and the province accepts the ADI-R done virtually but not the ADOS (PROGRAM POLICY). The Canadian Paediatric Society notes only that "in some jurisdictions, the ADOS and ADI-R are required," without itself mandating them. In Australia the Autism CRC National Guideline (second edition 2023, NHMRC-approved) calls for comprehensive assessment without naming a required tool, and NDIS access turns on a DSM-5 diagnosis with functional-impact evidence. Interview interpretation and integration time is billed under the psychological and developmental testing evaluation family by bare code number (for example 96130 and 96131 in the United States, which are time-based and count data integration and report writing), with structured administration under 96136 through 96139 and the relevant MBS item numbers in Australia, as payer policy with plan-specific rules. The format is a convention; the classification, the narrative, and the integration are the requirement. Edition currency and the qualification boundary are the ADI-R's live defensibility questions. As of July 2026 the 2003 edition remains current on the WPS product page, with no successor released, though an ADI Third Edition is reported to be in publisher-funded validation, so an edition claim should carry its date. More consequential for interpretation is that the diagnostic algorithm was built around DSM-IV Autistic Disorder, which is why DSM-5-based research algorithms now exist, Kim and Lord's toddler algorithms for children under four and Lampinen and colleagues' algorithms for ages four to seventeen, framed by their authors as "an alternative method for summarizing ASD symptoms in a DSM-5-compatible manner" rather than a replacement for the published scoring. Read the classification with that gap in mind. Two boundaries belong in the write-up when they apply. First, who may use it: the ADI-R is qualification level C, and WPS states it "only qualifies individuals for test purchase, not for test use," which licensing boards and scope of practice govern; clinical-use training and the separate research-reliability standard run by the test authors are different things, and published sensitivity can exceed real-world clinic performance because of that gap. Second, administration mode: WPS holds that conducting the ADI-R remotely "has not been specifically validated," so a phone or videoconference interview is a stated limitation, even where a program such as BC accepts it. ADI-R is a trademark of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common ADI-R write-up errors reviewers flag There is no payer audit series for autism interview write-ups; the accountability here is ethical and psychometric, and it is pointed. The ADI-R has not been reviewed in the Buros Mental Measurements Yearbook , most likely because it "provides categorical results rather than scales or norms," so clinicians expecting a Buros review should be told none appears to exist. The peer-reviewed evidence is real but bounded: a 2021 meta-analysis put ADI-R sensitivity at .75 and specificity at .82, with specificity higher in research samples (.85) than clinical ones (.72), and the observation-based ADOS-2 outperforming it. The errors below are write-up errors, not administration errors. The BastionGPT Clinical Advisory Board sees the same ones most often in ADI-R report reviews: Classification written as diagnosis. Reporting the ADI-R classification as the diagnosis itself, or a result not consistent with autism as ruling autism out. The ADI-R is one input; a defensible diagnosis integrates the caregiver history, direct observation, and clinical judgment under DSM-5-TR. Reproducing protected content. Printing the interview questions or coding conventions, domain totals, algorithm scores, or cutoff numbers. This breaches APA Ethics Standard 9.11 and WPS policy and degrades the instrument; report the categorical classification and a narrative description instead. Mixing the two algorithm families. Presenting a Current Behavior Algorithm result, meant for treatment and educational planning, as diagnostic, or blending it with the developmental-history Diagnostic Algorithm. Name which algorithm produced the classification and keep the diagnostic and planning outputs distinct. Ignoring the DSM-IV-to-DSM-5 gap. Reporting the classification as though it maps cleanly onto DSM-5-TR criteria. The published algorithm reflects DSM-IV Autistic Disorder, so the write-up should say the classification supports the DSM-5-TR determination rather than equals it. Glossing discordance or informant recall. Failing to explain an interview result that conflicts with the observation, or to state the informant's recall confidence. Agreement between caregiver interview and direct observation is often only fair, and for adults the developmental history may rest on distant, retrospective recall, so both deserve a sentence rather than silence. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Draft a results-section narrative from a pasted summary of the developmental-history findings and the classification, organized for your review, with no interview questions or item content required. Cross-check a draft for the gaps reviewers flag: a classification written as a diagnosis, a current-behavior result presented as diagnostic, an interview-versus-observation disagreement left unreconciled, or protected content that slipped in. Produce a plain-language summary of the findings for parents and referrers that keeps the classification as one input and the diagnosis as clinical judgment. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do ADI-R scores mean? The ADI-R does not produce scores in the usual sense. WPS states that because it focuses on behaviors rare in unaffected individuals, it "provides categorical results rather than scales or norms," so there are no standard scores, percentiles, or severity metrics. What the developmental-history Diagnostic Algorithm yields is a categorical result: whether the pattern the caregiver reported met the instrument's threshold for autism. Report that classification and describe the history in your own words, and leave domain totals and cutoff values out, both because they are protected content and because the result is not a number to be graphed. This is a key difference from norm-referenced measures such as a cognitive battery , where standard scores and percentiles are the output. Is the ADI-R required for an autism diagnosis? Not in most places. No United States or Australian authority requires it: DSM-5-TR governs the diagnosis, and the American Academy of Pediatrics affirms that no single instrument is mandated (LAW and guidance). Payers and programs name it only as an example among validated tools, which is PAYER POLICY, not a requirement. The clear exception is jurisdictional: in British Columbia, a private diagnosis eligible for Autism Funding requires both the ADOS-2 and the ADI-R (PROGRAM POLICY), and the Canadian Paediatric Society notes only that "in some jurisdictions" they are required. So the honest write-up names the ADI-R as one validated caregiver-history instrument that adds standardized developmental data, and routes the requirement question to the specific payer or program, most often through the psychological evaluation report that carries the diagnosis. ADI-R or ADOS-2: what is the difference? They measure different things and are designed to be used together. The ADI-R is a caregiver interview about developmental history and current behavior; the ADOS-2 is a direct, standardized observation of the person. Because they draw on independent sources, they make additive contributions, which is why the pairing is often called the research gold standard, and why specificity tends to improve when they are combined. They also disagree: agreement between caregiver interview and direct observation is often only fair, and the disagreement is greater for older and more subtle presentations. A report that uses both should state where they converge and reconcile where they diverge, rather than averaging them or treating a disagreement as a failure. The ADI-R carries particular weight when the observation understates difficulty, as with camouflaging. Is the ADI-R outdated, and is there an ADI-3? The 2003 edition is current as of July 2026, and WPS lists no successor on its product page, though an ADI Third Edition is reported to be in publisher-funded validation. The live issue is not the edition date but the diagnostic framework: the published algorithm was built around DSM-IV Autistic Disorder, so it predates the DSM-5 two-domain model. Researchers have developed DSM-5-based algorithms, Kim and Lord's for toddlers and Lampinen and colleagues' for ages four to seventeen, but their authors describe these as "an alternative method for summarizing ASD symptoms in a DSM-5-compatible manner," not the official scoring. The practical consequence for a write-up is to state that the classification supports the DSM-5-TR determination rather than mapping onto it one to one, and to date any edition claim. Can I use the ADI-R with adults whose parents are unavailable? The diagnostic algorithm depends on early developmental history, so it needs an informant who was present in childhood and can recall that period. For many adults, no such informant is available, and in that case the diagnostic algorithm cannot be validly completed. The right move is to document the limitation explicitly, shift evidentiary weight to current observation, self-report, and any available records, and state a clear caveat about diagnostic confidence rather than reporting a classification the data cannot support. Even where an informant is available, adult and older-adolescent histories often rest on distant recall, so note the informant's recall confidence. Some teams turn to adult-oriented structured interviews for this situation; whatever the approach, the report should name what the missing early history means for the conclusion. Is remote or telehealth ADI-R administration valid? WPS states plainly that conducting the ADI-R by phone or videoconference "has not been specifically validated," and recommends a secure platform that mimics face-to-face contact if it is done remotely. That is not the same as a prohibition: some programs accept it, and British Columbia explicitly accepts the ADI-R done virtually even though it does not accept a remote ADOS. The write-up rule is to disclose the mode: if the interview was conducted remotely, say so, describe the method, and note that the norms and validation base assume in-person administration. Treat remote administration as a documented limitation rather than a silent detail, and check the specific program's stance before relying on a remote interview for a funding decision. How much ADI-R detail belongs in the report? Domain-level narrative and the classification, integrated into the formulation, and nothing more granular. Describe what the developmental history and current-behavior information showed across the social, communication, and restricted or repetitive domains, name which algorithm was used, and state whether the diagnostic classification was consistent with autism. Do not transcribe the interview questions, list item-level codes, or print domain totals or cutoff values. Reproducing that content breaches APA Ethics Standard 9.11 and WPS policy and erodes the instrument, and it does not make the report stronger. Then integrate the ADI-R findings with direct observation and records in the evaluation report , so the classification reads as one converging line of evidence rather than a standalone result. What questions are on the ADI-R? This page does not reproduce them, and a report should not either. The ADI-R is a secure test: its items, prompts, and coding conventions are protected under APA Ethics Standard 9.11 and the publisher's agreements, and their value depends on informants not having seen them in advance. It is fair to describe the instrument in general terms, a 93-item, semi-structured caregiver interview covering early development, language, social development and play, and restricted or repetitive behavior across three domains, and to name what each domain assesses. It is not appropriate to publish the actual questions or the scoring keys, whether in a report, a template, or a web page. Clinicians who need the materials obtain them through WPS at the required qualification level. How does BastionGPT help with ADI-R write-ups? BastionGPT drafts and checks the results section, not the clinical judgment. Paste a summary of the developmental-history findings and the classification, and it returns an organized narrative for your review, with no interview questions or item content required. It can cross-check an existing draft for the gaps reviewers flag, a classification written as a diagnosis, a current-behavior result presented as diagnostic, or an interview-versus-observation disagreement left unreconciled, and it can turn the findings into a plain-language summary for parents and referrers. It is HIPAA-compliant and keeps the classification as one input while you make the diagnosis. See how clinicians use it day to day on the AI therapy notes page. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Western Psychological Services, ADI-R product page : 2003 publication, authors Rutter, Le Couteur, and Lord, 93 items across three functional domains, eight content areas, categorical results rather than scales or norms, the two Diagnostic and three Current Behavior Algorithms, qualification level C, 90 to 150 minute administration with scoring, the 16 published translations, the 16-hour Training Video Program, and the purchase-not-use statement quoted on this page. WPS, Assessment Telepractice and Telehealth Resources : the statement that remote ADI-R administration "has not been specifically validated." Lebersfeld, Swanson, Clesi, and O'Kelley, Systematic Review and Meta-Analysis of the Clinical Utility of the ADOS-2 and the ADI-R , Journal of Autism and Developmental Disorders 51(11), 4101-4114 (2021): ADI-R sensitivity .75 and specificity .82, with research and clinical specificity of .85 and .72. Lampinen and colleagues, DSM-5 based algorithms for the Autism Diagnostic Interview-Revised for children ages 4-17 years , Journal of Child Psychology and Psychiatry 66(9), 1403-1413 (2025): the DSM-IV lineage of the original algorithm and the DSM-5-based alternatives, developed in 2,905 cases. de Bildt and colleagues, Interrelationship between ADOS-G, ADI-R, and DSM-IV-TR classification , Journal of Autism and Developmental Disorders 34(2), 129-137 (2004): the finding that interview and observation agreement was only fair, with a substantial younger-versus-older difference. American Academy of Pediatrics, Identification, Evaluation, and Management of Children With Autism Spectrum Disorder (Hyman, Levy, Myers), Pediatrics 145(1) (2020), reaffirmed October 2025: no stand-alone diagnostic instrument. Province of British Columbia, How autism is diagnosed : the Autism Funding requirement for both the ADOS-2 and the ADI-R, and acceptance of the ADI-R done virtually but not the ADOS. Canadian Paediatric Society, Standards of diagnostic assessment for autism spectrum disorder (2019): the "in some jurisdictions" framing and the caution that no single tool diagnoses ASD alone. Autism CRC, National Guideline for the Assessment and Diagnosis of Autism in Australia (second edition 2023, NHMRC-approved): the instrument-agnostic, comprehensive-assessment position. Buros Center for Testing, Mental Measurements Yearbook test-review index: the absence of an ADI-R review, consistent with its categorical, non-normed output. ## Related templates ADOS-2 Report Write-Up SRS-2 Report Write-Up Vineland-3 Report Write-Up BASC-3 Report Write-Up Psychological Evaluation Report CARS-2 Report Write-Up MIGDAS-2 Report Write-Up Autism Spectrum Evaluation Report Child & Adolescent Developmental Assessment M-CHAT-R/F Score Documentation Sensory Profile 2 Report Write-Up GARS-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cars-2-report All note templates # CARS-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The CARS-2 (Childhood Autism Rating Scale, Second Edition) is a clinician-completed rating scale that quantifies autism symptom severity from direct observation, record review, and collateral information. Psychologists and other trained clinicians use it as one severity input within a broader autism evaluation, never as a standalone diagnosis. This page covers how to write up CARS-2 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, physicians, school psychologists, and other trained clinicians; publisher qualification level C Audience Autism evaluation teams, physicians, IEP and early-intervention teams, payers naming CARS-2 such as Nevada Medicaid and TRICARE, parents Typical length 300 to 550 words for the CARS-2 section · rating 5 to 10 minutes after observation and record review Format family Clinician-rated autism severity rating scale When it's used Autism diagnostic evaluations, ages 2 and up; ABA authorization, school, and support-service requests where a severity rating supports the picture Standards context Published by Western Psychological Services (2010); current edition as of July 2026; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CARS-2? The Childhood Autism Rating Scale, Second Edition (CARS-2) is a clinician-completed rating scale that quantifies autism symptom severity from direct behavioral observation combined with record review and collateral information. Western Psychological Services published it in 2010, and it is the second edition of the original CARS, one of the most widely used and studied autism rating scales. It has three forms: the Standard Version (CARS2-ST), for individuals younger than 6 and for older individuals with communication difficulties or below-average estimated IQ; the High-Functioning Version (CARS2-HF), for verbally fluent individuals aged 6 and older with estimated IQ above 80; and an unscored Questionnaire for Parents or Caregivers (CARS2-QPC) that gathers developmental information to inform the clinician's rating. Each scored form has 15 items covering functional areas, rated by the clinician on a graded response scale, and takes about 5 to 10 minutes to complete once the observation and background information are in hand. It spans ages 2 and up. As of July 2026 the 2010 edition remains current, with no CARS-3 announced. Beneath the total score sits the fact most competing pages get wrong: the CARS-2 standard scores and percentiles are referenced to a clinical sample of 1,034 individuals with autism spectrum disorders, not a nationally representative population sample. A percentile therefore expresses where a person falls relative to other diagnosed or referred individuals, not relative to the general population, and the write-up should phrase it exactly that way. The second load-bearing point is scope: the CARS-2 is a severity rating, not a standalone diagnosis, and the authors and publisher position it as informing a diagnosis rather than making one. Only the clinician completes the ST or HF rating booklet; parents and caregivers complete the QPC, never the scored forms. A defensible autism diagnosis under DSM-5-TR integrates the CARS-2 with a direct observation, often the ADOS-2 , a developmental-history interview such as the ADI-R , an informant trait scale such as the SRS-2 , and cognitive and adaptive testing, all feeding a whole evaluation . This page reproduces none of the test's items, rating anchors, or cutoff values. ## Who uses CARS-2 write-ups and when Autism diagnostic clinics, developmental-behavioral pediatricians, child and school psychologists, speech-language pathologists, occupational therapists, and early-intervention teams use the CARS-2 as the clinician-rated severity component of an autism evaluation, choosing the Standard or High-Functioning form by the person's age, verbal fluency, and estimated cognition. Its short rating time and modest training burden make it usable across a range of professions and settings, including places where access to an ADOS-2 -trained specialist is limited. It rarely stands alone: convention pairs it with a direct observation, a developmental-history interview such as the ADI-R , an informant trait scale such as the SRS-2 or a behavior rating scale , and cognitive and adaptive measures such as the Vineland-3 . The whole set anchors a psychological evaluation report . One boundary belongs in the report itself: the commercial CARS-2 was not validated for use without live, in-person observation, so a remotely gathered rating belongs in a limitations line, not a silent footnote. ## How to structure a CARS-2 results section No statute, payer, or publisher mandates a format for the CARS-2 section of an autism evaluation. The sequence below is the convention experienced evaluators converge on because it survives review, and because it is built around the rules the top search results routinely break: report the severity level in your own words and keep the item ratings, rating anchors, and band cutoff values out, and state plainly that a CARS-2 percentile is referenced to a clinical sample of individuals with autism, not the general population. Leaving the protected content out is a test-security obligation under APA Ethics Standard 9.11 and WPS policy, not an oversight. Each section below carries the pitfall that most often undermines it. Instrument, form, and administration. Name the CARS-2 and its 2010 edition, the form used (Standard or High-Functioning) and why it was chosen (age, verbal fluency, and estimated IQ), who completed the rating and their training, and that it was one component of the evaluation. Pitfall: naming the CARS-2 with no form, or writing it up as though the rating alone produced the diagnosis. Sources and basis of the rating. State who observed the individual, over what period and in what setting, and which collateral informed the rating: records, the caregiver questionnaire, and teacher or parent report. Pitfall: a severity number with no stated basis, so a reader cannot tell whether it rests on a brief clinic look or a rich multi-source picture. Severity result, reported in words. State the severity level in plain language tied to observed behavior and support needs. If you cite the standard score or percentile, say plainly that it is referenced to a clinical sample of individuals with autism rather than the general population. Pitfall: reporting a percentile as if it were population-based, or printing the item ratings, rating anchors, or band cutoff values. Integration with observation and history. Place the CARS-2 next to the direct observation, developmental history, and any rating scales, and say where the sources agree and where they diverge. Pitfall: when the CARS-2 and an ADOS-2 disagree, averaging them or picking one; reconcile the divergence in words, for example a masking child who presents below threshold on office observation but shows a clear developmental history. Diagnostic formulation under DSM-5-TR. State the diagnostic determination and the clinical judgment behind it, with the CARS-2 named as one severity input among several. Pitfall: a diagnosis that rests on the CARS-2 alone. The publisher and authors position the scale as informing a diagnosis, not making one. Limitations and validity. Note anything that bears on interpretation: a form choice made at the age or IQ boundary, rater training and the scale's inherent subjectivity, single-setting sampling, masking, in-person versus remote gathering, the thinner evidence base for the High-Functioning form, and limited validation in adults and in women and girls. Pitfall: silence on limitations, especially for high-masking presentations where the rating can under-detect autism. Recommendations linkage. Tie each recommendation, whether supports, services, or further assessment, to a specific finding. Pitfall: recommendations that could follow any evaluation and trace back to nothing in the rating. ### Blank template (copy and adapt) CARS-2 RESULTS SECTION Instrument/edition: CARS-2 (2010) Form: [Standard (CARS2-ST) / High-Functioning (CARS2-HF)] Form selected because: [age + verbal fluency + estimated IQ] Rated by: [name, credentials, training] Date: [ ] BASIS OF THE RATING [Who observed, over what period and setting; collateral used: records, caregiver questionnaire (CARS2-QPC), teacher/parent report] SEVERITY RESULT (narrative, in your own words) [Severity level tied to observed behavior and support needs. If a percentile is cited, state it is referenced to the scale's clinical sample of individuals with autism, not the general population. No item ratings, anchors, or band cutoff values.] INTEGRATION WITH OBSERVATION AND HISTORY [Direct observation, developmental history, rating scales; state where they converge or diverge, and reconcile divergence rather than averaging it] DSM-5-TR FORMULATION [The diagnostic determination and the clinical judgment behind it; the CARS-2 is one severity input, not the diagnosis] LIMITATIONS AND VALIDITY [Form choice at the boundary, rater subjectivity, single setting, masking, in-person vs remote, thinner HF evidence] RECOMMENDATIONS LINKAGE [Each recommendation tied to a specific finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample CARS-2 write-up (fictional) Scenario: a verbally fluent 10-year-old is referred by their school team for social difficulty and rigidity, with average estimated cognition and a developmental history of subtle but persistent social-communication differences. The High-Functioning form was rated as the clinician-severity component of a broader autism evaluation. This is the CARS-2 section only, and it deliberately reports the severity level and narrative behavior without item ratings, anchors, or band cutoff values, which are protected. All details are fictional. Client: M.R., 10 · Referral: school team, social difficulty and rigidity · Evaluator: L. Okafor, PhD, Licensed Psychologist · Rating date: 07/13/2026 · Report date: 07/17/2026 Instrument, form, and administration: The Childhood Autism Rating Scale, Second Edition (CARS-2) was completed as one component of an autism evaluation. The High-Functioning form (CARS2-HF) was selected because M.R. is a verbally fluent 10-year-old with average estimated cognition, the population the HF form was designed for. The rating was made by the examining psychologist from a clinic observation, record review, and the completed Questionnaire for Parents or Caregivers. Basis of the rating: Ratings drew on a 60-minute clinic session, a classroom report, and the caregiver questionnaire. M.R. was cooperative and articulate. Conversation ran largely one way, with limited reciprocal give-and-take, an intense and detailed interest that recurred across topics, discomfort with unexpected changes to routine, and reported sensitivity to certain sounds. Severity result: The CARS2-HF placed M.R. in the moderate range of autism-related behaviors. The converted percentile fell near the middle of the distribution relative to the scale's clinical sample of individuals with autism, not the general population, so it describes where M.R. sits among referred and diagnosed peers rather than among all same-age children. Consistent with test-security convention, item ratings, anchors, and band cutoff values are not reproduced here. Integration with observation and history: The rating converges with the developmental and informant data. The caregiver history documents early and persistent difficulty sustaining friendships, and a completed social-communication rating scale fell in the elevated range across home and school. The observation-based severity is best read as a floor rather than a ceiling: subtle, well-compensated presentations can produce lower office-based ratings that must be weighed against pervasive real-world impairment. DSM-5-TR formulation: Integrating the CARS-2 severity rating with the developmental history, the informant rating scale, and clinical judgment, M.R. meets DSM-5-TR criteria for autism spectrum disorder, requiring support (Level 1). The CARS-2 is one input to this determination, not the determination itself. Limitations: The rating reflects behavior sampled largely in a single clinic session and rests on clinician judgment, which the scale shares with other autism measures. The High-Functioning form has a thinner published evidence base than the Standard form, and subtle presentations in cognitively able children can be under-detected, so the finding is read alongside history and collateral report rather than in isolation. No remote administration was involved. Recommendations linkage: The reciprocal-communication findings support recommendation 1 (social-communication support at school with structured peer opportunities). The rigidity and sensory findings support recommendation 2 (classroom predictability and sensory accommodations). The diagnosis supports recommendation 3 (eligibility and service discussions by the school and payer teams applying their own criteria). This sample is fictional and for educational purposes. It does not describe a real person or record, and the severity level, percentile, and details are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The severity level is reported and its meaning explained, but no item ratings, anchors, or band cutoff values appear, so the write-up respects test security and the norms. The percentile is framed correctly, as standing relative to the scale's clinical sample of individuals with autism rather than the general population, which is the single most common error on competing pages. The form is named along with why it was selected, the person's age, verbal fluency, and estimated cognition, so a later reader can judge whether the right form was used. The rating is integrated with developmental history and an informant scale, and the summary states where the sources converge, which is what turns a severity number into evidence. The diagnosis is attributed to DSM-5-TR criteria and clinical judgment, not to the CARS-2 alone, and the masking caveat and the thinner High-Functioning evidence base are stated plainly, with every recommendation traced back to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the CARS-2 section knowing which decision it will feed, and label the strength of each requirement honestly. No jurisdiction legally requires the CARS-2, and no authority makes a CARS-2 score alone sufficient for a diagnosis, special-education eligibility, or insurance authorization. DSM-5-TR governs the diagnosis, and the AACAP practice parameter (Volkmar and colleagues, 2014) frames autism diagnosis as a comprehensive, multidisciplinary process rather than any single instrument. What names the CARS-2 is payer or program policy. In the United States, Nevada Medicaid's ASD diagnosis certification form FA-11F (updated June 2025) lists the CARS-2 alongside the ADOS-2 and GARS-3 as a diagnostic tool whose score a provider may enter, and North Carolina's 2026 draft Clinical Coverage Policy 8F proposed adding the CARS-2 standard and high-functioning forms to its accepted list, though that draft closed public comment in June 2026 and is not final. Aetna's clinical policy bulletin and TRICARE's Autism Care Demonstration both name the CARS-2 among diagnostic tools used with clinical assessment, all of which is PAYER POLICY, not a clinical mandate. Educational eligibility is separate again: under IDEA a school does not require a medical diagnosis or any specific instrument, and 34 CFR 300.304(b)(2) bars any single measure as the sole criterion, which is LAW. In Canada, the Canadian Paediatric Society (2019) treats CARS-2 findings as one contributor that cannot be used alone to diagnose autism (CONVENTION), and in Australia the Autism CRC National Guideline (2018, updated 2023 and NHMRC-approved) recommends standardized assessment integrated by a skilled team without mandating a specific tool (POLICY and CONVENTION). Testing and evaluation time is billed under the psychological and developmental testing code families by bare code number, for example 96112, 96113, 96130, 96131, 96132, 96133, 96136, and 96137 in the United States, as payer policy with plan-specific rules. The format is a convention; the severity rating, the narrative, and the integration are the requirement. Edition currency and the qualification boundary are the CARS-2's live defensibility questions. As of July 2026 the 2010 edition remains current, with no CARS-3 announced; this is worth re-checking each year, because publishers revise instruments without long notice. Two facts about the norms belong in the write-up. First, the standard scores and percentiles rest on a clinical sample of 1,034 individuals with autism assembled for the 2010 edition, which predates DSM-5 and is not a representative population sample, so a percentile is standing among referred individuals, not a population percentile. Second, who may use the scale: the CARS-2 is qualification level C, and WPS qualifies people to purchase the test, not to use it, which is set by licensing boards and scope of practice. The High-Functioning form is newer and less validated than the Standard form, its specificity drops against non-autism clinical comparison groups, and validation in adults and in women and girls is limited, so reports leaning on the HF form should say so. The commercial CARS-2 is also not validated for administration without live, in-person observation; research adaptations for remote use exist but are not the published instrument, so a remote rating is a stated limitation. CARS and CARS-2 are trademarks of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CARS-2 write-up errors reviewers flag There is no payer audit series for autism severity write-ups; the accountability here is psychometric and ethical, and it is pointed. The CARS-2 was reviewed in The Nineteenth Mental Measurements Yearbook (2014, test review by K. K. Malcolm), the correct Buros volume to cite. The peer-reviewed evidence is usable but bounded: a 2019 systematic review and meta-analysis of 24 studies and 4,433 participants found the CARS internally consistent and its sensitivity acceptable, but its specificity was not, and the authors concluded it should be used alongside other confirmatory tools. The errors below are write-up errors, not administration errors. The BastionGPT Clinical Advisory Board sees the same ones most often in CARS-2 report reviews: Percentile read as population-based. Reporting a CARS-2 percentile as though it ranked the person against the general population. The conversions rest on a clinical sample of 1,034 individuals with autism, so a percentile is standing among referred and diagnosed individuals; say so, or the number misleads. One cut score applied across both forms. Treating a single historical identification cut, such as the 30 used in DSM-IV-era standard-form research, as if it applied universally. The Standard and High-Functioning forms are scored separately and are not interchangeable, and a standard-form cut does not transfer to the HF form. Severity reported as diagnosis. Writing the CARS-2 severity level as the diagnosis itself, or resting the diagnosis on the CARS-2 alone. The scale informs a diagnosis; a defensible determination integrates observation, history, and clinical judgment under DSM-5-TR. Wrong form, or the wrong rater. Using the HF form for a minimally verbal child or the ST form for a fluent, cognitively able older child, or letting a parent or teacher complete the scored rating. Only the clinician completes the ST or HF booklet; parents complete the unscored questionnaire. Protected content in, discordance glossed. Printing item ratings, anchors, or band cutoff values, or averaging a CARS-2 and an ADOS-2 that disagree instead of reconciling them. Report the severity in words, keep the scoring materials out, and explain divergence rather than splitting the difference. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Draft a severity-section narrative from a pasted summary of the form used, the severity level, and your behavioral observations, organized for your review, with no item content required. Cross-check a draft for the gaps reviewers flag: a percentile written as population-based, a severity level reported as a diagnosis, a standard-form cut applied to the High-Functioning form, or integration with history left unstated. Produce a plain-language summary of the findings for parents and referrers that keeps the CARS-2 as one severity input and the diagnosis as clinical judgment. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do CARS-2 scores mean? The CARS-2 is a clinician-rated severity scale, not a diagnostic test. The examining clinician rates functional areas from direct observation, record review, and collateral information, and the ratings sum to a total that converts to a standard score and percentile. The essential thing to understand, and to write down, is that those conversions are referenced to a clinical sample of individuals already diagnosed with autism, not the general population, so a percentile describes where a person falls among referred and diagnosed peers, not among all same-age children. Report the severity level in words tied to observed behavior, and keep the item ratings and cutoff values out of the document. A CARS-2 result is one input a clinician integrates into a whole evaluation under DSM-5-TR. Which form do I use, Standard or High-Functioning, and what if the person is at the boundary? The manual's rule is age, verbal fluency, and estimated cognition. The Standard form (CARS2-ST) fits individuals younger than 6, and older individuals with communication difficulties or below-average estimated IQ; the High-Functioning form (CARS2-HF) fits verbally fluent individuals aged 6 and older with estimated IQ above 80. The two forms are scored separately and are not interchangeable. At the margin, a child who is 6 with emerging language, or whose estimated IQ sits near the boundary between the forms, name the judgment in the report and note that the form chosen can shift the severity classification. No authority prescribes a tie-breaker, so document the reasoning. Is the CARS-2 required to diagnose autism? No law in the United States, Canada, or Australia requires the CARS-2, and no authority makes a CARS-2 score alone sufficient for a diagnosis or for eligibility. DSM-5-TR governs the diagnosis. What names the CARS-2 is payer or program policy: Nevada Medicaid's diagnosis form lists it as a diagnostic tool whose score a provider may enter, a 2026 North Carolina Medicaid draft proposed adding both CARS-2 forms to its accepted list, and Aetna and TRICARE name it among tools used with clinical assessment. Educational eligibility is separate, since under IDEA a school requires no specific test. The honest sentence for a report is that the CARS-2 is convention and, for some payers, policy, not a legal mandate. It pairs naturally with the ADOS-2 . Is a single cut score like 30 enough to identify autism? No, for two reasons. First, a single historical cut score, such as the 30 that circulated for decades in DSM-IV-era research on the standard form, was an identification threshold, not a diagnosis, and it does not transfer to the High-Functioning form, which is scored separately. Second, the CARS-2 is a severity rating that informs a diagnosis rather than making one. A 2019 meta-analysis found the scale's sensitivity acceptable but its specificity not, and concluded it should be used alongside other confirmatory tools. Report the severity in words and integrate it with a direct observation and developmental history. CARS-2, ADOS-2, ADI-R, or SRS-2, which do I use? They do different jobs and are strongest together. The CARS-2 is a brief clinician severity rating built from observation and collateral; the ADOS-2 is a semi-structured direct observation and is often the payer or research benchmark; the ADI-R is a structured caregiver interview about developmental history; and the SRS-2 is a quantitative trait scale that helps flag masking when observation sits near the threshold. The CARS-2 is faster and cheaper and usable by a wider range of professionals, but best practice uses it as a supplement, not a substitute, for a structured observation plus history. Each is an input; none is the diagnosis. Can I include the actual CARS-2 items or scores in a report? No. The rating anchors, item content, and band cutoff values are protected test materials: psychologists are ethically obligated to maintain test security under APA Ethics Standard 9.11 , and WPS enforces the same through its purchase agreements. Reproducing the anchors or band tables also degrades the norms, which assume the person has not been coached on them. A report carries the severity level, your narrative description of behavior, and the interpretation, never the item ratings or cutoff values. Raw test data, as distinct from test materials, are releasable only under Standard 9.04 with an appropriate release, and still route through your test-security obligations. Is a telehealth CARS-2 valid? Not with the published instrument. The commercial CARS-2 was not designed or validated for use without live, in-person observation, so a remotely gathered rating should be described as a limitation rather than reported as a standard result. Research adaptations exist: Sanchez and Constantino validated an observation-only CARS-2 protocol in person, and a remote version has been piloted against the in-person ADOS-2 in a small sample, with promising but preliminary agreement. Those are research adaptations, not the published scale. When a remote observation is unavoidable, name the method and its limits in the report. Is the CARS-2 still current, or is there a CARS-3? As of July 2026 the CARS-2, published in 2010, is the current edition, and no CARS-3 has been announced by WPS or found in any credible source. Because the standard scores rest on a clinical sample assembled for the 2010 edition, which predates DSM-5, interpret percentiles as standing among referred individuals and re-check for an edition change periodically. Test on the current edition, name it and the year in the report, and treat any future edition change as a stated limitation. Can BastionGPT draft my CARS-2 write-up? Yes. Paste a summary of the form you used, the severity level, and your behavioral observations, and it drafts the CARS-2 section for your review: the severity in narrative form, the percentile framed against the clinical sample rather than the population, and the integration with observation and history laid out so you can confirm it. It can also cross-check a draft you wrote for the gaps reviewers flag, a percentile written as population-based, a severity level treated as a diagnosis, a standard-form cut applied to the High-Functioning form, or protected content that slipped in, and produce a plain-language summary for parents and referrers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and because you paste only your summary and observations, no item content leaves your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Western Psychological Services, CARS-2 product page : 2010 publication, ages 2 and up, qualification level C, the three forms (CARS2-ST, CARS2-HF, and the unscored CARS2-QPC), the 5 to 10 minute rating time, and the standard-score and percentile conversions based on a clinical sample of 1,034 individuals with autism spectrum disorders. Lee A. Wilkinson, Best Practice Autism review of the CARS-2 : scores are interpreted relative to a clinical sample of individuals diagnosed with autism, not the typical individual, and the High-Functioning form reports a discrimination index of .93 with sensitivity .81 and specificity .87. Moon and colleagues, 2019, Developmental Medicine and Child Neurology : systematic review and meta-analysis of 24 studies and 4,433 participants concluding the CARS is a supplementary diagnostic tool with acceptable sensitivity but weaker specificity. Dawkins, Meyer, and Van Bourgondien, 2016, Journal of Autism and Developmental Disorders : high diagnostic agreement of the CARS2 across DSM-IV-TR and DSM-5 criteria in a UNC TEACCH sample. Samadi and colleagues, 2025, Autism Research : an Iranian validation reporting high reliability, a one-factor structure for the Standard form and a three-factor structure for the High-Functioning form, and weaker discrimination from intellectual disability. American Psychological Association, Ethical Principles of Psychologists and Code of Conduct : Standard 9.11 on maintaining test security and Standard 9.04 on the release of test data. Buros Center for Testing, Mental Measurements Yearbook : the CARS-2 was reviewed in the Nineteenth Mental Measurements Yearbook (2014), test review by K. K. Malcolm. Autism CRC, National Guideline for the Assessment and Diagnosis of Autism in Australia (2018; updated 2023, NHMRC-approved): standardized assessment recommended, with no specific instrument mandated. ## Related templates Psychological Evaluation Report ADOS-2 Report Write-Up SRS-2 Report Write-Up ADI-R Report Write-Up BASC-3 Report Write-Up Vineland-3 Report Write-Up MIGDAS-2 Report Write-Up Autism Spectrum Evaluation Report Child & Adolescent Developmental Assessment Sensory Profile 2 Report Write-Up GARS-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/bayley-4-report All note templates # Bayley-4 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The Bayley-4 (Bayley Scales of Infant and Toddler Development, Fourth Edition) is an individually administered developmental assessment for children 16 days to 42 months old, published by Pearson in 2019. Psychologists, therapists, and early-intervention teams use it to measure cognitive, language, and motor development and to inform eligibility decisions. This page covers how to write up Bayley-4 results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, physical and occupational therapists, and speech-language pathologists; publisher qualification level B Audience Early-intervention and IFSP teams, NICU high-risk follow-up clinics, pediatricians, courts and payers, parents Typical length 400 to 800 words for the results section · administration 30 to 70 minutes by age and cooperation Format family Norm-referenced infant and toddler developmental battery When it's used Early-intervention (IDEA Part C) eligibility, NICU high-risk follow-up, developmental-delay and autism evaluations, re-evaluations Standards context Published by Pearson (2019, current edition; norms revised 2023); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Bayley-4? The Bayley-4 (Bayley Scales of Infant and Toddler Development, Fourth Edition) is a norm-referenced developmental assessment for children 16 days to 42 months old, published by NCS Pearson in September 2019 as the successor to the Bayley-III (2006), with Nancy Bayley and Glen P. Aylward as authors of record. It measures development across five scales. Three are administered directly to the child through structured play interaction: Cognitive; Language, with Receptive Communication and Expressive Communication subtests; and Motor, with Fine Motor and Gross Motor subtests. Two are collected from a primary caregiver through a questionnaire: Social-Emotional and Adaptive Behavior. Directly administered subtests yield scaled scores (mean 10, standard deviation 3), and each scale yields a composite standard score (mean 100, standard deviation 15) with a percentile rank, a 95% confidence interval, and a descriptive classification; the reports also produce developmental age equivalents and growth scale values. It is often called simply the Bayley. Administration is paper or fully digital on Q-global, and completion time runs about 30 to 70 minutes depending on the child's age and cooperation. As of July 2026 the Bayley-4 is the current edition, and no Bayley-5 has been announced. Two facts separate a defensible Bayley-4 write-up from the sample reports that rank in search. First, the norm version matters. In 2023 Pearson revised the Cognitive, Language, and Motor norms after its own postpublication research identified "70 cases...with performance concerns" that, in Pearson's words, "significantly impacted the normative information in the average to extremely high" score range; those cases were replaced, not deleted, with demographically matched cases from the Bayley-4 UK validation study, and Q-global scoring was updated to match. Any Cognitive, Language, or Motor score generated before that update derives from the original 2019 norms, so a report relied on for eligibility or longitudinal comparison should record which norm version produced it (the fastest check is the manual's copyright and normative-copyright dates). Second, the caregiver-report boundary is easy to blur: the Social-Emotional and Adaptive Behavior scales are questionnaire-based, not directly observed, and the write-up should label them that way and corroborate them against observation and history. The Bayley-4 also sits inside a family of measures a report frequently names together, the Vineland-3 or ABAS-3 for a fuller adaptive picture and the WPPSI-IV once a child can sustain a more cognitively demanding battery. ## Who uses Bayley-4 write-ups and when Psychologists, physical therapists, occupational therapists, and speech-language pathologists write up Bayley-4 results inside developmental evaluations, most often for early-intervention eligibility under IDEA Part C, where services run from birth to age 3 and physical therapists commonly administer the Motor scale. NICU high-risk follow-up is the other core setting: the Bayley is the standard developmental measure there, and many programs assess at fixed corrected ages such as 6, 12, 18, 24, and 36 months rather than on a fixed calendar interval. It also appears as a developmental-level adjunct in autism evaluations, sitting next to observation and interview measures such as the ADI-R and CARS-2 rather than diagnosing on its own, and as a neurodevelopmental endpoint in clinical research. Age drives instrument choice at the upper edge: the Bayley-4 stops at 42 months, so between roughly 30 and 42 months a team that needs an IQ estimate or a broader personal-social profile may move to the WPPSI-IV (from 2:6) or the BDI-3 (birth to 7:11). A Bayley-4 section rarely travels alone; it anchors the developmental portion of a broader evaluation next to adaptive data from the Vineland-3 or ABAS-3 , caregiver history, and the behavioral observations that infant and toddler validity rests on. ## How to structure a Bayley-4 results section No statute, payer, or publisher mandates a results-section format. The sequence below is the convention experienced infant and toddler evaluators converge on because it survives review: it states chronological and corrected age and the norm version before any score, gives behavioral observation the weight infant validity actually rests on, leads with standard scores over age equivalents, labels which data came from the caregiver, and phrases delay in the metric the eligibility rule uses. Each section carries the pitfall that most often undermines it. Measures, edition, norm version, and ages. Name the instrument and edition, the norm version (original 2019 or the 2023 revision for the Cognitive, Language, and Motor scales), the administration format (paper or Q-global), the scales given, and both the child's chronological age and, for a preterm child, the corrected age used to derive scores. Pitfall: a silent norm version and a single age. A Cognitive, Language, or Motor score means different things under the 2019 and 2023 norms, and a score table that does not say whether corrected or chronological age was used is uninterpretable at re-evaluation. Behavioral observations and validity. Document state and arousal, attention, cooperation, the role of caregiver participation, and any factor such as fatigue, hunger, or illness that could depress performance, then close with a validity statement tied to what you observed. Infant results are only as good as the state the child was in. Pitfall: a boilerplate validity sentence. "Results are considered valid" with no observational basis reads as template text, and with infants the observed state is the validity evidence. Domain-by-domain composite results. Report the Cognitive, Language, and Motor composites, each with its standard score, 95% confidence interval, percentile rank, and one named descriptor set applied consistently, and note receptive-versus-expressive or fine-versus-gross patterns in prose. Pitfall: leading with age equivalents or a developmental quotient. Standard scores, intervals, and percentiles carry the interpretation; an age equivalent is not a standard score and must not stand in for one. Caregiver-report domains. Report the Social-Emotional and Adaptive Behavior composites and label them explicitly as caregiver questionnaire data, corroborated against observation and history, with any observed-versus-reported discrepancy noted. Pitfall: presenting questionnaire scores as directly observed. These two scales come from a caregiver rating form, and a report that blurs the source overstates what was directly measured. Age equivalents and growth scale values. If age equivalents or growth scale values appear, present them only with their published cautions and only for their intended purpose. Pitfall: comparing values that do not compare. Age equivalents are not an equal-interval scale and, in Pearson's words, growth scale values "cannot be meaningfully compared between subtests or subdomains"; they track one child over time, they do not rank a child against peers. Corrected age and delay severity. State the corrected-age rule you applied and translate the composite into the exact metric the jurisdiction uses, standard-deviation cuts for SD-rule programs and percent-delay language for percent-delay programs. Pitfall: a single unstated correction rule. The 24-month default and the domain-differential alternative give different scores between 24 and 36 months, and a delay described in the wrong metric can qualify or disqualify a child by accident. Interpretive summary and recommendations linkage. Answer the referral question, state what the scores do and do not establish, place the eligibility decision with the team, and tie each recommendation to a specific finding and a re-evaluation interval. Pitfall: a low score read as a fixed verdict. An early Bayley score is a current estimate, not a diagnosis of cognitive impairment, and recommendations unmoored from findings tell the reader the data did not drive them. ### Blank template (copy and adapt) MEASURES, EDITION, NORM VERSION, AND AGES Instrument/edition: Bayley-4 (2019) Norms: [2019 / 2023 revision] Format: [paper / Q-global] Scales given: [Cog, Lang, Motor, SE, Adaptive] Chronological age: ____ Corrected age (if preterm): ____ Basis: [corrected / chronological] BEHAVIORAL OBSERVATIONS AND VALIDITY [State/arousal, attention, cooperation, caregiver participation, fatigue/hunger/illness; validity statement tied to observed state] DOMAIN-BY-DOMAIN COMPOSITE RESULTS (directly administered) [Cognitive / Language / Motor: standard score, 95% CI, percentile, one named descriptor set; receptive vs expressive, fine vs gross in prose] CAREGIVER-REPORT DOMAINS [Social-Emotional and Adaptive Behavior: labeled as caregiver questionnaire; corroborated vs observation; discrepancies noted] AGE EQUIVALENTS AND GROWTH SCALE VALUES (if reported) [With published cautions; not compared across subtests; not a substitute for standard scores] CORRECTED AGE AND DELAY SEVERITY [Correction rule applied (24-month default / domain-differential); delay stated in the jurisdiction's metric: SD cut or percent delay] INTERPRETIVE SUMMARY (answer the referral question) [What converges; what the scores do NOT establish; eligibility is the team's decision under jurisdiction criteria] RECOMMENDATIONS LINKAGE [Each recommendation tied to a finding; re-evaluation interval] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample Bayley-4 write-up (fictional) Scenario: a 24-month-old born at 27 weeks, seen in a NICU high-risk follow-up clinic. Because the child was about 3 months premature, all scores use corrected age (roughly 21 months), stated prominently, the detail most sample reports bury. An internally consistent pattern shows Cognitive in the low-average range with Language and Motor lower. This is the developmental results section only, condensed but structurally complete. All details are fictional. Child: M.R., chronological age 24 months (born at 27 weeks; corrected age 21 months) · Referral: NICU high-risk infant follow-up, developmental surveillance · Evaluator: J. Okafor, PT, DPT · Testing date: 07/09/2026 · Report date: 07/15/2026 Measures and administration: The Bayley Scales of Infant and Toddler Development, Fourth Edition (Bayley-4) was administered in paper format using United States norms (2023 revision for the Cognitive, Language, and Motor scales). The Cognitive, Language, and Motor scales were administered directly to the child through structured play; the Social-Emotional and Adaptive Behavior scales were completed by the child's mother on the caregiver questionnaire. All scores were derived using corrected age for prematurity per the clinic's convention, and both chronological and corrected age are reported. Composites are reported with 95% confidence intervals (CI), percentile ranks, and the descriptor set printed in the publisher's score reports, named here so the labels are interpretable. Behavioral observations and validity: M.R. arrived rested and fed, separated easily onto his mother's lap, and warmed to the examiner within a few minutes. He engaged well with the play materials, reached and manipulated readily with either hand, and vocalized in single words and jargon. Attention faded near the end of the session; one break restored engagement, and standardized procedures were maintained throughout. Because an infant's observed state is what the validity of these results rests on, the results below are considered a valid estimate of current developmental functioning at his corrected age. Scale Standard score 95% CI Percentile Classification Basis Cognitive (COG) 85 79-92 16 Low Average Directly administered Language (LANG) 74 68-82 4 Below Average Directly administered Motor (MOT) 76 70-84 5 Below Average Directly administered Adaptive Behavior (ADAP) 78 72-85 7 Below Average Caregiver report Social-Emotional (SE) 92 84-100 30 Average Caregiver report Domain results: M.R.'s Cognitive composite (85, Low Average, 16th percentile) places his play, problem-solving, and object exploration at the lower edge of the average range for his corrected age. His Language composite (74, Below Average, 4th percentile) is his lowest directly administered score, and within it receptive communication was stronger than expressive, consistent with his single-word output during testing. His Motor composite (76, Below Average, 5th percentile) reflects gross-motor skills weaker than fine-motor, a pattern common in children born very preterm. Standard scores, intervals, and percentiles carry these statements; the age-equivalent and growth-scale values from the score report are held for progress tracking rather than for ranking him against peers. Caregiver-report domains: The Adaptive Behavior composite (78, Below Average, 7th percentile) and the Social-Emotional composite (92, Average, 30th percentile) come from his mother's ratings on the caregiver questionnaire, not from direct observation, and are labeled that way here. The below-average adaptive score is consistent with the directly observed language and motor findings; the average social-emotional score matches the warm, well-regulated engagement seen in the session. No large observed-versus-reported discrepancy was noted. Corrected age and the reporting decision: Every score above uses corrected age, stated so a later evaluator can compare like with like. One caveat is documented rather than hidden: under the domain-differential rule Aylward (2020) proposes, Language and Motor composites would be corrected to 36 months at all degrees of prematurity, not stopped at 24, which would raise those two scores somewhat relative to a chronological-age reference. The clinic's convention is the 24-month default aligned with the American Academy of Pediatrics, so that is what the scores reflect; the alternative is named for the reader rather than applied silently. Delay severity is described here so the early-intervention team can map it onto this state's eligibility metric, which the results section does not itself decide. Summary and linkage to recommendations: Testing at 21 months corrected shows low-average cognitive development (COG 85) with below-average language (LANG 74) and motor (MOT 76) composites and concordant below-average adaptive behavior by caregiver report (ADAP 78), alongside age-typical social-emotional functioning (SE 92). The language finding supports recommendation 1 (speech-language evaluation and early-intervention services). The motor finding supports recommendation 2 (physical and occupational therapy referral). The stability caveat supports recommendation 3 (re-evaluation at the next scheduled corrected-age visit, comparing same-norm scores). The full profile feeds recommendation 4: the early-intervention team should apply this state's developmental-delay criteria to the whole evaluation, not to any single score. This sample is fictional and for educational purposes. It does not describe a real child or record, and the scores are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works The edition, norm version (2023 revision), administration format, and both chronological and corrected age are named before any score, so the numbers stay interpretable at re-evaluation and no reader has to guess which age or norms produced them. The validity statement is tied to the infant's observed state, rested, fed, engaged, with one break, which is the linkage infant and toddler reviewers look for in place of a boilerplate sentence. Every composite appears with its confidence interval, percentile, and one named descriptor set; standard scores lead, and the age-equivalent and growth-scale values are held for progress tracking rather than peer ranking. The two caregiver-report scales are labeled as questionnaire data and corroborated against observation, so the report never overstates what was directly measured. The corrected-age decision is made in the open, with Aylward's domain-differential alternative named rather than applied silently, and the eligibility call is left to the team under the state's metric, with every recommendation traced to a specific finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the results section knowing which decision framework will read it, and label each requirement's strength honestly. For infants and toddlers the primary framework is IDEA Part C early intervention, and its structure is law but its threshold is not. Federal regulation defines an eligible child as one "experiencing a developmental delay...in one or more of the following areas" of cognitive, physical, communication, social or emotional, and adaptive development, or with a diagnosed condition carrying a high probability of delay ( 34 CFR 303.21 ). The level of delay, though, is state-defined: each state must adopt its own "rigorous definition of developmental delay" and "specify the level of developmental delay...or other comparable criteria" ( 34 CFR 303.111 ), so the ECTA Center's compilation documents criteria families as different as 2 standard deviations below the mean in one area, 1.5 standard deviations in two areas, and 25, 30 to 40, or 50 percent delay, and established-condition lists vary just as much (Barger and colleagues counted 620 unique qualifying conditions across 49 states, DC, and 4 territories, with 89.3% listed by fewer than 10 jurisdictions). The practical move: phrase delay severity in the exact metric the child's jurisdiction uses, because the same Bayley-4 profile can qualify a child in one state and not the next. NICU high-risk follow-up is convention and program policy rather than statute (California's CCS High Risk Infant Follow-Up program, for example, reimburses Bayley-or-equivalent developmental assessment to age 3). US testing time bills under the developmental-testing code family 96112 and 96113 as payer policy with plan-specific rules. Norm currency is the Bayley-4's quiet defensibility question. As of July 2026 the Bayley-4 (2019) is the current edition and no Bayley-5 has been announced, but the 2023 revision of the Cognitive, Language, and Motor norms means a score's meaning depends on which norms produced it: name the norm version, and never compare a pre-revision score with a post-revision score as if they sat on the same metric. Editions are not interchangeable for longitudinal monitoring either. Because the Bayley-4 tracks close to the Bayley-III, in Pearson's technical report "0.1 scaled score points and 0.5 and 1.3 standard score points" lower across the three scales, the older edition's documented tendency to underestimate delay carries into interpretation, so switching editions mid-follow-up is a clinical-judgment call, not a clerical one. Access is the publisher's: qualification level B, with the Cognitive, Language, and Motor scales requiring in-person standardized administration, since per Pearson's own guidance those subtests "cannot be administered in a standardized format via" telepractice, and only the Social-Emotional and Adaptive Behavior questionnaires carry a validated remote path, so a hybrid workflow should document which components were in person. Outside the US, early intervention runs through provincial Infant and Child Development Programs in Canada (generally birth to age 3, with no single national cut score) and, in Australia, through the NDIS early childhood approach for children under 6 with developmental delay or developmental concerns, where "developmental delay" is defined in the NDIS Act 2013; clinicians outside the US should also confirm which normative version their Q-global account and print kit carry. Bayley, Bayley-4, and Pearson are trademarks, in the US and other countries, of Pearson plc or its affiliates; Bayley-4 materials are copyrighted by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Bayley-4 write-up errors reviewers flag There is no payer audit series for infant developmental write-ups; the accountability literature here is psychometric, and it cuts close to everyday practice. The Bayley-4 is reviewed in the Buros Center's Twenty-Second Mental Measurements Yearbook , and the sharper caution comes from the edition-comparison literature: an influential study of the Bayley-III found control-group composites running "between 0.55 and 1.23 SD above the normative mean" and concluded that proportions of children with delay "were grossly underestimated" (Anderson and colleagues, 2010), a concern that carries into the Bayley-4 because the two editions track closely. The mechanics of administration are the publisher's problem; the errors below are write-up errors, and they are yours. The BastionGPT Clinical Advisory Board sees the same ones most often in Bayley-4 report reviews: The norm version left unstated. Reporting a Cognitive, Language, or Motor score without saying whether it came from the original 2019 norms or the 2023 revision, or comparing a pre-revision score with a post-revision one at re-evaluation as if they were on the same metric. Pearson replaced 70 normative cases in 2023 and the change concentrated in the average-and-above range; name the norm version, and compare like with like. Corrected age handled silently or inconsistently. A score table that never says whether corrected or chronological age was used, correction applied to some scores but not others, or the 24-month default asserted as settled with no nod to the domain-differential debate. Aylward (2020) argues cognitive should correct to 24 months but language and motor to 36 at all degrees of prematurity, which changes those two composites between 24 and 36 months; state the rule you applied on every score. Age equivalents and growth scale values reported without their cautions. Leading with an age equivalent, treating a developmental quotient as if it were a standard score, or comparing growth scale values across subtests. Age equivalents are not an equal-interval scale, and Pearson states growth scale values "cannot be meaningfully compared between subtests or subdomains"; lead with standard scores, intervals, and percentiles. Caregiver-report scales presented as directly observed. Reporting the Social-Emotional and Adaptive Behavior composites without labeling them as caregiver questionnaire data, or failing to corroborate them against observation and history. These two scales come from a caregiver rating form; a report that blurs the source overstates what was directly measured. Delay flattened, and a low score over-read. Collapsing state-specific eligibility into a generic "below 70" or "25% delay," or treating a low Bayley score as a diagnosis of cognitive impairment. Predictive accuracy is asymmetric: a Bayley-III cutoff of 70 detected only 18% of children with later IQ below 70 (Mansson and colleagues, 2021), so a low score raises risk more than a passing score rules it out, and severity belongs in the jurisdiction's exact metric. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Paste your score summary (scales, standard scores, CIs, percentiles, descriptors, corrected or chronological age) and get a drafted developmental results section organized domain by domain, with the behavioral-observation linkage and the caregiver-report labeling framed for your review. Cross-check a finished draft for the gaps reviewers flag: an unstated norm version, corrected age applied inconsistently, age equivalents reported without cautions, caregiver-report scales presented as directly observed, and delay described outside the jurisdiction's metric. Translate the results section into a plain-language summary for parents and early-intervention teams that keeps scores as current estimates and eligibility as the team's decision. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Bayley-4 scores mean? Two levels of scores carry the interpretation. At the subtest level the Bayley-4 uses scaled scores with a mean of 10 and a standard deviation of 3; at the scale level the Cognitive, Language, Motor, Social-Emotional, and Adaptive Behavior scales each yield a composite standard score with a mean of 100 and a standard deviation of 15, reported with a percentile rank and a 95% confidence interval. As a rough map, composites near 100 are average, and scores drop into monitoring and then significant-delay territory as they fall one and then two standard deviations below the mean, though the exact eligibility line is set by the jurisdiction, not the test. The reports also print developmental age equivalents and growth scale values, both of which carry published cautions and should never lead. Report the standard score, interval, and percentile together, name the descriptor set you use, and integrate the result into a whole evaluation . Was the Bayley-4 renormed, and is there a Bayley-5? Yes to the first, no to the second. In 2023 Pearson revised the normative, reliability, and validity information for the Cognitive, Language, and Motor scales after its postpublication research found 70 normative cases with performance concerns; those cases were replaced with demographically matched cases from the Bayley-4 UK validation study, and Q-global scoring was updated so it now applies the revised norms automatically. The change concentrated in the average-and-above score range. Any Cognitive, Language, or Motor score generated before the update derives from the original 2019 norms, so a report used for eligibility or longitudinal comparison should record which norm version produced it; the quickest check is the manual's copyright and normative-copyright dates. As of July 2026 the Bayley-4 (2019) is the current edition and no Bayley-5 has been announced. How long should I correct for prematurity on the Bayley-4? State the rule you used on every score, because the field is genuinely unsettled. The dominant convention corrects to 24 months of chronological age across all domains and then stops, which aligns with the American Academy of Pediatrics, whose 2023 clinical report frames screening as "adjusted for a child's corrected age (if under 24 months)" (Davis and colleagues), and with Q-global's default. The Bayley-4 co-author Aylward (2020) argues a domain-differential rule instead: correct cognitive composite scores for the first 2 years, but correct "language and motor composite scores...to 3 years at all degrees of prematurity," concluding that not correcting at 3 years and younger places preterm infants at a distinct disadvantage. The practical write-up move is to use the 24-month default, and for language and motor composites in preterm children between 24 and 36 months, report both corrected and uncorrected scores and cite the rationale. Is the Bayley-4 required for early-intervention (IDEA Part C) eligibility? No. IDEA Part C is instrument-neutral: federal regulation requires a developmental delay measured by "appropriate diagnostic instruments and procedures" but names no specific test, and it leaves the level of delay to each state. So the requirement is structure, not a mandated tool: each state sets its own definition and criteria, and states approve lists of acceptable instruments on which the Bayley-4 commonly appears alongside measures such as the BDI-3. The result is real variation, since the same profile can qualify a child in one state and not the next. The honest sentence for a report is that the Bayley-4 is convention and, on many state instrument lists, accepted policy, not a legal mandate, and eligibility is a team determination under the jurisdiction's own delay criteria. Describe severity in that jurisdiction's metric, and let the team apply it. Bayley-4, WPPSI-IV, or BDI-3, which should I use? Age and question drive the choice. The Bayley-4 covers 16 days to 42 months and gives a fine-grained infant and toddler developmental profile, so it carries most early-intervention and NICU follow-up referrals. The WPPSI-IV starts at 2:6, so at the 30-to-42-month overlap the convention is to move to it when a child can sustain a more cognitively demanding battery and an IQ estimate is the question. The BDI-3 spans birth to 7:11 and adds a personal-social domain, so it is often chosen when a child straddles the toddler-to-preschool boundary or when that broader profile is needed. Pearson's own comparison offers an anchor for the transition: at about 36 months a mean Bayley-4 Cognitive scaled score of 9.7 equated to a standard score of 98.5 while the mean WPPSI-IV Full Scale IQ was 103.3. Name the instrument, edition, and norms whichever you pick. Can the Bayley-4 be administered by telehealth? Only in part. Pearson supports remote administration of the Social-Emotional and Adaptive Behavior questionnaires through Q-global, either by a secure email link that does not require teleconferencing or by reading items aloud over video. The Cognitive, Language, and Motor scales are a different matter: per Pearson's own guidance those subtests "cannot be administered in a standardized format via" telepractice, because they require hands-on, in-person structured play with the child. So a valid Bayley-4 cognitive, language, or motor result comes from an in-person session. When a workflow is hybrid, document which components were in person and which questionnaires were completed remotely, and treat any departure from standardized administration as a stated limitation on the scores it touches. Should a Bayley-4 report include age-equivalent scores? Only with an explicit caution, and never as the headline number. Age equivalents are intuitive to parents, which is exactly why they mislead: they are not an equal-interval or ratio scale, so, in Pearson's own guidance, they "cannot be added, subtracted, or averaged," and their reliability is poorer near the top of the range. Growth scale values carry a separate caution, since Pearson states they "cannot be meaningfully compared between subtests or subdomains" and are meant to track one child's progress over time rather than to rank a child against peers. Both have a legitimate niche, describing function for a child who scores below the standard-score floor, but they belong after the standard scores, intervals, and percentiles, framed as supplementary. Lead with the standard scores; keep age equivalents in a clearly labeled, cautioned place. How predictive is an early Bayley-4 score of later IQ? Moderately, and asymmetrically, which is the part reports most often get wrong. In the Bayley-III literature, cognitive and language scores at age 2 correlated around .81 and .78 with preschool IQ and most children kept their broad classification from 2 to 4 years (Bode and colleagues, 2014), but a low score is far better at flagging risk than a passing score is at ruling it out: at a cutoff of 70, one study detected only 18% of children whose later IQ fell below 70 (Mansson and colleagues, 2021). Scores below 12 months are especially unstable. The write-up implication is to present an early Bayley-4 score as a current estimate rather than a prediction, to avoid trajectory language a single administration cannot support, and to state the re-evaluation plan, comparing same-norm scores at follow-up. Can BastionGPT help with Bayley-4 write-ups? Yes. Paste your score summary (scales, standard scores, CIs, percentiles, descriptors, and the corrected or chronological age used) along with your behavioral observations, and it drafts the developmental results section for your review: the domains organized with standard scores leading, the caregiver-report scales labeled as such, and the corrected-age decision framed so you can confirm it. It can also cross-check a draft you wrote for the gaps reviewers flag, an unstated norm version, corrected age applied inconsistently, age equivalents reported without cautions, questionnaire scores presented as directly observed, or delay described outside the jurisdiction's metric, and it can produce a plain-language summary for parents and early-intervention teams. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and because you paste only your summary and observations, no test content leaves your records. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Pearson, Bayley-4 product page : September 2019 publication, ages 16 days to 42 months, completion time 30 to 70 minutes, qualification level B, the growth-scale-value non-comparability statement, and the Q-global prematurity default of 24 chronological months with an override option. Pearson, Bayley-4 Technical Update FAQs (2023): the 70 normative cases with performance concerns, their replacement with matched Bayley-4 UK validation study cases, and the negligible US-UK raw-score differences. Aylward, 2020, Journal of Developmental and Behavioral Pediatrics 41(2):128-133: the domain-differential correction argument for cognitive to 24 months and language and motor to 36 months at all degrees of prematurity. Davis and colleagues, 2023, Pediatrics 152(1):e2023062511: the American Academy of Pediatrics framing of developmental screening adjusted for corrected age if under 24 months. Pearson, Interpretation Problems of Age and Grade Equivalents : age and grade equivalents are not a ratio or interval scale and cannot be added, subtracted, or averaged. Aylward and Zhu, The Bayley Scales: Clarification for Clinicians and Researchers (Pearson technical report): Bayley-4 scores track close to the Bayley-III, and the WPPSI-IV comparison anchor at about 36 months. Anderson and colleagues, 2010, Archives of Pediatrics and Adolescent Medicine : control-group composites 0.55 to 1.23 SD above the normative mean and the resulting underestimation of developmental delay on the Bayley-III. Mansson and colleagues, 2021, Acta Paediatrica 110(11):3030-3039: at a Bayley-III cutoff of 70, sensitivity to detect later IQ below 70 was 18%, and cognitive scores accounted for 38% of IQ variance. Bode and colleagues, 2014, Journal of Developmental and Behavioral Pediatrics 35(9):570-575: Bayley-III cognitive and language correlations with preschool IQ of .81 and .78, with 81% retaining developmental classification from 2 to 4 years. US Department of Education, 34 CFR 303.21 and 303.111 : the IDEA Part C definition of developmental delay across five areas and the requirement that each state adopt its own rigorous definition and delay criteria. ECTA Center, state eligibility summary : the range of Part C developmental-delay criteria across states, including standard-deviation and percent-delay families. Buros Center for Testing, Tests Reviewed in the Twenty-Second Mental Measurements Yearbook : the Bayley-4 is reviewed in the 22nd MMY. ## Related templates WPPSI-IV Report Write-Up Vineland-3 Report Write-Up ABAS-3 Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up Psychological Evaluation Report Child & Adolescent Developmental Assessment ASQ-3 Score Documentation DAYC-2 Report Write-Up Battelle Developmental Inventory (BDI-3) Report Write-Up PLS-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mcmi-iv-report All note templates # MCMI-IV Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The MCMI-IV (Millon Clinical Multiaxial Inventory-IV, 2015) is a 195-item true/false personality inventory for adults 18 and older who are already in mental health assessment or treatment, published by NCS Pearson. Psychologists use it to clarify personality-pattern questions and plan treatment, reading results as prevalence-anchored Base Rate scores rather than T scores. This page covers how to write up MCMI-IV results, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinical and forensic psychologists; publisher qualification level C Audience Treating clinicians and psychiatrists, courts and attorneys, disability reviewers, treatment teams, referral sources Typical length 300 to 700 words for the results section · administration 25 to 35 minutes Format family Clinically normed self-report personality inventory When it's used Personality-pattern and diagnostic-clarification questions, treatment planning, disability and forensic evaluations of adults already in clinical care Standards context Published by NCS Pearson (2015, current edition as of July 2026); described here for write-up purposes; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MCMI-IV? The Millon Clinical Multiaxial Inventory-IV (MCMI-IV) is a 195-item true/false self-report inventory of personality patterns and clinical syndromes for adults 18 and older, published by NCS Pearson in 2015 and bylined by Theodore Millon (1928-2014), Seth Grossman, and Carrie Millon, the co-authors who completed the edition after Millon's death. It is the fourth edition in the Millon inventory line, and the authoritative Handbook of Personality Assessment describes it as "a new test rather than a revised version of the MCMI-III," with 75 new items (39.5% of the pool) and a new Inconsistency scale. Its output spans 30 scales: 25 clinical scales, which the Millon Personality Group lists as 12 Clinical Personality Patterns (including the new Turbulent scale), 3 Severe Personality Pathology scales (Schizotypal, Borderline, Paranoid), 7 Clinical Syndromes, and 3 Severe Clinical Syndromes, plus an Invalidity check, the Inconsistency scale, and the Disclosure, Desirability, and Debasement modifying indices; 45 Grossman Facet Scales (three per personality scale) sit underneath to localize findings. The publisher's Profile Report graphs Base Rate scores for 28 of them. Administration runs 25 to 35 minutes at a fifth-grade reading level on Q-global, Q Local, or mail-in scoring, at Pearson qualification level C. Two boundaries define this instrument for write-up purposes, and both are the publisher's own. First, the norms are clinical: Base Rate scores reference a clinical adult population of 1,547 patients from inpatient and outpatient settings, and Pearson's own interpretive report warns that respondents who do not fit that normative population, or who took the test for nonclinical purposes, "may have inaccurate reports." The MCMI-IV describes people already in mental health assessment or care; scored against anyone else, it tends to make health look like pathology. Second, its Base Rate (BR) metric is not a T score: BR values are anchored to how common each condition is in the clinical norm group, so they are not equal-interval, are not normally distributed, and cannot be compared across scales the way the T scores of the MMPI-3 or the PAI can. An elevation is a prevalence-calibrated hypothesis about a pattern, not a diagnosis; the results section it feeds belongs inside a broader evaluation that corroborates it. ## Who uses MCMI-IV reports and when Clinical psychologists administer the MCMI-IV inside psychological evaluations when the referral question involves personality structure: diagnostic clarification for an adult already in care, treatment planning where the therapy needs to work with a personality pattern rather than around it, and re-evaluations where a syndrome picture keeps reasserting itself. It typically enters the chart alongside a psychiatric diagnostic evaluation and symptom-specific measures, with the personality inventory carrying the dispositional context. Forensic and disability evaluators also reach for it: the 2022 Journal of Personality Assessment review of the Millon inventories in legal settings notes the family has "joined the canon of commonly-used psychological instruments in several forensic settings," while concluding in the same breath that the MCMI-IV-specific peer-reviewed base is thin and its family-court use is problematic, so forensic users carry a heavier justification burden than clinical ones. Against its neighbors, the choice logic is stable across the practitioner literature: choose the MMPI-3 when litigation is likely and a broadband, heavily researched validity architecture matters most; choose the PAI for general or higher-functioning clients where community norms answer the referral question; choose the MCMI-IV when the person is genuinely a clinical patient and the question is which personality pattern the presentation is built on. The one population rule is absolute: adults 18 and older only, with adolescents 13 to 18 routed to the MACI-II, which the publisher states "should not be used with nonclinical populations" either. ## How to structure an MCMI-IV results section No statute, payer, or publisher mandates a results-section format. The sequence below follows the order the publisher's own Interpretive Report uses: validity and response style before anything substantive, personality findings before clinical syndromes, and the computer narrative treated as a licensed input rather than a finished report. Each section carries the pitfall that most often undermines it. Measures, edition, norms, and administration. Name the instrument and edition (MCMI-IV), the platform (Q-global, Q Local, or mail-in scoring), the administration modality (in office or remote, with the arrangement described), and the reference population in one plain sentence: Base Rate scores calibrated to an adult clinical sample of 1,547 patients. State why this examinee fits that population, because the publisher's norms assume a person already in assessment or treatment. Pitfall: "an MCMI" with no edition or norms sentence. The MCMI-III is still on sale and the Handbook of Personality Assessment calls the MCMI-IV a new test, so an edition-less score list is ambiguous at re-evaluation and easy to attack on review. Protocol validity and response style, before anything else. Report the Invalidity and Inconsistency checks, then characterize the test-taking approach through the three modifying indices: how openly the person disclosed (Disclosure), how favorably they presented themselves (Desirability), and how self-critically (Debasement). End with a plain conclusion about interpretability, because the 2022 legal-settings review found the modifying indices have "questionable utility in the detecting of response bias," which means response style limits what the profile can carry rather than being something the software corrects. Pitfall: interpreting a favorable-presentation protocol as substance. A defensive profile with low elevations is common in high-stakes evaluations and is mostly uninformative, not evidence of health. The Base Rate frame, stated once. Tell the reader what the metric is in one or two sentences: BR scores are anchored to the prevalence of each condition among patients in the clinical norm group, by convention a BR at or above 75 indicates the presence of a trait or syndrome and 85 or above indicates prominence, and the publisher's profile graphs mark exactly those two anchors. Then spend the prose on the pattern, not the numbers. Pitfall: importing T-score habits. BR units are not equal-interval and not normally distributed, so a BR of 80 on one scale is not "the same amount" of anything as a BR of 80 on another, and percentile talk borrowed from cognitive testing misstates the metric. Personality patterns before syndromes. Follow the instrument's own logic: check the Severe Personality Pathology scales, describe the elevated Clinical Personality Patterns and how they hang together, and only then read the Clinical and Severe Clinical Syndrome scales inside that personality context, since the same anxiety syndrome means different things on an avoidant versus a narcissistic foundation. Pitfall: a syndrome-only summary. Reporting the clinical syndromes as a freestanding symptom list discards the personality architecture the instrument exists to describe. Facet scales to localize, not to headline. Where a personality scale is elevated, use its three Grossman facets to say which functional domain drives the elevation and where treatment can grip. Pitfall: facet scores promoted to standalone findings. Facets refine an elevated parent scale; quoted alone they over-slice the data and invite over-interpretation. Hedged pattern language, never a test-made diagnosis. Write "the profile is consistent with a prominent avoidant pattern," tied to interview and history, and let the diagnostic section of the evaluation decide. The publisher's report states its prototypes "correspond conceptually to formal diagnostic categories" while differing from DSM-5 criteria, that several DSM-5 syndromes are not assessed at all, and that definitive diagnoses require biographical, observational, and interview data. Pitfall: "the MCMI-IV diagnosed borderline personality disorder." No self-report inventory diagnoses anyone, and the concordance literature against structured interviews is modest. Interpretive-report integration, summary, and recommendations linkage. Paraphrase the computer narrative into case-anchored language, excerpting at most the licensed minimum, reconcile it with everything else you know, answer the referral question as hypotheses the profile supports, and tie each recommendation to a named finding: pattern-informed psychotherapy targets to the elevated personality scales, corroboration steps to what the inventory cannot settle. Pitfall: pasted boilerplate and free-floating recommendations. Verbatim canned paragraphs breach the report's own excerpting license and hand your interpretive responsibility to software, and a recommendation list that would fit any profile tells the reviewer the linkage is missing. ### Blank template (copy and adapt) MEASURES AND ADMINISTRATION Instrument/edition: MCMI-IV Platform: [Q-global / Q Local / mail-in] Modality: [in office / remote, arrangement noted] Norms: BR scores, adult clinical sample; fit stated: [why this examinee matches the clinical normative population] PROTOCOL VALIDITY AND RESPONSE STYLE (before any substance) Invalidity / Inconsistency: [within limits or not, in prose] Modifying indices: [Disclosure / Desirability / Debasement pattern described; open, defensive, or self-critical] Conclusion: [interpretable / interpretable with stated limits / not interpretable, and what that means here] BASE RATE FRAME (state once) [BR anchored to clinical prevalence; 75 presence / 85 prominence convention; not T scores; no cross-scale comparisons] PERSONALITY FINDINGS Severe Personality Pathology: [elevations or "unremarkable"] Clinical Personality Patterns: [elevated patterns and how they cohere, in prose] Facets: [Grossman facets localizing each elevated pattern] CLINICAL SYNDROMES (read inside the personality context) [Syndrome elevations tied to the pattern; severe syndromes noted] DIAGNOSTIC FORMULATION (multi-source) [Hedged pattern language; corroboration from interview, history, records; what the inventory cannot establish] INTERPRETIVE NARRATIVE INTEGRATION [Computer report paraphrased; licensed minimum excerpts only] INTERPRETIVE SUMMARY (answer the referral question) RECOMMENDATIONS LINKAGE [Each recommendation tied to a named finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample MCMI-IV write-up (fictional) Scenario: a 38-year-old warehouse supervisor, self-referred to outpatient therapy for longstanding anxiety and low mood, referred by his treating therapist for psychological evaluation to clarify diagnosis and guide treatment planning. He is a genuinely help-seeking clinical patient, the population the MCMI-IV is normed on, and the protocol shows open disclosure with a coherent avoidant-melancholic pattern, the treatment-planning shape where the write-up's job is to turn pattern findings into workable therapy targets. This is the MCMI-IV results section only, condensed but structurally complete. All details are fictional. Client: R.M., 38 · Referral: diagnostic clarification and treatment planning, referred by treating therapist · Evaluator: J. Okafor, PsyD, Licensed Psychologist · Testing date: 07/10/2026 · Report date: 07/16/2026 Measures and administration: The Millon Clinical Multiaxial Inventory-IV (MCMI-IV) was administered in the office on Q-global as part of a psychological evaluation that also included a diagnostic interview, collateral information from the treating therapist, and record review, reported in their own sections. MCMI-IV results are Base Rate (BR) scores calibrated to an adult clinical normative sample; R.M. fits that reference population as an adult engaged in outpatient mental health treatment. By the instrument's convention, a BR score at or above 75 suggests the presence of a trait or syndrome and 85 or above suggests prominence; BR scores are not T scores and are not compared across scales in this report. The computer interpretive report was reviewed and its findings are paraphrased and integrated here with interview and history rather than quoted. Protocol validity and response style: R.M. answered all 195 items. The Invalidity and Inconsistency checks were within acceptable limits. The modifying indices describe an open, non-defensive test-taking approach: disclosure was moderately high (Disclosure BR 64), favorable self-presentation was low (Desirability BR 42), and self-critical responding was mildly elevated (Debasement BR 61), a configuration consistent with a distressed patient describing himself candidly. The substantive profile is interpretable without qualification. Scale (family) Fictional BR score Disclosure, X (Modifying Index) 64 Desirability, Y (Modifying Index) 42 Debasement, Z (Modifying Index) 61 Avoidant, 2A (Clinical Personality Pattern) 86 Melancholic, 2B (Clinical Personality Pattern) 78 Dependent, 3 (Clinical Personality Pattern) 68 Anxiety (Clinical Syndrome) 81 Persistent Depression (Clinical Syndrome) 77 Personality findings: The Severe Personality Pathology scales (Schizotypal, Borderline, Paranoid) were unremarkable. The profile is organized around a prominent avoidant pattern (Avoidant BR 86) with a co-elevated melancholic pattern (Melancholic BR 78); the dependent scale approached but did not reach the presence threshold (BR 68), and no other personality scale was elevated. The Grossman facets localize the avoidant elevation in the interpersonal and self-image domains (Interpersonally Aversive BR 84, Alienated Self-Image BR 79), with intrusive-content responding lower (Vexatious Content BR 62); within the melancholic pattern, the self-evaluative facet carried the elevation (Worthless Self-Image BR 76). Functionally, this is the self-description of a man who expects criticism and rejection, keeps relationships at a distance he does not want, and reads his own performance as evidence of inadequacy. Clinical syndromes: Read inside that personality context, the syndrome elevations are coherent rather than freestanding: generalized anxious apprehension (Anxiety BR 81) and a chronic, low-grade depressive presentation (Persistent Depression BR 77), with the severe syndrome scales (Schizophrenic Spectrum, Major Depression, Delusional Disorder) unelevated. The pattern suggests the anxiety and dysthymic picture ride on the avoidant-melancholic foundation: apprehension rises where scrutiny is expected, and mood sags where self-worth is engaged. Diagnostic formulation and interpretive summary: The profile is consistent with a prominent avoidant personality pattern with melancholic features, accompanied by generalized anxiety and persistent depressive symptoms. These are prevalence-calibrated hypotheses from a self-report inventory, not diagnoses: the MCMI-IV's prototypes align only partly with DSM-5-TR criteria, and the diagnostic section of this evaluation weighs these findings together with the clinical interview, the treating therapist's observations, and history, which in this case corroborate the avoidant pattern (longstanding social self-protection since adolescence, promotion refused twice to avoid visibility) and the chronic mood component. Recommendations linkage: The avoidant and melancholic elevations, localized by the self-image and interpersonal facets, support psychotherapy that works the pattern directly: graded interpersonal exposure with explicit work on self-evaluative themes, coordinated with the existing therapy rather than replacing it. The anxiety and persistent-depression findings support tracking symptom change with brief rating scales as treatment proceeds. No finding on this administration indicates severe personality pathology or psychotic-spectrum process, and no recommendation rests on the inventory alone. This sample is fictional and for educational purposes. It does not describe a real client or record, and the scores are invented for illustration and correspond to no real person or record. ↑ Back to the template and downloads ### Why this sample works The clinical fit is stated, not assumed. One sentence establishes that the examinee belongs to the population the norms describe, which is the MCMI-IV's own precondition for meaning anything. Response style is settled before substance. The Invalidity and Inconsistency checks and the three modifying indices are disposed of first, with a plain interpretability conclusion, so every later sentence stands on a declared foundation. The Base Rate frame is stated once, then respected. The 75 and 85 anchors are declared as the instrument's convention, no BR value is treated as equal-interval, and no score is compared across scales. Syndromes are read inside the personality context. The anxiety and depressive elevations are interpreted on the avoidant-melancholic foundation, with the facets saying where treatment can grip, which is how the instrument is built to be read. Pattern language, never a test-made diagnosis. Every conclusion is a hypothesis tied to interview, collateral, and history, and the write-up says what the inventory cannot establish alone. Edition, platform, norms, and integration are named. MCMI-IV, Q-global, clinical reference population, interpretive report paraphrased: one administration another evaluator could reconstruct at re-evaluation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The decision contexts around the MCMI-IV mix law, payer policy, and convention, and the write-up should signal which is which. LAW (US courts): no statute names the MCMI-IV, and its admissibility record is an unresolved published debate, not a settled ruling. Rogers, Salekin, and Sewell (1999) concluded that "fundamental problems in the scientific validity and error rates" preclude MCMI-III admissibility under Daubert for personality-disorder assessment; Dyer and McCann (2000) answered that the critique was methodologically flawed, and Rogers and colleagues stood by their concerns . For the fourth edition specifically, the 2022 Journal of Personality Assessment review (whose authors span both sides of that debate, including the MCMI-IV's own co-author) finds "the peer-reviewed literature is virtually non-existent" and that normative and response-style issues "cause problems" in family court, where instructed fake-good profiles closely match custody litigants' elevations. A forensic write-up therefore discloses the clinical-norming limitation, foregrounds response style, and never rests an opinion on this inventory alone; the wider admissibility frame lives on the psychological evaluation report page. PAYER POLICY (US): the professional work is billed under 96130/96131 with administration and scoring under 96136/96137 (or 96138/96139 by a technician), and Medicare contractor article A57481 directs code 96146 for automated testing via an electronic platform, the lane a standalone Q-global MCMI-IV administration can occupy; remote administration runs through the publisher's Q-global On-Screen, Remote On-Screen, and Video Proctoring pathways per Pearson's telepractice guidance , which makes documenting the arrangement "a professional responsibility." Outside the US the funding frame changes: Australia's Better Access items rebate psychological therapy services and a July 2026 MBS search returns no item for personality-inventory administration, so MCMI-IV testing there is typically privately, insurer, or medico-legally funded; Canadian practice likewise runs through provincial, insurer, employer, and medico-legal channels rather than a national fee schedule. Edition and access questions decide defensibility more often than any scoring detail. As of July 2026 the MCMI-IV (2015) is the current edition with no successor announced, and the MCMI-III (1994, 175 items) remains on sale with hand-scoring kits the MCMI-IV catalog does not offer; since the Handbook of Personality Assessment calls the MCMI-IV "a new test rather than a revised version of the MCMI-III," name the edition in every report and do not carry MCMI-III validity evidence or interpretive habits across editions silently. Purchase and use sit at Pearson qualification level C (the widely copied "B-level" claim is wrong, as is the "8th-grade reading level" one: the publisher lists fifth grade), which is why the write-up names a qualified interpreter. The instrument is sold on a per-administration digital model (Profile and Interpretive reports priced per use), and the Interpretive Report's license permits excerpting only the minimum text necessary for your core conclusions, so the narrative is paraphrased and the printout is not the report of record. Item-level content stays out of the write-up entirely: the publisher's own sample reports omit item text, and raw responses belong in the test file under test-security ethics rather than in the released document. Retention of the protocol and report follows the record-set rules on the psychological evaluation report page. MCMI and Millon are registered trademarks of DICANDRIEN, Inc.; the MCMI-IV is published and distributed by NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, DICANDRIEN, Inc. or Pearson. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common MCMI-IV write-up errors reviewers flag There is no payer audit series for personality-test write-ups; the accountability pressure comes from courts, opposing experts, licensing boards, and the publisher's own license terms, and for the MCMI family it is unusually well documented because the instrument's metric and norming invite specific misreadings. The 2022 Journal of Personality Assessment legal-settings review concludes the MCMI-IV's own peer-reviewed literature is "virtually non-existent" and its modifying indices have "questionable utility in the detecting of response bias"; the base-rate metric itself drew a published methodological critique from Grove and Vrieze (2009) , who argued the manual's documentation of the transformation is incomplete and that "a single base rate per diagnostic category is ill-advised" because base rates differ across settings; and the custody literature documents a recurring defensive profile that faking-good studies reproduce almost exactly . The BastionGPT Clinical Advisory Board sees the same errors most often in MCMI-IV write-up reviews: BR scores written as if they were T scores. Reports that call a BR of 80 "two standard deviations above the mean," rank scales against each other by BR value, or convert BR to percentile talk are describing a metric the MCMI-IV does not use. Base Rate units are prevalence-anchored to the clinical norm group and are not equal-interval; state the 75 and 85 anchors once as the instrument's convention and describe the pattern in prose. An elevation reported as a diagnosis. The publisher's own report says its prototypes "correspond conceptually to formal diagnostic categories" while differing from DSM-5 criteria, and the concordance evidence against structured diagnostic interviews is modest (a Persian-version study reported criterion validity of 0.13 to 0.40 against SCID-5-based diagnosis). "The profile is consistent with a prominent avoidant pattern" is defensible; "the MCMI-IV shows avoidant personality disorder" is not. Non-clinical examinees scored on clinical norms without the boundary stated. The norms describe adults already in assessment or treatment, and Pearson warns that respondents outside that population "may have inaccurate reports." In custody and employment-adjacent contexts the documented result is a defensive configuration (elevated Desirability with Histrionic, Narcissistic, and Compulsive bumps and low Debasement) that instructed fake-good responders reproduce, so a write-up that pathologizes that profile without a response-style caveat reads as error to any informed reviewer. The interpretive report pasted instead of integrated. The computer narrative is licensed for minimum-necessary excerpting, does not correct for response style the way evaluators assume, and leaves interpretation your professional responsibility. Verbatim canned paragraphs are a copyright problem, an ethics problem, and a tell that opposing experts search for. Edition and scale-count drift. MCMI-III language, norms, or validity claims imported into an MCMI-IV report describe a different instrument ("a new test," per the Handbook of Personality Assessment), and scale counts copied from blogs (25, 28, and 30 all circulate) read as confusion unless the report says what it counted: 25 clinical scales, 28 graphed Base Rate values, 30 scales in total, 45 facets beneath. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Paste your scale summary (validity and modifying-index findings plus elevated scales with BR values) and get a drafted results-section narrative organized validity-first, with the Base Rate convention stated once, syndromes read inside the personality context, and the referral question kept in frame. Cross-check a finished draft for the gaps reviewers flag: BR values written as if T scores, an elevation phrased as a diagnosis, a missing clinical-fit sentence, response style discussed after substance, and an unstated edition, platform, or norm population. Translate the results section into a plain-language summary for the referral source, the treating clinician, or the client that keeps pattern findings as corroborated hypotheses, not verdicts. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do MCMI-IV Base Rate scores actually mean? Base Rate (BR) scores are the MCMI-IV's own metric, and they are not T scores. Instead of placing a person on a normal curve, each scale's BR values are anchored to how common that trait or syndrome is among the adult clinical patients the test is normed on, a logic that traces to Meehl and Rosen's 1955 work on base rates and cutting scores. By the instrument's convention, a BR at or above 75 suggests the trait or syndrome is present and 85 or above suggests it is prominent, and the publisher's Profile Report graphs BR scores for 28 scales with exactly those anchors marked. Because each scale is calibrated to its own prevalence, BR units are not equal-interval and the same number on two scales does not mean the same thing, so describe the pattern in prose rather than ranking scores. An elevation is a prevalence-calibrated hypothesis to corroborate, not a finding that stands alone. Is an elevated MCMI-IV scale a personality-disorder diagnosis? No. The publisher's own interpretive report describes its output as prototypes that "correspond conceptually to formal diagnostic categories" while using criteria and items that differ from the DSM-5, notes that several DSM-5 syndromes are not assessed at all, and states that definitive diagnoses must draw on biographical, observational, and interview data. The independent evidence says the same thing: a structured diagnostic interview remains the criterion standard for a categorical personality-disorder diagnosis, and the published MCMI-IV concordance work is thin, with a Persian-version study reporting criterion validity of only 0.13 to 0.40 against SCID-5-based diagnosis. The defensible write-up convention is hedged, dimensional pattern language, "the profile is consistent with a prominent avoidant pattern," tied to interview and history in a broader psychological evaluation , with the diagnostic conclusion made by the clinician from all sources, never by the inventory. Can the MCMI-IV be used with someone who is not in mental health care, or with adolescents? The norms say no, and the publisher says so directly: MCMI-IV reports are normed on patients in the early phases of assessment or psychotherapy, and respondents outside that population, or tested for nonclinical purposes, "may have inaccurate reports." Scored against clinical norms, ordinary self-protection in a high-stakes setting looks like pathology, which is exactly what the custody literature documents: litigants produce elevated Desirability with Histrionic, Narcissistic, and Compulsive bumps and low Debasement, a configuration that instructed fake-good responders reproduce almost exactly . If a non-clinical examinee must be tested, the write-up leads with that limitation and a community-normed alternative like the PAI deserves first consideration. For ages, the boundary is absolute: 18 and older take the MCMI-IV; adolescents 13 to 18 take the MACI-II, which Pearson likewise states "should not be used with nonclinical populations"; ages 9 to 12 take the M-PACI. MCMI-IV, MMPI-3, or PAI: which inventory should I use? The practitioner convention is stable. The MMPI-3 (335 items, census-matched community norms, non-gendered T scores) carries the deepest validity-scale architecture and research base, so it leads when litigation is likely or response style is the central question. The PAI (344 items, community, clinical, and college T-score norms, lower reading demand) suits general and higher-functioning clients and settings where community norming answers the referral question. The MCMI-IV is the specialist: 195 items, clinical-only norms, prevalence-anchored Base Rate scores, and a scale architecture organized around personality patterns, so it earns its place when the examinee is genuinely a clinical patient and the referral question is which personality structure the presentation is built on. Many evaluators administer a broadband inventory first and add the MCMI-IV only when a personality question emerges. Whichever you choose, name the instrument, edition, and norm population in the report. Is the MCMI-IV required for any evaluation, and does it hold up in court? No law, payer, or agency mandates the MCMI-IV for any evaluation; its presence in clinical and forensic batteries is convention. Admissibility is a genuinely contested record rather than a settled one: Rogers, Salekin, and Sewell (1999) argued the MCMI-III fails Daubert for personality-disorder assessment, Dyer and McCann (2000) answered the critique, and the exchange closed with both sides unmoved. For the MCMI-IV specifically, the 2022 Journal of Personality Assessment review found the edition's peer-reviewed literature "virtually non-existent," which tempers any general-acceptance claim. A court-facing write-up discloses the clinical-norming limitation, leads with response style, avoids resting any opinion on the inventory alone, and keeps the computer narrative paraphrased; the wider admissibility and testimony frame lives on the psychological evaluation report page. Does the MCMI-IV have 25, 28, or 30 scales? All three numbers circulate because they count different things, and a report reads as confused unless it says which it means. The Millon Personality Group lists 25 clinical scales: 12 Clinical Personality Patterns (including the Turbulent scale, new in this edition), 3 Severe Personality Pathology scales, 7 Clinical Syndromes, and 3 Severe Clinical Syndromes. Add the response-style measures (the Disclosure, Desirability, and Debasement modifying indices plus the Invalidity and Inconsistency checks) and the total is 30. The publisher's Profile Report, in its own words, "provides base rate scores for all 28 scales": the 25 clinical scales plus the three modifying indices, since Invalidity and Inconsistency carry raw counts rather than BR values. Beneath all of it sit 45 Grossman Facet Scales, three per personality scale, used to localize elevated patterns. The cleanest formulation for a write-up: 30 scales in total, 25 clinical, 28 graphed as Base Rate scores, 45 facets beneath. Can the MCMI-IV be administered remotely, and how is it billed? Yes, through the publisher's own pathways: Pearson's telepractice guidance lists the MCMI-IV among the questionnaires delivered on Q-global by On-Screen Administration, Remote On-Screen Administration, or Q-global Video Proctoring, and states that documenting the considerations, procedures, and conclusions "remains a professional responsibility," so the report names the modality and arrangement. On the payer side, the US psychological-testing family applies: evaluation services under 96130 and 96131, administration and scoring under 96136 and 96137 (or 96138 and 96139 by a technician), and Medicare contractor article A57481 directs code 96146 for automated testing via an electronic platform, the lane a standalone Q-global administration can occupy. Outside the US, a July 2026 MBS search returns no Australian item for personality-inventory administration, and Canadian funding runs through provincial, insurer, and medico-legal channels. Can I quote MCMI-IV items or the interpretive report in my write-up? Neither, beyond a licensed minimum for the report narrative, and no items ever. Test items are copyrighted trade secrets and protected by test-security ethics; Pearson's own sample reports print "Item content omitted" where responses are flagged, and a released write-up that quotes item text compromises the instrument for every future examinee, which is why leaked copies circulating online are both infringing and clinically corrosive. Describe what an endorsed pattern means clinically instead of what any item says. The computer Interpretive Report sits under its own printed license: the qualified user may excerpt only the minimum text necessary to describe its significant core conclusions in a written evaluation, so the working convention is to paraphrase the narrative into case-anchored language, reconcile it with interview and history, and attribute the instrument. Raw item responses and the protocol stay in the test file, not the released report; retention follows the record rules on the psychological evaluation report page. How does BastionGPT help with MCMI-IV report write-ups? BastionGPT is built for the report side of testing, where the hours actually go. Paste your validity and modifying-index findings and elevated scales with BR values, and it drafts a results-section narrative organized the way this page teaches: clinical fit stated, response style first, the Base Rate convention declared once, syndromes read inside the personality context, and hedged pattern language throughout. It will cross-check an existing draft for the errors reviewers flag, BR values treated as T scores, elevations phrased as diagnoses, or a missing edition-and-norms sentence, and it can translate a finished results section into a plain-language summary for the referral source or client. It is HIPAA-compliant, and it drafts while you keep interpretive responsibility, consistent with the ethics of automated scoring and interpretation services. Start from this page's template, or see how clinicians use it daily on the AI therapy notes page. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: NCS Pearson, MCMI-IV product page : 2015 publication, authors, 195 items, 25 to 30 minutes, ages 18+, 5th-grade reading level, qualification level C, clinical norm sample of 1,547, 28 scales, Q-global/Q Local/mail-in scoring, per-administration pricing, Spanish edition; and the MCMI-IV comprehensive brochure : the Profile Report's 28 base-rate-scale graph, the new Turbulent scale, and the Grossman Facet Scale structure. Pearson, MCMI-IV Interpretive Report sample : the clinical-norming caveat, the excerpting license, the DSM-5 correspondence caveats, the validity-first report order, the BR 75/85 profile anchors, the omitted-item-content practice, and the DICANDRIEN, Inc. trademark line. Millon Personality Group, MCMI-IV overview : the 25-clinical-scale listing by family, the modifying indices, 25 to 35 minute completion, fifth-grade terminology, and Theodore Millon (1928-2014) biography pages. Pearson, MACI-II product page (ages 13 to 18, clinical-only, MCMI-IV directed for 18+, M-PACI for 9 to 12) and MCMI-III product page (1994, 175 items, handscore kits still sold). Pearson, Telepractice and the MCMI-IV : Q-global OSA/ROSA/Video Proctoring delivery and the documentation-responsibility statement. Sellbom, Flens, Gould, Ramnath, Tringone & Grossman, Journal of Personality Assessment (2022) : the MCMI-IV and MACI-II in legal settings; the virtually-non-existent peer-reviewed base, modifying-index limits, and family-court cautions. Rogers, Salekin & Sewell, Law and Human Behavior (1999) ; Dyer & McCann, Law and Human Behavior (2000) ; Rogers, Salekin & Sewell, Law and Human Behavior (2000) : the Daubert admissibility exchange. Grove & Vrieze, Psychological Assessment (2009) : the base-rate-transformation critique; Meehl & Rosen, Psychological Bulletin (1955) : antecedent probability and cutting scores. Lenny & Dear, Journal of Personality Assessment (2009) : faking good on the MCMI-III and the custody-litigant profile (citing McCann et al., 2001); Mohammadi et al., Iranian Journal of Psychiatry (2021) : Persian MCMI-IV criterion validity 0.13 to 0.40 against SCID-5. Weiner & Greene, Handbook of Personality Assessment, 2nd ed., ch. 9 (Wiley) : the new-test characterization, 75 (39.5%) new items, and the new Inconsistency scale; Buros Center for Testing, Tests reviewed in the Twentieth Mental Measurements Yearbook : the MCMI-IV's independent review listing. CMS Medicare coverage database, billing article A57481 : the psychological-testing code family and the automated-testing code 96146. MBS Online, item search : no Australian MBS item for personality-inventory administration (search verified July 2026). ## Related templates MMPI-3 Report Write-Up PAI Report Write-Up Psychological Evaluation Report Psychiatric Diagnostic Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/migdas-2-report All note templates # MIGDAS-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 The MIGDAS-2 (Monteiro Interview Guidelines for Diagnosing the Autism Spectrum, Second Edition) is a sensory-based diagnostic interview that yields a qualitative three-domain profile instead of scores. Psychologists and evaluation teams use it within autism evaluations, almost always paired with a scored instrument. This page covers how to write up MIGDAS-2 findings, with a fictional sample and a results-section template. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, school psychologists, and multidisciplinary autism evaluation teams; publisher qualification level C Audience Autism evaluation teams, IEP teams, physicians, funding and eligibility reviewers, parents and the person assessed Typical length 500 to 900 words for the MIGDAS-2 narrative · interview 30 to 90 minutes, depending on format Format family Qualitative, sensory-based diagnostic interview When it's used Autism diagnostic evaluations across ages; verbally fluent, masking, adult, and minimally verbal presentations; paired with scored instruments for payer and district acceptance Standards context Published by Western Psychological Services (2018); current edition as of July 2026; described here for write-up purposes, no interview content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MIGDAS-2? The Monteiro Interview Guidelines for Diagnosing the Autism Spectrum, Second Edition (MIGDAS-2) is a sensory-based diagnostic interview process for gathering and organizing the qualitative information needed to evaluate autism in children, adolescents, and adults. Marilyn J. Monteiro, PhD, and Sheri Stegall, PhD, are the authors, and Western Psychological Services (WPS) published the second edition in 2018. The system is a manual plus five protocols: a Parent/Caregiver Questionnaire, a Teacher Questionnaire, and three diagnostic interviews matched to verbal fluency, one for individuals with limited to no verbal fluency, one for verbally fluent children and adolescents, and one for verbally fluent adults. The interview runs 30 to 90 minutes depending on format, uses everyday sensory-based toys and materials that are purchased separately from the kit, and requires publisher qualification level C. The output is a qualitative description of the person across three areas, Language and Communication, Social Relationships and Emotional Responses, and Sensory Use and Interests, summarized in a strengths-and-differences profile the materials call a brain style profile. As of July 2026 the 2018 edition remains current, with no successor announced. The load-bearing fact for the write-up: the MIGDAS-2 produces no scores, no norms, and no cutoffs, by design. WPS's own product description positions the result as a behavioral profile "supplementing assessment scores and informing diagnosis and treatment planning." That makes it the structural opposite of the scored instruments it usually sits beside, the ADOS-2 with its algorithm classification, the CARS-2 severity rating, and the SRS-2 T-scores. Where a scored instrument gives you a number to interpret, a MIGDAS-2 section carries the entire diagnostic argument through organized description mapped to DSM-5-TR criteria, which is why the writing itself matters more here than on any other autism measure, and why acceptance varies sharply by payer and program, as covered below. This page reproduces none of the interview content or protocol materials. ## Who uses MIGDAS-2 write-ups and when Psychologists, school psychologists, and multidisciplinary evaluation teams use the MIGDAS-2 across schools, private practice, and hospital settings, and Texas's state autism resource lists it as an evaluation tool for ages 3 to 99. It earns its place in three referral situations. First, verbally fluent and high-masking presentations, where a structured observation can sit near threshold and a low-demand, interest-led conversation gives the person room to show how they actually process the social world. Second, adult and late-identification evaluations, which have a dedicated adult interview protocol. Third, minimally verbal individuals, through the limited-to-no-verbal-fluency protocol with a caregiver present. Its qualitative output also fits teams that value strengths-forward feedback, since the process is built to support narrative reporting in language parents understand. What it does not do is stand alone: WPS itself markets the MIGDAS-2 as the qualitative half of a pairing with the ADOS-2 , and in practice evaluators pair it with a scored measure such as the SRS-2 or CARS-2 , plus a developmental history such as the ADI-R , inside a full psychological evaluation report . Where funding rules name specific scored instruments, the pairing is not optional, a point the considerations section takes up. ## How to structure a MIGDAS-2 results section Because the MIGDAS-2 yields narrative rather than numbers, the results section is longer than a score write-up and its organization carries the evidentiary weight. The convention that reads best to reviewers is a short methods statement, an observation paragraph, one paragraph per domain, then integration and formulation. Name the three areas consistently using the publisher's wording, since even WPS materials vary the order and phrasing across pages. Measures and context statement. Name the instrument in full with the edition and year, the specific protocol used (limited-to-no verbal fluency, child and adolescent, or adult), the questionnaires collected, and the scored measures administered alongside it. State in one sentence that the MIGDAS-2 is a qualitative interview that yields a descriptive profile rather than scores. Pitfall: leaving the reader to assume a score is coming; a reviewer who reaches the end of the section hunting for a number reads the report as incomplete rather than the instrument as qualitative. Interview conditions and behavioral observations. Describe the setting, who was present, the sensory-based materials offered, and how the person engaged over the session arc, including rapport, transitions, and response to the low-demand format. Pitfall: generic observation boilerplate; the sensory-based, interest-led format is the method, so describe how the person used it, not just that they "engaged well." Language and Communication. Describe conversational flow on preferred versus non-preferred topics, reciprocity, scripting or precise phrasing, prosody, and how language held up when the agenda shifted, each claim anchored to a described moment from the interview. Pitfall: adjectives without evidence; "pedantic speech" persuades no one, while a described exchange does the diagnostic work for you. Social Relationships and Emotional Responses. Describe how the person initiated and sustained interaction, read and used nonverbal signals, described relationships and emotions, and what changed when structure dropped away. Pitfall: glossing surface sociability; for masking presentations the load-bearing observation is often the contrast between rehearsed social openings and what happened in unstructured moments, so write the contrast explicitly. Sensory Use and Interests. Describe responses to the sensory materials, seeking and avoiding patterns, and the depth, intensity, and role of focused interests, including strengths they produce. Pitfall: cataloguing sensory behaviors as oddities; tie each pattern to DSM-5-TR's restricted, repetitive behavior criterion or to the person's self-regulation so the material argues rather than lists. Informant input and integration with scored measures. Summarize the parent, caregiver, and teacher questionnaire narratives, then state where the qualitative profile and the scored instruments converge and where they diverge, and reconcile the divergence rather than averaging it. Pitfall: letting a below-threshold score and a positive qualitative profile sit unreconciled; explain the discrepancy, for example camouflaging during brief structured observation, or revisit the conclusion. Diagnostic formulation and recommendations linkage. Map the described evidence criterion by criterion to DSM-5-TR, state the diagnostic conclusion as clinical judgment integrating all sources, summarize the strengths-and-differences profile in plain language, and tie each recommendation to a specific described finding. Pitfall: a strengths-forward summary that never lands on a clear diagnostic statement; affirming language and diagnostic clarity are not in tension, and funding reviewers need the conclusion in one findable sentence. ### Blank template (copy and adapt) MIGDAS-2 RESULTS SECTION Instrument/edition: MIGDAS-2 (2018) Protocol: [limited-to-no verbal fluency / child-adolescent verbal fluency / adult verbal fluency] Nature of measure: [one sentence stating the MIGDAS-2 is a qualitative interview yielding a descriptive profile, not scores or norms] Administered by: [name, credentials] Date(s): [ ] Paired measures: [scored instrument(s), e.g. SRS-2 / ADOS-2 / CARS-2, plus developmental history and records] INTERVIEW CONDITIONS AND OBSERVATIONS [Setting, who was present, sensory materials offered, engagement arc, rapport, transitions, response to the low-demand format] LANGUAGE AND COMMUNICATION [Described evidence: reciprocity, preferred-topic vs shifted-agenda speech, scripting, prosody; anchor each claim to a moment] SOCIAL RELATIONSHIPS AND EMOTIONAL RESPONSES [Described evidence: initiations, nonverbal signals, relationships, the structured-vs-unstructured contrast for masking presentations] SENSORY USE AND INTERESTS [Described evidence: seeking/avoiding patterns, focused interests, their intensity, role, and strengths; tie to criteria] INFORMANT INPUT AND INTEGRATION [Parent/caregiver and teacher questionnaire narratives; convergence and divergence with scored measures, reconciled not averaged] DSM-5-TR FORMULATION AND PROFILE SUMMARY [Criterion-by-criterion mapping of described evidence; diagnostic conclusion as clinical judgment; strengths-and-differences summary in plain language; limitations incl. qualitative, non-normed method] RECOMMENDATIONS LINKAGE [Each recommendation tied to a specific described finding] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. ## Sample MIGDAS-2 write-up (fictional) Scenario: a verbally fluent 9-year-old girl is referred by her school team for peer-relationship concerns alongside strong academics, a profile where masking is a live question. The MIGDAS-2 child and adolescent interview was used as the qualitative component of a broader evaluation, paired with the SRS-2 and a developmental history. This is the MIGDAS-2 section only, and it deliberately carries the argument through described behavior, with no interview content or protocol material reproduced. All details are fictional. Client: J.T., 9 · Referral: school evaluation team, peer relationships and intense interests · Evaluator: R. Vasquez, PhD, Licensed Psychologist · Interview date: 07/09/2026 · Report date: 07/16/2026 Measures and context: The Monteiro Interview Guidelines for Diagnosing the Autism Spectrum, Second Edition (MIGDAS-2, 2018) was administered using the diagnostic interview for children and adolescents with verbal fluency, with the parent/caregiver and teacher questionnaires collected beforehand. The MIGDAS-2 is a qualitative, sensory-based interview that yields a descriptive profile rather than scores; it was paired with the Social Responsiveness Scale, Second Edition (SRS-2) and a developmental history so that scored and descriptive evidence could be integrated. Interview conditions and observations: J.T. was seen in a quiet office stocked with everyday sensory materials, with her mother present for the opening minutes. She entered warily, answered early questions in brief rehearsed-sounding phrases, and settled visibly once the conversation moved to materials she could handle. Engagement was strongest when she directed the topic and thinned when the examiner introduced changes of agenda. She stayed the full session and accepted transitions with advance notice. Language and Communication: J.T. spoke in full, precise sentences, with vocabulary above age expectation. On her preferred topic, marine animals, she produced long, structured monologues with rehearsed phrasing and little checking of the listener's interest. When the examiner shifted to open-ended social topics, her responses shortened to a sentence or two, latency increased, and she twice returned the conversation to marine animals within a few turns. Conversational repair was effortful: asked to clarify an ambiguous answer, she repeated the same words more slowly rather than rephrasing. Social Relationships and Emotional Responses: J.T. described classmates by role rather than relationship, naming one friend "because we line up together." Her social openings with the examiner were polite and formulaic, and she monitored whether she was "doing it right," asking twice whether her answers were what the examiner wanted. When structure dropped away in casual moments, she oriented to the materials rather than the person. She described emotions with precision when given concrete scenarios but described her own feelings in learned, rule-like terms, for example that recess is "for making friendship attempts." Sensory Use and Interests: J.T. gravitated immediately to the textured items, sorted them by feel, and described strong aversions to cafeteria noise and clothing tags, which her mother confirmed shape morning routines. Her interest in marine animals is deep, factual, and organizing: she reads adult-level references, catalogues species by classification, and structures free time around it. The interest is a genuine strength, and it also crowded out reciprocal exchange whenever it was available as a topic. Informant input and integration: The parent questionnaire described early language that was fluent but script-like, sensitivity to sound and texture from toddlerhood, and friendships that fade without adult scaffolding. The teacher questionnaire described a capable student who follows rules closely, prefers predictable routines, and spends most unstructured time alone by choice. The SRS-2 total score fell in the moderate range (T = 71, invented for this illustration), consistent with the interview profile. The qualitative profile and the scored measure converge; the earlier screening picture had been mixed, which is consistent with a child who manages rehearsed, structured demands well and differs most in unstructured, reciprocal moments. DSM-5-TR formulation and profile summary: Across the interview, questionnaires, history, and the SRS-2, J.T. shows persistent differences in social-emotional reciprocity, in nonverbal and relationship understanding, and in restricted, repetitive patterns including sensory sensitivity and an intense focused interest, present from early development and impacting unstructured social functioning. These findings meet DSM-5-TR criteria for autism spectrum disorder, requiring support, as a clinical judgment integrating all sources; no single measure made the determination. Her profile pairs strong factual language, deep focused expertise, and rule-based learning with differences in reciprocal conversation, flexible social reading, and sensory processing. Recommendations linkage: The reciprocity findings support recommendation 1 (structured peer opportunities built around shared interests). The sensory findings support recommendation 2 (noise and clothing accommodations at school). The masking pattern supports recommendation 3 (adult check-ins that do not rely on J.T. self-reporting distress). Eligibility decisions rest with the school team applying its own criteria. This sample is fictional and for educational purposes. It does not describe a real person or record. All names, quotations, and the single rating-scale value are invented for illustration and correspond to no real child or record. ↑ Back to the template and downloads ### Why this sample works It states up front that the MIGDAS-2 is qualitative and non-normed, so the absence of scores reads as the instrument's design, not a gap in the evaluation. Every domain claim is anchored to a described moment, the shortened replies after an agenda shift, the formulaic openings, the sorting by feel, which is what lets narrative evidence do the work a score table does elsewhere. The qualitative profile is triangulated with a scored measure, informant questionnaires, and history, and the report says where they converge, which is what most acceptance and funding reviewers are checking for. The formulation maps described evidence to DSM-5-TR criteria and attributes the conclusion to clinical judgment across sources, never to the MIGDAS-2 alone. Strengths-forward language coexists with one clear, findable diagnostic statement, and every recommendation traces to a specific described finding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Acceptance is the decision that shapes a MIGDAS-2 write-up, and it splits cleanly by authority. LAW: no United States statute or regulation names any autism instrument; under IDEA, 34 CFR 300.304(b)(2) requires that a school evaluation "not use any single measure or assessment as the sole criterion," which is precisely the opening that lets a MIGDAS-2-anchored educational evaluation qualify when other data sources are present, and Texas's state autism resource (TSLAT) lists the MIGDAS-2 as an evaluation tool. PAYER POLICY: the picture inverts. Aetna's autism clinical policy bulletin names the ADI-R, ADOS-2, CARS-2, and Asperger Syndrome Diagnostic Scale as diagnostic tools used with clinical assessment, and does not name the MIGDAS-2, so acceptance of a MIGDAS-2-inclusive evaluation turns on whether the whole evaluation meets multi-source medical-necessity criteria rather than on the instrument. British Columbia is the hard exclusion: for Autism Funding, a private diagnosis "requires completion of both ADOS-2 and ADI-R" and other assessments will not meet eligibility, so a MIGDAS-2 profile cannot anchor a funding-eligible private diagnosis there, and the Canadian Paediatric Society notes that in some jurisdictions the ADOS and ADI-R are required. CONVENTION: Australia's National Guideline, whose 66 recommendations were approved by the NHMRC in December 2023, mandates no instrument and frames assessment as multi-method, the most room the MIGDAS-2 gets anywhere. The practical rule for the report: pair the qualitative profile with a scored instrument such as the ADOS-2 , CARS-2 , or SRS-2 unless you have confirmed the receiving system accepts qualitative-only evidence, and name the receiving system's requirement in the file. Evidence-base honesty belongs in the write-up posture. As of July 2026 the MIGDAS-2 has not been reviewed in the Buros Mental Measurements Yearbook, its state-resource research listing cites only the manual, and we located no independent peer-reviewed validation study, which is unsurprising for a deliberately qualitative, non-normed process but matters when a report calls it "standardized" or "evidence-based," claims a reviewer can test and a cross-examiner will. Describe it as what it is: a structured qualitative interview whose value is systematic organization of descriptive evidence, used alongside scored measures. Purchase requires WPS qualification level C, and your license and training govern use. On remote work, WPS's telepractice guidance treats the interview as an interactive social interaction that distancing and videoconference formats can affect, while the parent and teacher interviews adapt well to secure videoconferencing; if any component was gathered remotely, say so and note the effect on the interaction. Where the interview forms part of a psychological testing evaluation service, interview integration and report writing are typically reported within the usual testing-evaluation codes such as 96130 and 96131, subject to payer policy. Keep the three domain names consistent with the publisher's wording throughout the report, and date your edition statement, since instrument facts change. MIGDAS-2 is a trademark of Western Psychological Services (WPS). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ## Common MIGDAS-2 write-up errors reviewers flag The error pattern around this instrument is distinctive: because most assessment write-ups are organized around numbers, clinicians and the pages they consult keep trying to force the MIGDAS-2 into a scored mold. The publisher's own materials describe a qualitative process yielding descriptive results, it has never been reviewed in the Buros Mental Measurements Yearbook, and its state-resource research listing cites only the manual, yet reports and marketing copy still speak of MIGDAS-2 "scores" and "standardized" findings. The BastionGPT Clinical Advisory Board sees the same errors most often in MIGDAS-2 write-up reviews: Inventing score language for a scoreless instrument. Reports that cite a MIGDAS-2 "score," "cutoff," or "clinical range" describe something the instrument does not produce, and a reviewer who checks the publisher's description will read the rest of the report skeptically. Report a described profile, and let the paired scored measures carry the numbers. Calling it standardized, norm-referenced, or evidence-based. The interview is structured, but it is not standardized in the psychometric sense, has no norms, and has no located independent validation literature. Any of those labels hands an opposing reviewer an easy correction; describe it as a systematic qualitative interview used alongside scored measures. Letting the narrative stand alone where the receiving system requires scores. A MIGDAS-2-only evaluation submitted toward British Columbia Autism Funding fails eligibility outright, and payers whose policies name only scored tools will scrutinize a report with no standardized instrument. Verify the destination's requirements before the evaluation, not after the denial. Adjective-heavy, evidence-light domain sections. Writing "restricted interests" or "poor reciprocity" without a described moment wastes the instrument's entire advantage, which is the behavioral evidence it elicits. Each domain paragraph should let the reader see the behavior, then interpret it. A strengths narrative that never reaches a conclusion, or inconsistent domain labels. Strengths-forward framing is a feature of this system, but reviewers flag reports whose affirming summary buries or omits the diagnostic statement, and reports that rename or reorder the three areas mid-document. State the conclusion plainly, map it to DSM-5-TR, and use the publisher's domain wording consistently. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Draft the three-domain narrative from your interview and observation notes, organized into the report structure above for your review, with no protocol content required. Cross-check a draft for the gaps reviewers flag: score language applied to a scoreless instrument, "standardized" or "evidence-based" labels, domain claims with no described evidence, an unreconciled gap between the qualitative profile and a paired scored measure, or a missing DSM-5-TR mapping. Produce a plain-language summary of the findings for parents, teachers, and referrers that keeps the strengths-and-differences framing and the clear diagnostic statement together. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do MIGDAS-2 results look like if there are no scores? A MIGDAS-2 result is a qualitative description of the person across three areas, Language and Communication, Social Relationships and Emotional Responses, and Sensory Use and Interests, drawn from a sensory-based interview plus parent/caregiver and teacher questionnaires. The publisher describes the output as a behavioral profile that supplements assessment scores, summarized as a strengths-and-differences profile. In a report, that means one organized narrative paragraph per area, each claim anchored to described behavior, integrated with history and any scored measures, and mapped to DSM-5-TR criteria. The write-up should say explicitly that the instrument yields descriptive results rather than scores, so no reader goes hunting for a number that does not exist. The full structure is on this page, inside a whole evaluation report . Is the MIGDAS-2 standardized or norm-referenced? No on both counts, and reports should not use either word for it. The MIGDAS-2 is a structured qualitative interview: the protocols organize what the evaluator gathers and how it is documented, but there is no standardized administration yielding scores, no normative sample, and no cutoff. That is a design choice, not a defect, and the honest description, a systematic qualitative interview used alongside scored measures, is also the defensible one. Pages and reports that call it a standardized or scored measure are the most common factual error around this instrument. When a referral needs norm-referenced evidence, pair it with a scored tool such as the SRS-2 or CARS-2 . Is the MIGDAS-2 accepted by insurers, schools, and funding programs? It depends entirely on the destination, so check before evaluating. LAW: under IDEA, no instrument is named and no single measure may be the sole criterion, so a school evaluation can include the MIGDAS-2 within a multi-source assessment, and Texas's state autism resource lists it as an evaluation tool. PAYER POLICY: Aetna's autism policy names the ADI-R, ADOS-2, CARS-2, and Asperger Syndrome Diagnostic Scale, not the MIGDAS-2, so commercial acceptance rides on the whole evaluation's medical-necessity documentation. British Columbia's Autism Funding requires both the ADOS-2 and ADI-R for private diagnoses and accepts no substitute. CONVENTION: Australia's NHMRC-approved national guideline mandates no instrument and frames assessment as multi-method. The safe default everywhere: pair it with a scored instrument. MIGDAS-2 or ADOS-2, and where do the ADI-R, CARS-2, and SRS-2 fit? They answer different questions and are strongest combined. The ADOS-2 is a standardized direct observation with an algorithm classification, and it is the tool payers and funding bodies most often name. The ADI-R is a structured caregiver interview about developmental history. The CARS-2 is a clinician severity rating, and the SRS-2 is a norm-referenced trait scale. The MIGDAS-2 is the qualitative complement to all of them: a sensory-based, interest-led interview that surfaces how the person actually processes conversation, relationships, and sensory experience, which is where masking presentations often show most clearly. WPS itself markets the ADOS-2 and MIGDAS-2 as a pairing of quantitative data with lived experience. None of the five is the diagnosis; DSM-5-TR and clinical judgment are. Can adults be assessed with the MIGDAS-2? Yes. The system includes a dedicated Diagnostic Interview for Adults With Verbal Fluency, one of its five protocols, and the publisher's specifications span toddlers through adults. That makes it one of the few autism interview tools with an adult-specific protocol, which is why it appears often in late-identification evaluations, where a lifetime of camouflaging can mute what brief structured observation detects. Peer-reviewed work has quantified camouflaging across dozens of studies and validated self-report measures of it, so an adult write-up can pair the MIGDAS-2 narrative with a camouflaging measure and a scored autism instrument, then reconcile the sources. For adults, collateral history still matters; document what early-development information was available and its limits. What is a brain style profile, and should that phrase appear in my report? In the MIGDAS-2 materials, the brain style profile is the summary of the person's strengths and differences across the three areas, expressed in plain, non-deficit language, and the publisher's training emphasizes describing clients in positive and descriptive terms. It is a communication device, not a score, and it is often what families remember from feedback. In the report itself, use it as the plain-language summary paragraph after the DSM-5-TR formulation: describe what the person's processing style is good at and where it differs, in words a parent or teacher can act on. Keep the clinical terminology in the formulation, keep the profile summary consistent with it, and never let the friendly summary replace the diagnostic statement. Has the MIGDAS-2 been validated or reviewed in the Buros Mental Measurements Yearbook? As of July 2026, no. The MIGDAS-2 does not appear in the recent Buros Mental Measurements Yearbook tests-reviewed lists, the research section of its Texas state-resource listing cites only the manual, and we located no independent peer-reviewed validation, reliability, or diagnostic-accuracy study. For a deliberately qualitative, non-normed process this is unsurprising, since there are no scores whose accuracy a study could test in the usual way, but it has two write-up consequences. First, never describe the instrument as validated, evidence-based, or gold standard; those claims are checkable and false. Second, carry the evaluation's defensibility with the scored instruments and the convergence across sources, and treat the MIGDAS-2 as the organized qualitative evidence within that structure. If independent validation or a Buros review appears, this changes; date any claim you make. Can I include MIGDAS-2 interview questions or protocol content in my report? No. The protocols, interview guides, and manual content are the publisher's copyrighted test materials, and psychologists are ethically obligated to maintain test security under APA Ethics Standard 9.11 ; WPS enforces the same through its purchase qualifications. Copies of the protocols circulate on document-sharing sites, and using or citing them creates the same problem from the other direction. None of this restricts your report, because what belongs there was never the protocol: your report carries your description of the person's behavior, your organization of it by domain, and your interpretation. Describe what happened in the interview in your own words, at whatever level of behavioral detail the argument needs, and leave the instrument's internal content out of the record. Can BastionGPT draft my MIGDAS-2 write-up? Yes. Paste your interview and observation notes, the questionnaire summaries, and any scored results, and it drafts the MIGDAS-2 section for your review: three evidence-anchored domain narratives, the integration with informant and scored data, the DSM-5-TR mapping, and a plain-language strengths-and-differences summary. It can also cross-check a draft you wrote for the flags reviewers raise, invented score language, standardized or evidence-based labels, adjectives without described evidence, or an unreconciled gap with a paired measure, and produce a parent-friendly version of the findings. BastionGPT is HIPAA-compliant with a signed BAA on every plan, your data is never used to train models, and because you paste only your own notes and summaries, no protocol content ever enters the draft. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified July 2026: Western Psychological Services, MIGDAS-2 product page : 2018 publication, authors, the five protocols, qualification level C, 30-to-90-minute administration, ages spanning toddlers through adults, and the qualitative three-area description of results. Western Psychological Services, Combine ADOS-2 and MIGDAS-2 infographic : the publisher's own pairing of quantitative ADOS-2 data with the MIGDAS-2's qualitative, strengths-based behavioral profile. Western Psychological Services, telepractice guidance : the MIGDAS-2 interview as an interactive social interaction, effects of distance formats, and videoconference use of the parent and teacher interviews. Texas Statewide Leadership for Autism Training, MIGDAS-2 evaluation-tool listing : qualitative descriptions in three areas, ages 3 to 99, settings, and a research section whose only reference is the manual. U.S. Department of Education, 34 CFR 300.304 : IDEA evaluation procedures, including the bar on any single measure as the sole criterion. Aetna, Clinical Policy Bulletin 0648, Autism Spectrum Disorders : the diagnostic tools named in a major commercial policy, a list that does not include the MIGDAS-2. Government of British Columbia, How autism is diagnosed : the Autism Funding requirement that a private diagnosis include both the ADOS-2 and ADI-R. Canadian Paediatric Society, Standards of diagnostic assessment for autism spectrum disorder : diagnostic-assessment standards, including that findings from an assessment tool cannot alone diagnose ASD. Autism CRC, National Guideline for the assessment and diagnosis of autism in Australia : 66 NHMRC-approved recommendations, approved 8 December 2023, with a multi-method, instrument-agnostic framing. Buros Center for Testing, Mental Measurements Yearbook tests-reviewed lists : the MIGDAS-2 absent from the recent volumes checked (19th, 20th, and 21st). Cook, Hull, Crane, and Mandy, 2021, Clinical Psychology Review : systematic review of 29 studies quantifying camouflaging in children and adults, the evidence base behind low-demand qualitative methods for masking presentations. Hull and colleagues, 2019, Journal of Autism and Developmental Disorders : development and validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q). ## Related templates Psychological Evaluation Report ADOS-2 Report Write-Up SRS-2 Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up Autism Spectrum Evaluation Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/substance-use-assessment All note templates # Substance Use Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A substance use assessment is the comprehensive clinical evaluation that establishes whether a substance use disorder is present, how severe it is, and which level of care fits. Behavioral health clinicians and alcohol and drug counselors complete one at entry to treatment, usually organized around the six ASAM Criteria dimensions. A typical substance use assessment runs 400 to 1,200 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Alcohol and drug counselors, therapists, clinical social workers, SUD program staff Audience Treatment team, medical directors, payers, utilization reviewers, auditors Typical length 400 to 1,200 words · 30 to 60 minutes by hand (clinical team estimate) Format family Intake and assessment documentation (compare: biopsychosocial assessment, intake note) When it's used At entry to substance use disorder treatment, and when a level-of-care decision needs current clinical support Standards context No federal law prescribes a format; state SUD licensure rules, Medicaid waivers, and payer contracts set the requirements, most of them built on the ASAM Criteria On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a substance use assessment? A substance use assessment is the initial clinical evaluation that establishes whether a substance use disorder is present, documents the history behind it, and matches the client to a level of care. Clinicians also call it an SUD assessment, substance abuse assessment, chemical dependency evaluation, alcohol and drug assessment, or, where it drives placement, simply the ASAM assessment. That last name points to the organizing framework: the American Society of Addiction Medicine first published its patient placement criteria in 1991, renamed the work The ASAM Criteria with the 2013 Third Edition, and released the Fourth Edition in 2023 . The criteria are copyrighted professional guidance, not law, yet an HHS review of state regulation found 45 states specifically use them for placement in SUD treatment, which is why most assessments walk the same six dimensions from withdrawal risk to recovery environment. The most misunderstood rule around this document is 42 CFR Part 2 . Part 2 is a confidentiality rule, not a content rule: it governs who may see SUD records created by a federally assisted program that "holds itself out as providing, and provides, substance use disorder diagnosis, treatment, or referral" (42 CFR 2.11), and it restricts use of those records in civil, criminal, administrative, or legislative proceedings against the patient. It says nothing about what the assessment must contain; content requirements come from state licensure rules and payers. The document itself is a specialized cousin of the biopsychosocial assessment : it covers the same biological, psychological, and social ground, then adds a substance-by-substance use history, an explicit withdrawal-risk review, and a level-of-care recommendation that the rest of the record must support. ## Who uses substance use assessments and when The substance use assessment is the entry document wherever addiction treatment is organized as a program: outpatient and intensive outpatient SUD programs, residential treatment, withdrawal management, opioid treatment programs, and the SUD tracks of community behavioral health agencies. Alcohol and drug counselors write most of them, with therapists, clinical social workers, and program medical staff completing them where state rules allow. Referral sources vary widely, from a primary care screen to a family push, an employer, or a court, and the referral shapes what collateral the record needs. Two boundaries keep the neighbors straight. Upstream, a primary care or emergency setting runs a screening and brief intervention, and a positive screen refers out; the comprehensive assessment is this document, completed at the treatment program. Downstream from a general mental health intake, use this assessment when substance use is the presenting concern and a placement decision is needed; a practice whose client discloses moderate drinking inside therapy documents it in the intake note or biopsychosocial assessment instead, and a positive trauma screen along the way may call for a separate trauma assessment . ## Substance use assessment structure: what goes in each section Identifying information and referral context. Client identifiers, date, clinician and credentials, service type, start and stop times, and who sent the client: a primary care screen, a family member, an employer, a court. Pitfall: leaving the referral source vague; it determines what collateral belongs in the record and who will be asking for a report. Presenting problem in the client's frame. Why the client is here now, in their own words, plus what changed to prompt contact. Pitfall: recording only the referral reason; a client sent by someone else still has a frame of their own, and readiness planning starts from it. Substance use history, substance by substance. For each substance: age of first use, current pattern with amount and frequency, route, date and amount of last use, longest abstinence, and prior treatment episodes with outcomes. Pitfall: collapsing everything into "polysubstance use"; without per-substance last use and amounts, neither the withdrawal-risk review nor the diagnosis has support. Withdrawal risk and medical review. Current intoxication or withdrawal signs, any history of complicated withdrawal such as seizures or delirium, medical conditions, medications, and a standardized scale score where indicated. Pitfall: rating withdrawal risk low against a documented daily-use history with no scale score or vitals behind the rating. Mental health and co-occurring screen. Psychiatric history and treatment, current symptoms with a screening score, brief mental status observations , and a suicide and violence screen with findings written out. Some state rules require this screen by name, with an approved tool. Pitfall: leaving the screen implicit when findings are negative; write the negative finding so the record shows you asked. Readiness, relapse risk, and recovery environment. Stage of change anchored to a client statement or behavior, craving pattern, prior relapse circumstances, and the environment the client returns to: housing, work, relationships, supports, and exposure to use. Pitfall: recording readiness as a bare label; "contemplation" with nothing behind it tells the next clinician nothing about what to do with it. Diagnostic impression with criteria. For each substance, the DSM-5-TR criteria met and the count, the severity that count supports, and the matching ICD-10-CM code, with remission specifiers where they apply. Pitfall: naming a severity with no documented criteria; the count is the evidence, and 2 to 3 criteria is mild, 4 to 5 moderate, 6 or more severe. Level-of-care recommendation and dimensional summary. A rating or short narrative for each of the six dimensions, the recommended level of care that follows from them, alternatives considered, and the initial plan with referrals and consents. Pitfall: a placement no documented dimension supports; the medical-necessity chain for the whole episode starts at this paragraph. Signature, credentials, and date. The assessing clinician's dated signature with credentials, plus co-signature where a trainee wrote it. Pitfall: treating the signature as a formality; under rules like Minnesota's, the assessment is not complete until a qualified staff member has dated and signed it. ### Blank template (copy and adapt) Client: [initials] DOB/Age: Date: Clinician/credentials: Referral source: Service: substance use assessment Start/stop time: PRESENTING PROBLEM (client's words; why now): SUBSTANCE USE HISTORY (repeat per substance) Substance: Age of first use: Pattern / amount / frequency / route: Last use (date and amount): Longest abstinence + prior treatment and outcome: WITHDRAWAL RISK & MEDICAL REVIEW Current signs / scale score (e.g., CIWA-Ar): Complicated withdrawal history (seizures, delirium): Medical conditions / medications / allergies: MENTAL HEALTH & CO-OCCURRING SCREEN Psychiatric history / current symptoms + tool score: Suicide and violence screen (write negative findings): Mental status observations: READINESS (stage of change + client statement/behavior): RELAPSE RISK (craving pattern, prior relapse circumstances): RECOVERY ENVIRONMENT (housing, work, supports, exposure): STRENGTHS (what has worked; abstinence periods): DIAGNOSTIC IMPRESSION (per substance) DSM-5-TR criteria met + count; severity; ICD-10-CM code: DIMENSIONAL SUMMARY (ratings or narrative, Dimensions 1-6): LEVEL-OF-CARE RECOMMENDATION + rationale: Alternatives considered / barriers: INITIAL PLAN (referrals, consents, next appointment): Part 2 consent for disclosures reviewed/signed: [Y/N/NA] Consent to treatment reviewed and signed: [Y/N] Clinician signature/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample substance use assessment Scenario: adult referred by his primary care provider to an outpatient SUD program after an elevated alcohol screen at an annual visit. All details are fictional. Client: D.R., 34 · Date: 07/14/2026 · Clinician: M. Okafor, LADC · Service: Substance use assessment, 68 min, in office · Referral: Primary care provider · Start/stop: 09:03 to 10:11 Presenting problem: Referred after an AUDIT-C of 9 at his June physical. Client's frame: "the drinking crept up on me after my partner moved out." States he does not like where it is headed. Decided to follow through on the referral after canceling a morning with his daughter because he felt too rough to drive. Substance use history: Alcohol: first use at 15; weekend use through his 20s; daily since early 2025. Current pattern 5 to 7 standard drinks most evenings, beer and whiskey, starting about 8 PM. Last use last night, 6 drinks, ending around 11 PM. Longest abstinence 6 months in 2022, self-directed, ended during a stressful job change. No prior formal SUD treatment. Reports morning shakiness after heavier nights, twice eased by a morning drink in the past month. Cannabis: first use at 17; currently vapes 2 to 3 evenings weekly; last use three days ago; has twice tried to cut back to weekends only and not held it; sees it as "not the problem." Other: denies tobacco, opioids, stimulants, sedatives, and hallucinogens; no injection use ever. Caffeine: two coffees daily. Withdrawal risk and medical review: Mild morning tremor and sweats after heavier nights; no history of withdrawal seizures, hallucinations, or delirium. CIWA-Ar administered this morning = 4. BP 138/88, pulse 84. Hypertension diagnosed 2025, on amlodipine 5 mg; no other medications or allergies; last physical June 2026. No current intoxication observed. Ambulatory setting judged appropriate with daily check-ins during the first week of reduction; medical questions routed to his PCP. Mental health and co-occurring screen: No prior psychiatric treatment or hospitalization. PHQ-9 today = 11, driven by sleep, energy, and mood items; item 9 = 0. Denies current suicidal ideation, intent, or plan; denies history of attempts and self-harm; denies thoughts of harming others. Mental status: on time, groomed, cooperative; speech normal rate; mood "flat lately"; affect congruent; thought process linear; insight fair and improving. Depressive symptoms to be re-screened after early abstinence rather than diagnosed today. Readiness, relapse risk, and recovery environment: Preparation stage, anchored to conduct rather than a label: he called to schedule within two days of the referral and states "I don't want to be the dad who misses Saturdays." Craving strongest between 8 and 10 PM alone at home. Prior 6-month abstinence shows capacity; it ended without a support structure in place. Lives alone; brother 10 minutes away and aware; employed as an electrician with early starts; agreed to remove alcohol from the apartment this week. Exposure risk: solitary evenings, a poker group that centers on drinking. Diagnostic impression: Alcohol use disorder, severe (F10.20): 6 criteria documented above (larger amounts over longer period, unsuccessful efforts to cut down, craving, tolerance, withdrawal, continued use despite the hypertension conversation with his PCP and the missed morning with his daughter). Cannabis use disorder, mild (F12.10): 2 criteria (unsuccessful efforts to cut down, larger amounts than intended). DSM-5-TR counts support each severity. Dimensional summary and level of care: Dimension 1 (withdrawal): moderate; CIWA-Ar 4, no complicated-withdrawal history, manageable ambulatory with monitoring. Dimension 2 (biomedical): mild; hypertension, PCP engaged. Dimension 3 (emotional/behavioral): moderate; depressive symptoms to re-screen. Dimension 4 (readiness): engaged, preparation stage. Dimension 5 (relapse risk): high; nightly pattern, evening craving window. Dimension 6 (recovery environment): moderate; lives alone, supportive brother, stable work. Recommendation: intensive outpatient program, three evenings weekly, which places structure directly inside the high-risk window. Weekly individual outpatient was considered and set aside as insufficient against a nightly pattern. Client concurs; schedule fits his work hours. Initial plan: IOP intake 07/16/2026. Part 2 consent for disclosure to his PCP reviewed and signed; program medical provider to discuss medication options for alcohol use disorder at intake; labs requested through PCP. Re-administer PHQ-9 and AUDIT-C at week 4. Brother identified as a support contact with client consent. Limits of confidentiality, program rules, and fees reviewed; consent to treatment signed. Assessment complete upon signature below. M. Okafor, LADC, signed 07/14/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every substance has its own history with amount, pattern, and dated last use, so the withdrawal review and the diagnosis both trace to documented facts. Withdrawal risk is assessed rather than asserted: a scale score and vitals sit behind the rating, and the ambulatory disposition matches them. Each diagnosis names the criteria that support it, and the counts match the severity labels, which is exactly what a reviewer checks first. The level-of-care recommendation follows from the six-dimension summary, and the alternative considered is documented with the reason it was set aside. The co-occurring and risk screens record negative findings explicitly, so the chart proves the questions were asked. Consent, the disclosure authorization, times, credentials, and the dated signature that completes the assessment are all present. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Be precise about what 42 CFR Part 2 does, because most summaries are not. Part 2 protects the confidentiality of SUD records held by part 2 programs: federally assisted programs that hold themselves out as providing SUD diagnosis, treatment, or referral ( 42 CFR 2.11 ), which HHS has noted covers most SUD treatment programs. It controls consent and disclosure, and it restricts use of the records in civil, criminal, administrative, or legislative proceedings against the patient, protection that goes beyond HIPAA. The 2024 final rule (effective April 16, 2024, compliance required by February 16, 2026) lets patients sign a single consent for all future treatment, payment, and health care operations disclosures and gives HHS civil enforcement authority, including civil money penalties. What Part 2 does not do is prescribe one word of the assessment's clinical content, and a clinician outside a part 2 program is not under Part 2 at all, though HIPAA and state confidentiality law still apply. Write the assessment to clinical and payer standards; handle the finished record under Part 2's consent rules. Content and deadline requirements live in state licensure rules and payer policy, and they are specific. LAW: Minnesota requires the comprehensive assessment face-to-face within five calendar days of service initiation for a residential program, requires a co-occurring mental health screen with an approved tool, requires risk ratings across the dimensions with an ASAM level-of-care recommendation, and deems the assessment complete only upon a qualified staff member's dated signature ( Minn. Stat. 245G.05 ). LAW: opioid treatment programs must complete the full initial psychosocial assessment, care plan included, within 14 calendar days of admission ( 42 CFR 8.12 ). STATE AND PAYER POLICY: 45 states specifically use the ASAM Criteria for SUD placement, many through Medicaid 1115 demonstration terms ( HHS ASPE, 2021 ), and with the Fourth Edition released in 2023, contracts can reference different editions; confirm which edition each state rule and payer contract names before you build the form around one. The general shape holds here as everywhere: the format is a convention; the content is the requirement. The assessment then hands off to the treatment plan , and when the admission interview is billed as a psychiatric diagnostic evaluation , this document is the record behind the claim. ↑ Back to the template and downloads ## Common substance use assessment errors auditors flag SUD documentation audits produce some of the worst compliance numbers in behavioral health. In a 2021 audit of Colorado's Medicaid opioid treatment programs ( HHS-OIG A-07-20-04118 ), providers had billed almost 1.5 million services worth about $22.2 million over two years; of 100 sampled services, 21 complied with federal and state requirements and 79 did not, and OIG estimated about 79 percent of services statewide fell short. The deficiencies were documentation deficiencies: missing treatment plans, missing required signatures, missing signed consents, and services with no medical evaluation or initial assessment on file at all. The BastionGPT Clinical Advisory Board sees the same errors most often in substance use assessment reviews: Per-substance detail collapsed into a summary phrase. "Polysubstance use, daily" supports nothing. Each substance needs amount, frequency, route, and a dated last use; those facts are what the withdrawal review, the diagnosis, and the placement all stand on. Withdrawal risk asserted rather than assessed. A low risk rating above a documented daily-drinking history, with no scale score, vitals, or complicated-withdrawal history recorded, reads as unexamined and is a safety finding waiting to be written up. A screening score standing in for the assessment. An AUDIT-C or DAST-10 identifies risk; it does not establish a diagnosis or a level of care. A chart where the score is the whole workup fails reviewers on both, and it confuses the billing: screening and brief intervention are their own coded services, separate from the comprehensive assessment. A diagnosis with no criteria behind it. "Alcohol use disorder, severe" is a conclusion. Reviewers look for the documented criteria and the count that supports the severity: 2 to 3 mild, 4 to 5 moderate, 6 or more severe. A level of care no dimension supports. The placement recommendation is the medical-necessity anchor for the whole episode. When the dimensional summary is missing, thin, or contradicts the recommendation, every subsequent claim inherits the problem; rules like Minnesota's make the dimensional ratings and the ASAM level-of-care recommendation explicit requirements. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on substance use assessments. Draft a complete substance use assessment from interview bullets, a dictation, or a pasted transcript. Organize screening scores, prior records, and collateral into per-substance histories and a six-dimension summary, so the placement rationale arrives structured. Check the draft before you sign: a substance with no dated last use, an unassessed withdrawal risk, a criteria count that does not match the severity, or a level of care the findings do not support. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a substance use assessment be? Most substance use assessments run 400 to 1,200 words and take 30 to 60 minutes by hand. The time range is a clinical team estimate: no published study benchmarks completion time for this document specifically. No regulation in the US, Canada, or Australia sets a word count or page length; where rules exist, they set completion deadlines and content requirements instead, program by program. Write to what the document must support: the diagnosis with its criteria, the withdrawal-risk finding, and the level-of-care recommendation. Do I have to use the ASAM Criteria for a substance use assessment? No federal law requires it. The requirement, where it exists, comes from your state and payers: an HHS ASPE review found 45 states specifically use the ASAM Criteria for placement in SUD treatment, many through Medicaid 1115 demonstration terms, while New York developed its own level-of-care tool, LOCADTR . Where ASAM is named, document every dimension: research on structured administration reported an intraclass correlation of .77 for level-of-care assignment , and the dimensional narrative is what utilization reviewers actually read. Does 42 CFR Part 2 change what I write in a substance use assessment? No. Part 2 governs consent and disclosure for records held by federally assisted SUD programs; it imposes no content requirements on the assessment itself. The 2024 final rule modernized the consent side: one signed consent can now cover all future treatment, payment, and health care operations disclosures, compliance was required by February 16, 2026, and HHS holds civil enforcement authority that includes civil money penalties. Write the clinical content to state and payer standards; apply Part 2 when the finished record moves. What is the difference between a substance use assessment and a biopsychosocial assessment? The substance use assessment is the specialized version. It covers the same biological, psychological, and social ground as a biopsychosocial assessment , then adds a substance-by-substance use history, an explicit withdrawal-risk review, and a level-of-care recommendation tied to the six dimensions. General behavioral health settings lead with the biopsychosocial document and expand the substance section when a screen flags it; SUD programs lead with this one. In many agencies a single document serves both purposes, and reviewers care whether the required content is present, not which title sits at the top. How is a substance use assessment billed? The document is not the code; the billed service is. Where the admission interview qualifies as a psychiatric diagnostic evaluation , it is commonly billed as 90791, and many state Medicaid programs pay the assessment under H0001. Keep the screening track separate: Medicare's timed structured-assessment and brief-intervention codes G0396 and G0397, and Medicaid's H0049 and H0050 where states adopt them, describe early-intervention services , not the comprehensive assessment. In Australia, MBS online lists no item for a standalone alcohol and other drug assessment as of July 2026; assessment work reaches Medicare there inside attendance and mental health items. How soon after admission must a substance use assessment be completed? There is no general federal deadline; deadlines belong to program and state rules. Minnesota requires the comprehensive assessment face-to-face within five calendar days of service initiation for a residential program, excluding the day of initiation, and by the end of the fifth day on which a treatment service is provided in a nonresidential program. Opioid treatment programs must complete the full initial psychosocial assessment within 14 calendar days of admission under 42 CFR 8.12 . Check your own state's SUD licensure chapter: the deadline attaches to your program type, not to the document's name. Who can write a substance use assessment, and does it need a co-signature? State licensure decides, and Minnesota's rule shows the common shape: the comprehensive assessment may be administered by an alcohol and drug counselor, by a qualifying mental health professional with specific substance use disorder training, or by a clinical trainee practicing under supervision ( Minn. Stat. 245G.05 ). Co-signature expectations follow the trainee and billing rules in your state rather than any national standard. The signature itself is not a formality: under rules like Minnesota's, the assessment is complete only upon a qualified staff member's dated signature. Which edition of the ASAM Criteria should I document against? The Third Edition (2013) established the six-dimension format most programs still use, and the Fourth Edition was released in 2023 . Adoption runs through state rules, waiver terms, and payer contracts, and those references do not all update at once, so confirm which edition each state rule and each contract names, and document against that one. Whichever edition applies, the underlying content carries over: the per-substance history, the dimensional findings, and a placement rationale the findings support. Can BastionGPT draft a substance use assessment from a transcript? Yes. Paste the assessment interview transcript, dictate, or give it bullets, and it produces a structured draft with per-substance histories, the six-dimension summary, and the level-of-care rationale ready for your review. It can also fold screening scores, prior records, and collateral into the right sections. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Intake Note Biopsychosocial Assessment Psychiatric Diagnostic Evaluation Trauma Assessment Mental Status Exam (MSE) Note Child & Adolescent Developmental Assessment Cultural Formulation Interview (CFI) Intake Relapse Prevention Plan 42 CFR Part 2 Consent Fitness-for-Duty Evaluation Eating Disorder Assessment COWS Protocol Documentation CIWA-Ar Protocol Documentation CAGE Questionnaire Documentation AUDIT-C Score Documentation SASSI Documentation DAST-10 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/developmental-assessment All note templates # Child & Adolescent Developmental Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A child and adolescent developmental assessment documents a child's developmental history and milestone attainment across motor, language, cognitive, social-emotional, and adaptive domains, then ties the findings to eligibility or diagnosis. Child psychologists, developmental-behavioral teams, and early intervention programs complete one when a screen raises concern or a school, pediatric, or Part C referral asks what is going on. Most run 500 to 1,500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Child psychologists, developmental-behavioral pediatric teams, early intervention evaluators Audience Caregivers, early intervention programs, schools, pediatricians, payers Typical length 500 to 1,500 words · 60 to 120 minutes by hand (clinical team estimate) Format family Intake and assessment documentation (compare: biopsychosocial assessment, psychoeducational report) When it's used When a screen raises developmental concern, and for early intervention, school, and diagnostic referrals Standards context No statute prescribes the report's sections; IDEA Part C sets evaluation and timeline rules, and payer testing policies set the documentation bar On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a child and adolescent developmental assessment? A child and adolescent developmental assessment is the structured write-up of a child's developmental history and current functioning: pregnancy and birth history, milestone attainment across gross motor, fine motor, language, cognitive, social-emotional, and adaptive domains, family context, and what it all means for eligibility or diagnosis. Clinicians also call it a developmental history, a developmental-behavioral evaluation, a multidisciplinary evaluation (MDE), or, in early intervention, simply the Part C evaluation. No single body standardized it; three frameworks shaped the modern document. Federal early intervention law, 34 CFR 303.321 , requires a "timely, comprehensive, multidisciplinary" evaluation of every referred infant or toddler suspected of having a disability. The CPT testing codes 96112 and 96113 define the billable service the report documents, and the American Academy of Pediatrics coding guidance expects an interpretive report behind them. And pediatric surveillance frameworks feed the referrals: the AAP Bright Futures periodicity schedule calls for developmental surveillance at every well visit and standardized screening at the 9-, 18-, and 30-month visits, using tools aligned to the CDC and AAP milestone checklists, which were revised in 2022 to list milestones most children, defined as at least 75 percent, reach by each visit age. The structure clinicians treat as required, birth history first, then milestone tables, then domain summaries, is professional convention; no US statute specifies the report's internal sections. What the law fixes is narrower. Under 34 CFR 303.321, evaluation means "the procedures used by qualified personnel to determine a child's initial and continuing eligibility", qualified personnel must use informed clinical opinion, and, least known, a child's existing medical and other records can establish Part C eligibility without any new evaluation when they already document a qualifying delay. The document itself is a specialized cousin of the biopsychosocial assessment : the same whole-child sweep, but organized around milestones and developmental domains, and often produced as a standalone evaluative report outside any treatment episode, closer in life cycle to a psychological evaluation report than to an intake. ## Who uses child and adolescent developmental assessments and when Child psychologists, developmental-behavioral pediatric teams, and early intervention evaluators write developmental assessments; speech-language pathologists, occupational therapists, and early childhood special educators contribute domain findings inside multidisciplinary versions. Three referral streams keep the document busy. Early intervention first: a Part C referral starts a 45-day clock that must contain the evaluation, the child and family assessments, and the initial IFSP meeting, so the report is built to answer an eligibility question on a deadline. Second, diagnostic clinics: autism, global developmental delay, and language disorder workups all sit on a developmental history, and instrument-level write-ups such as a Bayley-4 report plug into this document rather than replace it. Third, pediatric and school referrals: a failed screen at a well visit, or a preschool asking what supports a child needs, arrives as a request for exactly this report. Choose the developmental assessment when the question is early development, roughly birth to five, told through milestones and domains. When the question is school-age learning, cognitive ability against academic achievement, that is the psychoeducational evaluation report . A broad diagnostic interview that opens a treatment episode belongs in an intake note or biopsychosocial assessment , and a full-battery write-up with identity-level conclusions is a psychological evaluation report . ## Child & adolescent developmental assessment structure: what goes in each section Identifying information and referral question. Child identifiers, date of birth and chronological age (with corrected age for children born preterm, where the practice uses it), caregivers present, clinician names and credentials, contact dates, and the referral source with the question asked. Pitfall: a report that never states the referral question; eligibility teams and payers read the whole document against it. Sources of information and procedures. Every procedure and informant: the records reviewed by name, parent or caregiver interview, observation settings, and instruments administered. Part C requires multiple procedures and informed clinical opinion, and it lets existing medical and other records carry eligibility on their own. Pitfall: listing instruments but not the records reviewed; an unnamed record supports nothing, and under 34 CFR 303.321 the records themselves can establish eligibility. Birth and early medical history. Pregnancy and delivery course, gestational age, birth weight, NICU time, early illnesses and hospitalizations, hearing and vision screen results, current health and medications. Pitfall: leaving hearing and vision status blank; a language delay with an unchecked hearing screen invites the wrong conclusion. Developmental milestones by domain. Attainment ages against expected ranges for gross motor, fine motor, language, cognitive, social-emotional, and adaptive skills, with the informant and basis for each (caregiver recall, records, direct observation). The revised CDC and AAP checklists benchmark ages most children, at least 75 percent, reach a milestone. Pitfall: recording "delayed" with no informant, date, or benchmark behind the word. Family and social context. Household composition, languages spoken, caregiving arrangements, family history of developmental or learning conditions, stressors and strengths. In Part C evaluations this is where the family-directed assessment of the family's resources, priorities, and concerns lives. Pitfall: writing this section as demographics only; the family's priorities are a required finding in early intervention, not color. Education and child care history. Settings attended, attendance pattern, supports already in place, prior IFSP, IEP, or therapy services and the response to them. Pitfall: omitting how the child responded to prior services; the next team inherits a plan with no evidence about what worked. Behavioral observations. What the child did across settings: engagement, attention, communication attempts, play, transitions, and caregiver-child interaction, with a note on the language or mode of assessment. Pitfall: observations that only restate scores; the observation section is where informed clinical opinion gets its evidence. Instruments, scores, and interpretation. Each instrument named with scores, the interpretation in plain terms, and a statement on validity of results. Keep screening and testing straight: a caregiver-report screen is a screen, administered standardized testing is testing, and the two carry different billing codes. Pitfall: presenting a parent-report screening instrument as if it were administered testing; payers group 96110 and 96112 differently and auditors check. Impressions: eligibility and diagnosis. The synthesis: what the history, observations, and scores together show, the eligibility statement tied to the state's criteria or the diagnosis with its basis, and informed clinical opinion where instruments missed what the record shows. Under Part C, informed clinical opinion can establish eligibility on its own but can never negate instrument results. Pitfall: an eligibility conclusion that rests on a single score with the rest of the record silent. Recommendations, signatures, and time. Prioritized recommendations that each trace to a finding, the service or referral plan, testing time with start and stop where billed, and dated signatures with credentials from each assessor, plus any signature the program requires from a parent or guardian. Pitfall: missing assessor or parent signatures; a federal audit in Maine flagged exactly this in children's assessment records. ### Blank template (copy and adapt) Child: [initials] DOB / chronological age: Corrected age (if preterm): [N/A] Caregivers present: Date(s) of assessment: Clinician(s)/credentials: Referral source + question: Service: developmental assessment Testing start/stop time: SOURCES OF INFORMATION & PROCEDURES Records reviewed (name each): Interviews / observation settings: Instruments administered (screen vs. testing labeled): Language(s)/mode of assessment: BIRTH & EARLY MEDICAL HISTORY Pregnancy / delivery / gestational age / birth weight: NICU, early illnesses, hospitalizations: Hearing screen result + date: Vision screen result + date: Current health / medications: DEVELOPMENTAL MILESTONES BY DOMAIN (age attained + informant/basis) Gross motor: Fine motor: Language (receptive/expressive): Cognitive / play: Social-emotional: Adaptive / self-help: FAMILY & SOCIAL CONTEXT Household, languages, caregiving arrangements: Family history (developmental/learning conditions): Family resources, priorities, and concerns (Part C): EDUCATION / CHILD CARE HISTORY (settings, supports, prior services + response): BEHAVIORAL OBSERVATIONS (across settings; caregiver-child interaction): INSTRUMENTS, SCORES & INTERPRETATION Instrument / score / interpretation (repeat per instrument): Validity of results statement: IMPRESSIONS: ELIGIBILITY / DIAGNOSIS Synthesis across sources; informed clinical opinion where used: Eligibility statement (state criteria) or DSM-5-TR / ICD-10-CM code: RECOMMENDATIONS (each tied to a finding; prioritized): Referrals / next steps / re-screen date: Parental consent for evaluation documented: [Y/N] Parent/guardian signature (if required): Assessor signature(s)/credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample child & adolescent developmental assessment Scenario: a 34-month-old referred by her pediatrician after a below-cutoff communication screen at the 30-month well visit. The developmental assessment report example below is condensed but structurally complete. All details are fictional. Child: S.M., 34 months · Dates: 07/09/2026 and 07/16/2026 · Assessors: L. Whitfield, PhD (licensed psychologist); R. Vega, MS, CCC-SLP · Referral: Pediatrician, for evaluation of language development · Testing time (96112): 09:05 to 10:12 Referral question: Whether S.M.'s language development is delayed relative to age expectations, whether delays extend to other domains, and what services fit. Referred after the communication domain of an ASQ-3 caregiver screen fell below cutoff at the 30-month visit. Sources and procedures: Review of pediatric records (well-visit notes 2023 to 2026, audiology report 06/24/2026); caregiver interview with mother, conducted in Spanish and English; developmental observation in clinic playroom; Bayley-4 administered 07/09/2026; M-CHAT-R with follow-up interview; assessment conducted bilingually by a Spanish-speaking examiner. Informed clinical opinion applied across sources. Birth and medical history: Born at 39 weeks, 3.4 kg, uncomplicated delivery, no NICU stay. Four episodes of otitis media between 12 and 24 months. Newborn hearing screen passed; repeat audiology 06/24/2026 showed hearing within normal limits bilaterally. No current medications. Vision screen at 30-month visit unremarkable. Milestones by domain (caregiver report anchored to records): Gross motor on time (walked at 13 months). Fine motor on time (tower of blocks by 20 months, scribbles). Language delayed: first words at 19 months, two-word combinations emerging only in the last two months, current vocabulary about 40 words across Spanish and English by caregiver count. Receptive language stronger: follows two-step directions in Spanish at home. Social-emotional typical: points to show, seeks comfort, plays alongside cousins. Adaptive age-typical: feeds self with spoon, helps with dressing. Family and social context: Lives with mother, father, and 6-year-old brother; Spanish is the primary home language, English at the family child care home three days weekly. No family history of developmental disorder; brother received speech services at age 4. Family priorities: help S.M. communicate wants without frustration, and guidance on two-language use at home. Family resources and concerns documented for the IFSP team. Education and child care: Family child care home since age 2, attends consistently; no prior early intervention services or IFSP. Provider reports S.M. gestures more than talks and has begun tantrums when not understood. Behavioral observations: Warm, referenced mother throughout, joint attention and pointing intact, functional and early pretend play observed (fed a doll). Communication attempts were frequent but mostly gesture plus single words in Spanish; imitated two-word phrases inconsistently. Attention adequate for tabletop tasks with breaks. Observations consistent across both visits and both languages. Instruments and interpretation: Bayley-4 (07/09/2026): Cognitive composite 92 (30th percentile, average range); Language composite 74 (4th percentile, well below the average range for age), receptive stronger than expressive; Motor composite 96 (39th percentile, average range). M-CHAT-R with follow-up: negative. ASQ-3 (30-month visit, caregiver report) is a screening result and is interpreted here only as the referral trigger. Results are considered valid: bilingual administration, caregiver present, engagement adequate. Impressions: History, observation, and testing converge on an expressive-predominant language delay in a bilingual context, with cognitive, motor, social-emotional, and adaptive development in expected ranges and peripheral hearing normal. Bilingual exposure does not account for the degree of delay across both languages. DSM-5-TR language disorder (F80.2). In this examiner's clinical opinion S.M. also meets this state's Part C eligibility criteria for developmental delay in the communication domain; the eligibility determination itself rests with the early intervention team. Recommendations: (1) Referral to the local Part C early intervention program for speech-language services; report to be shared with the family's consent. (2) Home language strategies in Spanish and English, modeled with mother at feedback; continuing both languages is supported. (3) Child care provider strategies shared with consent. (4) Re-screen expressive vocabulary at the 42-month visit; return sooner if two-word combinations do not consolidate. Each recommendation ties to the findings above. Parental consent for evaluation documented 07/02/2026. Feedback session held 07/16/2026 with plain-language summary provided in Spanish. L. Whitfield, PhD, signed 07/16/2026. R. Vega, MS, CCC-SLP, signed 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral question is stated and answered. The report opens with what the pediatrician asked and closes with impressions and recommendations that answer it, which is how eligibility teams and payers read. Every finding has a named source. Records are listed by date, each milestone carries an informant and basis, and informed clinical opinion is applied to the whole record, the way 34 CFR 303.321 expects, not as a substitute for it. Hearing was closed out before language conclusions. The audiology result is dated and in the record, so the language disorder impression cannot be undercut by an unchecked alternative explanation. Screening and testing stay in their lanes. The ASQ-3 is labeled a screening result and referral trigger, the Bayley-4 is administered testing with scores and interpretation, and testing time has a start and stop, which is what 96112 documentation reviews look for. Signatures, credentials, dates, and consent are complete. Both assessors sign with credentials, parental consent is documented, and the feedback session with a plain-language summary is recorded, the exact items federal auditors flagged as missing in children's assessment records. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A child's assessment outlives the childhood it describes, and the retention rules are stricter than most clinicians assume. HHS says the Privacy Rule sets no medical-record retention period and state law governs, and the state rules diverge hardest on minors: Washington's psychology rule keeps a minor's record until the client turns 22 or for 8 years, whichever is longer ( WAC 246-924-354 ); Ontario's college standard runs 10 years past the day the client turns 18 ( CPBAO Standards , in force July 2024); Australia's Psychology Board's record standards run 7 years from the last entry or until a child client turns 25; and the APA Record Keeping Guidelines , a professional convention rather than law, suggest 7 years after the last service or 3 years past majority, whichever is later. Plan for the report's audience, too: caregivers read it, copies travel to early intervention programs, child care settings, and schools with consent, and a plain-language feedback conversation is part of competent practice with families. The compliance chain runs on a small set of rules worth naming by strength. LAW: for Part C referrals, 34 CFR 303.310 puts screening, evaluation, the child and family assessments, and the initial IFSP meeting inside 45 days from referral, and 303.321 sets what the evaluation must be, not what headings the report uses. LAW: Medicaid's EPSDT benefit ( 42 U.S.C. 1396d(r) ) entitles children under 21 to periodic screening and to interperiodic screening when medically necessary, which is the coverage floor under most publicly insured referrals. PAYER POLICY: developmental testing documentation follows the billing rules, an interpretive report, time counted only for the reporting provider, and 96112 reported once per date of service , and CMS's testing article states plainly that "coverage of the service is not based on diagnosis" . CONVENTION: the section order, the milestone table, the domain summaries. The format is a convention; the content is the requirement, so an agency form that reorders sections loses nothing as long as the required findings are in it. ## Common child & adolescent developmental assessment errors auditors flag Documentation failures around children's assessments are now measured in recoupments. Between December 2024 and February 2026, HHS-OIG audits of Medicaid ABA services for children found improper or potentially improper payments in four states: Indiana (at least $56 million), Wisconsin (at least $18.5 million), Maine (at least $45.6 million), and Colorado (an OIG estimate of at least $77.8 million, against ABA spending that grew from $60.1 million in 2019 to $163.5 million in 2023). In Maine, every one of 100 sampled enrollee-months contained at least one improper or potentially improper claim line, and OIG specifically cited services to children who "did not receive the required comprehensive assessments" or whose assessments lacked the signatures of the staff who conducted them or of parents or guardians. The BastionGPT Clinical Advisory Board sees the same errors most often in developmental assessment reviews: Screening written up as testing. A caregiver-report instrument is a screen; administered standardized testing is testing with an interpretive report behind it. Blurring the two miscodes the service (96110 versus 96112) and leaves the report claiming work the record does not show. Missing time documentation. Testing codes are time-based: 96112 is reported once per date of service and only the reporting provider's time counts. A report with no start and stop, or with team time silently pooled, fails the first documentation check. Unsigned or credential-less reports. The assessor's dated signature with credentials, and a parent or guardian signature where the program requires one, are exactly the items the OIG series flagged as missing. An unsigned assessment reads as an incomplete assessment. Milestones without sources or benchmarks. "Speech delayed" with no informant, date, or benchmark is a conclusion without evidence. Name who reported each milestone, when, and against what expectation, especially since the 2022 checklist revision moved the benchmarks themselves. Conclusions detached from the evaluation record. An eligibility statement resting on one score, informed clinical opinion used to override instrument results rather than integrate them, or recommendations that appear from nowhere. Every conclusion should trace back to a documented finding, and under Part C the existing records themselves may already carry eligibility. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on child and adolescent developmental assessments. Draft a complete developmental assessment from a parent interview transcript, a dictation, or bullets, and produce a plain-language summary for caregivers alongside it. Merge pediatrician records, prior screens, therapy notes, and child care reports into one milestone chronology, organized by domain with each source named. Check the draft before you sign: a domain with no informant behind it, a hearing or vision status left open, a screen written up as testing, a recommendation with no finding to support it. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a developmental assessment be, and how long does it take? Most child and adolescent developmental assessments run 500 to 1,500 words and take 60 to 120 minutes to write by hand. The time range is a clinical team estimate: no published study benchmarks writing time for this document specifically. No regulation in the US, Canada, or Australia sets a length. The deadline that does exist is procedural: for Part C referrals, 34 CFR 303.310 requires screening, evaluation, the child and family assessments, and the initial IFSP meeting within 45 days of referral, which in practice sets the report's due date. What law actually requires a developmental assessment? For early intervention, 34 CFR 303.321 requires a timely, comprehensive, multidisciplinary evaluation of every referred child suspected of a disability, plus a family-directed assessment of the family's resources, priorities, and concerns. Two details from the same section rarely make it into practice: qualified personnel must use informed clinical opinion, which may establish eligibility on its own but may never negate instrument results, and a child's existing medical and other records may establish eligibility with no new evaluation at all. Outside Part C, no statute requires the document or its sections; expectations come from payers and professional convention. What is the difference between developmental screening and developmental testing? Screening asks a caregiver or other observer to report the child's skills on a standardized, validated instrument; testing measures what the child actually does on a psychometric instrument administered by the clinician, with an interpretive report behind it. The AAP coding fact sheet draws the same line for billing: screens report under 96110, administered developmental testing under 96112 and 96113. A failed screen is a referral trigger, not a finding, and the report should label it that way. Does billing developmental testing require a specific diagnosis? No. CMS's billing article for psychological and neuropsychological testing lists no ICD-10 codes because "coverage of the service is not based on diagnosis" (Article A57481). What the documentation must show is medical necessity and the service itself: the instruments, the time (96112 is reported once per date of service, counting only the reporting provider's time), and the interpretive report. For publicly insured children, Medicaid's EPSDT benefit ( 42 U.S.C. 1396d(r) ) separately entitles those under 21 to periodic screening and medically necessary follow-up. How is a developmental assessment different from an intake note or biopsychosocial assessment? An intake note or biopsychosocial assessment opens a treatment relationship and orients ongoing care. The developmental assessment is usually a standalone evaluative report, produced to answer an eligibility or diagnostic question, often with no treatment episode attached, and organized by milestones and developmental domains rather than the three biopsychosocial pillars. If a child entering therapy needs both, they stay separate documents: the intake carries the treatment plan, the developmental assessment carries the evaluation and its evidence. When do I write a developmental assessment versus a psychoeducational report? By age and question. The developmental assessment centers early development, roughly birth to five, told through milestones and domains, and typically feeds early intervention eligibility or a first diagnosis. The psychoeducational evaluation report centers school-age learning, cognitive ability against academic achievement, and feeds IEP eligibility and instruction planning. The payer line follows the purpose: school-eligibility testing is usually excluded from health-plan coverage, while developmental testing is billed to health plans on medical necessity, so the same child can legitimately generate different documents on different funding paths. How long do I keep a child's developmental assessment? Longer than adult records, and the "7-year HIPAA rule" is a myth: HHS confirms HIPAA sets no medical-record retention period, so state law governs. Minors' rules run long: Washington keeps a minor's psychology record until age 22 or 8 years, whichever is longer ( WAC 246-924-354 ); Ontario's college standard runs 10 years past the client's 18th birthday; Australia's Psychology Board expects 7 years from the last entry or until a child client turns 25. The APA guideline , a convention, suggests 7 years after last service or 3 years past majority, whichever is later. Follow the longest rule that applies to you. How do Australia and Canada handle developmental assessments? Through payer policy and professional standards rather than a dedicated statute. In Australia, the MBS complex neurodevelopmental disorder items (135 for a paediatrician, with parallel items for other specialists) fund assessment and a treatment and management plan for patients under 25, claimable once per lifetime, and the explanatory note AN.0.24 says diagnosis should be "evaluated in the context of both a physical and developmental assessment". The national autism guideline ( Autism CRC, second edition, 2023 ) carries 66 recommendations approved by the NHMRC CEO in December 2023. In Canada, provincial psychology college standards govern the document and the record, such as the CPBAO Standards of Professional Conduct in Ontario. Can BastionGPT draft a developmental assessment from my interview notes? Yes. Paste a parent interview transcript, dictate, or provide bullets plus scores, and it produces a structured draft with the milestone chronology by domain, sources named, and impressions and recommendations ready for your clinical judgment, plus a plain-language caregiver summary if you ask. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and data is never used to train models. Try it on the next Bayley-4 feedback write-up. ## Related templates Intake Note Biopsychosocial Assessment Psychoeducational Evaluation Report Psychological Evaluation Report Bayley-4 Report Write-Up Substance Use Assessment Cultural Formulation Interview (CFI) Intake NDIS Functional Capacity Assessment Report IEP Psychoeducational Input Parent & Caregiver Collateral Note School & Teacher Collateral Note ASQ-3 Score Documentation M-CHAT-R/F Score Documentation PSC-17 Score Documentation DAYC-2 Report Write-Up Battelle Developmental Inventory (BDI-3) Report Write-Up Beery VMI Report Write-Up GFTA-3 Report Write-Up PLS-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cultural-formulation-interview All note templates # Cultural Formulation Interview (CFI) Intake: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A cultural formulation interview (CFI) intake documents what the 16-question CFI from DSM-5-TR elicits: the client's own definition of the problem, explanatory model, stressors and supports, cultural identity, and preferences for care. Psychologists, therapists, and psychiatrists fold it into initial evaluations when cultural context shapes diagnosis or engagement. The interview takes about 20 minutes; the documented section usually runs 200 to 500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, therapists, psychiatrists, counselors, and clinical teams at intake Audience Treating clinician and care team; payers and auditors review the surrounding intake evaluation Typical length 200 to 500 words · 15 to 25 minutes by hand (clinical team estimate) Format family Structured interview addendum to intake documentation (compare: intake note, biopsychosocial assessment) When it's used During initial evaluations, when cultural context shapes presentation, engagement, or diagnostic risk Standards context A DSM-5-TR Section III convention; no US, Canadian, or Australian law, payer, or accreditor requires it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a cultural formulation interview (CFI) intake? A cultural formulation interview (CFI) intake is the documented record of the CFI, which the American Psychiatric Association describes as "a set of 16 questions" clinicians may use during a mental health assessment to understand how culture shapes a client's presentation and care. Clinicians also call the write-up a cultural formulation, a cultural assessment, or simply the CFI section of the intake. The lineage runs through three editions of the DSM: contributors to DSM-IV created the Outline for Cultural Formulation in 1994 to refocus interviews on the client's own explanatory model of illness, the DSM-5 Cross-Cultural Issues Subgroup turned that outline into an actual interview protocol for DSM-5 in 2013, and a six-country field trial with 318 patients and 75 clinicians found it feasible, acceptable, and useful. The protocol covers four domains: how the client defines the problem, what they see as its causes, context, and supports, how they have coped and sought help before, and what shapes their help seeking now. An Informant Version gathers the same picture from family or caregivers, and 12 supplementary modules go deeper on each subtopic and on specific populations, all published free on psychiatry.org . Two boundaries define the document. First, the CFI is a professional convention, not a mandate: it sits in Section III of DSM-5-TR among the emerging measures the APA offers to support clinical judgment, "not as the sole basis for making a clinical diagnosis" (APA). No US, Canadian, or Australian statute, payer policy, or accreditation standard requires a CFI, and no billing code attaches to it. Second, it is not a standalone note type: the cultural formulation almost always lives inside the intake note , biopsychosocial assessment , or psychiatric diagnostic evaluation it enriches. DSM-5-TR (2022) kept the interview itself intact: the 16 questions in the APA's posted DSM-5-TR edition match the 2013 version word for word, while the surrounding text adopted updated terminology such as "racialized" and "ethnoracial" per the APA's culture and racism fact sheet . ## Who uses CFI intakes and when Psychologists, therapists, psychiatrists, and counselors add the CFI to initial evaluations, and community mental health, refugee health, student counseling, and transcultural psychiatry services use it most. It earns its minutes when cultural context is doing real clinical work: when the client's explanation of the problem differs from the clinician's, when a presentation could be misread across cultures (a spiritual experience that might be mislabeled psychosis, distress expressed through the body), when engagement keeps failing, or when family expectations, migration, faith, or discrimination sit close to the presenting problem. The developers built it for any client, not only for visible cultural difference, and it has been evaluated internationally in the United States, Canada, Kenya, Peru, the Netherlands, India, and Mexico. Timing matters: the same review notes the CFI is harder to conduct with acute psychosis, suicidal behavior, aggression, or significant cognitive impairment, so in a crisis presentation clinicians stabilize first, complete the mental status exam and risk assessment, and hold the CFI for a calmer session. Where a focused concern dominates, a specialized assessment such as a substance use assessment or a developmental assessment leads, and the CFI supplies the cultural layer inside it. ## CFI intake structure: what goes in each section The section order below follows the four CFI domains, paraphrased for documentation. The interview itself is free to reproduce for patient care (see the FAQ), but the note should synthesize what the client said, not reprint the question set. Client and interview context. Client identifiers, date, clinician and credentials, the language the interview was conducted in, interpreter used or declined, which CFI components were administered (core interview, Informant Version and the informant's relationship, any supplementary modules), and the evaluation the addendum belongs to. Pitfall: leaving interview language and interpreter status off the record; language access is the one culture-adjacent element carried by law and accreditation, and it must be documentable. Cultural definition of the problem. The problem in the client's own words, including what they call it when talking to family or community, quoted briefly where the phrasing matters. Pitfall: translating the client's words straight into diagnostic vocabulary; the clinical label belongs in the assessment, this section preserves the client's frame. Causes and explanatory model. What the client believes causes the problem, what people close to them believe, and where those views agree or differ, including somatic, spiritual, moral, or situational explanations. Pitfall: recording the explanatory model as a curiosity and never letting it touch the formulation or the treatment conversation. Stressors and supports. The client's account of what makes things worse and better: relationships, work and money, housing, migration circumstances, faith community, discrimination experiences. DSM-5-TR frames these as social determinants of the presentation. Pitfall: labels without function, such as writing that family is supportive without saying what the support does clinically. Role of cultural identity. How the client describes their identity, which aspects they consider relevant to the problem, and any tension between identities or settings. Pitfall: assigning salience from demographics; the CFI asks rather than assumes, and the note should carry the client's answer, not the clinician's inference from group membership. Self-coping and past help seeking. What the client has already tried: their own strategies, family remedies, faith practices, traditional or community healers, and prior professional care, with what helped and what did not. Pitfall: dismissing non-clinical help sources instead of documenting them; they predict engagement and often belong in the plan. Barriers and preferences for current care. What has kept the client from care (cost, time, stigma, language, past experiences with services) and what kind of help they want now. Pitfall: eliciting preferences and then writing a plan that ignores them without comment; if the plan departs from the stated preference, document the shared reasoning. Clinician-client relationship. The client's concerns about being understood across difference, anything they want the clinician to know about working together, and interpreter or communication arrangements going forward. Pitfall: skipping the section because it feels awkward; the interview reserves a domain for it precisely because clients rarely volunteer it. Integration with formulation and plan. What the cultural findings change: how they bear on differential diagnosis, engagement, and the treatment plan, written as clinical reasoning that connects back to the history, mental status exam, risk assessment, and diagnosis documented in the parent evaluation. Pitfall: an orphaned CFI; auditors and payers read the diagnostic evaluation's required elements, and a cultural section that never joins the formulation adds length without clinical force. ### Blank template (copy and adapt) CULTURAL FORMULATION INTERVIEW (CFI) INTAKE ADDENDUM Client: _______________ DOB: ____________ Date: ____________ Clinician / credentials: ______________________________________ Part of (check): [ ] Intake / biopsychosocial assessment [ ] Psychiatric diagnostic evaluation (90791) Interview language: ____________ Interpreter: [ ] None needed [ ] Used (name/service): _____________ [ ] Offered, declined Components administered: [ ] Core CFI [ ] Informant Version (informant + relationship: _________) [ ] Module(s): _______ CULTURAL DEFINITION OF THE PROBLEM Problem in the client's own words; how they describe it to family or community: _______________________________________________________________ CAUSES AND EXPLANATORY MODEL What the client believes causes it; what family or community believe; where views differ: _______________________________________________________________ STRESSORS AND SUPPORTS What makes it worse; what helps (relationships, work, housing, migration, faith, discrimination experiences): _______________________________________________________________ ROLE OF CULTURAL IDENTITY Identity as the client describes it; which aspects the client ties to the problem, and how: _______________________________________________________________ SELF-COPING AND PAST HELP SEEKING Own strategies, family or community remedies, healers, prior professional care; what helped, what did not: _______________________________________________________________ BARRIERS AND PREFERENCES FOR CURRENT CARE What has kept the client from care; the help they want now: _______________________________________________________________ CLINICIAN-CLIENT RELATIONSHIP Client concerns about being understood; what would help the working relationship; communication arrangements: _______________________________________________________________ INTEGRATION WITH FORMULATION AND PLAN How the findings above change the differential, engagement, and the treatment plan in the parent evaluation: _______________________________________________________________ Clinician signature / credentials: ____________ Date: ________ Supervisor co-signature (if required): ________ Date: ________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample CFI intake note Scenario: a 38-year-old engineer, referred by his primary care physician for low mood, whose clinician added the core CFI to the diagnostic evaluation. The cultural formulation interview example below shows the documented addendum, condensed but structurally complete. All details are fictional. Client: A.T., 38 · Date: 07/14/2026 · Clinician: M. Okafor, PsyD · Context: CFI addendum to psychiatric diagnostic evaluation (90791) · Interview language: English, no interpreter needed · Components: Core CFI Cultural definition of the problem: Client calls the problem "burnout" and says he is "running on empty." With his parents he frames it as work stress; he does not use mental health language with them, which he says would not be understood as a health matter in his family. Causes and explanatory model: Client attributes the problem to two years of long hours combined with obligations he holds as the eldest son, including financial and caregiving support to his parents. He has read about depression and suspects it applies. He expects his parents would read the same symptoms as a lapse in discipline, which he identifies as the main reason he delayed seeking care. No somatic or spiritual attribution offered. Stressors and supports: Stressors: father's stroke last year with client coordinating his care; mortgage; a project schedule he describes as unforgiving. Supports: spouse, who encouraged this appointment and knows he is here; younger sister, the one family member he confides in; the temple community his parents belong to, where he attends on holidays and finds the atmosphere calming though not central to his life. Role of cultural identity: Client identifies as a second-generation Vietnamese American, English-dominant with conversational Vietnamese, and moves comfortably between settings. Asked which aspects of identity bear on the problem, he named the duty expectations attached to being the eldest son and family views of mental health care, and said language and religion do not. Self-coping and past help seeking: Has managed by cutting sleep to keep commitments, an exercise routine now lapsed, and herbal sleep remedies his mother provides, which he continues out of respect and finds mildly helpful. One employee-assistance session three years ago felt generic and he did not return. No prior psychiatric treatment. Barriers and preferences for current care: Barriers: family stigma as above, schedule, and a concern that counseling records could reach his employer, addressed today with a plain-language review of confidentiality and its limits. Preferences: a practical, skills-focused, time-limited approach; open to weekly therapy; hesitant about medication, citing his father's view of it as weakness and his own side-effect concerns. Clinician-client relationship: Client asked whether the clinician would treat his family's expectations as something to argue him out of. Clinician reflected the obligations as both a value and a load; client agreed with that framing and said it fit. No communication barriers identified; client comfortable proceeding with this clinician. Integration with formulation and plan: Findings are consistent with the major depressive disorder, moderate, documented with full history, mental status exam, and risk assessment in the diagnostic evaluation of today's date. The explanatory model (overload plus obligation, not illness) and stated preferences support cognitive behavioral therapy with behavioral activation planned around, not against, his family responsibilities. Medication education provided in neutral terms; decision deferred to the four-week review by agreement. Spouse to join a session with client's consent. Herbal remedies documented; no interaction concerns at present. Weekly sessions scheduled. M. Okafor, PsyD, signed 07/14/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The client's frame survives on the page. The problem is documented in his words, with the clinical label living where it belongs, in the diagnostic evaluation's assessment, so the record shows both how he understands it and how the clinician does. The explanatory model does work. His account of cause shapes the therapy choice, the pacing of the medication conversation, and the engagement plan; it is used clinically, not filed as background. Identity content is person-specific. Salience comes from his answers, including what he says does not matter (language, religion), so nothing rests on assumptions from group membership. The addendum stays tied to the billable evaluation. History, mental status exam, risk assessment, diagnosis, and plan sit in the 90791 note, referenced by date, so the cultural section enriches the evaluation without displacing any element payers require. Access and preference details are actionable. Interview language and interpreter status are recorded, the confidentiality concern was answered the same day, and each stated preference is either honored in the plan or given a documented revisit point. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations CFI content is ordinary clinical-record content: it belongs to the designated record set, clients can access it, and in US practice they increasingly read it through patient portals, so write the cultural sections in language you would stand behind with the client reading over your shoulder, anchored in their own words. It is not protected the way psychotherapy notes are, and it follows the retention rules of the evaluation it sits in. On retention, the "HIPAA 6-year rule" is a durable myth: 45 CFR 164.316(b)(2) requires keeping HIPAA's own compliance documentation (policies, procedures, required records of actions) for 6 years; clinical records follow state law, commonly 5 to 10 years and longer for minors. Ontario's psychology college standard keeps client records at least 10 years past the client's 18th birthday or last contact, and Australia's Psychology Board expects 7 years from the last entry, or until age 25 for child clients. Pre-licensed clinicians and trainees administering the CFI document under their supervisor's co-signature rules like any other intake content. Around the optional interview sits a short ladder of obligations that are not optional, and conflating the levels is the compliance mistake that matters. LAW: Title VI and Section 1557 require meaningful language access for people with limited English proficiency ( 45 CFR 92.201 ), which is why the interpreter line in this template is not decorative. ACCREDITATION: Joint Commission standards introduced in 2011 have hospitals identify each patient's preferred language and record race and ethnicity in the medical record ( R3 Report, Issue 1 ). GUIDANCE: the HHS Office of Minority Health's National CLAS Standards , 15 action steps HHS describes as "a blueprint," operate at the organization level. CONVENTION: the CFI itself. Payer requirements run the same direction: the Medicare LCD for psychiatric diagnostic evaluations ( L33252 ) expects a complete history including family and social history, a full mental status exam, a working diagnosis, and the client's ability and willingness to engage in the proposed treatment plan, and says nothing about cultural formulation. The intake elements are the requirement; the CFI is the craft you layer onto them, inside the intake note or biopsychosocial assessment . ## Common CFI intake errors auditors flag No behavioral health audit we could locate has ever faulted a clinician for the cultural content of an intake, and the word cultural does not appear once in the OIG's audit of first-pandemic-year psychotherapy ( A-09-21-03021 , May 2023). What that audit did find: of roughly $1 billion Medicare paid for psychotherapy, an estimated $580 million was improper, with providers failing requirements on 128 of 216 sampled enrollee days, for reasons like undocumented time, incomplete or missing treatment plans, and missing signatures. That is the risk map around a CFI intake: the danger is never the cultural formulation, it is what the cultural formulation displaces or decorates. The BastionGPT Clinical Advisory Board sees the same errors most often in CFI intake reviews: Demographic labels instead of formulation. Race, religion, and language recorded as data points that never touch diagnosis or plan. That is the exact failure mode the CFI was built to fix in its DSM-IV predecessor, and it survives the upgrade when the note stops at labels. Stereotype instead of inquiry. Group-level generalizations the client never said, written as findings. The CFI is person-specific by design: document what this client answered, including the aspects of identity they told you do not matter. A transcript instead of a synthesis. Sixteen question-and-answer pairs pasted into the record. The APA lets you reproduce the interview for clinical use, but the note auditors and colleagues need is the synthesis: what was learned, organized by domain, connected to the formulation. The displaced evaluation. A rich cultural section sitting where the history, mental status exam, risk assessment, or treatment plan should be. Those are the elements the Medicare LCD names and the deficiency categories that drive recoupments; a CFI cannot substitute for any of them. Unrecorded language services. An interview conducted through an interpreter, or in the client's second language, with no line saying so. Language access is the one culture-adjacent obligation with legal force, and the record is where you prove it happened. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on cultural formulation interview intakes. Draft the CFI addendum from your dictation, bullets, or session transcript, organized by the four domains with the client's own phrasing preserved where it carries meaning. Extract the cultural content scattered through a long intake conversation and place it in the right sections: definition of the problem, explanatory model, supports, identity, help seeking, preferences. Check the full evaluation before you sign: required intake elements present alongside the cultural section, interview language and interpreter status recorded, every stated preference either reflected in the plan or explicitly deferred, and a plain-language client summary on request. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a cultural formulation interview required at intake? No, anywhere. No US, Canadian, or Australian statute, payer policy, or accreditation standard requires a CFI or a documented cultural formulation, and no billing code attaches to it. The APA offers it as a Section III measure to support clinical judgment, "not as the sole basis for making a clinical diagnosis." The obligations that do bind sit next to it: language access for limited-English-proficient clients is federal law ( 45 CFR 92.201 ), accredited hospitals record race, ethnicity, and preferred language, and Medicare's intake documentation requirements govern the evaluation the CFI sits inside. Administering it is a clinical-quality choice, not a compliance box. How long does a CFI take, and how long is the note? "With practice, the CFI takes approximately 20 minutes to complete," per the developers' 2020 update in Focus, and coded sessions in the DSM-5 field trial averaged 24.56 minutes. Speed comes quickly: in the field trial , clinician feasibility ratings improved significantly after a single administration and subsequent interviews took less time. The documented section usually runs 200 to 500 words inside the parent evaluation. No regulation sets a length in any of the three countries. Do I have to ask all 16 CFI questions word-for-word? No. The CFI is a semi-structured interview: the published instrument pairs the questions with an italicized interviewer guide, and the peer-reviewed literature consistently describes it as semi-structured. Ask in your own words, follow up where the answers lead, and use the components that fit the client, including the Informant Version when family can add context. No DSM instruction or payer rule requires verbatim administration, completion in a single sitting, or repeating the interview at intervals. What earns its place in the record is documenting which components you used and in what language, then synthesizing the answers rather than transcribing them. Does Medicare or private insurance require a CFI to bill 90791? No. The Medicare LCD for psychiatric diagnostic evaluations ( L33252 ) expects a complete history including family and social history, a full mental status exam, a working diagnosis, and an evaluation of the client's ability and willingness to participate in the proposed treatment plan. Cultural formulation appears nowhere in the LCD or its billing article, so the CFI can only add to a psychiatric diagnostic evaluation , never substitute for a required element. One adjacent code is worth knowing: the interactive complexity add-on, 90785, covers specified communication factors that complicate delivery, and the LCD's examples include evaluations mediated by an interpreter. Should the CFI be its own note or part of the intake? There is no mandated location, and the CFI's developers left the placement question open, so practice convention decides. Most clinicians document it as a labeled section or addendum of the intake note , biopsychosocial assessment , or diagnostic evaluation, which keeps the cultural findings next to the formulation they inform. Wherever it lives, it is ordinary clinical-record content in the designated record set, readable by the client and reachable by the treatment team; it does not qualify for psychotherapy-notes protection. Pick one placement for your practice and keep it consistent, so reviewers always know where to find it. Can I copy the CFI into my EHR or intake forms? Yes, for patient care. The APA's posted instrument grants the right on its face: it can be "reproduced without permission by researchers and by clinicians for use with their patients," and the current DSM-5-TR edition is free to download from psychiatry.org , alongside the Informant Version and the supplementary modules. Any use beyond that, republishing, training products, commercial distribution, needs APA permission. So build the official wording into your EHR template if you like, but keep the note itself a synthesis of what the client said; a pasted worksheet is the transcript error auditors and supervisors flag. What is the difference between a CFI intake and a biopsychosocial assessment? Breadth versus lens. The biopsychosocial assessment is the wide container: biological, psychological, and social domains covered systematically to open an episode of care. The CFI is a focused instrument that goes deep on one axis, the client's own understanding: how they define the problem, what they believe causes it, which parts of identity matter, how they have coped and sought help, and what care they want. In practice the CFI is the strongest way to fill the social and cultural sections a biopsychosocial already contains, and its findings should flow into the same formulation, not run parallel to it. The intake note relates the same way, at lower depth. Do clinicians in Canada and Australia use the CFI? Yes, and the evidence base is partly Canadian: Canada was a field-trial country, and the developers' update lists Canada among the places the CFI has been evaluated. In neither country is it mandated. Canada's Truth and Reconciliation Commission Calls to Action 23 and 24 direct governments and medical and nursing schools to provide cultural competency training, obligations aimed at institutions rather than individual chart documentation, and Ontario's psychology college expects client records kept at least 10 years past age 18 or last contact. In Australia, a July 2026 search of the MBS for cultural formulation returns no items; the funded planning documents are the GP Mental Health Treatment Plan items (2700, 2701, 2715, 2717), and hospital accreditation carries separate cultural-safety and identity-recording actions under the NSQHS Standards. Australian records are kept 7 years, or to age 25 for child clients. Can BastionGPT draft a CFI intake note from a session transcript? Yes. Dictate after the session, paste a transcript, or give it bullets, and it drafts the addendum organized by the four CFI domains, preserves the client's own phrasing where it carries clinical meaning, writes the integration paragraph that connects the findings to formulation and plan, and flags anything the parent evaluation still needs before you sign, such as an undocumented interpreter or a preference the plan never addresses. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Intake Note Biopsychosocial Assessment Psychiatric Diagnostic Evaluation Mental Status Exam (MSE) Note Substance Use Assessment Child & Adolescent Developmental Assessment Aboriginal & Torres Strait Islander Culturally Safe Assessment Indigenous & First Nations Wellness Assessment Addendum Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/violence-risk-assessment All note templates # Violence & Homicide Risk Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A violence and homicide risk assessment is the clinical record of how a clinician evaluated a client's risk of harm to others: the threat and its context, target identifiability, intent and access to means, risk and protective factors, a formulation, and the protective actions taken or reasonably ruled out. Clinicians document one when a client threatens an identifiable person or homicidal ideation emerges. Most run 300 to 800 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, psychiatric prescribers, forensic and emergency clinicians Audience Treating team, supervisors, crisis and emergency services, auditors, licensing boards, attorneys and courts Typical length 300 to 800 words · 15 to 30 minutes by hand (clinical team estimate) Format family Risk assessment documentation (compare: suicide risk assessment, safety plan, crisis note, mandated report) When it's used When a client threatens an identifiable person, when homicidal ideation or other violence indicators emerge, and at intake, discharge, and forensic evaluations Standards context No mandated note format; the duty to protect it evidences is state law that varies from mandatory to none, so the documented analysis carries the weight On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a violence and homicide risk assessment? A violence and homicide risk assessment is the clinical record of how a clinician evaluated a client's risk of harm to another person: the threat or indicator that triggered the evaluation, the inquiry into intent, target, and access, the risk and protective factors weighed, the risk formulation, and the protective decisions made, including a reasoned decision that no protective disclosure was indicated. No professional body, statute, or payer has ever standardized this note as a document format. What has been standardized are two separate things the note draws on: the assessment instruments, such as the structured professional judgment tool now in its third version, the Historical, Clinical, Risk Management-20 ( HCR-20 V3 ), and actuarial scales such as the Violence Risk Appraisal Guide ( VRAG ); and the legal duty the note evidences, which descends from Tarasoff v. Regents of the University of California (17 Cal.3d 425, 1976), where Justice Tobriner wrote that "The protective privilege ends where the public peril begins" ( opinion text ). Clinicians and charts also say Tarasoff assessment, duty-to-warn or duty-to-protect note, danger-to-others (DTO) assessment, and HI assessment, charting shorthand for homicidal ideation that parallels SI. The distinction that carries the most weight is warn versus protect. California reframed the obligation as a duty to protect, discharged by "reasonable efforts to communicate the threat" to both the victim and a law enforcement agency, and the statute's own 2013 amendment text states that it only changed the name of the duty ( Civil Code 43.92 ): warning is one way to discharge the duty, alongside hospitalization, means restriction, and intensified treatment. The duty itself is state law that genuinely varies, from mandatory through permissive to none at all, so the note's job is to show which standard applied and how the clinician's analysis tracked it. The assessment note records that evaluation and reasoning; when a report or warning is actually made, the notification record is its own discipline, covered on the mandated reporting and duty-to-protect documentation page. Risk a client directs at themselves belongs in a suicide risk assessment , which shares the same documentation logic with a different subject and no third-party duty. ## Who uses violence and homicide risk assessments and when Every licensed mental health discipline writes them when the trigger arrives: therapists, psychologists, counselors, and clinical social workers in outpatient practice, psychiatric prescribers, emergency and crisis teams, and forensic clinicians for whom structured violence risk evaluation is routine work. In outpatient care the trigger is usually a statement in session, a disclosure of homicidal ideation, or a collateral warning from a family member, which counts as a treatment communication in California under Ewing v. Goldstein ( 2004 ); in emergency and inpatient settings the assessment anchors admission, discharge, and disposition decisions. The epidemiology argues for taking the outpatient case seriously: in the landmark MacArthur follow-up of discharged psychiatric patients, violence was "most frequently targeted at family members and friends" and most often took place at home, with substance use symptoms raising rates sharply ( Steadman et al., 1998 ), which means the identifiable-victim analysis is rarely hypothetical. This page covers the evaluation record; when the encounter itself is an emergent, time-based service, the crisis note carries the billing, a psychiatric diagnostic evaluation carries an intake, and the assessment reasoning lives inside or alongside them. ## Violence risk assessment structure: what goes in each section No statute prescribes these headings. They mirror what a later reader, a supervisor, a licensing board, or a court reconstructing your judgment, will look for: what was said, how identifiable the person at risk was, how the clinician weighed it against the jurisdiction's standard, and what was done or deliberately not done. The through-line is that the formulation and the duty analysis must explain the actions. Reason for assessment and context. What triggered the evaluation (a statement in session, disclosed homicidal ideation, a collateral report, a referral question) and the clinical context around it: precipitants, symptom course, substance use, current stressors. Pitfall: recording the trigger without the context. A threat statement stripped of its precipitant and course reads more alarming or less alarming than the clinician's actual judgment, and the reader cannot tell which. Threat inquiry. What the client actually communicated, in their own terms: the statement and its context, whether it was conditional or unconditional, when and to whom it was made, and whether it has been repeated or escalated. Pitfall: paraphrase so loose the threshold cannot be reconstructed. "Made threatening statements" supports nothing; record the substance of what was said and the questions you asked about it. Target and identifiability. Who the concern involves, how identifiable they are (a named person, a reasonably ascertainable person or group), the relationship, and current proximity or contact. Pitfall: skipping the identifiability analysis. Most duty-to-protect standards turn on a reasonably identifiable victim, and a note that never says whether one exists has skipped the pivotal legal fact. Intent, capability, and access to means. Stated intent as distinct from anger or ideation, capability, and access to lethal means asked about directly, with any restriction steps agreed and verified. Pitfall: conclusions with no conversation shown. "No access to means" reads as unsupported unless the note records what was asked, what was answered, and what was confirmed. History. Prior violence with recency, severity, and context, legal history, protective orders, and the sources consulted: the client's account, available records, collateral information. Pitfall: history from self-report alone, silently. Note what was and was not available to you; a reviewer treats "no history reported" and "records reviewed, no history found" very differently. Risk and protective factors. Acute and stable risk factors, and protective factors that are specific and current for this client, not a generic list. Pitfall: boilerplate copied forward unchanged between notes. Reviewers read identical factor lists as evidence the assessment was not individualized, and boilerplate cuts both ways in a courtroom. Risk formulation. The clinical synthesis: a risk characterization (many settings use low, moderate, or high, for acute and longer-term risk) with the rationale connecting the findings above to it. Any instrument used (HCR-20 V3, VRAG) is an input with its result and date, not the formulation itself. Pitfall: a score or a level standing in for reasoning. Instrument discrimination is modest at the individual level, with a 2025 systematic review reporting median AUCs of 0.68 to 0.71 across four established tools, so the defensible note shows judgment, never prediction language like "client will not act". Duty analysis, consultation, and decision. Which jurisdiction's standard applies, whether its threshold is met on today's facts, consultation obtained (supervisor, colleague, attorney, risk-management line) with who and when, and the decision reached, expressly including a reasoned decision that no disclosure is indicated. Pitfall: the analysis that only exists when the answer is yes. A decision not to warn that is never documented looks, years later, like a question never asked. Actions, disposition, and reassessment. Protective actions taken with dates, times, and content (any notification made, hospitalization, means restriction, treatment intensification), the disposition, the dated next contact, and when risk will be reassessed. Pitfall: "will monitor" with no date. Elevated risk documented once and never revisited reads as abandoned, not managed. ### Blank template (copy and adapt) Client: [initials] Date: Setting/context: Reason for assessment (statement, ideation, collateral report, referral): Threat inquiry (what was communicated, context, conditional/unconditional, when, to whom, repeated or escalated): Target & identifiability (named/ascertainable, relationship, contact): Intent, capability & access to means (asked directly; steps agreed/verified): History (prior violence, legal history, protective orders; sources consulted): Risk factors (acute / stable): Protective factors (specific to this client, today): Instruments & results (HCR-20 V3, VRAG), with dates, as inputs: Risk formulation (level + rationale connecting findings to decisions): Duty analysis & consultation (standard applied, threshold met or not, who was consulted, decision incl. reasoned decision not to disclose): Actions & disposition (dated/timed notifications, means steps, level of care): Follow-up & reassessment plan (dated next contact, triggers for re-review): Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample violence risk assessment Scenario: adult client in weekly outpatient therapy after a job loss makes an angry statement in session about a former supervisor, and the clinician completes and documents a full risk assessment, consultation, and a reasoned decision that no protective disclosure is indicated. All details are fictional, and the risk content is deliberately generalized: it shows what a defensible note records, not how risk presents. Client: D.R., 41 · Date: 07/21/2026 · Setting: Outpatient office, scheduled session · Reason for assessment: Statement about a former supervisor made in session Context and precipitant: Client is in week 7 of weekly therapy for an adjustment disorder with depressed mood following termination from a nine-year position. Client reports escalating rumination about the termination over the past two weeks, disrupted sleep, and drinking above baseline on several evenings. A wrongful-termination claim was declined by an attorney last week, which client describes as "the door closing." Threat inquiry: While describing the termination, client made a statement to the effect that the former supervisor should be afraid of what client might do. Asked directly about the statement, client described it as venting, denied wanting to harm the supervisor, and denied telling anyone else anything similar. Client reports recurrent angry thoughts about the supervisor over two weeks, brief and intrusive, without rehearsal or imagery of acting. The statement today was conditional and made in frustration; client withdrew it when asked to elaborate. No prior statements of this kind in treatment. Target and identifiability: The person referenced is a specific named individual known to the client through the former workplace. Client has had no contact since the termination, has no reason to be at the former workplace, and reports no current proximity. Because the person is reasonably identifiable, the duty threshold analysis below was completed and documented. Intent, capability, and access: Client denies intent to harm, denies any plan, and denies preparatory or rehearsal behavior. Access to means was reviewed directly with the client. One category of concern was identified, and a specific restriction step was agreed with the client and the client's spouse, with verification arranged and recorded in the plan below. Client engaged in the conversation without resistance. History: No adult history of violence toward persons. One verbal workplace altercation approximately six years ago without physical contact or disciplinary action, per client report. No arrests, no protective orders, no weapons-related incidents reported. Records available to this practice contain no violence history; no collateral records were available at the time of assessment, which is noted as a limit of today's data. Risk factors: Acute: humiliation-framed loss with a fresh legal setback, escalating rumination, disrupted sleep, alcohol use above baseline, recurrent angry ideation about a specific person. Stable: none identified; no violence history, no prior psychiatric hospitalization. Protective factors: Engaged in weekly treatment and disclosed spontaneously in session; married with a supportive spouse who is aware of the job loss; financially stable for the near term; active job search with two interviews scheduled; states the supervisor "is not worth my life" and connects acting on anger with losing their family and the career they are rebuilding; no intoxication in session. Risk formulation: Conditional angry statement about an identifiable person, without stated intent, plan, preparatory behavior, or history of violence, in the context of an acute loss, rumination, and increased alcohol use. Acute risk of harm to the named individual is judged low to moderate and manageable in outpatient care with the steps below; longer-term risk is low. Rationale: the statement was conditional and withdrawn on direct inquiry, intent and planning are denied with a consistent account, means access is being addressed with verification, protective factors are current and specific, and the client engaged openly in the assessment. Foreseeable destabilizers, contact initiated with the former workplace, an escalation in drinking, or a further legal setback, were named with the client and are covered by the reassessment plan. Duty analysis, consultation, and decision: This state's duty-to-protect standard requires a serious threat of physical violence against a reasonably identifiable victim. On today's facts the threshold is not met: the person is identifiable, but the statement was conditional and withdrawn, intent is denied, and no plan or preparatory behavior is present. Case reviewed same day with clinical supervisor, J.M., PhD, and the practice's risk-management consultation line; both concurred that no notification is indicated on the present facts. Decision: no protective disclosure at this time; intensify treatment and reassess. If intent, a plan, or an unconditional threat emerges, the duty analysis will be repeated immediately and the discharge options under this state's statute will be followed, with the analysis documented. Actions, disposition, and follow-up: Sessions increased to twice weekly for two weeks; next appointment 07/24/2026, phone check-in 07/22/2026 at 4:30 pm. Means-restriction step to be verified by spouse call on 07/22/2026, with client's written consent obtained in session. Coping plan for rumination and anger arousal reviewed; client agreed to hold alcohol at or below baseline while symptoms are acute and to review this next session. Client verbalized agreement to contact the practice or crisis services if thoughts of harming anyone intensify. Reassessment plan: Threat content, intent, means status, and alcohol use reviewed at every contact while active; full reassessment at the next session or immediately upon any change, including contact with the former workplace, an intoxicated presentation, or any new statement about the named individual. This sample is fictional and for educational purposes. It does not describe a real patient, and it intentionally omits the clinical specifics a real record would contain. ↑ Back to the template and downloads ### Why this sample works The inquiry is shown, not summarized: what was said, what was asked, what the client endorsed and denied, and how the statement changed on direct questioning, so the formulation sits on visible facts. Identifiability gets its own analysis: the note states that the person is reasonably identifiable and shows that the duty threshold was therefore examined, which is the pivotal legal fact most notes skip. The formulation states a risk level with the rationale connecting the findings to it, in judgment language rather than prediction language. The decision not to warn is documented as a decision: the standard applied, why the threshold is not met on today's facts, and the conditions that would reopen the analysis. Consultation is recorded with who, when, and outcome, turning reasonable professional judgment into visible evidence. Every action is dated and verifiable: the means-restriction step with consent and a verification call, the increased session frequency, the phone check-in, and a reassessment plan with named triggers. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write this note knowing the standard it will be held to: not whether the outcome was predicted, but whether a reasonable assessment process ran and is visible on the page. The confidence gap in the field is documented: in a survey reported in the American Psychological Association's duty-to-protect volume, 90 percent of psychologists were confident in their knowledge of their legal responsibilities with dangerous clients, while "as many as 75 percent of the sample were misinformed" about their legal duties ( APA, The Duty to Protect ), which is exactly why the note should name the standard applied rather than assume it. The note sits in the ordinary record: the client generally holds access rights to it, it is reachable by subpoena, and the separately kept psychotherapy notes category never covers it (see the psychotherapy notes authorization page). Under the information-blocking rules, withholding it from an access request requires the narrow preventing-harm exception, a reasonable belief that release would "substantially reduce a risk of harm" being prevented, judged case by case ( 45 CFR 171.201 ), so write facts, reasoning, and actions rather than speculation. After a bad outcome, never revise the original entry; anything learned later goes in a clearly dated late entry or addendum. In telehealth, the prevailing compliance guidance is to analyze the duty under the law of the state where the client is located during the session, which can change the standard mid-caseload. When a threat situation, a warning, or any legal process touches your documentation, consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). Keep the legal layers straight, because they carry different weight. The duty is LAW that varies by state: a 2022 National Conference of State Legislatures review classifies roughly 29 jurisdictions as imposing a mandatory duty, 19 as permissive, and 3 as recognizing no duty ( NCSL ), and the details diverge sharply: California discharges the duty by "reasonable efforts to communicate the threat" to both the victim and law enforcement ( Civil Code 43.92(b) ), Texas recognizes no duty and makes protective disclosure permissive, to medical, mental health, or law enforcement personnel only, never directly to the victim ( Health and Safety Code 611.004 ), Florida's statute reaches psychiatrists specifically and requires the threat be communicated to a law enforcement agency, which then acts ( 456.059 ), and New York's SAFE Act pathway is a separate mandatory report to the director of community services, not to the victim ( Mental Hygiene Law 9.46 ). Federal privacy law is PERMISSIVE, never a duty: HIPAA lets a covered entity disclose in good faith to "prevent or lessen a serious and imminent threat", including to the target, and presumes the good faith ( 45 CFR 164.512(j) ). Two consequences are worth documenting awareness of: in California a qualifying report to law enforcement triggers a five-year firearm prohibition for the client ( Welfare and Institutions Code 8100(b)(1) ), and in Canada the Supreme Court's public safety exception in Smith v. Jones permits disclosure on clarity, seriousness, and imminence grounds while leaving open whether it is ever required ( [1999] 1 S.C.R. 455 ); Australia's APS Code likewise permits disclosure for "an immediate and specified risk of harm to an identifiable person" ( A.5.2(c) ) alongside the Privacy Act's serious-threat permission (s 16A). The note itself is CONVENTION everywhere, which is why it matters: it is the only evidence you control that the analysis ran and the duty, where one existed, was met. On the payer side there is no violence-note code to defend; the assessment is reviewed inside whatever service carried it, a crisis encounter or an evaluation, and the action side of a report or warning has its own documentation discipline on the mandated reporting and duty-to-protect page, with client-facing stabilization work in a safety plan . ↑ Back to the template and downloads ## Common violence risk assessment errors reviewers flag The audit numbers around this document are generic to psychotherapy records, and they are bad enough: a 2020 HHS Office of Inspector General audit of a psychotherapy practice found 111 of 120 sampled claims non-compliant and estimated at least $3.3 million in Medicare overpayments, with treatment-plan documentation deficiencies on 111 claims and notes signed with digital images of signatures on 109 ( A-02-19-01012 ). The legal exposure is sharper here than for any other note type because most clinicians misjudge their own state's rules: the confidence-versus-accuracy gap reported in APA's duty-to-protect volume ran 90 percent confident against roughly 75 percent misinformed ( APA ). The BastionGPT Clinical Advisory Board sees the same errors most often in violence risk assessment reviews: "Denies HI" as the entire assessment. A denial is one finding, not a formulation. It says nothing about what was asked, the statement that triggered the assessment, target identifiability, means access, or why the clinician credited the denial, and it is indefensible standing alone after a threat statement or a collateral warning. No identifiability analysis. The note records a threat and a response but never says whether a reasonably identifiable victim exists, which is the fact most duty-to-protect standards turn on. A reviewer reconstructing the threshold analysis finds the pivotal variable missing. A risk label with no reasoning, or prediction language. "Low risk" without findings creates false precision, and "client will not act" promises what no assessment can deliver: a 2025 systematic review put median instrument AUCs at 0.68 to 0.71 ( Swiss Medical Weekly ). Pair any level with an individualized rationale, and let an instrument result be an input, never the formulation. The analysis that only exists when the answer is yes. A warning is documented with no threshold reasoning behind it, or a decision not to warn is never written down at all, so the chart is silent exactly where judgment was exercised. Document the standard applied, the facts against it, the consultation obtained with who and when, and the decision, in both directions. Template text that contradicts the narrative. The narrative records an angry statement about a named person while an auto-populated line says "denies HI", or the history names an altercation the checklist denies. Reviewers treat an internally contradictory record as unreliable everywhere, so reconcile the fields before signing. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on violence and homicide risk assessments. Draft a structured assessment from your bullets or dictation, with the threat inquiry, identifiability analysis, risk and protective factors, formulation, duty analysis, and actions in the right places for your review. Check a finished note before you sign: flag a formulation with no rationale, a missing identifiability analysis, an undocumented means conversation, a consultation left unrecorded, or actions with no dated follow-up. Summarize risk documentation across sessions for supervision, consultation, or a records request, and keep the summary to what the receiving reader should see. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a violence risk assessment legally required, and is there a required format? No statute in the US, Canada, or Australia prescribes a violence risk assessment template, and no professional body has standardized the note as a document format. The obligations are layered instead: the duty to protect that the note evidences is state law that varies from mandatory to none, licensing boards require adequate records everywhere, and the clinical standard of care fills the rest. Most run 300 to 800 words, and the test any format is held to is whether the formulation and the duty analysis explain the actions taken, or the reasoned decision not to act. Do all 50 states require a Tarasoff warning? No, and the widely repeated claim that they do is false. A 2022 National Conference of State Legislatures review classifies roughly 29 jurisdictions as imposing a mandatory duty, 19 as permissive, and 3 as recognizing no duty, and published counts differ by methodology. Texas is the marquee counterexample: its supreme court declined to adopt Tarasoff in 1999, and its statute makes disclosure permissive, to medical, mental health, or law enforcement personnel only, never directly to the victim ( Health and Safety Code 611.004 ). Read your own state's current statute rather than a national summary, and name the standard you applied in the note. What is the difference between the duty to warn and the duty to protect? Warning is one option; protecting is the duty. California's statute is the clearest illustration: it imposes a duty to protect, discharged by "reasonable efforts to communicate the threat" to both the victim and a law enforcement agency, and the 2013 amendment's own text says it only changed the name of the duty ( Civil Code 43.92 ). Depending on the jurisdiction, protective options can include hospitalization, means restriction, intensified treatment, and involving supports, which is why the note documents the protective reasoning, not just whether a phone call was made. How do I document a decision not to warn? As a decision, not an omission. Record the standard that applies in your jurisdiction, the facts weighed against its threshold (identifiability of the person at risk, the seriousness and conditionality of what was communicated, intent, means, history), the consultation obtained with who, when, and outcome, the conclusion that the threshold was not met, and the specific changes that would reopen the analysis. The sample above shows the pattern. Documented reasoning matters equally in both directions: years later, an undocumented negative decision looks like a question that was never asked. Do I need the HCR-20 or another instrument? No regulation reviewed for this page mandates a violence risk instrument in any of the three countries. Structured professional judgment tools such as the HCR-20 V3 organize the factor review, and actuarial scales such as the VRAG estimate group-level recidivism probabilities, but discrimination is modest: a 2025 systematic review found median AUCs of 0.68 to 0.71 across four established instruments, with predictive value hinging on local base rates. Your employer or setting can require a named tool by policy. Either way, record the result as an input with its date, and let the formulation carry the reasoning. What should the note record when I do make a warning or notification? First the duty analysis that led there, then the mechanics: who was contacted, when, by what means, the substance of what was communicated, and any report or reference number law enforcement provided. Your state's discharge formula controls the recipients: California expects efforts to reach both the victim and law enforcement, Florida channels the required notification to a law enforcement agency, which then acts, and New York's separate SAFE Act pathway runs to the director of community services. Enter it the same day, contemporaneously. The notification record itself has a full walkthrough on the mandated reporting and duty-to-protect documentation page. How is this different from a mandated report, a safety plan, or a crisis note? The violence risk assessment is the evaluation record: the inquiry, the factors, the formulation, and the duty analysis. A mandated report documents a statutory report, child or elder abuse reporting with its own triggers, recipients, and deadlines, and covers the action side when a duty-to-protect notification is made. A safety plan is a client-facing stabilization intervention, not an assessment. A crisis note records an emergent, time-based encounter, which may contain risk assessment reasoning inside it. Risk a client directs at themselves belongs in a suicide risk assessment , the closest sibling of this document. Can the client see this note? Generally yes. It lives in the ordinary record, where clients hold access rights under HIPAA and the information-blocking rules, and the separately kept psychotherapy notes category never covers it. Withholding it from an access request requires the preventing-harm exception, which demands a reasonable belief that withholding will "substantially reduce a risk of harm", applied case by case and narrowly ( 45 CFR 171.201 ). Write every line knowing the client, their attorney, and a board may read it: facts, reasoning, and actions, without speculation. Can BastionGPT draft a violence risk assessment? Yes. Give it bullets or a dictation from the session, and it drafts a structured assessment with the threat inquiry, identifiability analysis, risk and protective factors, formulation, duty analysis, and actions in the right places for your review. It can also check a finished note before you sign: a formulation with no rationale, a missing identifiability analysis, an undocumented consultation, or actions with no dated follow-up. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance claims on this page trace to these authorities, last verified July 2026: Tarasoff v. Regents of the University of California, 17 Cal.3d 425 (1976) : the duty-to-protect holding and the Tobriner opinion quoted above. California Civil Code 43.92 : the trigger, the discharge formula, and the 2013 amendment's name-change language. Ewing v. Goldstein (Cal. Ct. App. 2004) : family-member communications as patient communications under 43.92. California Welfare and Institutions Code 8100(b)(1) : the five-year firearm prohibition following a qualifying psychotherapist report. Texas Health and Safety Code 611.004 : permissive disclosure, authorized recipients, and the absence of victim notification, following Thapar v. Zezulka (Tex. 1999). Florida Statutes 456.059 : the psychiatrist-specific trigger and law-enforcement notification channel. New York Mental Hygiene Law 9.46 : the SAFE Act reporting pathway to the director of community services. 45 CFR 164.512(j) : the permissive serious-and-imminent-threat disclosure and its good-faith presumption. 45 CFR 171.201 : the information-blocking preventing-harm exception governing access requests. National Conference of State Legislatures, Mental Health Professionals' Duty to Warn (updated 2022) : the state-by-state mandatory, permissive, and no-duty classification. Smith v. Jones, [1999] 1 S.C.R. 455 : Canada's public safety exception and its clarity, seriousness, and imminence factors. APS Code of Ethics, A.5.2(c) and the OAIC APP 6 guidelines (s 16A) : Australia's permissive disclosure grounds. HHS Office of Inspector General, A-02-19-01012 (2020) : the psychotherapy documentation audit figures in the errors section. American Psychological Association, The Duty to Protect (front matter) : the confidence-versus-accuracy survey figures. Steadman et al., Archives of General Psychiatry (1998) and the 2025 Swiss Medical Weekly base-rates review : the epidemiology and instrument-performance evidence, with the HCR-20 V3 described at PMID 27422806 . ## Related templates Suicide Risk Assessment Mandated Reporting & Duty-to-Protect Documentation Safety Plan Crisis Psychotherapy Note Psychiatric Diagnostic Evaluation Self-Harm (NSSI) Assessment Fitness-for-Duty Evaluation Forensic Psychological Evaluation Report General Risk Screening Note PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/self-harm-assessment All note templates # Self-Harm (NSSI) Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A self-harm assessment, also called an NSSI assessment, is the clinical documentation of how a clinician evaluated non-suicidal self-injury: the behavior's history and severity, the suicide risk screen that separates it from suicidal self-injury, its function, and the risk formulation with the response. Clinicians document one when self-injury is disclosed or discovered. Most run 300 to 800 words, and no regulator prescribes a format. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, clinical social workers, school and college counselors, psychiatric prescribers, adolescent and crisis teams Audience Treating team, supervisors, crisis and emergency services, auditors, licensing boards, legal reviewers Typical length 300 to 800 words · 10 to 25 minutes by hand (clinical team estimate) Format family Risk assessment documentation (compare: suicide risk assessment, safety plan, crisis note) When it's used When self-injury is disclosed, observed, or reported, at intake when history is positive, after an episode, and on a set cadence while the behavior is active Standards context No mandated format; everything NSSI-specific is professional convention, and accreditor suicide-screening standards govern once suicide risk indicators appear On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a self-harm (NSSI) assessment? A self-harm assessment, in clinical shorthand an NSSI assessment, is the clinical record of how non-suicidal self-injury was evaluated: how the behavior came to light, its history and severity, the suicide risk screen and the intent determination that separate it from suicidal self-injury, the function the behavior serves, and the risk formulation with the response. The construct it documents has a clear lineage. Psychiatrist Armando Favazza built the foundational clinical account of self-injury in Bodies Under Siege (1987) and defined it as "the deliberate, nonsuicidal destruction of one's own body tissue" ( Favazza, 1998 ). DSM-5 placed nonsuicidal self-injury disorder in Section III in 2013 as a condition requiring further study rather than an official diagnosis ( Zetterqvist, 2015 ), and the VA built the Self-Directed Violence Classification System with the CDC to give clinicians one vocabulary for behavior that "may or may not reflect suicidal intent" ( VA Rocky Mountain MIRECC ). Clinicians and records also say self-injury, deliberate self-harm (common in the UK and Australia), and nonsuicidal self-directed violence (VA and DoD settings). The distinction that carries the compliance weight: nobody ever standardized the assessment as a document. The construct's authors defined the behavior; no professional body, payer, or regulator in the US, Canada, or Australia prescribes an NSSI assessment form, so the structure on this page is professional convention assembled from validated research interviews and inventories and from what reviewers reconstruct after an adverse event. Two documentation acts do the load-bearing work. First, the note shows the intent determination being assessed, never assumed: non-suicidal self-injury is defined by the absence of suicidal intent, and that absence is a clinical finding with a basis, not a default. Second, the note always contains a suicide risk screen, because the two problems co-occur: a history of self-injurious thoughts and behaviors is consistently cited as one of the strongest predictors of future suicidal behavior ( Ribeiro et al., 2016 ). When the screen turns positive, the document that governs is the suicide risk assessment ; this page covers the assessment written when the determination is non-suicidal and the reasoning that supports it. ## Who uses self-harm assessments and when Every discipline that treats the populations where the behavior concentrates: therapists, psychologists, counselors, and clinical social workers in outpatient and school-linked care, adolescent and college mental health teams, and inpatient and emergency clinicians when self-injury surfaces during other care. The population logic is stark: pooled lifetime prevalence runs 17.2 percent among adolescents and 13.4 percent among young adults against 5.5 percent in adults ( Swannell et al., 2014 ), so adolescent-serving clinicians write these assessments most. The trigger is usually a disclosure after direct inquiry, a screening result, a collateral report from a parent, school, or roommate, or findings noticed during medical care. The document also has clear neighbors: an acute, unstable presentation is documented as a crisis note with the risk reasoning inside it; a positive suicide screen converts the work into a suicide risk assessment ; and when self-injury emerges inside a broader history of adversity, the trauma assessment carries the developmental context while this document carries the behavior-specific evaluation. ## Self-harm assessment structure: what goes in each section No statute prescribes these headings. They mirror what a later reader, a supervisor, an auditor, a board, or a court, tries to reconstruct: how the behavior surfaced, how suicidal intent was ruled in or out, what the behavior does for the client, and what the clinician did about it. The through-line is that the intent determination and the formulation must both be shown, not asserted. Reason for assessment and context. What brought the behavior into view (a disclosure after direct inquiry, a screening result, a collateral report, findings noticed during medical care) and the clinical context around it: presenting problems, current stressors, substance use. Pitfall: an assessment with no visible trigger. A note that opens with findings and never says who noticed what, or why the assessment happened today, reads as template output rather than an evaluation. Self-injury history and characterization. Onset, most recent episode, a frequency band, the method category, whether the pattern is stable or changing, and medical severity: whether any episode needed medical attention. Record what care requires, at the level of detail care requires. Pitfall: minimizing shorthand. "Superficial" written as a conclusion, with no documented severity assessment behind it, reads to reviewers as dismissal rather than assessment, and it forecloses the question the note exists to answer. Suicide risk screen and intent determination. The load-bearing section. Direct inquiry about suicidal ideation, plan, intent, and behavior history, a validated screen where your setting uses one, the client's own account of what the behavior is and is not, and the stated basis for the determination that this is non-suicidal self-injury. Pitfall: intent inferred from the injury itself. The distinction between non-suicidal and suicidal self-injury is a finding the note must show being assessed; a determination assumed from how an injury looks is the least defensible sentence this document can carry. Function, antecedents, and triggers. What the behavior does for the client, in their words and yours: research reduces the many reported functions to two factors, intrapersonal (regulating internal states) and social ( Klonsky et al., 2015 ). Add the antecedents and settings in which urges rise, and anything that has interrupted the behavior before. Pitfall: function skipped. If the note never records what the behavior regulates, the treatment plan that follows cannot target it, and a reviewer reads "coping skills" as generic. Risk and protective factors. Acute and chronic risk factors, and protective factors that are specific and current for this client. Include the factor research puts at the center: the self-injury history itself elevates risk for later suicidal thoughts and behavior. Pitfall: protective factors copied forward unchanged session after session. Reviewers read that as boilerplate, and boilerplate cuts both ways in a later review. Risk formulation. The clinical synthesis: what the findings above add up to, why the chosen response fits them, and how the co-occurring suicide risk question stays under surveillance. Guidance genuinely splits on risk tiers, so lead with the narrative reasoning and add the overall risk level your setting requires. Pitfall: a tier with no reasoning, or prediction language ("client will not escalate"). The note documents a reasonable assessment process; it never promises an outcome. Interventions and response plan. What was done in this encounter: skills introduced and rehearsed, a written response plan for high-urge windows completed with the client (a close cousin of the safety plan ), access-reduction steps agreed, supports involved with consent recorded, crisis resources provided. Pitfall: interventions listed as intentions. "Will develop coping strategies" with nothing completed, dated, or handed to the client documents a plan to intervene, not an intervention. Diagnosis and coding linkage. The underlying diagnosis first; R45.88, nonsuicidal self-harm, added as a symptom code where your setting codes the behavior. NSSI is not an official DSM diagnosis, so the diagnostic weight rests on the underlying condition. Pitfall: R45.88 standing alone as the principal diagnosis. It sits in the symptoms chapter, and coding it in place of a definitive diagnosis undercuts both the claim and the record. Disposition, follow-up, and reassessment. The level of care decided, the dated next contact, escalation instructions given, any consultation obtained, and the cadence at which the behavior and the suicide screen will be reassessed while the behavior is active. Pitfall: "will monitor" with no date. An identified behavior with no reassessment plan reads as noticed and then abandoned. ### Blank template (copy and adapt) Client: [initials] Date: Setting/context: Reason for assessment (disclosure, screen, collateral report, medical finding): Self-injury history (onset; most recent episode; frequency band; method category; medical severity / care required; stable or changing): Suicide risk screen (ideation, plan, intent, behavior history; screen used) and intent determination for this behavior, with its basis: Function and antecedents (what the behavior regulates; triggers; what has interrupted it before): Risk factors (acute / chronic, incl. the self-injury history itself): Protective factors (specific to this client, today): Instruments and scores, if used (ISAS, SITBI, DSHI), with dates: Risk formulation (narrative reasoning; add the risk level your setting requires): Interventions and response plan (skills rehearsed; response plan completed, copy given; access-reduction steps agreed; supports involved w/ consent; crisis resources provided): Diagnosis and coding (underlying diagnosis; R45.88 as an added symptom code): Disposition and follow-up (level of care, dated next contact, escalation instructions): Consultation / supervision (who, when, outcome): Reassessment plan (behavior and suicide screen, at what cadence): Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample self-harm (NSSI) assessment Scenario: a 19-year-old university student in weekly outpatient therapy for anxiety discloses recurrent self-injury when her clinician asks directly, and the clinician completes and documents a structured assessment in the same session. All details are fictional, and the behavior content is deliberately generalized: the sample shows what a defensible record contains, not how self-injury presents. Client: E.J., 19 · Date: 07/22/2026 · Setting: Outpatient office, scheduled session · Reason for assessment: Self-injury disclosed in session after direct inquiry Context and precipitant: Client is in week 6 of weekly CBT for generalized anxiety disorder with depressive features. Routine review of coping since the last session included a direct question about self-harm; client disclosed recurrent self-injury without suicidal intent. Current stressors are academic pressure and a conflict with her roommate. No alcohol or drug involvement with the behavior reported. Self-injury history and characterization: Onset approximately 12 months ago; most recent episode four days before this session. Frequency has been stable at a few episodes per month, with no reported change in pattern. One method category was identified and recorded; the specifics are documented in the chart and deliberately generalized here. No episode has required medical attention, and medical severity is assessed as low to date. Suicide risk screen and intent determination: Asked directly. Client denies current suicidal ideation, plan, and intent, and denies ideation in the past month. C-SSRS screen completed: ideation items negative; behavior item positive for non-suicidal self-injury only, with no attempts, aborted or interrupted attempts, or preparatory acts. Client states the behavior has never carried a wish to die and describes its purpose consistently and without prompting. Determination: non-suicidal self-injury. Basis: direct inquiry, negative validated screen, a consistent account, and no behavioral indicators of suicidal intent. The determination is reassessed at every review while the behavior is active. Function, antecedents, and triggers: Client identifies relief of built-up distress as the primary function, most often after academically stressful days that end in isolation at her apartment, with urges strongest late in the evening. Texting her sister has interrupted the urge on several occasions. No interpersonal-influence function was endorsed on inquiry. Risk factors: Acute: ongoing academic pressure, roommate conflict, evening isolation, and the recurrent self-injury itself, which elevates risk for later suicidal thoughts and behavior. Chronic: anxiety disorder with depressive features; no attempt history; no family history of suicide reported. Protective factors: Voluntary disclosure once asked directly; engaged in weekly treatment; supportive sister with regular contact; client states the behavior worries her and she wants alternatives; no substance involvement; stable housing and enrollment. Risk formulation: Recurrent non-suicidal self-injury of stable frequency and low medical severity to date, serving an identified emotion-regulation function, in a client with a negative suicide screen today and no attempt history. The intent determination was directly assessed, not assumed. Because a self-injury history elevates risk for later suicidal thoughts and behavior, the suicide screen repeats at every session while the behavior is active. With that surveillance in place, acute risk of suicidal behavior is judged low today; the primary clinical concern is the behavior's role as the client's main response to distress, which treatment now targets. Interventions and response plan: Distress-tolerance alternatives introduced and rehearsed in session; client selected two to practice this week. A written response plan for high-urge windows was completed collaboratively (early warning signs, alternatives in order, people to contact, crisis resources), copy to client. One access-reduction step was agreed and recorded. With client's written consent, her sister is named in the plan as an evening contact. Crisis line contact saved to client's phone. Diagnosis and coding: Generalized anxiety disorder, F41.1, remains the primary diagnosis. R45.88 added as an additional symptom code for nonsuicidal self-harm. Session billed as individual psychotherapy with time documented. Disposition, follow-up, and consultation: Continue weekly outpatient therapy; treatment plan amended in session to add an emotion-regulation goal targeting the behavior's function. Next appointment 07/29/2026, with a secure-message check-in after the weekend. Escalation instructions reviewed: use the response plan, and contact the crisis line or the nearest emergency department if thoughts of suicide emerge. Case discussed same day with clinical supervisor, R.M., PhD, who concurred with the formulation and plan. Reassessment plan: Self-injury status, urges, response-plan use, and the suicide screen reviewed at each session while the behavior is active; full reassessment with any change in pattern, severity, or screening status. This sample is fictional and for educational purposes. It does not describe a real patient, and it intentionally omits the behavioral specifics a real record would contain. ↑ Back to the template and downloads ### Why this sample works The intent determination is shown with its basis (direct inquiry, a negative validated screen, a consistent account), and the note says when the determination will be reassessed, so "non-suicidal" is a documented finding rather than a label. The suicide risk screen sits inside the assessment with its result written out, which is the structural answer to the co-occurrence problem: the two findings stay distinct, and the screen repeats on a stated cadence. The behavior is characterized on exactly the dimensions that make the next reassessment comparable (onset, recency, frequency band, severity, stability) without narrative detail the record does not need. Function is documented in the client's terms, and the treatment plan is amended to target that function in the same encounter, so the plan follows from the assessment instead of running parallel to it. Every intervention is recorded as completed and dated, with the copy given, the consent obtained, and the named support written down, not promised for later. A reassessment plan closes the note, so an active behavior is tracked forward instead of assessed once and left behind. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A self-harm assessment is read by more audiences than the session notes around it: the treating team and supervisor first, then auditors, licensing boards, and, after a bad outcome, coroners and courts. It lives in the ordinary clinical record: clients generally hold HIPAA access rights to it, it is reachable by subpoena, and the separately kept psychotherapy notes category never covers it, so write clinically necessary detail rather than speculation. With adolescent clients, where the behavior concentrates, access is more layered: a parent or guardian's right to the record depends on state law and on the consent authority under which care is provided, so know your state's rule before the disclosure conversation and document what was shared, with whom, and why. Retention runs long, and minors extend it: state law sets the adult period (California seven years, New York six), the APA record-keeping guideline suggests seven years after the last service or three years after a minor reaches majority, whichever is later ( APA ), Ontario colleges expect ten years from the last contact or from age 18, and Australian practice keeps records of clients seen as minors until the 25th birthday. After an adverse event, never revise the original entry: anything learned later goes in a clearly dated late entry or addendum. When a death, serious injury, or legal process touches your documentation, consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). On the payer side there is no NSSI-specific code to defend: the assessment is reviewed as part of whatever service carried it, usually psychotherapy or a diagnostic evaluation, and the audit exposure is the generic kind. In the first pandemic year, Medicare paid an estimated $580 million improperly for psychotherapy services; for 128 of 216 sampled enrollee days requirements were not met, with failures as ordinary as psychotherapy time not documented ( HHS-OIG A-09-21-03021 ). CMS's 2024 reporting put outpatient psychiatry improper payments at 16.1 percent, $254.5 million, with insufficient documentation behind 78.3 percent of them ( CMS MLN ). The format is a convention; the content is the requirement: the NSSI-specific work, the intent determination, the function, the formulation, has to sit inside a note that also supports the billed service with time, medical necessity, intervention, response, and linkage to a treatment plan goal that targets the behavior's function. Code the underlying disorder first: R45.88, nonsuicidal self-harm, joined the code set with fiscal year 2022 (effective October 1, 2021) as a symptom code, so it supplements a definitive diagnosis rather than replacing one. In accredited facilities, suicide-risk screening standards fold a self-harm presentation into the validated-screen-and-assessment workflow ( Joint Commission R3 Report 18 ); in private practice no accreditor applies, and the clinical standard of care asks for the same distinction to be assessed and documented. ↑ Back to the template and downloads ## Common self-harm assessment errors reviewers flag The audit numbers around this document are generic to psychotherapy, and that is itself the finding: no payer policy we reviewed publishes an NSSI-specific documentation rule or denial code. What auditors enforce is the ordinary chain, and they enforce it hard: Medicare's first-pandemic-year psychotherapy review found $580 million in improper payments, $348 million of it for telehealth, with 128 of 216 sampled enrollee days failing requirements as basic as documented time ( HHS-OIG, 2023 ), and insufficient documentation drove 78.3 percent of outpatient psychiatry improper payments in CMS's 2024 reporting ( CMS MLN ). The BastionGPT Clinical Advisory Board sees the same errors most often in self-harm assessment reviews: Intent assumed rather than assessed. The note calls the behavior non-suicidal without showing the inquiry: what was asked, what the client said about a wish to die, what a screen found, and why the account was credited. Self-directed violence "may or may not reflect suicidal intent" ( VA Rocky Mountain MIRECC ), which is precisely why the determination is a documented finding, never a default. A missing or buried suicide screen. A history of self-injurious thoughts and behaviors is consistently cited as one of the strongest predictors of future suicidal behavior ( Ribeiro et al., 2016 ), so an NSSI assessment with no documented screen is an incomplete risk document on its face, and the first gap a reviewer finds after an adverse event. Characterization language doing the work of assessment. "Superficial" as a conclusion with no severity assessment behind it, or older terms like "gesture" and "manipulative," assert intent and function findings the note never actually made. Record what was assessed: severity in terms of care required, function in terms of what the client reported. R45.88 doing a job it cannot do. It is a symptom code, added in fiscal year 2022, and it belongs alongside the definitive diagnosis, not in its place. Claims and problem lists that carry R45.88 alone misstate the clinical picture and invite exactly the documentation review the note should survive. Function absent, so the plan is generic. The evidence base reduces the behavior's many reported functions to two factors, intrapersonal and social ( Klonsky et al., 2015 ). When the note never records which one is operating, the plan defaults to "coping skills," the goal linkage auditors look for is missing, and the treatment cannot be shown to target the behavior at all. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on self-harm assessments. Draft a structured self-harm assessment from your bullets or dictation, with the history, intent determination, suicide screen, function, formulation, and plan in the right places. Check a finished note before you sign: flag an intent determination stated without its basis, a missing suicide screen, an undated follow-up, or a formulation that does not support the disposition. Pull the risk-relevant thread out of months of session documentation for supervision, consultation, or a records request, and keep the summary to what the receiving reader should see. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a self-harm assessment legally required, and is there a required format? No statute, payer policy, or professional body in the US, Canada, or Australia prescribes an NSSI assessment document or form. The obligations are layered instead: general record-keeping rules from boards and colleges apply everywhere, accreditor suicide-risk standards define screening and assessment workflows in facilities, and payers expect the record to support the billed service. Everything NSSI-specific, the intent determination, the function, the formulation, is professional convention, which is why the defensible move is to document what reviewers reconstruct afterward. Most run 300 to 800 words. How is non-suicidal self-injury different from a suicide attempt? By intent, and intent is a finding, not a guess. Non-suicidal self-injury is deliberate injury without a wish to die, typically serving another purpose such as regulating unbearable internal states. The VA built its classification system with the CDC because self-directed behavior "may or may not reflect suicidal intent" ( VA Rocky Mountain MIRECC ) and clinical language kept blurring the two. Document what the client says the behavior is and is not, the basis for crediting that account, and reassess the determination while the behavior is active, because intent can change between episodes. Do I need a suicide risk assessment every time a client self-harms? A suicide risk screen belongs in every self-harm assessment; a full suicide risk assessment follows whenever the screen or the clinical picture calls for it. The co-occurrence numbers are the reason: a history of self-injurious thoughts and behaviors is consistently cited as one of the strongest predictors of future suicidal behavior ( Ribeiro et al., 2016 ), and in adolescents who self-harm, suicide attempts are markedly more common than in peers who do not ( Gillies et al., 2018 ). A documented negative screen is a finding in its own right; an absent screen is a gap. Is NSSI an official DSM diagnosis? No. DSM-5 placed nonsuicidal self-injury disorder in Section III in 2013 as a condition requiring further study, with a proposed frequency threshold of self-injury on five or more days in the past year, and it has stayed a study condition rather than graduating to an official diagnosis ( Zetterqvist, 2015 ). A 2025 expert proposal in Lancet Psychiatry would reclassify NSSI as a clinical specifier instead of a standalone disorder. For records and claims the practical consequence is the same: diagnose the underlying condition, and code the behavior as a symptom. What ICD-10 code do I use for self-harm without suicidal intent? R45.88, nonsuicidal self-harm, added with the fiscal year 2022 code set effective October 1, 2021. It sits in the symptoms chapter, so report it in addition to the definitive diagnosis, never alone as the principal diagnosis when a definitive diagnosis exists. The intent determination in your note is what justifies coding the behavior as nonsuicidal in the first place, which is one more reason the note must show that determination being made rather than assumed. Should I document a risk level like low, medium, or high? The guidance genuinely conflicts, and template pages rarely say so. The UK's NICE guideline NG225 instructs clinicians "Do not use global risk stratification into low, medium or high risk" for predicting repetition or allocating treatment, directing a risk formulation in every psychosocial assessment instead, while US accredited settings document an overall risk level under Joint Commission suicide standards. The defensible synthesis: lead with the narrative formulation, add the level your setting requires, and never let a tier stand without reasoning, because prediction from classification is weak everywhere it has been measured. Do I need a body map or a specific assessment instrument? None is mandated by any regulator or payer we reviewed; body maps and structured tools are facility policy or clinical preference. The validated research instruments, the SITBI structured interview and the ISAS and DSHI inventories, are optional structure for the inquiry, and the ISAS evidence base is why "function" is on the template at all: the many reported functions reduce to an intrapersonal factor and a social factor ( Klonsky et al., 2015 ). What convention does expect in every note: the intent determination, the suicide screen, and a formulation that supports the plan. What is the difference between a self-harm assessment, a safety plan, and a crisis note? The self-harm assessment is the clinician's evaluation record: findings, intent determination, formulation, response. A safety plan is a client-facing intervention document, the prioritized steps a client uses as urges or risk rise, and the response plan in this page's sample is its NSSI-focused cousin. A crisis note documents an urgent encounter as the service itself, with the risk reasoning inside it. In a typical disclosure session you produce the assessment and complete a response plan as one documented intervention, and you reserve the crisis note for acute presentations. Can BastionGPT draft a self-harm assessment? Yes. Give it bullets or dictation from the session, and it drafts a structured assessment with the history, intent determination, suicide screen, function, formulation, and plan in the right places for your review. It can also check a finished note before you sign: an intent determination stated without its basis, a missing suicide screen, or an undated follow-up. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The compliance and research claims on this page trace to these authorities, last verified July 2026: NICE, NG225: Self-harm: assessment, management and preventing recurrence (2022) : the risk-stratification prohibition and the formulation-in-every-assessment recommendation. VA Rocky Mountain MIRECC, Self-Directed Violence Classification System : the VA and CDC common nomenclature, adopted by VA and DoD. HHS Office of Inspector General, A-09-21-03021 (May 2023) : $580 million in improper Medicare psychotherapy payments in the first pandemic year; 128 of 216 sampled enrollee days failed requirements. CMS Medicare Learning Network, outpatient psychiatric care compliance tips : the 16.1 percent improper-payment rate and 78.3 percent insufficient-documentation share (2024 reporting). CMS, ICD-10-CM fiscal-year code files : R45.88, nonsuicidal self-harm, present from the FY2022 code set (effective October 1, 2021) and absent from FY2021. 42 CFR 482.61 : psychiatric-hospital records must permit determination of the degree and intensity of treatment. Joint Commission, R3 Report Issue 18 and NPG #8 : validated-tool screening from age 12 and the assessment workflow in accredited settings. Swannell et al., Suicide and Life-Threatening Behavior (2014) : pooled lifetime NSSI prevalence of 17.2 percent in adolescents, 13.4 percent in young adults, 5.5 percent in adults. Gillies et al., Journal of the American Academy of Child and Adolescent Psychiatry (2018) : suicide-attempt risk ratio of 9.14 in adolescents who self-harm, across 172 datasets and 597,548 participants. Ribeiro et al., Psychological Medicine (2016) : self-injurious thoughts and behaviors as predictors of later suicidal thoughts and behaviors, and the weakness of prediction overall. Nock et al., Psychological Assessment (2007) : the Self-Injurious Thoughts and Behaviors Interview (SITBI). Klonsky, Glenn et al., Child and Adolescent Psychiatry and Mental Health (2015) : converging two-factor structure of NSSI functions, with the ISAS (Journal of Psychopathology and Behavioral Assessment, 2009). Zetterqvist, Child and Adolescent Psychiatry and Mental Health (2015) : DSM-5 Section III placement and the proposed NSSI disorder criteria. Lengel, Muehlenkamp, Zetterqvist et al., Lancet Psychiatry (2026, online 2025) : the proposal to reclassify NSSI as a clinical specifier. Favazza, Journal of Nervous and Mental Disease (1998) and Bodies Under Siege (1987): the foundational clinical account and definition. APA, Record Keeping Guidelines , and ACSQHC, NSQHS Comprehensive Care Standard (Actions 5.31 and 5.32): retention guidance and the Australian self-harm and suicide-risk accreditation actions. ## Related templates Suicide Risk Assessment Safety Plan Crisis Psychotherapy Note Trauma Assessment Violence & Homicide Risk Assessment General Risk Screening Note C-SSRS Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/relapse-prevention-plan All note templates # Relapse Prevention Plan: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A relapse prevention plan is a written plan a clinician and client build together that names the client's triggers, early warning signs, coping strategies, support contacts, and the steps to take after a lapse. Substance use and mental health clinicians write one mid-treatment or before step-down, usually inside the treatment plan. Most run 300 to 800 words, built in a 20 to 45 minute session. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, and SUD clinicians, together with the client Audience The client and their chosen supports; the treatment team; payers read it inside the treatment plan Typical length 300 to 800 words · 20 to 45 minutes with the client (clinical team estimate) Format family Checklist plus narrative (also called RPP, relapse prevention worksheet, continuing care or aftercare plan) When it's used SUD treatment and relapse-prone mental health care: mid-treatment, at step-down, before discharge, and after any lapse Standards context Convention, not law: US rules fold it into the treatment plan; Australia's MBS mental health plan notes name it as a required element On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a relapse prevention plan? A relapse prevention plan is a written plan a clinician and client build together that records the specific situations and internal shifts that put this one person's recovery at risk, and what the client and their supports have agreed to do about each: triggers and high-risk situations, early warning signs, matched coping responses, people to contact, and the steps that follow a lapse. The format descends from two separate 1980s traditions that vendor pages routinely blur into one. The cognitive-behavioral model of G. Alan Marlatt and Judith Gordon ( Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors , Guilford Press, 1985) treats relapse as a process that begins before any use, driven by immediate determinants such as high-risk situations, coping skills, outcome expectancies, and the abstinence violation effect, plus covert antecedents such as lifestyle imbalance and cravings. The counseling lineage of Terence Gorski and Merlene Miller ( Staying Sober: A Guide for Relapse Prevention , Independence Press, 1986, built on the CENAPS model) popularized working from a personal list of relapse warning signs. Clinicians also call the document an RPP, relapse prevention worksheet, continuing care plan, or aftercare plan. No statute or payer rule we can point to requires a standalone document with this title; where relapse prevention appears in regulation, it lives inside the parent record. Minnesota's SUD licensure chapter requires each treatment plan to identify "maintenance strategy goals and methods designed to address relapse prevention" ( Minn. Stat. 245G.06 ), and Medicare's benefit manual builds partial hospitalization and intensive outpatient documentation around an individualized treatment plan whose stated purposes include preventing relapse or hospitalization. Australia goes furthest: the MBS explanatory notes make a written crisis-intervention or relapse-prevention plan a required element of preparing the GP mental health treatment plan. One boundary matters clinically: a relapse prevention plan is not a safety plan . The safety plan is a suicide-crisis intervention; the relapse prevention plan targets a return of substance use or of psychiatric symptoms. When a client carries both risks, they need both documents, not a merged one. ## Who uses relapse prevention plans and when Substance use disorder programs are the core users: outpatient and intensive outpatient counselors, residential programs, and opioid treatment programs build one mid-treatment, at step-down, or as part of discharge planning, then revisit it after any lapse. The same structure serves mental health care wherever a condition has a known relapse signature, including depression, bipolar disorder, psychosis, and eating disorders, where Australia's MBS notes apply the same planning element to eating disorder treatment plans. Peer-led models cover adjacent ground: the Wellness Recovery Action Plan (WRAP), founded in 1997 and developed by people living with mental health challenges, is a client-owned wellness tool rather than a clinical record. Reach for the treatment plan when a payer or licensure rule is the driver, the treatment plan review to update goals on schedule, a substance use assessment when the question is severity and level of care, and a safety plan whenever suicide risk, not relapse, is the concern. The relapse prevention plan earns its place when the client needs one page in their own words to carry out of treatment. ## Relapse prevention plan structure: what goes in each section Header and ownership. Client name, date, who prepared it (the client with the clinician), the next review date, and where the client's copy lives: wallet card, phone photo, fridge door. This is the client's document more than the chart's. Pitfall: no dated review target; a plan without a next date is a plan nobody reopens. Recovery goal in the client's words. One or two sentences of what the client is protecting: sobriety, stability, custody mornings, a job worth keeping. It anchors every later section. Pitfall: importing the treatment plan's clinical phrasing; "maintain abstinence" is a goal the client was handed, not one they said. Triggers and high-risk situations. The external half of the risk map: people, places, times, dates, and conflicts that raise risk, plus the internal states that ride along. Specific beats complete: the route home on Fridays, the custody call, the anniversary in September. Pitfall: generic worksheet lists ("people, places, things") with no situation this client actually faces; reviewers read individualization as the marker of real planning work. Early warning signs. The internal half: the thoughts, mood shifts, and behavior changes that come before use or symptom return, in the client's language. The best signs are ones the client or a named support can actually notice: skipped meetings, short sleep, going quiet. Pitfall: conflating this section with triggers; triggers live in the world, warning signs live in the client, and the plan responds to each differently. Coping responses, matched pair by pair. For each trigger and warning sign, the rehearsed response: "when X, I will Y." Delay-and-distract routines, urge surfing, calling before entering a risky situation, moving a session earlier. Pitfall: a wall of unmatched skills; ten signs and six strategies with no line connecting them is a handout, not a plan. Support contacts, with consent. Who the client will contact, their role, when to call them, and how they can tell something is off. In SUD programs, involving a family member or sponsor in care still requires 42 CFR Part 2 compliant consent, so note that consent on file. Pitfall: naming contacts the client never agreed to involve, or listing "sponsor" with no name, number on the client's copy, or when-to-call. Lapse response and the crisis handoff. Script the first hours after a lapse: leave, call, tell the clinician, book the extra session. Name the difference between a lapse and a relapse so the abstinence violation effect does not turn one drink into "I have already failed." Then hand crisis off cleanly: if distress crosses into suicide risk territory, the client uses their safety plan , which carries the crisis resources. Pitfall: merging crisis content into this document, or ending the escalation path at "call 911" with nothing between a craving and an emergency. Review, signatures, and the client copy. Tie reviews to the treatment plan review cadence, plus a revision after any lapse, step-down, or medication change. Client and clinician signatures follow your agency and payer policy; handing the client a copy is convention and good practice. Pitfall: a plan written once at intake and never touched; a dated revision history is what makes it credible at audit and useful in month four. ### Blank template (copy and adapt) RELAPSE PREVENTION PLAN Client: _____________________ Date: ___________ Prepared with: ______________ Next review: ___________ My copy is kept: _________________________________ MY RECOVERY GOAL (in my own words) __________________________________________________ __________________________________________________ MY TRIGGERS AND HIGH-RISK SITUATIONS 1. _______________________________________________ 2. _______________________________________________ 3. _______________________________________________ MY EARLY WARNING SIGNS (what shows up first) 1. _______________________________________________ 2. _______________________________________________ 3. _______________________________________________ WHEN I NOTICE ONE, I WILL (matched responses) 1. _______________________________________________ 2. _______________________________________________ 3. _______________________________________________ PEOPLE I CAN CONTACT (name / role / when to call) 1. _______________________________________________ 2. _______________________________________________ Consent to involve them documented: [ ] Yes IF A LAPSE HAPPENS, MY NEXT THREE STEPS 1. _______________________________________________ 2. _______________________________________________ 3. _______________________________________________ A lapse is a signal to act, not proof of failure. IF I AM IN CRISIS: I will use my safety plan. Safety plan location: ____________________________ KEEPING MY BALANCE (sleep, routine, meetings, meals) __________________________________________________ Client signature: ____________ Clinician: ____________ Reviewed/revised (date + initials): ______________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample relapse prevention plan Scenario: client finishing an intensive outpatient program for alcohol use disorder, with a history of recurrent depression, building the plan with his counselor in the final week before step-down to weekly outpatient sessions. The entries are in the client's own words. All details are fictional. RELAPSE PREVENTION PLAN Client: D.R., 41 · Date: 07/21/2026 · Prepared with: S. Whitfield, LPC Next review: 08/17/2026 (first weekly session of the month) · My copy: wallet card + phone photo My recovery goal. Stay sober, stay present for my daughter on our weekends, and get back to running three mornings a week. My triggers and high-risk situations. (1) Driving past the Riverside Tap on the late-shift route home on Fridays. (2) Custody schedule calls with my ex that turn into arguments. (3) Sunday afternoons alone after I drop my daughter off. (4) Deadline weeks when I skip lunch and sleep under six hours. (5) The first week of September, around my father's anniversary. My early warning signs. (1) Telling myself "one beer would settle me down." (2) Skipping the Tuesday meeting because "this week is too busy." (3) Up past 1 a.m. scrolling, then short sleep. (4) Going quiet with my sister; she notices before I do. (5) Losing interest in running, which was the first thing to go the last time my depression came back. When I notice one, I will. (1) Say it out loud to my sister or my sponsor the same day, not "handle it myself." (2) Run the 15-minute urge-surfing exercise from group, then reassess. (3) Take the alternate route home on Fridays; it adds six minutes and skips the Tap. (4) After a custody call, walk the block once before I decide anything. (5) If two signs show up in the same week, ask to move my session earlier. People I can contact. (1) K.R., my sister: daily text check-in; she calls if I sound flat two days running. Consent to involve her is on file. (2) J.M., my sponsor: I call before walking into any event with alcohol; we meet Tuesdays. (3) S. Whitfield, LPC: sessions Mondays; program after-hours line for between-session support (number on my copy). If a lapse happens, my next three steps. (1) Leave the situation and call J.M. within the hour; if no answer, call K.R. (2) Tell S. Whitfield the next business day and book an extra session that week. (3) Reread the card we wrote: a lapse is a signal to use this plan, not proof that I have failed. If I am in crisis. If I have thoughts of harming myself, I use my safety plan (updated today, kept with this card) and the contacts on it. Keeping my balance. Run Monday, Wednesday, Saturday with the 6 a.m. group. Meals at regular times on deadline weeks. Lights out by 11:30. One recovery meeting a week minimum, two on hard weeks. Review and signatures. Reviewed at each treatment plan review and within one week after any lapse, schedule change, or medication change. Client: D.R., signed 07/21/2026. Clinician: S. Whitfield, LPC, signed 07/21/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every trigger and warning sign has a matched response written as "when X, I will Y", which is the skill-rehearsal logic of the cognitive-behavioral relapse prevention model put on paper. The warning signs are observable, personal, and split from triggers: a named support can spot "going quiet" and short sleep, and the depression signature (losing interest in running) is tracked alongside the alcohol risk. Supports are listed with roles, when-to-call rules, and documented consent, so involving the sister and sponsor stays inside 42 CFR Part 2 boundaries. The lapse section scripts the first hour and names the lapse-versus-relapse distinction, taking the abstinence violation effect head on instead of pretending lapses never happen. Crisis is handed to the safety plan rather than duplicated here, so each document stays usable in its own moment. The plan carries a next review date, a revision trigger list, both signatures, and a note of where the client's copy lives. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Sort every "requirement" you have heard about this document into law, payer policy, or convention, because almost all of them land in the last two bins. No US federal statute or regulation we can point to requires a standalone document titled relapse prevention plan. Medicare's benefit manual ( Ch. 6 ) builds partial hospitalization and intensive outpatient documentation around an individualized treatment plan aimed at preventing relapse or hospitalization, federal opioid treatment program standards require an individualized treatment plan ( 42 CFR 8.12 ), and state licensure chapters like Minnesota's fold relapse prevention goals into that plan. The familiar rules of thumb (a client signature, an update every 90 days, a separate document, a copy handed to the client) are agency policy or good practice, set by your program, state, and payer rather than by federal law. Confidentiality law is the same story: HIPAA and 42 CFR Part 2 govern who may see an SUD record and what consent a disclosure requires, including Part 2's single consent for treatment, payment, and operations disclosures with compliance required by February 16, 2026 ( 2024 final rule ); neither says a word about what a relapse prevention plan must contain. For reimbursement, the parent document is the one that matters: reviewers ask whether the treatment plan is individualized, current, and signed, and the large SUD audit findings turn on exactly those failures rather than on any missing relapse prevention plan. Australia is the working exception: the MBS explanatory notes make "a plan for crisis intervention and/or for relapse prevention" a written element of preparing a GP mental health treatment plan under items 2700, 2701, 2715, and 2717 ( note AN.0.56 ), required again at review if not previously provided, and apply the same element to eating disorder treatment plans, so document it explicitly when your note supports those items. Retention follows the whole record wherever you practice: California requires 7 years after termination, New York at least 6, Ontario's psychology college 10, British Columbia 16, and Australia 7 years from the last entry or until the 25th birthday for minors. Two boundaries keep the document clean: routine updates to relapse-prevention content belong in the treatment plan review , and crisis content belongs in the safety plan , which carries the crisis resources so this document does not have to. ## Common relapse prevention plan errors auditors flag Auditors almost never cite a missing relapse prevention plan; they cite the treatment plan around it. HHS-OIG's February 2020 audit of New York opioid treatment programs ( A-02-17-01021 ) estimated at least $39.3 million in improper federal Medicaid claims across its 2014 to 2017 audit period because providers failed to maintain service documentation, delivered services out of step with beneficiaries' treatment plans, or did not maintain signatures. Denial data points the same direction: KFF's analysis of 2023 HealthCare.gov plans found insurers denied 20% of in-network claims, only 6% of those denials citing lack of medical necessity, and fewer than 1% of denials appealed. The exposure sits in plan adherence and signatures, not in richer prose. The BastionGPT Clinical Advisory Board sees the same errors most often in relapse prevention plan reviews: A plan about nobody. Trigger lists copied from a worksheet, goals in clinical boilerplate, nothing this client actually faces or said. Reviewers read individualization as the marker of real planning work, and clients do not use documents that are not theirs. Merged with the safety plan. One hybrid document tries to cover suicide risk and relapse risk at once, and both jobs suffer: crisis steps get buried under coping lists, and relapse content gets dropped when the crisis template takes over. When both risks are present, build both documents and have each point to the other. Warning signs without responses. Ten signs and six coping skills with no line connecting them. Each sign needs its matched "when X, I will Y" response, rehearsed, not just listed. Support contacts nobody agreed to. Family and sponsors named without documented consent (in SUD programs, a 42 CFR Part 2 problem), or contacts with no role, no number on the client's copy, and no when-to-call. Frozen at intake. The plan is produced once to satisfy an admission checklist, then never revised after a lapse, a step-down, or a medication change, and carries no next review date. A dated revision history is what separates a living plan from paperwork. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on relapse prevention plans. Draft a complete relapse prevention plan from session bullets, a dictation, or the transcribed planning conversation, keeping the client's own words where you flag them. Rewrite the finished plan in plain language at the client's reading level, ready to print as the copy the client carries out the door. Check a draft before you sign: warning signs without a matched response, supports without consent or contact details, a missing lapse step, and no next review date. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a relapse prevention plan be, and what format does it take? Most run 300 to 800 words across one or two pages, built with the client in a 20 to 45 minute conversation; the time is a clinical team estimate, since no published study benchmarks it. The working format is a checklist plus short narrative in the client's own words: triggers, early warning signs, matched coping responses, support contacts, lapse steps, and a next review date. No law in the US, Canada, or Australia sets a length or a required field set for it. Is a relapse prevention plan required by law? The document is a convention; the content is what rules require. No US federal statute or regulation we can point to names a standalone relapse prevention plan. Medicare's manual requires an individualized treatment plan for partial hospitalization and intensive outpatient care with relapse prevention among its aims, federal opioid treatment program standards require an individualized treatment plan, and Minnesota's licensure chapter requires every treatment plan to identify maintenance strategies addressing relapse prevention. Australia is the exception that proves the pattern: the element is required, in writing, inside the GP mental health treatment plan rather than as a separate document. Does it have to be a separate document from the treatment plan? No rule requires a separate document; regulations place relapse-prevention content inside the treatment plan , recovery plan, or discharge plan, and routine updates belong in the treatment plan review . Many clinicians still produce a standalone page for a practical reason: the client keeps a copy, and a one-page plan in the client's own words gets used in the moment in a way a chart document never will. What is the difference between a relapse prevention plan and a safety plan? A safety plan is a suicide-crisis intervention: the six-step Stanley-Brown format moves from warning signs through coping and contacts to making the environment safer. A relapse prevention plan targets a return of substance use or of psychiatric symptoms. The two are routinely confused, and the confusion is clinically consequential, because one does not substitute for the other: a client facing both risks needs both documents, each pointing to the other. Is a relapse prevention plan required for insurance or Medicare billing? Not in the US: the billing-relevant document is the treatment plan or plan of care, and audit findings turn on that plan's completeness, adherence, and signatures. Australia is different: preparing a GP Mental Health Treatment Plan under MBS items 2700, 2701, 2715, or 2717 requires a written plan for crisis intervention and/or relapse prevention ( note AN.0.56 ), the same element returns at review if not previously provided, and eating disorder treatment plans carry it too. Does the client have to sign it, and how often should it be updated? No federal rule answers either question for this document specifically. Signature and review requirements attach to the treatment plan under state and payer rules, and agency policy usually extends them here; the widely cited 90-day update is a payer or program cadence, not federal law. The clinically defensible rhythm: revise after any lapse, at step-down or discharge, at each treatment plan review, and whenever a warning sign or support changes. A client signature is convention, but a plan the client co-wrote and signed is one the client is likelier to use. Does relapse prevention planning actually work? The evidence is meaningful but modest, and it favors practice over paperwork. A meta-analysis of 26 studies with 9,504 participants found relapse prevention generally effective, particularly for alcohol problems and when combined with medication ( Irvin et al., 1999 ). A randomized trial of 286 adults in aftercare found mindfulness-based relapse prevention produced fewer substance-use days and less heavy drinking at 12-month follow-up than standard relapse prevention or treatment as usual ( Bowen et al., 2014 ). Writing the plan is the record; rehearsing the matched responses in it is the active ingredient. How long do I keep a relapse prevention plan? As long as the chart it belongs to: retention attaches to the whole record, not to the document. California requires 7 years after termination, New York at least 6, Ontario's psychology college 10 (a correction to the often-repeated 7), British Columbia 16, and Australia's Psychology Board 7 years from the last entry or until the client's 25th birthday for minors. HIPAA sets no chart retention period at all; its six-year rule covers compliance documentation. When the client keeps a copy, note that in the record; the chart copy is the one retention rules govern. Can BastionGPT draft a relapse prevention plan? Yes. Bring session bullets, a dictation, or the planning conversation itself, and it drafts the full plan with triggers, warning signs, matched responses, contacts, and lapse steps, keeping the client's phrasing where you mark it. It can also produce a plain-language client copy and check a draft for unmatched signs, missing consent notes, or a missing review date before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and data is never used to train models. ## Related templates Treatment Plan Safety Plan Treatment Plan Review & Update Substance Use Assessment Case Formulation Note Outcome Measure Review Note Better Access Mental Health Treatment Plan Referral Return-to-Primary-Care Summary 42 CFR Part 2 Consent CAGE Questionnaire Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/outcome-measure-note All note templates # Outcome Measure Review Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An outcome measure review note is the documentation entry of measurement-based care: it records which standardized instrument was administered, the score, the change from baseline, the clinical interpretation, the discussion with the client, and the treatment decision that follows. Therapists, psychologists, and collaborative care teams write one each time a measure like the PHQ-9 or GAD-7 is given. Most run 75 to 300 words beside a score flowsheet. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, and collaborative care teams Audience The clinician and client first; the care team, supervisors, payers, and auditors on request Typical length 75 to 300 words plus the score flowsheet · 3 to 10 minutes by hand (clinical team estimate) Format family Structured flowsheet plus short narrative (also called an MBC note, ROM note, or progress monitoring note) When it's used Every administration of a standardized measure: baseline at intake, every 2 to 4 weeks in many programs, and before each treatment plan review Standards context Accreditors, some payer codes, and Australian plan rules mandate the measurement; no rule anywhere sets the note's format On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an outcome measure review note? An outcome measure review note is the documentation entry for measurement-based care: it records that a standardized instrument was administered, the score it produced, how that score compares with baseline and prior administrations, what the change means clinically, that the result was discussed with the client, and the treatment decision that followed. Measurement-based care itself was defined in a widely cited review as the systematic use of symptom rating scales to drive clinical decision making for the individual patient ( Fortney et al., 2017 ), a practice built through the 1990s patient-focused research of Kenneth Howard, Michael Lambert's work on outcome feedback and not-on-track alerts, and the brief session-by-session scales of Scott Miller and Barry Duncan. The instruments carry precise pedigrees: the PHQ-9 was validated on 6,000 primary care and obstetrics patients ( Kroenke, Spitzer & Williams, 2001 ), the GAD-7 followed in 2006 ( Spitzer et al. ), and the PCL-5 is public domain and takes 5 to 10 minutes ( National Center for PTSD ). Clinicians also call the entry an MBC note, a routine outcome monitoring (ROM) note, a feedback-informed treatment note, or a progress monitoring note. Here is the fact that reframes every "required elements" list you will find online: the instruments are standardized; the note is not. No statute, CPT rule, accreditation standard, or professional college in the US, Canada, or Australia defines the fields, layout, or template of this note. Where requirements exist, they mandate the measurement, never the note format: the Joint Commission requires accredited behavioral health organizations to monitor progress with a standardized instrument ( standard CTS.03.01.09 , effective January 1, 2018), Medicare's collaborative care codes require validated rating scales and a registry, and Australia's MBS requires an outcome measurement tool when a GP prepares a mental health treatment plan. The gap between evidence and practice is the other defining fact: in a national survey of 504 clinicians, only 13.9% used standardized progress measures at least monthly and 61.5% never used them ( Jensen-Doss et al., 2018 ). The note sits downstream of two neighbors: it supplies the data that a treatment plan review acts on, and it is often written as a labeled section inside the psychotherapy progress note rather than as a separate document. ## Who uses outcome measure review notes and when Any clinician who administers a standardized measure and wants the administration to count writes one: outpatient therapists and psychologists reviewing a PHQ-9 or GAD-7 the client completed through the portal, group practices running symptom scales on an every-2-to-4-week cadence, collaborative care teams whose billing codes require validated rating scales tracked in a registry ( CMS MLN909432 ), Joint Commission accredited organizations meeting the standardized-instrument standard, child and adolescent services scoring the SDQ, and Australian clinicians working under a GP mental health treatment plan or in PHN-commissioned programs that report the K10+ into the national minimum data set. The cadence is the point: a baseline at intake, re-administration every few weeks, and a visible trend by the time the treatment plan review is due. Reach for the psychotherapy progress note when the session narrative is the deliverable and the score is one line inside it, and for the psychological evaluation report when the question calls for clinician-administered testing with interpretation and a written report, the way a full battery write-up like an MMPI-3 or BASC-3 report does. The outcome measure review note earns its place when a brief self-report scale needs to move treatment: score it, compare it, interpret it, discuss it, and change something. ## Outcome measure review note structure: what goes in each section Instrument, date, and administration context. Name the instrument and version, the administration date, and how it was completed: in session, through the portal beforehand, or in the waiting room. Method matters when scores shift with setting. Pitfall: "screen completed" with no instrument name; an unnamed measure supports neither the billing line nor the trend. Score and scoring integrity. The total score and any subscales, plus who scored it. Brief measures are self-report, but the scoring and documentation are yours. Pitfall: transcribing the number wrong or leaving a half-completed form unscored; a score you cannot reproduce from the responses is a score an auditor cannot credit. Trend against baseline and prior scores. The line that makes it measurement rather than paperwork: today's score next to baseline and the last two or three administrations, with dates. A flowsheet does this in one row per administration. Pitfall: a score with no comparison; a single number tells nobody whether treatment is working. Clinical interpretation beyond the number. Severity range, direction and size of the change, and whether the score matches what you see in session. Scores and presentation can disagree; say so when they do. Pitfall: restating the number as its label ("score 11, moderate") with no clinical read; interpretation is the part only a clinician can supply. The conversation with the client. That you shared the result, how you framed it, and the client's response. Feedback trials keep finding that measurement helps when results reach the clinical conversation, not the chart alone. Pitfall: no evidence the client ever saw the score; measurement-based care without the discussion is data collection. Item-level flags and risk follow-up. Note any item that changes your plan for the session, and always the risk item: an elevated PHQ-9 item 9 gets a same-day risk conversation, documented with its outcome, linking the suicide risk assessment or safety plan when one follows. Pitfall: filing a positive risk item with no documented follow-up; that single omission turns a routine note into the chart's weakest page. The treatment decision that follows. What the score changes or confirms: continue, adjust the intervention, change frequency, refer, or carry the target into the next treatment plan review , with one line of rationale linking score to decision. Pitfall: no linkage; a trend that never touches the plan is the pattern payers read as measurement theater. Signature, date, and billing linkage. Sign and date the entry, and make the billing line match the documentation: one unit per instrument administered, scored, and documented when you report 96127. Pitfall: units billed for instruments the note never names, or a signature that would not survive an authentication check. ### Blank template (copy and adapt) OUTCOME MEASURE REVIEW NOTE Client: _____________________ Date: ___________ Clinician: __________________ Session #: ______ INSTRUMENT AND ADMINISTRATION Instrument and version: __________________________ Administered by / method: ________________________ [ ] In session [ ] Portal before session [ ] Waiting room SCORE Total score: ________ Subscale scores (if any): ____________ Scored by: __________________ TREND VS BASELINE AND PRIOR SCORES Baseline (date / score): _________________________ Prior administrations (date / score): ____________ __________________________________________________ Change since baseline: ________ Since last: ______ CLINICAL INTERPRETATION Severity range and direction of change: __________ __________________________________________________ Consistency with presentation in session: ________ __________________________________________________ ITEM-LEVEL FLAGS AND FOLLOW-UP Items reviewed / elevated: _______________________ Risk follow-up (if indicated) and outcome: _______ __________________________________________________ DISCUSSION WITH CLIENT Scores shared and discussed: [ ] Yes [ ] Declined Client response: _________________________________ __________________________________________________ TREATMENT DECISION [ ] Continue current plan [ ] Adjust intervention [ ] Frequency change [ ] Referral / consult [ ] Goal or target update at next treatment plan review Rationale linking score to decision: _____________ __________________________________________________ Next administration due: ___________ Clinician signature / credentials / date: ________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample outcome measure review note Scenario: an outpatient psychologist reviews PHQ-9 and GAD-7 scores with a 34-year-old client at session 8 of weekly CBT for depression and anxiety, completed through the portal before the visit. Use it as a measurement-based care documentation example you can adapt to any instrument. All details are fictional. OUTCOME MEASURE REVIEW NOTE Client: M.T., 34 · Date: 07/22/2026 · Session: 8 of weekly CBT Clinician: R. Alvarez, PhD, Licensed Psychologist Instruments and administration. PHQ-9 and GAD-7, completed by client via secure portal 07/21/2026, evening before session. Both complete; scored automatically, verified by clinician. Scores and trend. 06/03/2026 (baseline, intake): PHQ-9 16 · GAD-7 15 06/17/2026: PHQ-9 13 · GAD-7 13 07/01/2026: PHQ-9 11 · GAD-7 12 07/22/2026 (today): PHQ-9 8 · GAD-7 11 Change: PHQ-9 down 8 points from baseline; GAD-7 down 4. Interpretation. Depression scores have moved from the moderately severe range at intake into the mild range and have now crossed below the 10 threshold, past the 5-point change our clinic flowsheet flags as meaningful improvement. The trajectory matches session presentation: brighter affect, resumed morning walks, back to full workdays. Anxiety is improving more slowly and remains in the moderate range, consistent with M.T.'s report that low mood has lifted while worry episodes continue most evenings. Sleep item on the PHQ-9 remains at 2, unchanged across all four administrations. Item-level review. PHQ-9 item 9 scored 0 at every administration to date, consistent with client's denial of suicidal ideation in session today. No risk follow-up indicated. Discussion with client. Reviewed the score graph on screen together. M.T.'s reaction: "the mood line finally looks like how I feel." Client expressed frustration that worry has not moved as far; we normalized the differential response and agreed the numbers support a shift in focus rather than a change in course. Treatment decision. Continue weekly CBT. Beginning next session, shift emphasis from behavioral activation (goals substantially met) to the worry exposure module targeting evening rumination, and add the sleep log given the static sleep item. Both measures to be re-administered in two weeks (due 08/05/2026). Goal 2 target on the treatment plan will be updated at the scheduled treatment plan review on 08/12/2026 to reflect the revised anxiety focus. Billing. 96127 x 2 units (PHQ-9, GAD-7) reported with today's psychotherapy session per payer contract; administration, scores, and review documented above. Signature. R. Alvarez, PhD, Licensed Psychologist, signed 07/22/2026, 4:35 PM. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every element an auditor would look for is present and linked: named instruments, administration method and date, verified scores, a dated trend, interpretation, client discussion, a treatment decision, and a billing line that matches what the note documents. The trend is shown as a four-administration flowsheet with dates, so the 8-point PHQ-9 change is visible at a glance and the differential response between depression and anxiety becomes the clinical story. Interpretation goes beyond the number: severity ranges, correlation with session presentation, and an item-level read (the static sleep item) that feeds directly into the plan. The risk item is reviewed and documented even though it is negative, showing the habit that protects the chart on the day an item 9 comes back positive. The client saw the scores, reacted to them, and shaped the decision, which is the difference between measurement-based care and data collection. The decision is specific and dated: a module shift, a re-administration date, and a hand-off to the scheduled treatment plan review where the goal target changes. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Completed instruments and their scores are ordinary clinical record material, and that cuts both ways. Clients can access them: the HIPAA right of access ( 45 CFR 164.524 ) reaches the designated record set, and scores cannot be tucked into psychotherapy notes because the psychotherapy-notes definition expressly excludes results of clinical tests and summaries of progress to date (45 CFR 164.501). Write every interpretation as a line the client may read. The same logic drives risk documentation: when an item-level response signals risk, the same-day conversation and its outcome belong in the record, with a suicide risk assessment or safety plan linked when one follows. In Canada no college requires outcome measures at all; Ontario's psychology and behaviour analysis college instead requires registrants to be "familiar with evidence-based tools and techniques" and to justify the decision when they choose not to use them ( CPBAO Standards 2024 , s. 10.4), while its retention standard keeps individual records at least 10 years after the last contact, or 10 years after a minor client turns 18. Australian records follow the whole-chart rule of 7 years from the last entry, or until age 25 for minors. On the payer side, sort each rule into its bin, because here the usual line runs backwards: the measurement is the requirement; the note format is the convention. The Joint Commission's standardized-instrument standard binds accredited organizations (accreditation policy), Medicare's collaborative care and behavioral health integration codes build validated rating scales and a registry into the service itself ( payer policy ), Australia ties an outcome measurement tool to preparing and reviewing the GP mental health treatment plan unless clinically inappropriate ( MBS note AN.0.56 ) and names specific instruments only in the PHN program data set ( PMHC MDS v5.0 : K10+, or K5 for Aboriginal and Torres Strait Islander clients, and the SDQ, at episode start and end at minimum), and US law reaches only coverage: screening recommendations graded A or B, including depression screening for adults and anxiety screening for adults 64 and younger ( USPSTF, June 2023 ), must be covered without cost sharing by non-grandfathered private plans under 42 USC 300gg-13. For the billing line itself: 96127 carries no physician work value in the fee schedule, is reported per instrument with no time threshold, and its Medicare per-day cap is 3 units with an appealable date-of-service edit; psychotherapy already includes continuing evaluation under NCCI policy, so same-day reporting with therapy codes is payer-specific; and since 2024 marriage and family therapists and mental health counselors enroll with Medicare directly (42 CFR 410.53, 410.54). Medicare's annual depression screening benefit runs under its own code, G0444. When the score triggers a goal change, document it in the treatment plan review so the measurement chain and the medical-necessity chain meet on paper. ## Common outcome measure review note errors auditors flag The audit record here is old, specific, and still the operative precedent. HHS-OIG's review of Medicare Part B mental health services ( OEI-03-99-00130 ) found 42 percent of psychological testing services inappropriate, almost half of them self-administered or self-scored screens such as the Beck Depression Inventory and Geriatric Depression Scale billed as testing when they belong inside the evaluation visit, and about a third of the inappropriate testing records carried no written interpretation of the results. The modern audits repeat the pattern at the treatment-plan layer: one New York provider's psychotherapy claims failed on all 100 sampled beneficiary days, largely on treatment plan and supervision requirements, for an estimated $1,118,789 in overpayments ( A-02-21-01006 ), and another was billed $3.9 million across 23,947 claims with findings that included treatment plan gaps and signatures stamped as digital images ( A-02-19-01012 ). The BastionGPT Clinical Advisory Board sees the same errors most often in outcome measure review note reviews: Self-scored screens billed as psychological testing. A PHQ-9 or Beck inventory reported under the timed testing codes instead of the brief-assessment code. NCCI policy is explicit that the testing codes cannot be reported for self-administered tests, and this exact substitution is what the OIG flagged. Screens ride 96127 or the visit itself. A score graveyard. Flowsheet cells fill up for months while the numbers never touch interpretation, the session narrative, or the plan. A trend nobody acts on documents that measurement changed nothing, which is worse at audit than no measurement at all. No evidence the client saw the result. The feedback literature keeps finding that one-time screening and results delivered outside the clinical encounter do not move outcomes, and a reviewer reads a chart with scores but no discussion the same way. One sentence of client response fixes it. Unit stacking past the cap. Four or five instruments reported on one date when the Medicare per-day edit for 96127 is 3 units, adjudicated per date of service; excess units need an appeal with documentation, not a modifier, and Medicaid programs publish their own separate edit tables. Signatures and score integrity that fail authentication. Unsigned entries, image-stamped signatures like the 109 claims in the On-Site audit, or a recorded total that cannot be reproduced from the completed instrument. The score is evidence; treat its chain of custody accordingly. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on outcome measure review notes. Extract instrument names, dates, and scores from a dictation, session transcript, or pasted portal export, and return them as a dated flowsheet line ready for the chart. Turn a scores flowsheet plus session bullets into the complete note: trend against baseline, interpretation in context, the client discussion, and the treatment decision, in your clinic's format. Check the note before you sign: a score with no comparison or interpretation, no documented client discussion, a positive risk item without follow-up, or a billing line the documentation does not support. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a required format for an outcome measure review note? No. No statute, CPT rule, accreditation standard, or professional college in the US, Canada, or Australia defines this note's fields or layout; every "required elements" list you will find online is a vendor synthesis. The requirements that exist mandate the measurement instead: the Joint Commission's standardized-instrument standard for accredited organizations, validated rating scales and a registry inside Medicare's collaborative care codes, and Australia's outcome-tool element in the GP mental health treatment plan. Most working notes run 75 to 300 words beside a score flowsheet and take 3 to 10 minutes, a clinical team estimate. What should an outcome measure review note include? Six elements cover what payers and auditors expect to find: the instrument name and version, the administration date and method, the score, the comparison with baseline and prior scores, your clinical interpretation, and the treatment decision that follows, plus one line showing the result was discussed with the client. The discussion line is the one most often missing, and the feedback evidence says results that never reach the clinical conversation do not change outcomes. How many units of 96127 can I bill per day? Medicare's practitioner Medically Unlikely Edit for 96127 is 3 units per date of service in the current quarterly table , with adjudication indicator 3: a per-day edit, so units past the cap are denied and recovery runs through appeal with documentation rather than a modifier. Medicaid programs publish separate MUE tables of their own, and commercial limits are contractual. Bill one unit per standardized instrument administered, scored, and documented in the note. Is 96127 time-based, and does it need a separate report? Neither. The code is reported per standardized instrument, not per minute; the minute thresholds that circulate on billing forums belong to the timed testing codes. It also carries 0.00 physician work RVUs in the Medicare fee schedule, which is the technical way of saying no separately valued interpretation or report exists: the clinician's read of the score is part of the visit it accompanies. A brief score-plus-interpretation line in the note is still good practice, and some payers ask for one. Can I bill 96127 with a psychotherapy session on the same day? Payer-specific. NCCI policy treats psychotherapy as including continuing psychiatric evaluation, and many payers bundle brief screens into the session on that logic; others pay separately. Check the current NCCI edits and your contract before building a workflow on the answer. Two related facts age quickly in vendor guidance: marriage and family therapists and mental health counselors have billed Medicare directly since 2024 (42 CFR 410.53 and 410.54), and Medicare's annual depression screening benefit runs under its own code, G0444, not 96127. How is this note different from a psychological testing report? Scale, administration, and interpretation. The testing codes 96130 to 96139 describe timed, clinician-administered services that include interpretation and a written report, and NCCI policy bars reporting them for self-administered tests; that work product is a psychological evaluation report . A brief self-report screen like the PHQ-9 is the opposite object: minutes long, self-scored, no separate report, documented in this note. The OIG's most on-point mental health audit found exactly this confusion, self-scored screens billed as testing, to be the largest source of inappropriate testing claims. Do I have to keep the paper questionnaire the client filled out? No regulation names the physical form; what the rules require is a record sufficient to support the care and the claim, retained on the whole-chart clock: commonly 5 to 10 years by state in the US, at least 10 years after the last contact (or after a minor client turns 18) under Ontario's college standard, and 7 years or until age 25 in Australia. If the completed instrument is the only place the item-level responses exist, keep it or its scanned image: scores are part of the accessible record, and clients can request them under HIPAA's right of access. Does Australia's Better Access program require an outcome measure? At the plan level, yes. The MBS explanatory notes make administering an outcome measurement tool a written element of preparing the GP mental health treatment plan unless clinically inappropriate, with the same tool re-administered at review ( note AN.0.56 ); the choice of tool is the clinician's, with the K10 and DASS-21 named as examples. The psychologist's session items carry no outcome-measure element; their documentation duty is the written report back to the referrer. The named-instrument mandate lives in PHN-commissioned services instead, where the PMHC minimum data set requires the K10+ (or K5) and the SDQ at episode start and end. How does BastionGPT help with measurement-based care documentation? Bring the scores however they exist: a portal export, a dictation, or session bullets. BastionGPT drafts the full review note with the trend, the interpretation, the client discussion, and the plan linkage, keeps the flowsheet current across administrations, and checks drafts for the gaps this page lists: a score with no comparison, an undiscussed result, a positive risk item without follow-up, or a billing line the documentation does not support. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Treatment Plan Treatment Plan Review & Update Relapse Prevention Plan Case Formulation Note Psychotherapy Progress Note Better Access Mental Health Treatment Plan Referral Collaborative Care (BHI/CoCM) Note Utilization Review Summary Psychological Testing Administration Note Psychological Testing Scoring Note NDIS Plan Reassessment & Progress Report Workers' Compensation Progress Report Diagnosis & Problem List Update (ICD) Eating Disorder Assessment PHQ-9 Score Documentation GAD-7 Score Documentation PCL-5 Score Documentation BDI-II Score Documentation DASS-21 Score Documentation Y-BOCS Score Documentation Insomnia Severity Index Documentation AUDIT-C Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation MADRS Score Documentation WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/better-access-treatment-plan All note templates # Better Access Mental Health Treatment Plan Referral: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A Mental Health Treatment Plan is the document an Australian GP or prescribed medical practitioner prepares under Better Access to assess a diagnosed mental disorder and support Medicare-rebated psychology sessions. A separate signed and dated referral, not the plan, authorises those sessions. GPs prepare both; the psychologist holds the referral and reports back after each course. Most plan-and-referral pairs run 600 to 1,200 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it A GP or prescribed medical practitioner; psychiatrists and paediatricians refer through their own pathway Audience The patient, the referred psychologist or allied mental health professional, the practice, and Medicare auditors Typical length 600 to 1,200 words across plan and referral · 25 to 45 minutes by hand (clinical team estimate) Format family Structured plan plus separate referral letter (also called MHTP, mental health care plan, GP mental health treatment plan) When it's used When a patient with a diagnosed mental disorder needs Medicare-rebated psychological treatment, then at reviews and each new course Standards context Medicare payer policy sets the required elements; no template, named tool, or plan format is mandated anywhere On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a Better Access Mental Health Treatment Plan? A Mental Health Treatment Plan (MHTP) is the structured document a GP or prescribed medical practitioner prepares to assess and manage a patient with a clinically diagnosed mental disorder under Better Access, the Medicare initiative that commenced on 1 November 2006. Preparation is billed under MBS items 2700, 2701, 2715, or 2717 (the higher-fee pair belongs to GPs with recognised mental health skills training), with video equivalents in the 92112 to 92123 range. You will also hear it called a mental health care plan, a GP MHTP, or informally "the 2700." The widely used templates come from the General Practice Mental Health Standards Collaboration (GPMHSC), established in 2002 and part of Better Access since 2006, and its guidance is blunt about their status: it is "not mandatory to use any particular form" , and it is not necessary to complete every field. The distinction this page exists to make: the plan and the referral are two separate documents, and clinicians conflate them constantly. MBS explanatory note MN.6.3 states that "a Mental Health Treatment Plan is not considered a referral." The plan documents the assessment and the shared roadmap; only the written, signed, and dated referral authorises Medicare-rebated psychology sessions, and the psychologist must hold that referral at the first consultation. Almost everything else clinicians treat as mandatory sits a rung lower: the required elements are Medicare payer policy set out in explanatory notes, while the template, the K10 specifically, and the patient signature block are convention. For the clinician-authored planning document used in US practice, see the treatment plan page; this page covers the Australian pathway pair. ## Who uses Better Access treatment plans and when Both sides of the referral use this documentation. On the referring side, a GP or prescribed medical practitioner at the patient's MyMedicare-registered practice, or the patient's usual medical practitioner, prepares the plan, issues the referral, and reviews progress through time-tiered attendance items; psychiatrists and paediatricians run a parallel pathway through the Psychiatrist Assessment and Management Plan (item 291, with a written report back to the GP within two weeks) or direct referral. On the receiving side, eligible clinical psychologists deliver psychological therapy, while eligible registered psychologists, mental health occupational therapists, and accredited mental health social workers deliver focussed psychological strategies; their documentation duties are holding the referral and writing the end-of-course report. Keep the neighbours straight: the psychologist's own goal-and-intervention document is a treatment plan , each session generates a progress note , and the scheduled re-check of goals is a treatment plan review . None of those replace the MHTP or the referral. ## Better Access plan and referral structure: what goes in each section Patient details and recorded agreement. Explain the service, record the patient's agreement to proceed, and note that a copy of the plan was offered to the patient (and carer where appropriate) before the item is claimed; a copy also goes in the record. These are claim conditions in note AN.0.56, not courtesies. Pitfall: billing the preparation item before the copy is offered or with no recorded agreement; the signature block on the GPMHSC template is convention, the recorded agreement is not. Assessment with an outcome tool. Administer an outcome measurement tool as part of the assessment unless it is clinically inappropriate, and record the score. The K10 and DASS 21 are the named examples; the choice of evidence-based tool is the practitioner's. The same tool is re-administered at review, so pick one you can repeat. Pitfall: skipping the tool without documenting why it was clinically inappropriate; the myth that no outcome measure is required reads the current note backwards. Diagnosis and formulation. A formulation of the mental disorder including a provisional or formal diagnosis, with the presenting history that supports it. The optional IAR-DST can inform the level-of-care decision. Pitfall: a symptom list with no formulation or diagnosis recorded; the item requires one, even provisional. Goals, treatment, and patient actions. Treatment options discussed, goals agreed with the patient, patient actions, and the education you provided, all recorded in writing in the plan. Pitfall: goal text identical from patient to patient; non-individualised plans are the recurring theme of Professional Services Review findings on these items. Crisis and relapse planning. Create a plan for crisis intervention and/or relapse prevention, plus the referrals, supports, review, and follow-up you arranged. A structured relapse prevention plan can sit behind this section. Pitfall: the heading left blank or marked N/A with no reasoning while the rest of the template is complete. The referral, a separate document. In writing, signed and dated by the referrer (electronic signature is fine), containing the patient's name, date of birth, and address; symptoms or diagnosis; a list of current medications; the number of services this course; and a statement that an MHTP or PAMP has been prepared. The plan can travel with it only where appropriate and with the patient's agreement. Pitfall: sending the plan and calling it the referral; MN.6.3 says it is not one, and the psychologist cannot process rebates against it. The report back. At the end of the initial course, the treating psychologist or allied mental health professional writes to the referrer: assessments carried out, treatment provided, and recommendations on future management. This report is what the GP weighs before referring a subsequent course. Pitfall: no written report on file at course completion; the obligation belongs to the treating clinician, not the GP. ### Blank template (copy and adapt) GP MENTAL HEALTH TREATMENT PLAN (Better Access) Patient: [initials] DOB: Plan date: Referrer: GP / PMP at MyMedicare-registered practice or usual practitioner [ ] Agreement to proceed recorded: [ ] Copy offered to patient: [ ] Outcome tool + score (or why clinically inappropriate): Formulation / diagnosis (provisional or formal): Presenting issues and relevant history: Goals agreed with the patient: Treatment, patient actions, education provided: Crisis intervention / relapse prevention plan: Referrals, supports, review and follow-up arranged: GP/PMP name and signature: Date: REFERRAL (separate document; the plan is not a referral) Patient name / DOB / address: Symptoms or diagnosis: Current medications: Number of services this course: [initial course maximum 6] MHTP or PAMP prepared: [ ] Plan attached with patient agreement: [ ] Referrer signature (electronic accepted): Date: END-OF-COURSE REPORT (treating clinician to referrer) Assessments carried out: Treatment provided: Recommendations on future management: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample Better Access treatment plan and referral Scenario: adult patient with generalised anxiety disorder, plan and referral prepared by the regular GP, initial course of six sessions with a registered psychologist. All details are fictional. GP Mental Health Treatment Plan. Patient: J.M., 34 · Plan date: 15/07/2026 · Referrer: Dr A.P., the patient's usual medical practitioner (MyMedicare-registered practice) Agreement and copy: Service explained; J.M.'s agreement to proceed recorded. Copy of the completed plan offered and accepted; copy filed in the practice record. Assessment: Six months of persistent worry, initial insomnia, muscle tension, and irritability, worse with workload; no prior mental health treatment. K10 administered today: 27 (high distress). Alcohol 2 to 4 drinks weekly; no other substances. No current thoughts of self-harm or harm to others; protective factors include stable housing and a supportive partner. Formulation and diagnosis: Generalised anxiety disorder (formal diagnosis). Longstanding worry style with recent occupational stressors; physical review and thyroid screen arranged to exclude contributors. Goals agreed with J.M.: (1) reduce K10 from 27 to below 20 by plan review; (2) return to full work duties without midweek absences; (3) restore sleep to 6.5 hours or more most nights. Patient actions: daily 20-minute walk, caffeine after midday ceased, worry-time exercise as taught. Education: anxiety cycle and treatment options discussed; written psychoeducation provided. Crisis and relapse arrangements: No current risk identified. Early warning signs agreed (two consecutive nights of minimal sleep, missed work days); J.M. will contact the practice or the after-hours GP service if they appear. Follow-up: referral below; plan review at this practice in approximately 10 weeks; the K10 will be re-administered at review. Referral for Better Access psychological treatment. To: Harbourside Psychology (registered psychologist) · Dated and signed electronically: Dr A.P., 15/07/2026 Patient: J.M., DOB 03/02/1992, 14 Example St, [suburb]. Symptoms and diagnosis: generalised anxiety disorder as formulated in today's Mental Health Treatment Plan. Current medications: nil regular. Referred for: 6 individual focussed psychological strategies sessions (initial course). A Mental Health Treatment Plan has been prepared and is held at this practice; a copy is attached with the patient's agreement. End-of-course report (psychologist to GP, excerpt). 25/09/2026 Sessions 1 to 6 completed. Assessments: K10 at intake 26, at session 6 it is 19. Treatment: focussed psychological strategies (CBT: cognitive restructuring, graded worry exposure, sleep hygiene). Recommendations: gains are consolidating but incomplete; a subsequent course of 4 sessions this calendar year is recommended, then review against plan goals at the practice. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The two claim conditions are visible on the face of the plan: the patient's agreement to proceed is recorded, and a copy was offered before the item is claimed. An outcome tool is named with its score at baseline, the review will re-administer the same tool, and the goals quantify against it, so change is auditable instead of asserted. The plan is unmistakably individualised: this patient's stressors, drinking pattern, sleep numbers, and work goals, not text a reviewer could find in another chart. The referral stands alone with every required element, including the session count and the plan statement, so the psychologist can bill the first session without chasing the practice. The end-of-course report closes the loop with the three required contents, giving the GP an evidence base for the subsequent-course decision. Crisis and relapse arrangements are completed with concrete early warning signs, not a heading left blank. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The record side is where audits live. The Department of Health, Disability and Ageing runs regular post-payment auditing of these items, and note AN.0.78 expects adequate and contemporaneous records naming the date, time, and who attended; only details recorded at the time count toward a time-tiered attendance. The plan must sit in the practice record with the offered-copy step documented. On the receiving side, the psychologist must hold the referral at the first consultation and retain it for 2 years from the first service, write the end-of-course report, and keep the whole record on the ordinary clinical clock: 7 years from the last entry in Australia, or until age 25 for clients who were minors. Between courses, the working documents are the session progress notes and the outcome measure review note tracking the same tool the plan named. The payer mechanics changed on 1 November 2025 and plenty of circulating guidance predates them. Plan preparation, review, and referral now belong to a GP or PMP at the patient's MyMedicare-registered practice or the patient's usual medical practitioner, and the dedicated review items are gone: reviews run through time-tiered general attendance items, no more than once in 3 months and not within 4 weeks of a plan claim, with most patients needing no more than 2 review attendances a year. Telehealth preparation requires a face-to-face service with the practitioner or practice in the prior 12 months, and a previous video or phone consultation does not count (note AN.1.1 ). Session arithmetic: 10 individual and 10 group services per calendar year, an initial course of at most 6, unused referred sessions rolling over but counting against the new year, and an eating disorder plan blocks a new or reviewed MHTP for 12 months absent exceptional circumstances. Keep the strength of each rule straight, because the format is a convention and the content is the requirement. Law sits in the Health Insurance Act instruments and privacy statutes; payer policy is the explanatory-note element lists that decide whether the benefit is payable; convention is the template itself, the K10 specifically, and the signature block. What no rule requires: a particular form, a minimum plan length, an annual plan remake, or sending the plan (rather than the referral) to the psychologist. ## Common Better Access documentation errors auditors flag The Department undertakes regular post-payment auditing of Better Access claims, and Australia's Professional Services Review has repeatedly made findings of inappropriate practice around the plan items where records were not individualised for the patient. The system also leaks patients: in the Better Access evaluation analysis by Tapp and colleagues (ANZJP, 2026), 58.8% of people referred in 2021 went on to any treatment session, down from 66.9% in 2018, with a median 22-day wait to the first session, so documentation that stalls a rebate has a real clinical cost. The BastionGPT Clinical Advisory Board sees the same errors most often in Better Access reviews: The plan billed or sent as if it were the referral. Sessions delivered with no compliant referral in hand cannot be rebated, and the repayment risk lands on the treating clinician. Template text not individualised. Identical goals, actions, and formulations across charts are the pattern Professional Services Review findings keep naming; the template is optional, the individualisation is not. No outcome tool and no reason. The assessment element requires an outcome measurement tool unless clinically inappropriate, and the inappropriateness has to be documented, not assumed. Session counts unmanaged. A referral missing its service number, a count above the course maximum, or rollover sessions quietly breaching the new year's cap all require contacting the referrer, not billing through. Reviews mistimed and plans remade annually. Reviews within 4 weeks of the plan claim or more than once in 3 months, and fresh plans written each January for a document that does not expire, both read as churning to an auditor. ### Better Access treatment plans in the US, Canada, and Australia Aspect United States Canada Australia Status No legal force; the nearest analogue is the payer-convention treatment plan that supports medical necessity No national equivalent; psychology access runs through provincial programs, employer benefits, and private pay Payer gateway: the plan plus a compliant referral decide whether Medicare rebates psychology sessions Terminology Treatment plan, individualized treatment plan (ITP), plan of care Care plan, service plan; physician referral letters are common but gate nothing federally Mental Health Treatment Plan (MHTP), mental health care plan, Better Access referral, Psychiatrist Assessment and Management Plan (PAMP) What changes No session cap tied to a physician gateway document; payer contracts and state rules shape plan content and review Provincial and college expectations shape planning documents; referral requirements vary by program and insurer Referrer must be the MyMedicare practice or usual practitioner; 10 individual plus 10 group sessions per calendar year, initial course capped at 6; written report back to the referrer Retention Part of the chart; state law governs CMPA recommends at least 10 years from last entry, 16 years in British Columbia, longer for minors Whole record 7 years since last entry, until age 25 for clients under 18; the psychologist keeps the referral itself at least 2 years An MHTP has no effect outside Australia: it cannot generate a US or Canadian payer entitlement, and a US or Canadian clinician meets one only as inbound context for an Australian client. What all three systems share is the logic underneath: an assessed need, an individualised plan, and documentation that lets a payer trace the thread. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on Better Access treatment plans and referrals. Draft the plan and the separate referral from your consult notes or dictation, with the assessment, agreed goals, and crisis section individualised to the patient in front of you. Check a referral against the required elements and the session arithmetic before the first rebate is processed, and flag a missing outcome tool or an undocumented agreement before you claim the plan item. Convert between the Australian pathway documents and US or Canadian treatment plan formats, and draft the end-of-course report back to the referrer from your session notes. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does a Mental Health Treatment Plan expire? No. The plan (and a psychiatrist's assessment and management plan) does not expire, a new plan is not needed each calendar year, and one should not be written at all absent exceptional circumstances such as a significant change in the patient's mental health. When referred sessions run out, the patient needs a new referral, not a new plan, and a new plan generally should not be prepared within 12 months of the previous one. Annual plan remakes are a billing habit, not a rule. Is the Mental Health Treatment Plan the same thing as the referral? No, and the distinction decides whether anyone gets paid. MBS note MN.6.3 states that "a Mental Health Treatment Plan is not considered a referral." The referral is a separate written document, signed and dated, carrying the patient's details, symptoms or diagnosis, current medications, the number of services, and a statement that a plan exists. The psychologist must hold it at the first consultation; the plan itself travels only where appropriate and with the patient's agreement. How many sessions does Better Access cover? Up to 10 individual and 10 group mental health treatment services per calendar year. The initial course of treatment is capped at 6 individual services; a subsequent course takes the patient up to the annual 10. Unused referred sessions roll into the new year without a new referral, but they count against that year's cap. The referring practitioner sets the number for each course, and the treating clinician must confirm it with the referrer when it is missing or above the maximum. Is the K10 or the official template required for the plan? Neither, and both myths cost clinicians time. The GPMHSC, which publishes the standard templates, states it is "not mandatory to use any particular form" and that completing every field is unnecessary. What the current MBS note does require is an outcome measurement tool at preparation and again at review, unless clinically inappropriate; the K10 and DASS 21 are named examples and the choice belongs to the clinician. Between plan reviews, track the scores with an outcome measure review note . Who can prepare, review, or refer after the November 2025 changes? A GP or prescribed medical practitioner at the patient's MyMedicare-registered practice, or the patient's usual medical practitioner, meaning the one providing the majority of care across a 12-month window. The dedicated review items are retired: reviews and ongoing mental health consultations now run through time-tiered general attendance items. Telehealth preparation requires a face-to-face service with the practitioner or practice in the prior 12 months, and a previous video or phone appointment does not satisfy that. Psychiatrists and paediatricians still refer directly or through an assessment and management plan. What does the psychologist have to keep and send? Three obligations. Hold a compliant referral at the first consultation, because rebates cannot be processed without it; retain that referral for 2 years from the date of the first service; and at the end of the initial course, send the referrer a written report covering assessments carried out, treatment provided, and recommendations on future management. The plan itself is optional on the psychologist's file. Session records remain ordinary progress notes on the usual clinical clock. What if the referral names the wrong number of sessions, or none? Contact the referrer, then bill. The allied health professional must confirm the intended number when the referral omits it or exceeds the course or annual maximum; if the referrer cannot be reached, the clinician may proceed on clinical judgment within the caps. Services Australia's provider enquiry line can confirm a patient's remaining allocation. Billing through a defective referral is the real error, because the repayment risk lands on the treating clinician. Do US or Canadian clinicians ever use a Mental Health Treatment Plan? Only as inbound context, for example an Australian client arriving with their records. The MHTP is a Medicare construct with no legal force in the US or Canada, and services delivered outside Australia cannot generate a Better Access rebate. The working analogue in US practice is the payer-facing treatment plan , which shares the medical-necessity logic but gates no access to a provider. How does BastionGPT help with Better Access documentation? Bring the consult however it exists: dictation, bullets, or the previous plan. BastionGPT drafts the individualised plan, the separate referral with every required element, and the end-of-course report back to the GP, then checks the pieces auditors read first: the recorded agreement, the outcome tool and score, the session arithmetic, and whether the goal text is actually this patient's. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Treatment Plan Treatment Plan Review & Update Psychotherapy Progress Note Outcome Measure Review Note Relapse Prevention Plan NDIS Functional Capacity Assessment Report Australia Telehealth & MBS Case Conference Note NDIS Plan Reassessment & Progress Report Aboriginal & Torres Strait Islander Culturally Safe Assessment Referral Letter GAD-7 Score Documentation DASS-21 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/adhd-evaluation-report All note templates # ADHD Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An ADHD evaluation report is the written product of a diagnostic assessment for attention-deficit/hyperactivity disorder. It integrates the referral question, history, observations, and multi-informant rating scales into a DSM-5-TR criteria mapping, a differential, a diagnosis, and recommendations. Psychologists, psychiatrists, pediatricians, and primary care prescribers write them for children and adults. Most run 3 to 10 pages and are shared with families, schools, and prescribers. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and neuropsychologists; psychiatrists, pediatricians, and primary care prescribers documenting their own diagnostic assessments Audience The referrer and the prescriber, the client and family, schools on request, payers and auditors; in Australia, the state unit behind a stimulant authority Typical length 1,000 to 3,500 words (3 to 10 pages) · 2 to 5 hours of interviews, scoring, and writing (clinical team estimate) Format family Interpretive assessment report (compare: psychological, psychoeducational, and neuropsychological reports) When it's used When attention, hyperactivity, or executive-function complaints need a diagnostic answer: first diagnosis, medication decisions, school or workplace supports, or a second opinion Standards context No law mandates the format; AAP, CADDRA, and AADPA guidelines set the convention, and stimulant-authority rules give the content legal teeth On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an ADHD evaluation report? An ADHD evaluation report is the written product of a diagnostic assessment for attention-deficit/hyperactivity disorder. It integrates a clinical and developmental interview, standardized rating scales from more than one informant, and any cognitive or achievement testing into an explicit mapping of findings onto DSM-5-TR (or ICD) criteria, then states a diagnosis, the differential that was considered, and recommendations. Clinicians and payers also call it an ADHD assessment report, an ADHD diagnostic assessment, or a comprehensive ADHD evaluation; when a full test battery drives it, it becomes a psychological evaluation report billed under the testing codes. It is a document family rather than a single instrument: the rating-scale tradition runs from the Conners scales of the 1960s to today's Conners 4 and CAARS 2 , and the guideline scaffolding comes from the American Academy of Pediatrics , which first published ADHD evaluation recommendations in 2000 and last revised them in 2019, Canada's CADDRA practice guidelines (4.1 Edition, January 2020), and Australia's NHMRC-approved AADPA guideline , developed from systematic reviews new and updated from NICE 2018. Here is the fact that organizes everything else on this page: no statute, regulator, or payer in the US, Canada, or Australia mandates the report's format, and yet this document has more legal teeth than almost any other note a mental health clinician writes, because it is what stands behind a controlled-substance prescription. Australian state law makes that explicit: NSW defines ADHD for stimulant-authority purposes as a diagnosis meeting DSM-5 or ICD-11 criteria, and prescribing runs through a class authority that names who may hold it. In the US the diagnosis itself needs no physician signature, but prescribing does, and federal telemedicine flexibilities for prescribing stimulants without a prior in-person exam currently run only through December 31, 2026. The report is also narrower than its neighbors: an interview-based first diagnosis at the start of routine care is a psychiatric diagnostic evaluation , a school-eligibility evaluation is a psychoeducational report , and this page covers the focused diagnostic document that answers one question: does this person meet criteria for ADHD, and what should happen next? ## Who uses ADHD evaluation reports and when Psychologists and neuropsychologists write them in private practice and clinics; psychiatrists, pediatricians, and primary care prescribers write leaner versions documenting their own diagnostic assessments, which is where most childhood ADHD is actually diagnosed. The numbers explain the volume: about 1 in 9 US children has ever received an ADHD diagnosis, and an estimated 15.5 million US adults had a current diagnosis in 2023, about half of them diagnosed in adulthood. The report is the right tool when attention, hyperactivity, or executive-function complaints need a diagnostic answer: a first diagnosis in a child whose teacher and parent see different things, an adult seeking evaluation after years of workplace difficulty, a medication decision that needs a documented basis, or a second opinion after an online diagnosis. For children the evidence spine is multi-informant rating data like the Conners 4 with a BRIEF-2 for executive function; for adults it is the CAARS 2 with collateral input where available; cognitive context from a WISC-V or WAIS-5 joins only when the referral question asks for it. When the question is school eligibility rather than diagnosis, the psychoeducational report takes over; when the developmental question is broader than attention in a young child, a developmental assessment comes first; and when the referral spans injury, decline, or a full battery, the neuropsychological or psychological evaluation report is the wrapper. The audience is wider than the chart: the referrer and the prescriber read it to act, parents and adult clients read it to understand, schools read it to plan supports, and in Australia a state pharmaceutical-services unit may read it before a stimulant authority issues. ## ADHD evaluation report structure: what goes in each section No authority mandates a single format. The sections below are the consensus skeleton of the AAP, CADDRA, and AADPA guideline tradition and the elements payers and stimulant-authority reviewers expect to find, each with the pitfall that most often undermines it. Identifying information and referral question. Client, date of birth, evaluation dates, evaluator and credentials, referrer, and the question in one sentence: who is asking, and what decision does the answer feed? Pitfall: a report with no referral question cannot show medical necessity, and payers read testing done purely for school accommodations as an educational service, not a covered one. Sources of information and consent. List every source: interviews and dates, informants, rating scales with versions, tests, records reviewed, and who consented. Note the modality if any part ran by telehealth. Pitfall: an unlisted source cannot support a criterion later; if a teacher scale never came back, say so and describe the workaround. History. Developmental, medical, psychiatric, family, and school or work history, including sleep, substance use, prior evaluations, and treatment. For adults, evidence of childhood symptoms: report cards, a parent's account, an old evaluation. Pitfall: onset undocumented. Symptoms must be traceable to before age 12, and "client recalls being distractible" carries less weight than a named source. Behavioral observations. What you saw across sessions: attention, activity, effort, rapport, and whether results looked valid. One line on interpretability protects every score that follows. Pitfall: observations that contradict the conclusion with no comment; a child who sustained attention for 90 minutes of one-on-one testing needs a sentence explaining why that does not rule out ADHD in a classroom of 25. Rating-scale results, by informant. Each instrument with version, informant, setting, and scores presented the way the manual scales them, then the pattern: where informants agree, where they diverge, and what the divergence means. Pitfall: one informant, one setting. The AAP guideline expects "documentation of symptoms and impairment in more than 1 major setting," and a parent-only profile leaves the second setting unevidenced. Cognitive and achievement context, when tested. If a WISC-V , WAIS-5, or achievement measure was given, report it here as context: no cognitive profile diagnoses ADHD, but the results answer rule-out questions and shape recommendations. Pitfall: implying IQ or achievement testing was required for the diagnosis; no guideline in the US, Canada, or Australia requires it, so justify the battery you gave. DSM-5-TR criteria mapping and differential diagnosis. The engine of the report: criterion by criterion, which symptoms are endorsed, by whom, in which settings, since when, with what impairment, and what else could explain them: sleep, anxiety, depression, learning problems, substance use, medical causes. Pitfall: scores treated as the verdict. A rating scale is evidence toward criteria; the mapping is the diagnosis, and a report that skips from T-scores to diagnosis will not survive a records review. Diagnosis and formulation. The conclusion with presentation and severity, the ICD-10-CM code payers will see (F90.0, F90.1, F90.2, F90.9), conditions ruled out, and a short formulation connecting history, data, and impairment. State it even when the answer is no ADHD; a documented negative protects the client and the next evaluator. Pitfall: hedged conclusions ("consistent with possible ADHD") that neither support treatment nor close the question. Recommendations, feedback, and signature. Ranked, specific, and matched to the findings: behavioral treatment, school or workplace strategies, medication discussion with the prescriber where indicated, and follow-up. Document the feedback session, who received the report, time for any time-based testing codes, and sign with credentials and date. Pitfall: a psychologist's report that reads as a prescription; diagnosis is yours, prescribing is the physician's, and in Australia the state authority's. ### Blank template (copy and adapt) ADHD EVALUATION REPORT Client: ______________________ DOB: ____________ Age: ______ Evaluation date(s): ______________ Report date: ____________ Evaluator / credentials: ______________________________________ Referred by: ____________________ Modality: [ ] In person [ ] Telehealth REFERRAL QUESTION _____________________________________________________________ SOURCES OF INFORMATION AND CONSENT Interviews / informants: ____________________________________ Rating scales (version, informant, setting): ________________ Tests administered: ______________ Records reviewed: ________ Consent obtained from: ______________________________________ HISTORY (developmental, medical, psychiatric, family, school/work) _____________________________________________________________ Evidence of symptoms before age 12: _________________________ BEHAVIORAL OBSERVATIONS / VALIDITY STATEMENT _____________________________________________________________ RATING-SCALE RESULTS (by informant and setting) _____________________________________________________________ COGNITIVE / ACHIEVEMENT CONTEXT (if tested) _____________________________________________________________ DSM-5-TR CRITERIA MAPPING AND DIFFERENTIAL Inattention: ____ of 9 endorsed. Sources: ___________________ Hyperactivity-impulsivity: ____ of 9. Sources: ______________ Settings with impairment: ______________ Onset: ____________ Alternatives considered / ruled out: ________________________ DIAGNOSIS AND FORMULATION Diagnosis (ICD-10-CM code): _________________________________ Formulation: ________________________________________________ RECOMMENDATIONS, FEEDBACK, AND FOLLOW-UP 1. ___________________________________________________________ 2. ___________________________________________________________ Feedback session / report shared with: ______________________ Testing time (if time-based codes billed): __________________ Signature / credentials: ______________ Date: ______________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample ADHD evaluation report Scenario: a licensed psychologist evaluates a 9-year-old referred by her pediatrician after two years of classroom attention concerns. A full report runs 3 to 10 pages; this version is condensed but structurally complete. Use it as an ADHD evaluation report sample you can adapt for children or adults. All details are fictional. ADHD EVALUATION REPORT (CONFIDENTIAL) Client: M.R., age 9, grade 4 · DOB: 03/2017 Evaluation dates: 07/07/2026 and 07/14/2026 · Report date: 07/21/2026 Evaluator: L. Chen, PhD, Licensed Psychologist · Referred by: A. Okafor, MD (pediatrician) Referral question. Dr. Okafor asks whether M.R. meets criteria for ADHD and what supports are indicated, ahead of a medication discussion the family has requested. Sources of information and consent. Two in-person sessions with M.R.; parent interview with mother; developmental and medical history questionnaire; Conners 4 (parent and teacher forms); BRIEF-2 (parent form); WISC-V (selected indexes); review of report cards, grades 1 to 4, and the pediatrician's referral note. Teacher forms completed by the grade 3 teacher in June, before the school break. Written consent from mother; assent from M.R. History. Unremarkable pregnancy and delivery; milestones on time. Teachers have noted incomplete work and difficulty sustaining attention since grade 1; report cards repeat "loses focus" and "needs reminders" across three years. Homework requires constant supervision at home. Hearing and vision screened normal 04/2026. Sleep is adequate on the family's account (about 10 hours, regular schedule). No medications, no prior mental health treatment, no significant medical history. Family history includes attention problems in father, never formally evaluated. No recent stressors or changes at home. Behavioral observations. M.R. was friendly and engaged. She required repetition of instructions on longer tasks, left her seat twice per session, and lost track of multi-step directions, but persisted with encouragement. Effort was adequate across both sessions; results below are considered interpretable. Sustained one-on-one attention here does not contradict classroom reports; demands differ sharply between settings. Rating-scale results. Conners 4 parent form: Inattention and Executive Function scales Very Elevated (T = 78 and 74); Hyperactivity T = 62. Conners 4 teacher form: Inattention Elevated to Very Elevated (T = 71); Hyperactivity within the average range. BRIEF-2 parent form: clinically meaningful elevations on Working Memory and Plan/Organize indexes. The two informants agree on inattention across home and classroom and agree that hyperactivity is not the primary picture. Cognitive context. WISC-V estimated overall ability in the average range (FSIQ 102); Working Memory Index 88, a relative weakness consistent with the rating-scale profile. Academic skills per records and curriculum measures are broadly at grade level, with written output the weakest area. No learning disorder is identified on this screening; fuller achievement testing was not indicated by the referral question. DSM-5-TR criteria mapping and differential. Inattention: 7 of 9 symptoms endorsed by parent and 6 of 9 by teacher, present in both home and classroom. Hyperactivity-impulsivity: 3 of 9, below threshold. Onset: documented from grade 1 (age 6). Impairment: incomplete classwork, homework supervision burden, rising frustration around school. Alternatives considered: anxiety (subclinical on interview and scales), learning disorder (not supported by screening), sleep disruption (not supported by history), medical causes (screened by pediatrician). None better explains the pattern. Diagnosis and formulation. ADHD, predominantly inattentive presentation, moderate (F90.0). A capable child whose working-memory weakness and inattention undermine performance as academic demands grow; strengths include verbal reasoning, warm family support, and willing engagement. Recommendations. (1) Share this report with the school, with parental consent, to support classroom strategies and consideration of a formal support plan. (2) Parent behavior-management training; resources provided. (3) Return to Dr. Okafor for the medication discussion; this report documents the diagnostic basis. (4) Re-evaluate response in 3 to 6 months; repeat teacher ratings once new-year placement settles. Feedback and signature. Findings reviewed with mother 07/21/2026; plain-language summary provided. Report released to Dr. Okafor with consent. L. Chen, PhD, Licensed Psychologist, signed 07/21/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral question gets a direct answer: a named diagnosis with its code, plus the "what happens next" the referrer asked for. Two informants, two settings: every criterion in the mapping traces to a source, and the sustained-attention observation that could contradict the conclusion is addressed rather than ignored. Onset and impairment are evidenced, not asserted: three years of report cards, not a recollection. The differential is written down, including what was ruled out and on what basis, which is what separates an evaluation from a score printout. The prescribing boundary is clean: the psychologist diagnoses and documents; the medication decision returns to the physician with the evidence in hand. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the report knowing who will read it. It sits in the designated record set, so the client, or a parent for most minor clients, holds a right of access under 45 CFR 164.524 , and a finalized electronic report is releasable by default through the portal, often before the feedback session. Release to a school is a disclosure the parent authorizes, and it changes the audience: a report headed for a student support meeting should carry its conclusions in plain language and keep raw data out of the narrative. The rules for releasing test data versus test materials are the same as for any testing report; the psychological evaluation report page covers them in detail. Retention runs on the whole-chart clock, which is state or provincial law rather than a single national rule: commonly 5 to 10 years by state in the US, with the APA's record-keeping guideline suggesting 7 years, or 3 years past majority for minors , as the default; at least 10 years after last contact, or 10 years past the 18th birthday, under Ontario's college standard; and 7 years for adults or until age 25 for minors in Australian state regimes. An ADHD report written for a child will be requested again a decade later, for a college accommodation or an adult re-evaluation, so write it to be read cold. The compliance pressure sits in two places, and neither is the format. First, payment. In the US the interview-based diagnostic evaluation bills 90791 or 90792, brief rating scales ride under 96127 where payers allow, and the timed testing codes (96130, 96131, 96136, 96137) apply only when standardized instruments are administered and interpreted; contractor policy is blunt that self-scored inventories are not testing, that testing "administered for educational or vocational purposes" without a medical-management purpose is excluded, and that time must be documented for time-based codes. In Australia there is no dedicated ADHD-assessment item; psychiatrist assessments run under attendance items like 291, where the explanatory note makes the document itself the deliverable: "The detailed report is a fundamental component of this item" (MBS note AN.0.30). Second, prescribing. The report is what a prescriber, and in Australia a state pharmaceutical unit, relies on before a stimulant is authorized: NSW requires a diagnosis meeting DSM-5 or ICD-11 criteria under its class authority, Queensland's December 2025 reform lets specialist GPs initiate psychostimulants for adults on the strength of a documented assessment, and US telemedicine prescribing without a prior in-person exam runs on DEA flexibilities that expire December 31, 2026 unless extended or replaced. Those rules change month to month, so date-stamp any prescribing statement in your report. The format is a convention; the content, and what it authorizes, is the requirement. ## Common ADHD evaluation report errors auditors flag ADHD evaluations are under enforcement scrutiny that most document types never see. In June 2024, federal agents arrested the founder and the clinical president of the telehealth company Done, whose platform, per the government's charges, arranged prescriptions of over 40 million pills of Adderall and other stimulants for more than $100 million in revenue on a monthly-subscription model with an auto-refill function; a jury convicted both in November 2025 of conspiracy to distribute controlled substances and health care fraud conspiracy. The lesson for legitimate practice is not that telehealth evaluation is suspect; it is that the evaluation document is the control point regulators read, and a thin one now carries real risk. The BastionGPT Clinical Advisory Board sees the same errors most often in ADHD evaluation report reviews: One informant, one setting. A parent-only or self-report-only profile with nothing from a teacher, partner, or workplace. The criteria demand evidence across settings, and single-source reports are the first thing a reviewer notices. Scores treated as the diagnosis. Rating-scale elevations reported, then a diagnosis, with no criterion-by-criterion mapping between them. Scales are evidence toward criteria, not verdicts, and no scale score is diagnostic by itself. Onset asserted, not evidenced. "Symptoms since childhood" with no report card, no parent account, no record. Adult evaluations fail here most: the before-age-12 thread is the criterion auditors and stimulant-authority reviewers can actually check. The missing differential. No documented consideration of sleep, anxiety, depression, learning problems, substance use, or medical causes. A report that never says what else was considered reads as a foregone conclusion. The purpose problem. Testing driven by an accommodation request billed to insurance as medically necessary. Payers exclude purely educational or vocational testing, and the referral question is where they look first. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on ADHD evaluation reports. Draft the full report from your interview notes, dictation, and score summaries: history, observations, criteria mapping, differential, and recommendations in your structure, ready for your clinical judgment. Extract multi-informant rating-scale results into a clean by-informant, by-setting summary, and turn the findings into a plain-language version for parents or the client. Check a draft before you sign: an unanswered referral question, a criterion with no named source, missing onset or second-setting evidence, or a battery the report never justifies. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a required format for an ADHD evaluation report? No. No statute, regulator, or payer in the US, Canada, or Australia prescribes the report's title, section order, or length; the familiar structure is professional convention set by the AAP guideline in the US, CADDRA's guidelines in Canada, and the NHMRC-approved AADPA guideline in Australia. The binding rules attach to content and consequences instead: payer documentation requirements, retention and access law, and the stimulant-authority rules that decide what the report must establish before prescribing. Most reports run 3 to 10 pages and take 2 to 5 hours of interviewing, scoring, and writing, a clinical team estimate. What does an ADHD evaluation report need to support a stimulant prescription? It depends on the jurisdiction, and the answers are moving. In the US a psychologist's diagnosis is valid, but prescribing needs an authorized prescriber, and DEA flexibilities allowing telemedicine stimulant prescribing without a prior in-person exam run only through December 31, 2026 unless extended or replaced. In Australia the report feeds a state authority: NSW defines ADHD as a diagnosis meeting DSM-5 or ICD-11 criteria under its class authority for psychiatrists, paediatricians, and neurologists, and Queensland lets specialist GPs initiate psychostimulants for adults from 1 December 2025 , with other states phasing in GP roles on their own timetables. In Canada, physicians prescribe; a psychologist's report informs but cannot authorize. Date-stamp any prescribing statement you make. Do I need a continuous performance test, IQ testing, or a computer-based test to diagnose ADHD? No guideline in the US, Canada, or Australia requires one for a routine clinical diagnosis. The strongest endorsement any of these tools has is as an add-on: NICE's 2024 guidance names QbTest an option for ages 6 to 17 and says "It should only be used with standard clinical assessment by a healthcare professional," with adult use still research-only, and the systematic review behind that decision concluded it "should not be used as stand-alone screening or diagnostic tool." The AAP guideline likewise reports that neuropsychological testing "has not been found to improve diagnostic accuracy in most cases." Cognitive testing joins when the referral question needs it, and high-stakes accommodation or licensing contexts effectively expect more, including validity data; a routine diagnosis needs a thorough clinical assessment, not a specific instrument. Are teacher rating scales required, and what about adult evaluations? No law requires a teacher scale for anyone. For children it is strong guideline convention: the AAP expects DSM-based rating scales from parents and school sources, because the criteria demand evidence from more than one setting, and for adolescents it suggests reaching at least two teachers or other adults who see the student work. For adults there is no teacher-scale analogue; observer input from a partner, parent, or colleague on an instrument like the CAARS 2 observer form is best practice rather than a requirement, and childhood evidence (report cards, a parent's account) does the onset work instead. When an informant never returns a form, document the attempt and what you relied on in its place. How is an ADHD evaluation billed: 90791 or the testing codes? By what was actually done. An interview-based diagnostic evaluation is 90791 (or 90792 with medical services); brief rating scales ride under 96127 where payers allow, with per-day unit caps; and the timed testing codes (96130, 96131, 96136, 96137) apply only when standardized instruments are administered and interpreted, with the time documented. Contractor policy adds the traps: self-scored inventories are not testing, and testing "administered for educational or vocational purposes" without a medical-management purpose is excluded, which is how school-accommodation referrals become denials. When a full battery drives the evaluation, the work product is a psychological evaluation report and its billing rules apply. In Australia there is no dedicated ADHD-assessment item; psychiatrist assessments run under attendance items such as 291, where the explanatory notes make the written report to the referrer the core deliverable. Does the report have to use DSM-5-TR criteria verbatim? No rule anywhere requires quoting criteria text or using the TR edition specifically. What reviewers look for is the mapping: which symptoms, endorsed by whom, in which settings, since when, with what impairment, and what else was ruled out. NSW makes the edition point explicit by accepting a diagnosis meeting DSM-5 or ICD-11 criteria for stimulant authority, and US claims use ICD-10-CM codes (F90.0, F90.1, F90.2, F90.9) regardless of which manual framed the assessment. Reproducing pages of criteria text adds copyright risk and no clinical value; the evidence under each criterion is what makes the report defensible. How is this different from a psychoeducational report or a full psychological evaluation? Same family, different questions. The ADHD evaluation report answers a focused diagnostic question and may rest on interview plus rating scales alone. A psychological evaluation report is the battery-driven wrapper billed under the testing codes; a psychoeducational report answers a school-eligibility question under education law, which is also why insurers treat school-purpose testing as not medically necessary; a psychiatric diagnostic evaluation is the interview-based start of routine care; and a young child with broader developmental questions starts with a developmental assessment . Autism evaluations use different instruments and criteria entirely. Pick the document that matches the referral question, and say so in the first paragraph. How long do I keep an ADHD evaluation report, and who can see it? Retention is state or provincial law on the whole-chart clock: commonly 5 to 10 years by state in the US, with the APA guideline suggesting 7 years, or 3 years past majority for minors; at least 10 years after last contact, or 10 years past the client's 18th birthday, under Ontario's college standard; and 7 years for adults or until age 25 for minors under Australian state rules. Access is broader than most evaluators expect: the report sits in the designated record set under 45 CFR 164.524 , parents generally hold a minor's right, portals release finalized reports by default, and schools see it whenever a parent authorizes release. A child's report resurfaces years later for college accommodations or an adult re-evaluation, so write every section to be read cold by a stranger. Can BastionGPT draft an ADHD evaluation report? Yes. Bring the evaluation however it exists: interview notes, dictation, score summaries, a records chronology. BastionGPT drafts the report in your structure, from referral question through criteria mapping, differential, and recommendations, tabulates multi-informant scores by setting, produces the plain-language version for parents or the client, and checks the draft for the failure points this page lists: an unanswered referral question, a criterion with no named source, missing onset evidence, or an unjustified battery. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Psychological Evaluation Report Psychoeducational Evaluation Report Conners 4 Report Write-Up CAARS-2 Report Write-Up BRIEF-2 Report Write-Up Section 504 Clinician Statement School & Teacher Collateral Note ASRS v1.1 Score Documentation Vanderbilt ADHD Scale Documentation Stroop Test Score Documentation T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/autism-evaluation-report All note templates # Autism Spectrum Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An autism spectrum evaluation report is the written product of a diagnostic evaluation for autism spectrum disorder. It integrates developmental history, caregiver interview, direct observation, and standardized instruments into a DSM-5-TR criteria mapping, a diagnosis with severity and support levels, and functional recommendations. Psychologists, developmental pediatricians, and multidisciplinary teams write them for children and adults. Most run 4 to 15 pages and feed schools, payers, and disability programs. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and neuropsychologists; developmental pediatricians, psychiatrists, and multidisciplinary developmental teams documenting their own diagnostic assessments Audience The referrer and family, schools and early intervention programs, payers and auditors, and the funding programs that act on the written diagnosis Typical length 1,500 to 5,000 words (4 to 15 pages) · 4 to 8 hours of interviews, observation, scoring, and writing (clinical team estimate) Format family Interpretive assessment report (compare: psychological, neuropsychological, and psychoeducational reports) When it's used When developmental or social-communication differences need a diagnostic answer: a first diagnosis in childhood, an adult evaluation, or the written evidence a school, payer, or disability program requires Standards context No law mandates the format; DSM-5-TR is the reference, guidelines set the convention, and funding programs decide what the written diagnosis must contain On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an autism spectrum evaluation report? An autism spectrum evaluation report is the written product of a diagnostic evaluation for autism spectrum disorder. It integrates developmental history, a structured caregiver interview, direct behavioral observation, and cognitive, language, and adaptive data into an explicit mapping of findings onto DSM-5-TR criteria, then states a diagnosis with severity and support levels, the differential that was considered, and functional recommendations. Clinicians, payers, and funding programs also call it a comprehensive diagnostic evaluation, an ASD diagnostic evaluation, or an autism diagnostic assessment report. It is the integrating document that instrument write-ups feed: an ADOS-2 observation, an ADI-R caregiver interview, SRS-2 informant ratings, a CARS-2 severity rating, or a MIGDAS-2 qualitative interview each produce findings, and this report is where they become a diagnosis. The diagnostic reference across the US, Canada, and Australia is DSM-5-TR. The 2013 fifth edition replaced four separate DSM-IV diagnoses with what the APA's work group called "a single umbrella disorder," and the text revision of March 18, 2022 tightened Criterion A so that all three social-communication characteristics must now be affirmatively present, a change the work group said was intended to "maintain a high diagnostic threshold." Here is the fact that organizes everything else on this page: no statute, regulator, or diagnostic standard in the US, Canada, or Australia prescribes the report's format or requires any specific instrument, and yet this document is the gate that services, school supports, and disability funding run through. Ontario's Autism Program funds services on a written diagnosis containing four data elements (the child's full name and date of birth, the assessment date, a statement that diagnostic criteria are met, and the professional's name and credentials) and names no instrument. British Columbia takes the opposite approach for private diagnoses: Autism Funding eligibility requires completion of both the ADOS-2 and the ADI-R, and "Other assessments will not meet eligibility for Autism Funding." Australia funds the assessment itself through once-per-lifetime Medicare items whose required deliverable is a documented treatment and management plan. The report is also narrower than its neighbors: a battery-driven evaluation of broad diagnostic questions is a psychological evaluation report , a young child's milestone and eligibility workup is a developmental assessment , and a focused attention question is an ADHD evaluation report . This page covers the document that answers: does this person meet criteria for autism spectrum disorder, with what support needs, and what should happen next? ## Who uses autism spectrum evaluation reports and when Psychologists and neuropsychologists write them in clinics and private practice; developmental pediatricians, psychiatrists, and multidisciplinary developmental teams write them in hospital and public systems, where team-based assessment is the convention for complex presentations. Demand keeps rising: the CDC's monitoring network found autism prevalence of 32.2 per 1,000 children aged 8 (about 1 in 31) in 2022, and identification is reaching more children earlier. Practice models differ by system: Canada's paediatric society standard describes a sole-practitioner path in which an experienced clinician diagnoses on clinical judgment and DSM-5 criteria "with or without data obtained using a diagnostic assessment tool," while Australia's NHMRC-approved National Guideline (66 recommendations, second edition 2023) frames assessment around clinical need rather than any named test. The report is the right tool when developmental or social-communication differences need a diagnostic answer: a first evaluation after screening concerns in a toddler, a school-age child whose needs became visible as demands grew, an adult seeking an explanation and access to supports, or a re-evaluation feeding school or disability documentation. The instrument spine is multi-source by design: an ADOS-2 observation and an ADI-R caregiver interview where indicated, SRS-2 ratings from parent and teacher, a CARS-2 severity rating or MIGDAS-2 qualitative interview in some settings, and a Vineland-3 for the adaptive picture. When the question is broad early delay, a developmental assessment comes first; when it is school eligibility, the psychoeducational report and the education-law process take over; when attention is the presenting question, the ADHD evaluation report is the focused sibling, and the two differentials cross often. The audience is wider than the chart: families read it to understand their child, schools read it to plan supports, payers read it to authorize services, and funding programs read it to open doors. ## Autism spectrum evaluation report structure: what goes in each section No authority mandates a single format. The sections below are the consensus skeleton of guideline practice and the elements payers, schools, and funding programs expect to find, each with the pitfall that most often undermines it. Identifying information and referral question. Client, date of birth, evaluation dates, evaluator and credentials, referrer, and the question in one sentence: who is asking, and what decision does the answer feed? Funding programs check the basics: Ontario's program, for example, requires the child's full name and date of birth, the assessment date, a criteria-met statement, and the professional's name and credentials. Pitfall: a report with no referral question cannot show medical necessity, and a purely school-driven purpose reads to payers as an educational service. Sources of information and consent. Every source, listed: interviews and dates, informants, instruments with versions and modules, records reviewed (early intervention, school, prior evaluations), who consented, and the modality of each component. Pitfall: an unnamed remote component. British Columbia accepts a virtual ADI-R but not a virtual ADOS-2; a report that hides which parts ran by video invites exactly that scrutiny. Developmental history and caregiver interview. Early milestones, language development, regression if any, medical history (hearing, vision, seizures, genetics), family history, and the structured caregiver interview ( ADI-R or equivalent) that anchors symptom onset in the early developmental period. Pitfall: onset asserted from memory alone. Records, early videos, and specific caregiver examples carry the onset criterion; masked presentations, documented especially in girls and women, make this section the report's backbone. Observations across settings. Direct observation, standardized ( ADOS-2 module and rationale, or a MIGDAS-2 qualitative interview) or structured-informal, plus what teachers and caregivers see where the clinician cannot. State whether results are interpretable and why. Pitfall: a single-context observation treated as the whole picture; one office session is one setting, and the criteria ask for more. Instrument results, by source. Each instrument with version, module or form, informant, and setting, reported the way its manual presents results: classifications as classifications, ratings with their band language, categorical results without invented scores. Then the pattern: where sources agree, where they diverge, and what the divergence means. Pitfall: classification treated as diagnosis. An instrument result is evidence toward criteria, and no single tool can carry the conclusion alone. Cognitive, language, and adaptive functioning. Cognitive estimate, current language level, and adaptive data (a Vineland-3 or equivalent) that ground the DSM-5-TR specifiers and the severity picture in daily-life evidence. Pitfall: severity and support levels assigned with no adaptive or functional data behind them; a level with no evidence reads as a label, not a finding. DSM-5-TR criteria mapping and differential diagnosis. The engine of the report: each of the three social-communication characteristics affirmatively evidenced (the 2022 text revision requires all three), the restricted-repetitive pattern documented, onset and impairment stated, and the alternatives considered in writing: language disorder, intellectual disability, ADHD, anxiety, trauma responses. Pitfall: Criterion A met on a subset, the most consequential documentation change reviewers now check. Diagnosis, severity, and formulation. The conclusion with the ICD-10-CM code payers will see (F84.0), the DSM-5-TR specifiers, severity recorded for each of the two domains with its functional basis, co-occurring conditions, and a short formulation that names strengths as well as needs. Document a negative finding just as clearly. Pitfall: hedged conclusions that neither open services nor close the question. Recommendations, feedback, and signature. Functional, ranked, and matched to the findings and the destination: school supports, therapies, family guidance and community resources, re-evaluation timing. Document the feedback session, who received the report, time for any time-based testing codes, and sign with credentials and date. Pitfall: a generic recommendation list; this report is reread for years, by schools, funding programs, and often by the autistic person it describes. ### Blank template (copy and adapt) AUTISM SPECTRUM EVALUATION REPORT Client: ______________________ DOB: ____________ Age: ______ Evaluation date(s): ______________ Report date: ____________ Evaluator / credentials: ______________________________________ Referred by: ____________________ Modality by component: ________ REFERRAL QUESTION _____________________________________________________________ SOURCES OF INFORMATION AND CONSENT Interviews / informants: ____________________________________ Instruments (version, module/form, informant, setting): _____ _____________________________________________________________ Records reviewed: ______________ Consent obtained from: _____ DEVELOPMENTAL HISTORY AND CAREGIVER INTERVIEW Milestones / language development: ___________________________ Onset evidence (early developmental period, with source): ____ Medical / family history: ___________________________________ OBSERVATIONS ACROSS SETTINGS / VALIDITY STATEMENT _____________________________________________________________ INSTRUMENT RESULTS (by source and setting) _____________________________________________________________ Agreement / divergence across sources: ______________________ COGNITIVE, LANGUAGE, AND ADAPTIVE FUNCTIONING _____________________________________________________________ DSM-5-TR CRITERIA MAPPING AND DIFFERENTIAL A (all three characteristics, with sources): _________________ B (patterns documented, of four): ____ Sources: ____________ Onset: ______________ Settings with impairment: ____________ Alternatives / co-occurring considered: _____________________ DIAGNOSIS, SEVERITY, AND FORMULATION Diagnosis (ICD-10-CM code) / specifiers: ____________________ Severity by domain (with functional basis): _________________ Formulation (strengths and needs): __________________________ RECOMMENDATIONS, FEEDBACK, AND FOLLOW-UP 1. ___________________________________________________________ 2. ___________________________________________________________ Feedback session / report shared with: ______________________ Testing time (if time-based codes billed): __________________ Signature / credentials: ______________ Date: ______________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample autism spectrum evaluation report Scenario: a licensed psychologist evaluates a 6-year-old referred by his pediatrician after classroom social-communication concerns. A full report runs 4 to 15 pages; this version is condensed but structurally complete. Use it as an autism evaluation report sample you can adapt for children or adults. All details are fictional. AUTISM SPECTRUM EVALUATION REPORT (CONFIDENTIAL) Client: J.T., age 6, grade 1 · DOB: 02/2020 Evaluation dates: 07/08/2026 and 07/15/2026 · Report date: 07/22/2026 Evaluator: R. Alvarez, PsyD, Licensed Psychologist · Referred by: S. Nguyen, MD (pediatrician) Referral question. Dr. Nguyen asks whether J.T. meets criteria for autism spectrum disorder and what supports are indicated, after his grade 1 teacher raised social-communication concerns and a pediatric visit found them consistent with earlier patterns. Sources of information and consent. Two in-person sessions with J.T.; ADI-R with his mother (in person); ADOS-2 Module 3 (fluent speech); SRS-2 School-Age forms from mother and teacher; Vineland-3 Comprehensive Interview Form with mother; brief cognitive estimate (WISC-V, selected subtests); review of kindergarten and grade 1 report cards, a 2024 preschool speech-language screening, and the referral note. Written consent from mother; assent from J.T. No component was administered remotely. Developmental history and caregiver interview. First words on time but first phrases after 30 months, with preschool speech-language support; no loss of skills at any age. Mother describes strong early routines, distress at transitions since toddlerhood, and limited interest in other children through preschool. Hearing and vision screened normal 03/2026. No significant medical history; no medications. Family history includes a father described as "similar as a child," never evaluated. ADI-R responses produced categorical results consistent with autism across all three domains, with specific examples dated before age 3 (lining up vehicles for long periods, no pretend play with peers, echoed phrases). Observations across settings. J.T. separated easily and engaged with materials. He initiated conversation about transit maps with detailed, fluent language, responded briefly or not at all to the examiner's bids on other topics, used few gestures, and made brief eye contact when excited. He followed test directions and persisted; results are considered interpretable. His teacher reports parallel patterns in class: strong academics, scripted approaches to peers, and distress when the schedule changes. Instrument results, by source. ADOS-2 Module 3: met the autism spectrum classification, with observed differences concentrated in social reciprocity and nonverbal communication and a comparison score in the moderate-to-high range. SRS-2: mother's Total T-score 77 (severe range); teacher's Total T-score 70 (moderate range); the two settings agree. Vineland-3: Adaptive Behavior Composite 76, with Socialization and Communication below Daily Living Skills. Cognitive estimate in the average range. The sources converge: fluent language and average-range cognition alongside consistent social-communication differences and inflexibility across home and school. DSM-5-TR criteria mapping and differential. All three Criterion A characteristics are documented: social-emotional reciprocity (one-sided exchanges in session and class), nonverbal communication (limited gesture and eye contact per ADI-R and observation), and developing and maintaining relationships (no reciprocal friendships per mother and teacher). Criterion B: two patterns documented, insistence on sameness (transition distress, fixed routines) and highly restricted interests (transit systems, unusual in intensity); sensory responses were explored and did not reach the documented threshold. Onset: before age 3 per ADI-R examples and preschool records. Impairment: present at home and school. Alternatives considered: language disorder alone (language now fluent; social differences persist beyond language), intellectual disability (average-range estimate), ADHD (attention adequate in session; teacher ratings not elevated for inattention), social anxiety (differences present from toddlerhood, not situational). None better explains the pattern; none is diagnosed as co-occurring today. Diagnosis, severity, and formulation. Autism spectrum disorder (F84.0), without accompanying intellectual impairment, without accompanying language impairment (with a history of early language delay). Severity recorded by domain per DSM-5-TR: Level 2 for social communication, Level 1 for restricted, repetitive behaviors, based on the adaptive data and observations above. J.T. is a curious, academically capable autistic child with deep interests, a strong memory, and a warm bond with his parents; his support needs center on peer interaction, flexibility, and communication in unstructured settings. Recommendations. (1) Share this report with the school, with parental consent, to inform a support plan; a separate educational evaluation may follow under the school's own process. (2) Speech-language services targeting social communication in natural settings. (3) Parent guidance and connection to community and neurodiversity-informed resources; written summary provided. (4) Review adaptive goals and supports with Dr. Nguyen in 12 months, or sooner if needs change. Feedback and signature. Findings reviewed with both parents 07/22/2026; plain-language summary provided. Report released to Dr. Nguyen with consent. R. Alvarez, PsyD, Licensed Psychologist, signed 07/22/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral question gets a direct answer: a named diagnosis with its code, severity recorded by domain, and the "what happens next" the referrer, family, and school each need. All three Criterion A characteristics trace to named sources across two settings, which is the bar the 2022 text revision set; nothing is inferred from a score alone. Onset and adaptive functioning are evidenced, not asserted: dated ADI-R examples and preschool records carry onset, and the Vineland-3 data stands behind the severity levels. Instrument results are treated as evidence: the classification, the ratings, and the interview converge in a written mapping, and the differential and co-occurring reasoning are documented. The person is described with strengths as well as needs, in language the family, the school, and one day the client himself can read; the prescriptive decisions stay with the systems that own them. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the report knowing who will read it, and for how long. It sits in the designated record set, so the client, or a parent for most minor clients, holds a right of access under 45 CFR 164.524 , and a finalized electronic report is releasable by default through the portal, often before the feedback session. Released to a school, it becomes a planning document read by non-clinicians, so the conclusions belong in plain language; the rules for releasing test data versus test materials are the same as for any testing report, and the psychological evaluation report page covers them in detail. One retention myth needs killing: HIPAA sets no retention period for the record itself; the six-year clock in 45 CFR 164.316 applies to compliance documentation. Retention runs on state or provincial law: commonly 5 to 10 years by state in the US, with the APA's record-keeping guideline suggesting 7 years, or 3 years past majority for minors ; at least 10 years after last contact, or 10 years past the 18th birthday, under Ontario's college standard; and 7 years for adults or until age 25 for minors in Australian state regimes. This report is reread for longer than almost any other: school support plans, funding applications, and adult re-evaluations all reach back to it, and many autistic adults eventually read their own childhood evaluation. Write with those readers in the room. The compliance pressure sits in two places, and neither is the format. First, payment for the evaluation. In the US the interview-based diagnostic evaluation bills 90791, and the timed testing codes (96130, 96131, 96136, 96137) apply only when standardized instruments are administered and interpreted, with the time documented; contractor policy excludes testing done for purely educational purposes, and the psychological evaluation report page carries the testing-code detail. In Australia, Medicare funds the assessment through consultant items 135 and 289 (paediatrician and psychiatrist): a patient aged under 25, an attendance of at least 45 minutes, claimable once per lifetime ($320.55 on the July 2026 schedule), and the deliverable is a treatment and management plan that must include the confirmed diagnosis, the findings of assessments performed, a risk assessment, and treatment options, with a copy to the referring practitioner. Second, what the report opens. Ontario's program checks its four written-diagnosis elements; British Columbia's private-diagnosis route requires both the ADOS-2 and the ADI-R and accepts the interview, but not the observation, done virtually ; and US ABA authorization typically runs on a comprehensive diagnostic evaluation meeting DSM-5-TR criteria, the standard Cigna's policy states outright while Kaiser's Mid-Atlantic policy spells out six documentation domains, from milestones and regression through sensory responses to a differential showing the presentation "cannot be explained by a diagnosis other than ASD." Neither names a required diagnostic instrument, and peer-reviewed work documents that "many insurers mandate use of a specific assessment tool(s)" in review criteria anyway. Check the destination program before you evaluate, and say in the report which door it was written for. The format is a convention; the content, and what it opens, is the requirement. ## Common autism spectrum evaluation report errors auditors flag The evaluation document is under direct audit scrutiny. In an ongoing OIG series on Medicaid autism services, Maine's January 2026 audit found at least $45.6 million in improper payments, including applied behavior analysis delivered to children who "did not receive the required comprehensive assessments," after Indiana (at least $56 million, December 2024) and Wisconsin (at least $18.5 million, July 2025); the comprehensive diagnostic evaluation is exactly the document those findings turn on. On the diagnosis side, the field's most consequential myth is that one instrument settles it: a JAMA Pediatrics study of 349 young children found diagnoses made with and without the ADOS agreed 90 percent of the time and concluded the tool is "generally not required for diagnosis of ASD in young children," while peer-reviewed work documents that "many insurers mandate use of a specific assessment tool(s)" in their review criteria regardless. The defensible report documents its instrument choices against its destination instead of assuming either extreme. The BastionGPT Clinical Advisory Board sees the same errors most often in autism evaluation report reviews: Criterion A met on a subset. Since the March 2022 text revision, all three social-communication characteristics must be affirmatively documented. A report that evidences two and infers the third is the first thing a trained reviewer checks now. Classification treated as diagnosis. An ADOS-2 classification, a CARS-2 range, or an elevated SRS-2 is evidence toward criteria, not a verdict. The written mapping is the diagnosis, and a report that jumps from instrument output to conclusion will not survive a records review. Onset and adaptive evidence asserted, not documented. "Symptoms since early childhood" with no dated example, record, or caregiver-interview anchor; severity levels assigned with no adaptive data behind them. Masked presentations, described especially in girls, women, and adults diagnosed late, fail here most. The missing differential and co-occurring reasoning. No documented consideration of language disorder, intellectual disability, ADHD, anxiety, or trauma responses. Kaiser's policy model is blunt: co-occurring conditions identified, and a presentation that "cannot be explained by a diagnosis other than ASD." The destination mismatch. A clinically sound report that fails the program it feeds: a British Columbia private diagnosis without both required instruments, an Ontario submission missing one of the four written-diagnosis elements, a virtual ADOS-2 sent to a program that will not accept one, or a medical evaluation offered where the school's own eligibility process was the actual question. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on autism spectrum evaluation reports. Draft the full report from your interview notes, observation notes, and score summaries: history, observations, criteria mapping with severity by domain, differential, and functional recommendations in your structure, ready for your clinical judgment. Merge multi-source findings into one coherent picture: caregiver interview, direct observation, teacher ratings, and adaptive results organized by source and setting, plus a plain-language version for the family or school. Check a draft before you sign: a Criterion A characteristic with no named source, missing onset or adaptive evidence, classification language standing in for a diagnosis, or an unanswered referral question. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a required format for an autism spectrum evaluation report? No. No statute, regulator, or diagnostic standard in the US, Canada, or Australia prescribes the report's title, sections, or length. DSM-5-TR supplies the criteria, and professional standards supply the convention: Canada's paediatric society standard describes sole-practitioner, shared-care, and team models, and Australia's National Guideline (second edition 2023, 66 recommendations approved by the NHMRC) frames assessment around clinical need. What binds is narrower: funding programs define what the written diagnosis must contain, payer policies define documentation expectations, and record law governs retention and access. Most reports run 4 to 15 pages and take 4 to 8 hours of interviewing, observation, scoring, and writing, a clinical team estimate. Is the ADOS-2 required for an autism diagnosis? Not by any diagnostic standard. DSM-5-TR names no instrument, and a JAMA Pediatrics study of 349 young children found diagnoses made with and without the ADOS agreed 90 percent of the time, concluding the tool is "generally not required for diagnosis of ASD in young children." The practical answer depends on the destination: Kaiser's and Cigna's policies name no required diagnostic instrument, peer-reviewed work documents that "many insurers mandate use of a specific assessment tool(s)" in review criteria anyway, Aetna's policy lists the ADOS-2 and ADI-R among accepted diagnostic tools, and British Columbia requires both for private Autism Funding eligibility. Check the program or plan the report feeds before you evaluate, and document why your instrument choices answer the question. Do I need to document every DSM-5-TR criterion? Best practice is yes, criterion by criterion. Since the March 18, 2022 text revision, all three social-communication characteristics must be affirmatively present, so a defensible report evidences each one with a named source and setting rather than stating "meets criteria" and moving on. Document the restricted-repetitive patterns you observed, onset in the early developmental period with its evidence, impairment, and the alternatives you ruled out, then record severity for each of the two domains with the functional data behind it. Reproducing the manual's criterion text is neither needed nor wise; the mapping of evidence onto each criterion is what reviewers read. What does the report need to open funding doors like the OAP or the NDIS? Each program defines its own gate, and format is never it. Ontario's Autism Program registers a child on a written diagnosis containing four elements: the child's full name and date of birth, the assessment date, a statement that diagnostic criteria are met, and the professional's name and credentials. British Columbia's private route requires both the ADOS-2 and the ADI-R and will not accept a virtual ADOS . In Australia, the once-per-lifetime Medicare assessment items require a documented treatment and management plan, and NDIS access runs on evidence of functional impact in daily life, which is why the adaptive and severity sections matter as much as the diagnosis line; a dedicated functional capacity assessment is often the companion document. In the US, ABA authorization typically requires a comprehensive diagnostic evaluation meeting DSM-5-TR criteria. Write the report knowing which door it is for, and it will usually satisfy the others. How is an autism evaluation billed? By what was actually done. In the US an interview-based diagnostic evaluation is 90791; the timed testing codes (96130 and 96131 for evaluation and report, 96136 and 96137 for administration) apply when standardized instruments are administered and interpreted, with time documented, and testing for purely educational purposes is excluded from coverage; the psychological evaluation report page carries the testing-code rules in detail. Australia funds the assessment directly: MBS items 135 (consultant paediatrician) and 289 (consultant psychiatrist) cover an attendance of at least 45 minutes for a patient under 25, claimable once per lifetime at $320.55 on the July 2026 schedule, with the treatment and management plan as the required deliverable, including the confirmed diagnosis, assessment findings, a risk assessment, and treatment options. Canadian billing is provincial, with public and private routes to the same written diagnosis. Who can diagnose autism, and can one clinician do it alone? It depends on the jurisdiction and the program reading the report. In the US, licensed psychologists and physicians diagnose, and payer policies set whose reports they accept. Ontario's program accepts a written diagnosis from a qualified professional. Canada's paediatric society standard supports an experienced sole practitioner diagnosing on clinical judgment and DSM-5 criteria "with or without data obtained using a diagnostic assessment tool," reserving shared-care and team models for complex presentations, and adds that findings from an assessment tool "cannot be used alone to diagnose ASD." Australia's National Guideline frames single-clinician versus consensus-team assessment by clinical need. A school's identification of autism for special education is a separate process under education law; it opens school services, not medical ones, which is one more reason the report should say which question it answers. How is this different from a developmental assessment, a psychoeducational report, or a full psychological evaluation? Same family, different questions. A developmental assessment maps a young child's milestones across domains and often comes first, sometimes ending in a referral for this evaluation. A psychoeducational report answers a school-eligibility question under education law, which is also why payers treat school-purpose testing as educational rather than medical. A psychological evaluation report is the battery-driven wrapper for broad diagnostic questions, and its page carries the testing-code rules. An ADHD evaluation report answers the focused attention question; the two conditions co-occur often, so each report should show the other was considered. This page covers the diagnostic document for autism specifically: multi-source by design, criteria-mapped, and written for the service systems that act on it. What changes in an adult autism evaluation report? The structure holds; the evidence work shifts. Onset must be reconstructed: school records, old report cards, family accounts where available, and the client's own developmental narrative, because the early developmental period is decades back. Masking matters: camouflaged presentations, described especially in women and in adults diagnosed late, can sit below instrument thresholds while the history still supports the diagnosis, so document the strategies the client describes and what they cost. Observer input from a partner, parent, or long-term friend strengthens the record where it exists; when it does not, say so and describe what carried the evidence instead. No instrument is required for adults any more than for children. And write with particular care: the reader most affected by this report is the adult it describes, who came asking a real question about their own life. Can BastionGPT draft an autism spectrum evaluation report? Yes. Bring the evaluation however it exists: interview notes, observation notes, score summaries, a records chronology. BastionGPT drafts the report in your structure, from referral question through criteria mapping, severity by domain, differential, and functional recommendations, merges multi-informant results into a clean by-source summary, produces the plain-language version for the family or school, and checks the draft for the failure points this page lists: a criterion with no named source, missing onset or adaptive evidence, or classification language standing in for a diagnosis. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates ADOS-2 Report Write-Up ADI-R Report Write-Up SRS-2 Report Write-Up CARS-2 Report Write-Up MIGDAS-2 Report Write-Up Psychological Evaluation Report NDIS Functional Capacity Assessment Report ASQ-3 Score Documentation M-CHAT-R/F Score Documentation Sensory Profile 2 Report Write-Up GARS-3 Report Write-Up CELF-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ndis-functional-capacity-assessment All note templates # NDIS Functional Capacity Assessment Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An NDIS functional capacity assessment report is an allied health report documenting how a person's disability affects everyday functioning across six domains: communication, social interaction, learning, mobility, self-care and self-management. It provides evidence for NDIS access and planning decisions and recommends supports with frequency and duration. Occupational therapists, psychologists and other qualified professionals write them. Most run 15 to 30 pages after 1 to 10 hours of assessment. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Occupational therapists most often; physiotherapists, psychologists, speech pathologists and medical specialists qualified to make the assessment Audience NDIA delegates and planners deciding access and funding, the participant and their supports, support coordinators, and review bodies when a decision is disputed Typical length 4,000 to 9,000 words (15 to 30 pages) · 1 to 10 hours of assessment (the NDIA's published range) plus 3 to 6 hours of writing (clinical team estimate) Format family Formal report (also called FCA, functional assessment report, psychosocial functional capacity assessment) When it's used NDIS access requests, plan reassessments and changes of circumstances, and high-stakes funding questions such as supported independent living, home modifications and assistive technology Standards context No law requires an FCA or sets its format; it is professional convention built around the NDIS Act's disability and funding tests On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an NDIS functional capacity assessment report? An NDIS functional capacity assessment report (FCA) is an allied health report that documents how a person's disability affects their ability to do everyday tasks, and what supports that evidence justifies, for Australia's National Disability Insurance Scheme. The NDIA describes a functional capacity assessment as an official assessment of the ability to perform daily tasks at home, at work and in the community, done by occupational therapists, physiotherapists, psychologists or medical specialists who are "qualified to make the assessment". No standards body created the format. It grew as professional convention around the NDIS Act 2013: the disability requirements ask for evidence of a "substantially reduced functional capacity" in one or more daily life activities (NDIS Act s 24), and the six domains every FCA works through, communication, social interaction, learning, mobility, self-care and self-management, mirror that activity list. The same document also answers to FCA, functional assessment report, and psychosocial functional capacity assessment when the impairment is psychiatric. The load-bearing fact, and the one most template pages bury: no law requires an FCA. The NDIA's own guidance states, "You don't have to do a functional capacity assessment to be eligible for NDIS." The report exists as evidence, not as a statutory form, so it succeeds by how well it proves functional impact and support need, never by matching a particular layout. That also separates it from a diagnostic document. A psychological evaluation report answers what condition explains the presentation; an FCA answers what the person can and cannot do on most days, with and without support, and what that means for access and planning decisions. The two often travel together, diagnosis first and function second, especially for autism and psychosocial disability, where an autism evaluation report or psychiatric evidence establishes the impairment and the FCA translates it into support needs. ## Who uses NDIS functional capacity assessment reports and when Occupational therapists write the most FCAs, and no rule reserves the document for them: the NDIA's list runs to physiotherapists, psychologists and medical specialists, with the only stated bar being qualification to make the assessment. Psychologists write them for psychosocial disability, autism and intellectual disability, where standardized adaptive measures such as the Vineland-3 and ABAS-3 carry much of the functional evidence, and speech pathologists and physiotherapists assess within their own scope. The report is commissioned at four moments: an access request that needs functional evidence, a plan reassessment, a change in circumstances, and high-stakes funding questions such as supported independent living, home modifications and assistive technology, where a current FCA is often the piece of evidence the decision turns on. Support coordinators arrange them; NDIA delegates and planners read them and act on them. Neighbours matter here: when the presenting question is a young child's milestones, a developmental assessment comes first; when the question is a diagnosis, an evaluation report answers it; the FCA is the document for the function and support question, and it assumes the diagnostic work already exists somewhere it can cite. ## NDIS functional capacity assessment report structure: what goes in each section No regulation prescribes the sections, the length or the field list. The skeleton below is the professional consensus, shaped by what NDIA delegates need in order to apply the access and funding tests, and the blank NDIS functional capacity assessment report template that follows uses the same bones. Each section carries the pitfall that most often undermines it. Identifying details and referral context. Participant name, date of birth, NDIS number if a participant, assessment and report dates, assessor name, discipline and registration, and who asked for the report and why: access evidence, plan reassessment, change of circumstances, or a specific support question such as supported independent living or assistive technology. State the question in one sentence. Pitfall: a report with no stated purpose. The delegate reading it cannot tell which decision it is meant to inform, and the recommendations float free of any test. Background: diagnoses, history and current supports. Diagnoses with their sources and dates, relevant medical and psychiatric history, medications with functional effects, current formal and informal supports, and living, work and study arrangements. Cite the diagnostic documents rather than re-arguing them; a psychological evaluation report or specialist letter carries the diagnosis so this report can carry the function. Pitfall: restating the diagnosis as if it were the evidence. A diagnosis names the impairment; it says nothing yet about what the person can do on most days. Assessment methods and presentation. Everything the findings rest on, listed: interview dates and settings, direct observation including any home or community visit, informants and their relationship to the participant, standardized tools with versions, records reviewed, and the modality of each component. Describe presentation during assessment: communication, engagement, fatigue, and anything that shaped the data. Pitfall: an unlisted setting or informant. Findings from one office visit, with no observation and no second source, read as a snapshot rather than an assessment. Standardized measures and scores. The tools, the scores, and what each score means in daily life. Adaptive behavior measures such as the Vineland-3 and ABAS-3 anchor self-care, communication and daily living evidence; the WHO's WHODAS 2.0 is widely used for whole-of-life functioning. No tool is mandated anywhere, so choose for the question and say why. Pitfall: scores without translation. A composite two standard deviations below the mean means nothing to a planner until the report says what it looks like at the kitchen bench, at the bus stop, or at work. Functional capacity across the six domains. The core of the report: communication, social interaction, learning, mobility, self-care and self-management, each described for what the person does on most days, in what settings, with and without support. Capability with prompting is not capability alone; say which one you observed. Pitfall: best-day reporting. A single observed success, generalised into "independent", collapses at review when every other source describes daily prompting. Informant reports and observation. What carers, family, support workers and teachers each report, tied to who said it, and what you directly observed. Where sources disagree, address the discrepancy and explain what you weighted and why. Pitfall: unexplained contradictions. A report that scores self-care as intact while the carer interview two pages later describes daily bathing support hands the review its reason to discount you. Strengths, informal supports and risk. What the person does well, what they value, and what informal supports already carry, then current risks and how they are managed. Strengths evidence is planner-credible and participant-respectful, and it shows what funded support adds rather than replaces. Pitfall: a deficits-only report. It reads as advocacy rather than assessment, and it fails the participant who will read their own record. Functional impact summary. The plain-language translation: which domains show substantially reduced functional capacity, on what evidence, and what that means for work, study and social participation. Write it so a non-clinician can quote it, because this is the section decisions cite. Pitfall: conclusory statutory language. "Substantially reduced functional capacity" asserted without the evidence chain behind it persuades no one; the phrase earns its place only when the domains above prove it. Recommendations and support needs. Support by support: type, frequency, duration, and the rationale that traces each one from the impairment through the functional evidence to the need. Address how the support relates to the disability, why it will be effective and beneficial, and how it works alongside informal and mainstream supports, because those are the criteria the delegate must apply. Recommend supports, not therapy goals; treatment planning belongs to the treating team. Pitfall: a recommendations list with no dosage. "Ongoing support with daily living" gives the planner nothing to fund; "6 hours per week of daily-living support for 12 months, reviewed at reassessment" does. ### Blank template (copy and adapt) NDIS FUNCTIONAL CAPACITY ASSESSMENT REPORT Participant: ______________________ DOB: ____________ NDIS no.: ____________ Assessor / discipline / registration: _________________________________________ Assessment dates and settings: ________________________________________________ Report date: ____________ Requested by / purpose: ____________________________ REFERRAL CONTEXT Decision this report informs (access / reassessment / specific supports): ________________________________________________________________________ BACKGROUND Diagnoses (source and date): ___________________________________________ Relevant history and medications (functional effects): _________________ Current supports, living, work and study arrangements: _________________ ASSESSMENT METHODS AND PRESENTATION Interviews / observation (incl. home or community visit): ______________ Informants and relationship: ___________________________________________ Standardized measures (tool and version): ______________________________ Records reviewed: ______________________________________________________ Presentation during assessment: ________________________________________ FUNCTIONAL CAPACITY BY DOMAIN (on most days; with and without support) Communication: _________________________________________________________ Social interaction: ____________________________________________________ Learning: ______________________________________________________________ Mobility: ______________________________________________________________ Self-care: _____________________________________________________________ Self-management: _______________________________________________________ INFORMANT REPORTS AND OBSERVATION (discrepancies addressed) ________________________________________________________________________ STRENGTHS, INFORMAL SUPPORTS AND RISK Strengths and what informal supports carry: ____________________________ Risks and current management: __________________________________________ FUNCTIONAL IMPACT SUMMARY (plain language; domains substantially reduced, and the evidence) ________________________________________________________________________ ________________________________________________________________________ RECOMMENDATIONS AND SUPPORT NEEDS (each: type / frequency / duration / rationale tracing impairment to need) 1. ______________________________________________________________________ 2. ______________________________________________________________________ 3. ______________________________________________________________________ Assessor signature / date: _____________________ Provider: ____________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample NDIS functional capacity assessment report Scenario: a psychologist completes a functional capacity assessment for a 21-year-old NDIS participant ahead of a plan reassessment, with a question about daily-living support hours. A full report runs 15 to 30 pages; this version is condensed but structurally complete, and it follows the same skeleton as the blank NDIS functional capacity assessment report template above. All details are fictional. NDIS FUNCTIONAL CAPACITY ASSESSMENT REPORT (CONFIDENTIAL) Participant: D.M., age 21 · NDIS participant (number withheld in sample) Assessor: L. Okafor, Registered Psychologist · Assessment dates: 6 and 13 July 2026 (home and clinic) · Report date: 20 July 2026 Requested by: D.M. and his mother, through his support coordinator, for plan reassessment Referral context. D.M.'s plan reassessment is due in September 2026. The question for this report: what daily-living and community-participation supports does his functional capacity evidence, now that he has finished school and his weekday structure has ended? Background. Autism spectrum disorder diagnosed at age 8 (paediatric assessment, 2013) and mild intellectual disability confirmed on cognitive assessment at age 16 (school psychologist report, 2021); both reports reviewed. No current medications. D.M. lives with his mother, who provides most daily prompting, and attends a community art group fortnightly. No paid supports are currently in place beyond a support coordinator. Assessment methods and presentation. Interview with D.M. at home (6 July); structured observation of morning routine and meal preparation at home (6 July); clinic session with Vineland-3 (Comprehensive Interview Form, mother as respondent) and WHODAS 2.0 self-report with support (13 July); interview with his mother; review of school-leaver and prior diagnostic records. D.M. engaged warmly across both sessions, communicated in short sentences, needed questions rephrased without idiom, and fatigued after about 40 minutes, so sessions were split. Standardized measures. Vineland-3 Adaptive Behavior Composite 68, with Daily Living Skills and Socialization the lowest domains; WHODAS 2.0 responses consistent with informant report, highest difficulty in self-care, getting along with people, and life activities. Scores converge with observation and history; no measure is treated as decisive on its own. Functional capacity by domain. Communication: expresses needs verbally in familiar settings; on most days needs written information simplified and verbal instructions broken into single steps. Social interaction: values contact and attends his art group when accompanied; misreads unfamiliar social cues and has not sustained peer contact without facilitation. Learning: learns practical routines through repeated supported practice over weeks, demonstrated with laundry sequence learned across a school term. Mobility: physically independent; travels one practised bus route; became distressed and returned home when a roadwork detour changed the route in May 2026 (mother's report, consistent with observed rigidity). Self-care: showers and dresses with daily verbal prompting to initiate and sequence; observed to leave hygiene steps incomplete without prompts. Self-management: no independent money handling beyond small cash purchases; does not initiate or track appointments; his mother currently structures each day. Informant reports and observation. D.M. self-rates his independence higher than his mother's report and my observation in self-care and community access. The discrepancy is consistent with his stated wish to be seen as independent and does not alter the observed evidence; both perspectives are recorded, and his goals are carried into the recommendations. Strengths, informal supports and risk. D.M. is punctual, follows practised routines reliably, produces detailed artwork, and wants to work "with animals or in a garden". His mother currently provides 2 to 3 hours of daily prompting and all transport beyond one bus route; she reports fatigue and reduced capacity to sustain this. Risk: episodes of distress when routines change, managed by prediction and preparation; no risk to others identified. Functional impact summary. On most days and across home, community and assessment settings, D.M. shows substantially reduced functional capacity in self-care, self-management and social interaction, with converging evidence from standardized adaptive measures, direct observation and informant report. Without daily prompting and facilitated community access he does not initiate or complete core daily-living tasks, and his participation in study, work and community life depends on support that one informal carer currently provides at an unsustainable level. Recommendations and support needs. 1. Daily-living skills support, 6 hours per week for 12 months, building shower, meal and money routines through repeated supported practice: relates directly to the documented self-care and self-management impacts, is likely to be effective given his demonstrated supported-practice learning, and reduces reliance on his mother rather than duplicating it. 2. Community participation support, 4 hours per week for 12 months, sustaining the art group and building one supported work-experience placement toward his stated goal. 3. Occupational therapy review of travel supports and low-cost assistive prompts within 6 months. 4. Reassess functional capacity at the next scheduled plan reassessment or on significant change of circumstances. Assessor: L. Okafor, Registered Psychologist · Signed 20 July 2026 This sample is fictional and shows structure and register only. It is not clinical, legal or funding advice, and no real person, provider or decision is described. ### Why this sample works Function is evidenced, not asserted: every domain finding traces to a named source, a standardized measure, direct observation or an identified informant, and the three converge. The statutory phrase earns its place: substantially reduced functional capacity appears only in the impact summary, after six domains of on-most-days evidence, with the affected domains named. Capability with support and capability alone are kept distinct throughout, which is exactly the distinction planning decisions need and the one most reports blur. Each recommendation carries type, frequency, duration and a rationale that traces the impairment through the functional evidence to the support need, and speaks to how the support works alongside the informal care already in place. Strengths, the participant's own goals, and the informant discrepancy are documented rather than smoothed over, which makes the report credible to a delegate and respectful to the person who will read their own record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations An FCA is a health record first and NDIS evidence second, and the record rules come from health law and professional regulation, not from NDIS law. Retention runs on state and territory health records legislation: in New South Wales, health service providers must keep records for 7 years from the last service, or until age 25 for information collected from a client under 18, and Victorian law sets the same floors, whichever ends later. Professional boards add their own record-keeping obligations, and registered NDIS providers carry information-management duties under the NDIS Practice Standards. The participant will read this report: it travels to the NDIA, into their record, and often into future assessments, so plain, respectful language is a compliance feature as well as a courtesy. Currency matters too. Nothing sets an expiry date on an FCA, but decisions are made on present function, and a report that predates a major change of circumstances is evidence about a person who no longer quite exists. If a decision goes to internal review, the request window is 3 months, the NDIA allows up to 28 days for further information, and external review at the Administrative Review Tribunal (the AAT's successor) must be sought within 28 days of the internal review decision; at each step, current functional evidence is usually what moves the outcome. The funding mechanics are payer policy and worth stating plainly. Participant-commissioned FCAs are typically paid from plan funding rather than Medicare, under the NDIS pricing arrangements : on the 2026-27 schedule effective 1 July 2026, the hourly price limit is $252.99 for psychologists, including provisionally registered psychologists working under supervision, and $193.99 for occupational therapists, and the schedule carries dedicated NDIA Requested Reports line items, so report writing is billable, priced work rather than unpaid goodwill. The deeper compliance frame: the format is a convention, the evidence is the requirement. What the law supplies is the tests, the disability requirements and the reasonable and necessary criteria; what the report supplies is the functional proof those tests consume. And the landscape is moving. The NDIA is introducing its own support needs assessments from April 2027, delivered by trained accredited assessors using the I-CAN v6 tool for participants 16 and over, with further targeted assessments for people with complex needs; the agency's stated aim is that providers spend their time "delivering NDIS supports rather than writing reports". Clinician-written functional evidence keeps its place at access, at review, and wherever a targeted question needs discipline-specific depth, but a report written in 2026 should be dated, purposeful and built to stand next to what is coming. For the neighbouring Australian pathway, Medicare-rebated psychology under Better Access, see the mental health treatment plan page; the two systems run on different laws, different budgets and different documents. ## Common NDIS functional capacity assessment errors reviewers flag There are no NDIS denial codes to study. Funding and access decisions are administrative decisions, reviewed on the evidence: a participant can seek internal review within 3 months and then the Administrative Review Tribunal within 28 days of the internal decision, and the NDIA's own guidance is blunt that an FCA is not required at all. A commissioned report therefore exists for exactly one reason, to be persuasive evidence against the statutory tests, and reports fail quietly, in review outcomes and planning decisions, rather than loudly in a remittance code. The BastionGPT Clinical Advisory Board sees the same errors most often in NDIS functional capacity assessment reviews: Diagnosis restated, function never evidenced. The report proves the impairment three ways and then asserts its impact in a sentence. The delegate already has the diagnosis; the FCA's entire job is the daily-life evidence, on most days, with and without support, across settings. Best-day snapshots. One clinic observation, one informant, no home or community data, and capability-with-prompting recorded as independence. Reviews read every source on file; a single-setting report contradicted by the carer interview discounts itself. The statutory phrase with no chain. Substantially reduced functional capacity appears in the summary, but no domain section shows the evidence that makes it true, and scores are reported without translation into tasks. Conclusory language persuades no one who is paid to test it. Recommendations without dosage or rationale. Supports listed with no type, frequency, duration or link from the impairment to the need, or written as a therapy treatment plan rather than support needs. A recommendation the planner cannot cost is a recommendation deferred. Stale or inconsistent evidence at the decision point. The report predates a change of circumstances, contradicts other reports on file without addressing them, or arrives at review restating the evidence the original decision already weighed. Reviews turn on what is new, current and reconciled. ### NDIS functional capacity assessment reports in the US, Canada, and Australia Aspect United States Canada Australia Status No NDIS equivalent; the nearest analogues are the Social Security residual functional capacity assessment and clinician medical source statements in disability claims No national equivalent; functional evidence flows through the CPP Disability medical report and each province's disability program forms Convention, not law: no instrument requires an FCA or sets its format; it exists as evidence for NDIS access and planning decisions Terminology Residual functional capacity (RFC), medical source statement, functional capacity evaluation (FCE) in workers' compensation and vocational contexts CPP Disability medical report, Disability Tax Credit certificate, provincial packages such as Ontario's Disability Determination Package Functional capacity assessment (FCA), functional assessment report, psychosocial functional capacity assessment What changes The agency, not the clinician, decides the RFC: 20 CFR 404.1545 defines it as the most a claimant can still do despite limitations, assessed from the whole record, and clinician input feeds that decision rather than making it CPP Disability turns on a "severe and prolonged" disability leaving the person incapable regularly of substantially gainful work; provincial programs set their own forms, thresholds and reviewers The report answers the NDIS Act's tests: substantially reduced functional capacity across six life domains, and supports that meet the reasonable and necessary criteria, with the NDIA deciding on the evidence Retention Part of the clinical record; state licensing rules govern, commonly 5 to 10 years and longer for minors Provincial colleges govern; Ontario's occupational therapy college requires 10 years after the last entry, with pediatric extensions Whole record kept 7 years from last service, or to age 25 for clients seen as minors, under state health records law; professional boards add their own duties An FCA is Australian by construction: it is written to the NDIS Act's tests and has no force anywhere else, while the US and Canadian documents answer work-capacity and benefit-eligibility questions under their own statutes. A clinician outside Australia meets an FCA only as inbound context, and relabelling an RFC opinion or a CPP-D medical report as an FCA, or the reverse, helps no one. What all three share is the discipline underneath: functional evidence, tied to a legal test, current enough to describe the person the decision is about. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on NDIS functional capacity assessment reports. Draft domain-by-domain findings from your assessment notes, observation records, informant interviews and score summaries, in your own report structure. Turn clinical findings into the plain-language functional impact summary that planners, participants and families all read, without losing the evidence chain. Check the draft before you sign: domains with no named source, capability-with-support recorded as independence, recommendations missing frequency or duration, and impact statements with nothing behind them. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there an official NDIS functional capacity assessment report template? No. No law, NDIS rule or NDIA standard prescribes a template, a section list or a length. The six functional domains, communication, social interaction, learning, mobility, self-care and self-management, come from the NDIS Act's disability requirements and the NDIA's access guidance, and professional convention builds the report around them. What decision-makers need is evidence they can apply to the statutory tests: function on most days, with and without support, across settings, and recommendations with type, frequency, duration and rationale. Most full reports run 15 to 30 pages and follow 1 to 10 hours of assessment time, the NDIA's own published range. The free NDIS functional capacity assessment report template on this page is a starting skeleton to adapt, not a mandated form. Is a functional capacity assessment required for NDIS eligibility? No. The NDIA's own guidance says directly that you do not have to do a functional capacity assessment to be eligible for the NDIS. What the scheme needs is evidence of how a permanent impairment reduces functional capacity, and an FCA is one strong way to provide it. In practice the document becomes close to indispensable at particular decision points: supported independent living, home modifications, assistive technology, and reviews where the original evidence did not carry the decision. Treat it as evidence commissioned to answer a question, never as a form that must be filled. Who can write an NDIS functional capacity assessment report? Does it have to be an OT? It does not. The NDIA names occupational therapists, physiotherapists, psychologists and medical specialists, with the only stated bar being that the assessor is qualified to make the assessment, and other allied health professionals assess within their scope. Occupational therapists write the most. Psychologists are often the right assessor for psychosocial disability, autism and intellectual disability, where adaptive measures such as the Vineland-3 and ABAS-3 carry much of the functional evidence. One 2026-27 pricing detail worth knowing: provisionally registered psychologists working under supervision bill under the psychologist line items, which makes supervised FCA work billable at the full psychologist price limit. Who pays for an FCA, and what does one cost? Usually the participant's NDIS plan, not Medicare. Under the NDIS pricing arrangements , therapy price limits are discipline-specific: on the 2026-27 schedule effective 1 July 2026, the national hourly limit is $252.99 for psychologists and $193.99 for occupational therapists, with higher remote loadings, and the schedule carries dedicated NDIA Requested Reports line items for report work the agency asks for. The NDIA's published range for assessment time is 1 to 10 hours depending on the disability and the professional, and scoring and writing add hours on top, so total cost scales with complexity. The Medicare pathway, Better Access, is a different system with different documents and does not fund FCAs. How is an FCA different from a psychological evaluation report? A psychological evaluation report answers a diagnostic question: what condition explains this presentation, established through interview and testing. An FCA answers a support question: what does this person do on most days, with and without help, and what funded supports does that evidence justify. The FCA cites diagnosis rather than making it, and a strong one leans on the diagnostic documents, an autism evaluation report for an autistic participant or a developmental assessment for a young child, then does its own distinct work across the six functional domains. Sending a diagnostic report to do an FCA's job, or the reverse, is one of the most common reasons NDIS evidence falls short. How current does an FCA need to be? No rule sets an expiry date. Decisions are made on present function, so the practical test is whether the report still describes the person in front of the delegate: a change of circumstances, a new diagnosis, finishing school or losing a carer ages a report faster than the calendar does. At review the timing rules are concrete: 3 months to request internal review, up to 28 days to give the NDIA further information during it, and 28 days after the internal decision to go to the Administrative Review Tribunal, and what usually moves a review is evidence that is new, current and reconciled with the rest of the file. Date the assessment sessions and the report separately, state what has changed since the last assessment, and build a reassessment trigger into your recommendations. How long do we keep FCA reports? Retention comes from health records law and professional rules, not NDIS law. In New South Wales and Victoria the floor is 7 years from the last service , or until age 25 for clients whose information was collected as minors, whichever runs longer, and the other states and territories run equivalent schemes. Registered NDIS providers also carry information-management obligations under the NDIS Practice Standards, and professional boards expect good records in any setting. Worth remembering while writing: the participant can read the report, the NDIA holds it in their record, and future assessors will build on it, three audiences that outlast the funding decision. Will the NDIA's new support needs assessment replace FCAs? Not now, and not wholesale. From April 2027 the NDIA is introducing support needs assessments delivered by trained, accredited assessors using the I-CAN v6 tool, developed with the University of Melbourne and the Centre for Disability Studies, for participants aged 16 and over, with further targeted assessments for people with more complex support needs; the surrounding framework rolls out in stages from mid-2026. The agency's stated aim is fewer participant-funded reports. Clinician-written functional evidence keeps its place at access, at review, and wherever a targeted question needs discipline-specific depth, but an FCA written now should be dated, purposeful, and ready to be read alongside an agency assessment. Can BastionGPT draft an NDIS functional capacity assessment report? Yes. Bring the assessment however it exists: interview and observation notes, informant summaries, Vineland-3, ABAS-3 or WHODAS 2.0 score summaries, prior reports. BastionGPT drafts the report in your structure, from referral context through domain-by-domain findings to the plain-language impact summary and dosed recommendations, and checks the draft for the failure points this page lists: a domain with no named source, capability-with-support recorded as independence, a recommendation missing frequency or duration. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Vineland-3 Report Write-Up ABAS-3 Report Write-Up Psychological Evaluation Report Autism Spectrum Evaluation Report Child & Adolescent Developmental Assessment Better Access Mental Health Treatment Plan Referral NDIS Plan Reassessment & Progress Report Aboriginal & Torres Strait Islander Culturally Safe Assessment Disability Evaluation & Medical Source Statement WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/collaborative-care-note All note templates # Collaborative Care (BHI/CoCM) Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A collaborative care note is the registry-backed record a behavioral health care manager keeps for each patient in the psychiatric Collaborative Care Model (CoCM) or general behavioral health integration (BHI). It logs contacts, validated symptom scores, psychiatric consultant recommendations, and clinical minutes across the calendar month. The billing practitioner's monthly claim under 99492, 99493, 99494, or 99484 stands on that time-and-registry trail. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The behavioral health care manager, under general supervision of the billing practitioner; consultant recommendations may be entered by either Audience The CoCM team, the billing practitioner, payers, and auditors reconstructing the month Typical length 75 to 250 words per registry entry · 3 to 8 minutes per entry by hand (clinical team estimate) Format family Registry-based monthly care management record (CoCM/BHI family) When it's used Every calendar month of a CoCM or BHI episode, from initiating visit through relapse-prevention planning and discharge Standards context Medicare payment policy dictates the content; no statute mandates a note format, and retention is the part that is law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a collaborative care (BHI/CoCM) note? A collaborative care note is the running record a behavioral health care manager (BHCM) keeps for each patient enrolled in the psychiatric Collaborative Care Model: dated registry entries for every contact, validated symptom scores, the psychiatric consultant's recommendations, and a log of clinical minutes that accumulates toward the month's billing threshold. The model traces to the IMPACT trial (Unützer and colleagues, JAMA, 2002), which randomized 1,801 older adults across 18 primary care clinics in 5 states; at 12 months, 45% of intervention patients had a 50% or greater reduction in depressive symptoms against 19% in usual care, and a follow-up analysis found lower total healthcare costs over 4 years. The University of Washington's AIMS Center, established by the trial's principal investigator, turned that protocol into the registry-and-caseload workflow the note now documents. CMS built the payment scaffolding in two steps: temporary codes G0502, G0503, G0504, and G0507 effective January 1, 2017, replaced by CPT 99492, 99493, 99494, and 99484 on January 1, 2018. You will also hear the note called a registry note, caseload review note, care manager contact note, BHCM note, or, for the lighter model, a general BHI note. Two boundaries do most of the work. First, this is not a generic care coordination note : CoCM documentation is registry-driven, measured with validated scales, reviewed weekly with a named psychiatric consultant, and banded to the calendar month, while general coordination notes follow no billing clock at all. (Tracking scores inside ordinary therapy belongs in an outcome measure review note .) Second, the "required fields" clinicians trade lists of are almost entirely payer policy, not law: CMS publishes no note template, and the registry, rating-scale, and weekly-review expectations live in the January 2026 MLN909432 booklet as conditions of Medicare payment. What is law: retention, 7 years from the date of service under 42 CFR 424.516(f), and the boundary in 45 CFR 164.501 that keeps this material out of protected psychotherapy-notes status, because registry entries, scores, medications, and session times belong to the ordinary medical record. ## Who uses collaborative care notes and when The daily author is the behavioral health care manager embedded in a primary care practice, though the model runs anywhere a billing practitioner with E/M scope operates: family medicine, internal medicine, pediatrics, and specialty clinics such as oncology or cardiology. The psychiatric consultant, typically remote, reviews the caseload weekly and leaves recommendations without seeing the patient; the treating practitioner bills the month. Clinical psychologists and clinical social workers keep the same style of record under G0323, with a psychiatric diagnostic evaluation (90791) as the initiating visit. Reach for this format when a patient is enrolled in CoCM or general BHI and the calendar month has to be reconstructable for a claim. Reach for a plain care coordination note when you are coordinating outside any billing model, and a discharge summary when the episode ends and care hands back to the referring or ongoing clinician. ## Collaborative care note structure: what goes in each section Team and month header. Patient identifiers, the calendar month, and the three named roles: treating (billing) practitioner, behavioral health care manager with credentials, and psychiatric consultant. CMS describes the BHCM as having "formal education or specialized training in behavioral health" (MLN909432), so record the credential you are relying on. Pitfall: a consultant who is never named in the record makes the three-member team unverifiable and the month's claim unsupportable. Consent and initiating visit. One line anchoring the consent conversation (date, who obtained it, cost sharing explained) and the visit that started services: an annual wellness visit, IPPE, comprehensive E/M, or TCM visit, or a 90791 evaluation for G0323. Consent may be verbal, is documented once, and is refreshed only if the billing practitioner changes. Pitfall: consent that lives in a clinician's memory instead of the record; auditors read absence as never obtained. Contact and activity log. The spine of the note: one dated entry per contact or care activity with the mode (phone, portal, video, in person), what clinically happened, and the minutes. Time spent documenting in the registry or EHR counts; scheduling, reminders, translation logistics, and travel do not. Pitfall: a single end-of-month block of minutes with no activities attached; time that cannot be reconstructed is time a reviewer will not credit. Validated measure tracking. The month's score on the scale the registry tracks (PHQ-9, GAD-7, or another validated instrument; CMS names no specific tool) beside the baseline and prior scores so the trajectory is visible at a glance. Pitfall: a filed score with no response to it; flat or worsening numbers over an unchanged plan invite the medical-necessity question. Psychiatric consultant review. The date the patient was covered in the weekly caseload review, the consultant's recommendations, and who entered them. The care manager may record the recommendations; the consultant is not generally expected to see the patient. Pitfall: a standing weekly meeting that never shows up in this patient's chart; for audit purposes it did not happen for this patient. Treatment-to-target response. What changed because of the numbers: a medication recommendation relayed to the prescriber, a brief evidence-based intervention such as behavioral activation or motivational interviewing, a stepped referral toward specialty care. Pitfall: describing activity without linking it to the target the registry tracks; "supportive contact" is not a treatment adjustment. Monthly time summary and code. Total countable clinical minutes for the calendar month and the code that total supports: 70 minutes for the first month, 60 for subsequent months, each additional 30 for the add-on unit, 30 for a partial month under G2214, 20 for general BHI. The month is the unit; minutes do not carry forward. Pitfall: quietly rolling leftover minutes into next month, or letting the billing practitioner's solo registry review pad the care manager's total. Next month and exit plan. What the team watches next, the re-administration date for the scale, and, as scores approach remission, relapse-prevention planning toward discharge from active treatment. An episode also ends after 6 consecutive months without CoCM services. Pitfall: episodes with no exit logic; a registry full of indefinite maintenance patients is a signal reviewers notice. ### Blank template (copy and adapt) COLLABORATIVE CARE (CoCM/BHI) MONTHLY NOTE Patient: ____________________ DOB: __________ Month/Year: __________ Billing practitioner: ________________________________________________ Behavioral health care manager (credentials): ________________________ Psychiatric consultant: ______________________________________________ Consent documented (date / verbal / cost sharing explained): _________ Initiating visit (type and date): ____________________________________ CONTACT AND ACTIVITY LOG (clinical minutes only) Date Mode Activity and clinical content Minutes _____ _______ _________________________________________ ______ _____ _______ _________________________________________ ______ _____ _______ _________________________________________ ______ _____ _______ _________________________________________ ______ _____ _______ _________________________________________ ______ VALIDATED MEASURES Scale: ________ Baseline: ____ Prior: ____ This month: ____ (date: _____) PSYCHIATRIC CONSULTANT REVIEW Caseload review date: ________ Recorded by: _________________________ Recommendations: _____________________________________________________ Relayed to billing practitioner (date / action taken): _______________ TREATMENT-TO-TARGET RESPONSE ______________________________________________________________________ MONTH TOTAL: ______ clinical minutes Code(s) supported: ___________ Clerical, travel, and non-BHCM time excluded from total: [ ] confirmed PLAN FOR NEXT MONTH / RELAPSE PREVENTION AND EXIT ______________________________________________________________________ Signature / credentials / date: ______________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample collaborative care note Scenario: a 58-year-old man enrolled in his primary care clinic's CoCM program for recurrent depression, month 3 of the episode. This is a complete month of collaborative care documentation: the registry contact log, the score trend, the consultant's recommendation, and the time total behind the claim. All details are fictional. Collaborative Care Monthly Note (CoCM). Patient: D.M., 58M · Month: July 2026 (month 3 of episode) · Dx: F33.1 · Billing practitioner: Dr. A. Patel (family medicine) · BHCM: L. Moreno, LCSW · Psychiatric consultant: Dr. R. Ellison, MD Consent and initiating visit: Verbal consent documented 05/06/2026 at comprehensive E/M visit with Dr. Patel; cost sharing explained; billing practitioner unchanged, no new consent required. Contact and activity log (clinical minutes): 07/02, phone: PHQ-9 administered, score 11; sleep log reviewed; behavioral activation plan reinforced. 14 min 07/08, video: behavioral activation session; activity scheduling around work shifts; barriers to walking plan addressed. 28 min 07/09, team: weekly caseload review with Dr. Ellison; case presented, recommendation below. 12 min 07/10, phone: consultant recommendation relayed to Dr. Patel; sertraline increased 50 mg to 100 mg, prescription filled 07/11. 8 min 07/16, clinic: brief in-person check; PHQ-9 8, GAD-7 7; medication tolerated, no side effects reported; walking goal met 4 of 7 days. 22 min 07/23, registry: flowsheet updated with treatment-to-target status; relapse-prevention items drafted for August. 8 min Validated measures: PHQ-9: 16 baseline (05/06), 12 (06/11), 11 (07/02), 8 (07/16). GAD-7: 9 (06/11), 7 (07/16). PHQ-9 item 9 scored 0 at both July administrations. Psychiatric consultant review (07/09): Partial response on sertraline 50 mg with PHQ-9 stalled near 11 for four weeks. Recommendation: increase to 100 mg, re-check score in 4 weeks, continue behavioral activation. Recorded by L. Moreno, LCSW; Dr. Ellison had no direct patient contact. Relayed to and actioned by Dr. Patel 07/10. Treatment-to-target response: Medication adjusted per consultant recommendation; behavioral activation continued with activity scheduling; PHQ-9 fell from 11 to 8 after the change. Month total: 92 clinical minutes. Supports 99493 (subsequent month) plus one add-on unit of 99494 for the additional 30 minutes. No clerical or travel time counted. No general BHI code reported this month. Plan: Re-administer PHQ-9 and GAD-7 first week of August; if PHQ-9 holds under 10, begin relapse-prevention planning toward discharge from active treatment around month 5; patient remains on the weekly caseload review list. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The team is named and roled. Billing practitioner, care manager with credentials, and psychiatric consultant are all identified, so the three-member model Medicare pays for is visible on the face of the note. Every minute has a clinical anchor. Six dated entries pair activities with minutes that sum to the 92-minute total; nothing clerical is counted, and a reviewer can rebuild the month line by line. The scores drive the care. A stalled PHQ-9 triggers a consultant recommendation, a same-week medication change, and a measured response: treatment to target, not passive tracking. The consultant's role is documented, not implied. Review date, recommendation, who recorded it, and the relay to the prescriber are all present, with no direct patient contact required. The month is self-contained with an exit in view. The codes match this month's arithmetic, nothing rolls forward, and the relapse-prevention plan shows the episode heading toward discharge rather than drifting. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Everything in the CoCM record, registry entries included, is ordinary medical record. 45 CFR 164.501 reserves psychotherapy-notes protection for a therapist's separated process notes and expressly excludes medication monitoring, session times, modalities, and clinical test results, which covers most of what a collaborative care note contains. It sits in the designated record set, patients can access it, and it is discoverable. Retention is the genuinely statutory layer: Medicare enrollment rules require the documentation behind orders, referrals, and payment requests to be kept for 7 years from the date of service (42 CFR 424.516(f)), CMS can revoke enrollment for failing to keep or produce it (42 CFR 424.535(a)(10)), Medicare Advantage contracts carry a 10-year federal audit window (42 CFR 422.504), and state medical-records law can extend the clock further. Hold the month's record for the longest rule that touches it. The payer chain is where collaborative care documentation actually fails. Reading Medicare's own claims data, Marcotte and colleagues found denials rose from 5% of CoCM and general BHI services in 2017 to 32% in 2018 while volume grew eightfold (Psychiatric Services, 2021), and early adopters interviewed by Carlo's team described the cumulative minute tracking and once-monthly claim entry as the arduous part of the model (Journal of General Internal Medicine, 2019). The conditions themselves are stable and knowable: an initiating visit on file, consent documented once, a named three-member team, a registry with a validated scale in active use, a documented consultant review, and a time log that clears the threshold for the code billed, with clerical, travel, and practitioner-solo time excluded and no general BHI code in the same month as CoCM for the same patient. The note's format is a convention; that evidence chain is the requirement. Keep the claim arithmetic beside the clinical story the way a billing note does, and keep the measurement thread consistent with the outcome measure review notes your therapists write outside the model. ## Common collaborative care note errors auditors flag There is no CoCM-specific OIG audit to point to: a July 2026 search of OIG's public reports returns none, and the audit heat in the care-management family sits on Chronic Care Management instead (Work Plan project OAS-26-09-007, announced March 16, 2026, covers CCM payments from 2019 through 2024). The warning lives in the denial data: Medicare denied 32% of CoCM and general BHI services in 2018, up from 5% in 2017, even as volume grew eightfold (Marcotte et al., Psychiatric Services, 2021). The BastionGPT Clinical Advisory Board sees the same errors most often in collaborative care note reviews: Time that cannot be reconstructed. A monthly total with no dated, activity-level entries behind it, or a total padded with scheduling, reminders, travel, or the billing practitioner's solo registry review, none of which count toward the threshold. No consultant fingerprints in the billed month. The weekly caseload review happened as a meeting but never reached this patient's chart: no date, no recommendation, no record that the case was covered. Registry and scale gaps. CoCM billed for months with no registry entry or no validated measure administered or reviewed, which severs the claim from the model's evidence spine. Consent nowhere in the chart. Verbal consent is enough, but only if documented; a missing consent line undermines every month that follows, and it is the cheapest fix on this list. Same-month conflicts and double counting. General BHI reported in the same month as CoCM for the same patient, add-on units the minutes do not support, or a licensed care manager's psychotherapy time counted toward both the therapy code and the CoCM threshold. ### Collaborative care notes in the US, Canada, and Australia Aspect United States Canada Australia Status Payer policy: a Medicare code family (99492, 99493, 99494, 99484) whose conditions of payment shape the note; no statute prescribes a format No CoCM billing analog; collaborative mental health care runs on provincially funded shared-care models and team programs No CoCM billing analog; the nearest MBS structures are episodic plan and case-conference attendance items Terminology CoCM note, registry note, caseload review note, BHCM note, general BHI note Shared care, collaborative mental health care (the CCMHI lineage), interprofessional team-based care GP Mental Health Treatment Plan, multidisciplinary case conference, team care arrangement What changes The calendar month is the unit: time thresholds, registry tracking, and weekly consultant review must be reconstructable for every month billed No time-banded month exists; physician billing is provincial fee-for-service and team funding flows through programs, so the record follows college documentation standards rather than a payment clock Items pay per attendance, not per month: a plan item such as 2715 pays a single preparation attendance, and case-conference items in the 735 family pay time-tiered conferences Retention 7 years from the date of service is law (42 CFR 424.516(f)); Medicare Advantage adds a 10-year audit window; state rules can run longer College standards govern; CPSO in Ontario requires 10 years from the last entry, or 10 years after a child patient turns 18 Whole record 7 years from last entry, until age 25 for clients under 18 The clinical model itself travels: measurement-based caseloads, a consulting psychiatrist, and treatment to target all exist inside Canadian shared-care programs and Australian integrated-care arrangements, and the logic of this note transfers with them. The billing scaffolding does not. Import the workflow rather than the US payment rules, and document to your province's college standard or to the MBS item actually being claimed. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on collaborative care notes. Draft the month's registry entries from bullets, dictation, or a call transcript, with the activity, clinical content, and minutes structured the way the claim needs them. Check the month before it goes out: time total against the code, consultant review present, scores tracked and responded to, consent and initiating visit anchored, no same-month code conflict. Convert the registry trail into whatever leaves the building: a referral summary, a handoff to specialty care, or a clean month narrative for the billing practitioner's record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How detailed does each collaborative care note entry need to be? Enough that a stranger can rebuild the month. Each registry entry carries the date, the mode of contact, what clinically happened, and the minutes, which usually lands between 75 and 250 words. The monthly total then has to be traceable to those entries: six short, dated entries that sum to 92 minutes beat one polished paragraph sitting over a bare total. The sample above shows the full pattern. Is there an official CMS template for CoCM documentation? No. CMS has never published a collaborative care note template, and no statute prescribes a format. What exists are conditions of payment in the MLN909432 booklet (January 2026): an initiating visit, documented consent, a registry in active use, validated rating scales, weekly psychiatric caseload review, and time thresholds per calendar month. Meet those on paper however your EHR allows; the format is yours, the evidence chain is not. Do I need written consent for collaborative care? No. CMS is explicit in MLN909432: "We don't require written consent." Verbal consent works, but it must be documented in the medical record, the patient must be told cost sharing applies, and a new consent is needed only if the patient changes billing practitioners. The recurring audit failure is not the consent conversation; it is the missing line documenting it. Whose minutes count toward the monthly threshold? Only the behavioral health care manager's clinical time: contacts with the patient, coordination with the team and outside providers, and time documenting in the registry or EHR, per the AIMS Center finance FAQ. Scheduling, appointment reminders, translation logistics, travel, and the billing practitioner's solo registry review do not count. A licensed care manager can also bill psychotherapy in the same month, but the same minutes can never serve both codes. Which billing codes does this note support? The psychiatric CoCM family: 99492 for the first calendar month (70 minutes of care manager time), 99493 for subsequent months (60 minutes), 99494 for each additional 30 minutes, and G2214 for a 30-minute partial month. General BHI runs on 99484 (20 minutes of clinical staff time), with G0323 as the clinical psychologist and clinical social worker version. CoCM and general BHI are never billed in the same month for the same patient, and the July 2026 CMS practitioner file lists a medically unlikely edit value of 4 for 99494, with payer variation, so confirm your own payer's limit. Does the psychiatric consultant have to write in the chart or see the patient? Neither. The care manager may record the consultant's recommendations, and CMS describes the consultant as not generally expected to have direct patient contact, prescribe, or deliver treatment directly. The consultant does not bill these codes either: the CMS FAQ published by the APA notes that "psychiatric work is defined as a sub-component of the psychiatric CoCM codes," so payment flows through the treating practitioner, who holds the consultant relationship by contract or employment. How is this different from a care coordination note? Scope and clock. A care coordination note records coordination work wherever it happens, with no registry, no consultant, and no monthly arithmetic. The collaborative care note is a payment-bearing record: registry-driven, scale-measured, consultant-reviewed, and banded to the calendar month. In between sits measurement-based care in ordinary therapy, which belongs in an outcome measure review note , and the end of any episode can feed a discharge summary . Do Canada or Australia have collaborative care billing codes? No. Neither country has a monthly, time-tracked CoCM code family. Canada funds collaborative mental health care through provincial shared-care and team programs, with documentation governed by college standards such as CPSO's 10-year retention rule in Ontario. Australia's nearest structures are episodic MBS attendances: Better Access plan items such as 2715 and time-tiered case-conference items in the 735 family. The clinical workflow transfers; the US billing rules do not. How does BastionGPT help with collaborative care documentation? Bring the month however it exists: call notes, a dictated recap, the registry export. BastionGPT drafts the dated entries with minutes and clinical content in place, then checks what reviewers read first: the time total against the code, the consultant review, the score trend and the response to it, and the consent anchor. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Care Coordination Note Outcome Measure Review Note Psychiatric Diagnostic Evaluation Discharge Summary Claim-Support Billing Note Multidisciplinary Team Meeting Note Case Conference Note Hospital Consult & Liaison Note Consultation Response & Report Chronic Care Management (CCM) Note Medication Management Follow-Up Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Australia Telehealth & MBS Case Conference Note 42 CFR Part 2 Consent Psychiatric Consultation Note PHQ-9 Score Documentation GAD-7 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/team-meeting-note All note templates # Multidisciplinary Team Meeting Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A multidisciplinary team meeting note records an interdisciplinary discussion of one client's care: who attended, what each discipline reported, what the team decided and why, and the action items with owners and due dates. Treatment teams in inpatient, rehabilitation, hospice, and community behavioral health programs write one after each team meeting or staffing. Most run 100 to 400 words per client discussed. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Team leads, therapists, case managers, nurses, and program staff across disciplines Audience The treatment team, program leadership, surveyors, payers, auditors Typical length 100 to 400 words per client · 5 to 15 minutes by hand (clinical team estimate) Format family Brief meeting summary (compare: case conference note, care coordination note) When it's used After each treatment team meeting, staffing, or care conference where a client is discussed Standards context The meeting is often mandated (IRF, hospice, SNF); the note's format is a convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a multidisciplinary team meeting note? A multidisciplinary team meeting note documents a case discussion held by clinicians from two or more disciplines: the assessment each discipline brought, the decisions the team made with their rationale, and the action items that left the room with an owner attached. It is a family of documents rather than a single form. The inpatient rehabilitation weekly team conference note, the hospice interdisciplinary group (IDG) note, the psychiatric hospital treatment plan review, the skilled nursing care conference note, the assertive community treatment (ACT) daily team log, and the tumor board summary are all members, and no professional body invented the form: the interdisciplinary team entered US regulation through the Medicare conditions of participation in the 1970s and 1980s and was consolidated in the 2007 to 2008 rewrites of the hospital and rehabilitation rules. You will also hear it called an MDT note, IDT note, treatment team note, staffing note, rounds note, or team conference note. The distinction that carries this page: where a team meeting is mandated, the mandate attaches to the meeting, not to the note's format. 42 CFR 412.622(a)(5) makes a documented weekly interdisciplinary team conference a condition of Medicare payment in inpatient rehabilitation facilities and names the required participants; hospice groups must review the plan of care at least every 15 calendar days under 42 CFR 418.56; skilled nursing teams build and revise the care plan under 42 CFR 483.21. None of these prescribes a template, a signature block, or a rule that every attendee sign. The nearest neighbor is the case conference note , a payer-defined event about one client with participant and time rules attached; the team meeting note covers the recurring internal meeting that may discuss several clients in one sitting. ## Who uses multidisciplinary team meeting notes and when Inpatient psychiatric units, rehabilitation facilities, hospices, skilled nursing facilities, partial hospitalization and intensive outpatient programs, ACT teams, and community mental health centers all run scheduled team meetings, and someone in the room, usually the team lead, the program nurse, or the client's primary clinician, writes the note. Group practices use a lighter version for weekly staffings. Write a team meeting note when the discussion is a recurring internal review. Switch to a case conference note when the meeting is a pre-arranged, often billable conference about one client, to a care coordination note for coordination that happens outside any meeting, and to a treatment plan review when the output is a formal revision of the plan itself. ## Multidisciplinary team meeting note structure: what goes in each section Meeting header. Date, program or unit, meeting type (scheduled review, staffing, discharge planning), and start and stop times wherever the meeting supports billing. Ontario's OHIP case conference codes require documented times and a minimum of 10 minutes of participation, and a surveyor reads an undated note as an undocumented meeting. Pitfall: a note that proves a discussion happened but not when; the date and times are the first things checked against the billing calendar. Attendees with disciplines. Name, credential, and discipline for every participant, plus who led. The IRF rule names the required roles: a rehabilitation physician who leads, a rehabilitation nurse, a social worker or case manager, and a therapist from each involved discipline. Pitfall: writing "team present"; a reviewer cannot verify composition requirements from a headcount. Client and participation. Which client the entry covers, and whether the client or family attended or were offered the chance. US skilled nursing rules and Ontario long-term care rules give residents the right to participate in care planning. Pitfall: silence on participation in settings where the opportunity is a documented right. Status by discipline. What each discipline reported: symptoms, scores, attendance, medication response, housing or benefits progress. Keep it specific to this client and this week. Pitfall: cloned text; the most-failed Joint Commission behavioral health standard involves treatment goals that are not individualized to the client. Decisions and rationale. What the team changed, kept, or escalated, and why, linked to treatment plan goals. Pitfall: recording the decision without the reasoning; a later reviewer, or the same team in three months, cannot reconstruct why the plan changed. Action items with owner and due date. One line each: task, named owner, timeframe. A Japanese hospital study found 7% of MDT recommendations were never implemented, so the follow-through trail matters. Pitfall: actions assigned to "team," which in practice means assigned to no one. Concurrence or sign-off. Who approved the outcome. In IRFs, CMS guidance requires the rehabilitation physician to document concurrence with team decisions at each meeting; in most other settings the leader or the author signs. Pitfall: chasing signatures from every attendee, which no US, Canadian, or Australian rule requires, while missing the one concurrence that is required. Next review. When the team looks at this client again, so the cadence is visible across entries. Hospice groups must revisit the plan of care at least every 15 calendar days; most other settings set a weekly or monthly rhythm. Pitfall: no next date; a reviewer cannot tell whether the client fell off the schedule or left the program. ### Blank template (copy and adapt) MULTIDISCIPLINARY TEAM MEETING NOTE Client: ______________ DOB: __________ Client ID: __________ Date: __________ Program/unit: _______________________________ Meeting type: [ ] scheduled review [ ] staffing [ ] discharge planning Start/stop time (if billed): ___________ Client/family present: [ ] yes [ ] no Opportunity to participate offered (LTC/SNF): [ ] yes [ ] n/a ATTENDEES (name, credentials, discipline; mark the leader) 1. _____________________________ 2. _____________________________ 3. _____________________________ 4. _____________________________ STATUS BY DISCIPLINE (this client, this period) _______________________________________________________________________ _______________________________________________________________________ DECISIONS AND RATIONALE (tie each to a treatment plan goal) _______________________________________________________________________ _______________________________________________________________________ ACTION ITEMS (task, named owner, due date) 1. ____________________________________________________________________ 2. ____________________________________________________________________ Physician/leader concurrence (where required): _________________________ Next team review date: __________ Author signature/credentials/date: _____________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample multidisciplinary team meeting note Scenario: weekly treatment team meeting at a community mental health center, reviewing a 29-year-old client in the adult coordinated care track. This is one client's entry from the meeting record. All details are fictional. Multidisciplinary Team Meeting Note. Client: R.B., 29 · Date: 07/21/2026 · Program: adult coordinated care, CMHC · Meeting: weekly treatment team, 2:00 to 2:14 pm for this client · Client present: no; invited 07/17, declined this week Attendees: S. Okafor, MD (psychiatry, team lead); J. Rivera, LPC (primary therapist, author); M. Chen, RN (nursing); T. Alvarez, MSW (case management); D. Boone, CPS (peer support). Status by discipline: Psychiatry: paliperidone injection given 07/09, next due 08/06; client reports improved sleep, no side effects at 07/14 med check. Therapy: attended 2 of 2 sessions since last review; PHQ-9 11 on 07/14, down from 14 on 06/23; practicing thought records. Nursing: metabolic labs pending, last drawn 04/2026. Case management: housing subsidy application submitted 07/10, waitlist confirmation received; client raised interest in part-time work. Peer support: met 07/16; client attended one community group. Decisions and rationale: 1) Continue current medication unchanged; symptom trend and adherence both improving (supports plan Goal 1, symptom stability). 2) Add a supported employment referral; client raised the work goal independently in therapy and with peer support, and the team judged readiness adequate (new objective under Goal 2, community functioning). 3) Order metabolic labs before the 08/06 injection per monitoring schedule. Action items: T. Alvarez: submit supported employment referral by 07/24. M. Chen: schedule labs so results precede 08/06. J. Rivera: review sleep log and introduce work-stress coping plan at 07/28 session. D. Boone: accompany client to next community group, week of 07/27. Concurrence and next review: Dr. Okafor reviewed and concurred with team decisions. Client to be offered the 08/04 meeting; next full team review 08/04/2026. Entry by J. Rivera, LPC, 07/21/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Composition is verifiable. Every attendee is named with credentials and discipline, the leader is identified, and the author is marked, so a reviewer can check the roster against any staffing requirement. The entry is client-specific. Scores with dates, an injection schedule, a housing application, and a named group: nothing here could be pasted into another client's chart, which is the cloning failure surveyors flag most. Decisions carry rationale and goal links. Each decision states why and names the treatment plan goal it serves, which is the medical-necessity thread an auditor follows. Every action has an owner and a date. Four tasks, four named owners, four deadlines; the note can be audited for follow-through at the next meeting. Concurrence and cadence close the loop. The physician's concurrence is documented and the next review is dated, covering the two elements that recur in payment rules. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Decide, before you write, whether the note is a chart entry or a committee document, because the legal consequences differ. An entry about one client's care belongs in that client's record, is part of the designated record set, and is releasable and discoverable like the rest of the chart. Minutes of a quality-improvement or peer-review discussion may be protected from discovery by state privilege statutes, but that protection is state-specific and fragile: in Siegel v. Snyder (New York, 2021), committee minutes that did not identify who said what lost protection for those statements. Keep per-client clinical entries and quality-committee analysis in separate documents, and know which one a records request reaches. When substance use disorder information from a Part 2 program is discussed, 42 CFR Part 2 limits what can be shared and documented outside that program. The payer layer is narrower than most teams assume. CPT codes 99366, 99367, and 99368 describe medical team conferences, but Medicare assigns them bundled status and pays nothing separately, and several commercial payers follow. Ontario pays physicians through OHIP case conference K-codes that require documented times and participation, and accept one shared chart note initialled by the physician participants. Australia pays through time-tiered MBS case conference items carrying contemporaneous record obligations. The meeting is the mandate and the note is the evidence; format everything else around your program's needs. Retention runs long: Canadian psychology and social work colleges generally require 10 years, Australian statutes 7 years for adults or to age 25 for minors, and US state rules vary around 5 to 10 years. When the meeting produces a formal plan change, record it in a treatment plan review ; when it ends an episode, the trail feeds the discharge summary . ## Common multidisciplinary team meeting note errors auditors flag The numbers here are unusually concrete. The Joint Commission's behavioral health treatment-planning standard CTS.03.01.03 was the single most-failed standard on 2020 behavioral health surveys, with a 61.69% noncompliance rate, driven by cloned and non-individualized goals. In hospice, where the interdisciplinary group note is a condition of participation, a 2026 OIG report found documentation for 45 of 100 sampled certification periods did not meet Medicare requirements and estimated $255.1 million in avoidable payments. The BastionGPT Clinical Advisory Board sees the same errors most often in team meeting note reviews: Cloned, interchangeable entries. The same status text and goals across clients or across weeks; surveyors specifically flag goals not stated in the client's own words and reviews that repeat by template. A meeting the record cannot prove. Missing date, missing times where billing needs them, or attendees listed without disciplines, leaving composition requirements unverifiable. Group discussion without per-client detail. One block note covering several clients with nothing individualized; in PHP and IOP audits, a group note that fails to address each participant individually is a recurring denial basis. The required concurrence is missing. IRF reviews target the rehabilitation physician's documented concurrence; teams chase every attendee's signature instead, which no rule requires. Recommendations without owners. Decisions recorded with no named person or due date; published MDT research found 7% of recommendations were never implemented, and an unowned action is where that starts. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on multidisciplinary team meeting notes. Draft the per-client entry from a dictated recap, the meeting agenda, or a transcript, structured into status by discipline, decisions with rationale, and owned action items. Check the note before it is filed: attendees with credentials, client-specific content, a rationale on every decision, an owner and date on every action, and concurrence where the setting requires it. Convert one meeting record into per-client chart entries, a treatment plan review draft, or a plain-language update for the client or family. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to write a note for every client discussed in the team meeting? It depends on the setting and the payer. In inpatient rehabilitation, skilled nursing, and hospice, and for any billed case conference, the discussion belongs in the discussed client's record. For internal staffings there is no universal rule: follow your program's policy, and be consistent, because an auditor will compare the meeting schedule to the charts. Does everyone at the meeting have to sign the note? No. No US, Canadian, or Australian instrument requires every attendee to sign. Ontario's OHIP accepts one shared chart note signed or initialled by the physician participants, and the US IRF rule requires the rehabilitation physician to document concurrence with team decisions, not to collect signatures. All-attendee signature blocks are convention, and chasing them delays filing. Can we bill for the team meeting itself? Sometimes, and the geography matters. CPT 99366 to 99368 describe team conferences, but Medicare bundles them and pays nothing separately. Ontario pays physicians through OHIP case conference K-codes with time and documentation rules. Australia pays through MBS case conference items 735 to 758. If you bill any of these, the note must carry that scheme's required elements: participants with disciplines, start and stop times, and the outcome. Are team meeting minutes discoverable, or releasable to the client? It depends on what the document is. A per-client entry in the chart is part of the record set: releasable on request and discoverable. Quality-assurance or peer-review minutes may be privileged under state law, but the protection varies widely and can be lost; in New York, minutes that failed to identify the speaker lost protection for those statements (Siegel v. Snyder, 2021). Ask your counsel which category your meeting record falls into before a request arrives. Do I need the client's consent to be discussed at a team meeting? Within a treatment team, generally no: sharing information for treatment among providers is permitted under HIPAA without separate authorization. The significant exception is substance use disorder information covered by 42 CFR Part 2, which follows its own consent rules. Long-term care settings add a different right: the resident must be given the opportunity to participate in care planning. What is the difference between a team meeting note and a case conference note? The team meeting note records a recurring internal meeting that may cover several clients; its obligations come from the care setting. A case conference note records a pre-arranged, single-client, multi-provider conference and is the version payers define and pay for, with participant minimums and time rules attached. If you intend to bill, write the case conference version; for coordination work that happens between meetings, use a care coordination note . How long do we keep team meeting notes? As long as the chart they live in. Canadian psychology and social work colleges generally require 10 years from last contact or from the client's 18th birthday. Australian health records statutes in the ACT, NSW, and Victoria require 7 years for adults, or to age 25 for records made while the client was a minor. US retention is state law, commonly 5 to 10 years. Minutes held as quality documents follow your organization's retention policy instead. How does BastionGPT help with team meeting documentation? Bring the meeting however it exists: a dictated recap, the agenda with margin notes, or a transcript. BastionGPT drafts the per-client entry with status by discipline, decisions with rationale, and owned action items, then checks the things reviewers look for first: attendees with credentials, client-specific content, and the concurrence line where your setting requires one. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Care Coordination Note Collaborative Care (BHI/CoCM) Note Discharge Summary Treatment Plan Review & Update Case Conference Note Hospital Consult & Liaison Note Consultation Response & Report Chronic Care Management (CCM) Note Australia Telehealth & MBS Case Conference Note Case Management Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/case-conference-note All note templates # Case Conference Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A case conference note documents a pre-arranged meeting held by two or more providers to discuss and direct one identified client's care. It records the participants, times, decisions, and allocated tasks, plus whatever elements the payer requires when the conference is billed. Conferences are billable in Australia and Ontario and in limited US contexts. Most notes run 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The organizing clinician; every provider billing a conference item keeps their own record Audience Conference participants, the referring team, the client's chart, payers, auditors Typical length 150 to 400 words · 10 to 15 minutes by hand (clinical team estimate) Format family Meeting note (compare: multidisciplinary team meeting note, care coordination note) When it's used After a scheduled multi-provider conference about a single client, especially when billed Standards context Billed conferences carry enforceable content rules (MBS, OHIP); unbilled ones are convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a case conference note? A case conference note is the record of a formal, pre-arranged meeting held to discuss and direct the care of a single identified client by two or more providers, sometimes with the client, family, or other supports present. Unlike an ordinary progress note, its subject is a multi-party meeting rather than a one-to-one encounter. The most codified version is Australian: the Health Insurance (General Medical Services Table) Regulations 2021 define the multidisciplinary case conference and make an enumerated content list a condition of billing MBS items 735 to 758, a lineage that runs back to the Enhanced Primary Care package of the late 1990s. North America codified it differently: as the care planning conference in US long-term care and hospice regulation (42 CFR 483.21 and 418.56), as a mandated conference with listed note contents in Ontario long-term care (O. Reg. 246/22, s. 27), and as OHIP case conference K-codes for physicians. You will also hear it called a case conference record or summary, care conference note, family meeting note, or case review note. Two boundaries matter. First, the multidisciplinary team meeting note covers the recurring internal meeting that may discuss several clients; the case conference note is single-client and, when billed, must satisfy the item's exact content rules. Second, the US special-education "case conference committee" under IDEA is a separate statutory instrument and not a clinical case conference. No universal template exists in any of the three countries: the enforceable content lists are jurisdiction- and payer-specific, and in Australia they bind only the practitioner who bills the item, a point the explanatory notes make expressly. ## Who uses case conference notes and when Therapists, psychologists, GPs and psychiatrists, allied health providers, and care coordinators use case conferences when one client's care crosses enough providers that decisions need to be made together: complex chronic conditions with behavioral health involvement, adolescents with school and family systems in play, discharge planning across agencies, and palliative coordination. In Australia the conference itself is a fundable service with organiser and participant items, including mental health case conferencing items introduced 1 July 2023; in Ontario, physicians bill OHIP K-codes; in the US, most clinicians document the conference inside care management work rather than billing it separately. Write a care coordination note for asynchronous coordination that is not a scheduled meeting, and a treatment plan review when the work product is a plan revision rather than a conference record. ## Case conference note structure: what goes in each section Purpose and pre-arrangement. State what the conference was convened to decide and that it was scheduled in advance. OHIP defines a case conference as a pre-scheduled meeting to discuss and direct the management of an individual patient, and expressly excludes educational rounds and journal clubs. Pitfall: writing up an ad hoc hallway discussion as a conference; the pre-arrangement is part of the definition. Participants and minimum team. Every participant with name, credentials, discipline, and role, and whether the minimum team rule is met. Australian GP items require at least three members, no more than two of whom may be medical practitioners, each providing a different kind of care; OHIP requires the most responsible physician plus at least two others. Pitfall: a conference of three physicians, which fails the Australian mix requirement outright. Consent and attendance. Record the client's agreement to the conference before it happens, and who attended. In Australia the client's agreement must be recorded and, for the mental health items, agreement to each participant; the client does not have to attend. In Ontario long-term care the resident and substitute decision-maker must be given the opportunity to participate. Pitfall: consent that exists in conversation but nowhere in the record. Times and mode. The day, start and end times, and each participant's mode (in person, video, phone). Under the Australian regulations, recording the day and the start and end times is a legal precondition of billing an organiser item; OHIP pays in time units with a 10 minute minimum. Pitfall: a note with no times; for billed conferences that is not a style problem, it is an unpayable item. Discussion: needs and outcomes. The clause the Australian definition enumerates travels well everywhere: the client's history, their multidisciplinary care needs, the outcomes to be achieved, and a review of previously identified outcomes. Pitfall: "discussed client's care" with no content; an auditor treats an empty discussion section as an undocumented service. Decisions and tasks allocated. What was decided and which named team member owns each task, with timeframes. Pitfall: outcomes with no owner; task allocation to named members is part of the Australian definition and the practical point of the meeting everywhere. Distribution. A copy in the client's record, a summary offered to the client and carer, and a summary to each team member, all of which the Australian organiser rules require. Pitfall: a conference no one else can see; the distribution trail is the difference between coordination and a private memo. Signature. The author signs with credentials. No regulation reviewed in the US, Canada, or Australia requires every participant to sign; OHIP accepts one shared chart note signed or initialled by the physician participants, and the Australian rules ask the organiser to record participants and distribute summaries instead. Pitfall: circulating the note for every attendee's signature; the trail reviewers check is participants recorded and summaries distributed, not a page of countersignatures. ### Blank template (copy and adapt) CASE CONFERENCE NOTE Client: ______________ DOB: __________ Date: __________ Purpose of conference (pre-arranged): __________________________________ Client agreement recorded (date, scope): _______________________________ Client/carer attending: [ ] yes [ ] no (most schemes require recorded consent, not attendance) PARTICIPANTS (name, credentials, discipline/role, mode: in person/video/phone) 1. _____________________________ 2. _____________________________ 3. _____________________________ 4. _____________________________ Minimum team check (payer-specific, e.g. 3+ members, mix of care types): [ ] Start time: ________ End time: ________ Total: ________ DISCUSSION History and current needs: _____________________________________________ Outcomes to be achieved: _______________________________________________ Previously identified outcomes reviewed: _______________________________ DECISIONS AND TASKS ALLOCATED (task, named member, timeframe) 1. ____________________________________________________________________ 2. ____________________________________________________________________ 3. ____________________________________________________________________ DISTRIBUTION Copy in client record: [ ] Summary offered to client/carer: [ ] Summary provided to team members: [ ] Next conference or review date: __________ Author signature/credentials/date: _____________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample case conference note Scenario: an outpatient behavioral health practice convenes a scheduled conference on a 47-year-old client with major depressive disorder and chronic low back pain whose recovery has stalled across providers. All details are fictional. Case Conference Note. Client: L.K., 47 · Date: 07/22/2026 · Purpose: coordinate care for depression complicated by chronic pain and work absence · Pre-arranged: yes, scheduled 07/10 · Client agreement recorded 07/10, including agreement to all participants; client elected not to attend Participants and mode: A. Whitfield, LCSW (primary therapist, organizer, in person); Dr. P. Nunez, MD (family medicine, video); R. Osei, DPT (physical therapy, video); J. Lam, RN (practice care coordinator, in person). Minimum team check: 4 participants, 3 disciplines beyond the organizer, met. Times: 12:05 to 12:39 pm (34 minutes). Discussion: History reviewed: PHQ-9 16 on 07/15, unchanged over eight weeks; sleep fragmented by pain; out of work since 04/2026. Needs identified: depression treatment is stalled while activity avoidance and deconditioning maintain the pain cycle, and the client is discouraged by conflicting advice. Outcomes to be achieved: a single graded-activity message across providers, behavioral activation targets tied to physical therapy milestones, and a medication review for sleep. Review of prior outcomes: the 05/2026 plan (therapy plus standalone home exercise) was not implemented; client reported not understanding how the pieces fit. Decisions and tasks: 1) R. Osei: restart supervised PT twice weekly with a written graded-activity ladder, share the ladder with the team by 07/29. 2) A. Whitfield: align behavioral activation homework to the PT ladder from the 07/28 session onward; re-administer PHQ-9 at four weeks. 3) Dr. Nunez: medication review for sleep at the 08/03 visit; assess fitness-for-work pathway. 4) J. Lam: send the conference summary to all participants and offer a copy to the client by 07/24; book the review conference. Distribution and next steps: Copy filed in the client record; summary offered to the client 07/23 and accepted; summaries sent to all participants. Review conference set for 09/09/2026. Entry by A. Whitfield, LCSW, 07/22/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The definition is satisfied on the face of the note. Pre-arranged, single client, stated purpose, and a decision-directing discussion: the four things that make it a case conference and not a hallway chat. Consent is recorded, attendance is not assumed. The client's agreement, including to the participants, is dated before the meeting, and the note is explicit that the client chose not to attend, which no rule requires. The team mix and times are auditable. Four named participants with disciplines and modes, and a 34-minute bracket, cover the participant-minimum and time-tier questions every payer scheme asks. Outcomes are reviewed, not just set. The note assesses why the prior plan failed before allocating new tasks, mirroring the review element the Australian definition enumerates. Every task has a name and a distribution trail exists. Tasks are allocated to named members with dates, and the record shows the summary went to the team and was offered to the client. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The case conference note is unusual: in two of the three countries, parts of its content list are set out in the instrument itself. Ontario's O. Reg. 246/22, s. 27 requires long-term care homes to hold a care conference within six weeks of admission and at least annually, and to keep a record of the date, the participants, and the results. Australia's General Medical Services Table Regulations make the recorded agreement, the day, the start and end times, the participants, the enumerated discussion activities, the copy in the record, and the summary offers legal preconditions of billing an organiser item, and the consultant physician notes add that a patient would not normally need more than five case conferences in 12 months. Only the practitioner who bills must meet the item's requirements; other participants document under their ordinary professional obligations. The format of the note is convention everywhere; those content lists are not. In the US there is no standalone Medicare professional fee for a routine case conference: CPT team conference codes are bundled, and conference work is usually documented inside chronic care management, transitional care management, or care planning, so the note supports medical necessity rather than a conference claim; the care coordination note covers that documentation chain. One US change worth knowing: the 42 CFR Part 2 final rule (compliance date 16 February 2026) lets a single patient consent cover future uses and disclosures for treatment, payment, and operations, which simplifies bringing substance use disorder information into a multi-provider conference. Retention follows the chart the note sits in: state-law periods in the US (Massachusetts, one representative example, requires at least seven years from the last encounter), 10 years from last contact or the client's 18th birthday under Ontario college standards, and seven years for adults or until age 25 for minors under the Australian Psychology Board framework, with some state rules running longer. ↑ Back to the template and downloads ## Common case conference note errors auditors flag Case conference billing is actively audited. Australia's Professional Services Review has published case outcomes involving item 758, including a practitioner who rendered the highest volume of the item nationally and could not show the requirements were met, and PSR outcomes repeatedly turn on records that fail one enumerated element. In the US, an OIG audit of transitional care management found almost $1.7 million in overpayments where multiple practitioners billed the same beneficiary's same 30-day period. Ontario reviews check the pre-scheduled character, the participant minimum, and the time records behind K-code claims. The BastionGPT Clinical Advisory Board sees the same errors most often in case conference note reviews: No start and end times. The most expensive omission: times select the time tier and are a billing precondition in Australia and the unit basis for OHIP claims. Team composition that fails the rule. Too few participants, or the wrong mix, such as three doctors and no other kind of care under the Australian minimum-team requirement. Consent recorded nowhere. The client's agreement, obtained before the conference, is a recorded element in Australia; a missing consent line can unwind the claim. A plan review dressed as a conference. Australian rules require the conference to be clinically separate from preparing or reviewing a care plan; billing both for one discussion is a flagged pattern. No distribution trail. No copy in the record, no summary to the team or client; the note exists but the coordination it claims cannot be shown to have left the room. ### Case conference notes in the US, Canada, and Australia Aspect United States Canada Australia Status No standalone Medicare conference fee; team conference CPT codes are bundled, and conference documentation lives inside care planning law (42 CFR 483.21, 418.56) and care management billing Payer policy plus one statute: OHIP K-codes (K121, K124, K700-series) for physicians, and O. Reg. 246/22 s. 27 mandating long-term care conferences with listed note contents Delegated legislation: the GMST Regulations define the conference and make the content list a billing precondition for items 735 to 758 and the mental health case conferencing items Terminology Care planning conference, family meeting, IDT conference Case conference (K-codes), care conference (long-term care) Multidisciplinary case conference; organiser and participant items What changes Little is separately billable; the note's job is supporting medical necessity inside CCM, TCM, and plan documentation Physician billing rules: pre-scheduled meetings only, most responsible physician plus two others, 10 minute minimum, documented times; one shared note initialled by physician participants suffices Recorded consent, day and start/end times, minimum team of three with no more than two doctors, enumerated discussion elements, copy in the record, and distributed summaries are all payment preconditions Retention State law varies (Massachusetts: at least 7 years from the last encounter); Medicare claim-supporting documentation carries its own multi-year periods College standards: 10 years from last contact or from the client's 18th birthday, whichever is later (CPBAO, CRPO) 7 years for adults; for minors, until age 25 (Psychology Board framework; some state rules run longer) The conference itself looks the same in all three countries: scheduled, multi-provider, single client. What changes is enforcement. Australian and Ontario billers work against enumerated content lists with audit programs attached, while US clinicians write the note to prove coordination and necessity inside other billed services. Write to the strictest list that touches your claim. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on case conference notes. Draft the conference record from your agenda, dictation, or a meeting transcript, with participants, times, decisions, and allocated tasks in their places. Check the note against the elements your payer enforces: consent recorded, minimum team met, start and end times, outcomes reviewed, and the distribution trail, before anything is claimed. Convert the record into the summaries the rules expect: a plain-language copy for the client and carer, and a team summary ready to send to each participant. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does the client have to attend the case conference? Usually not, and the rules are precise. In Australia the client must agree to the conference (and, for the mental health items, to each participant) but does not have to attend. In Ontario long-term care, the resident and any substitute decision-maker must be given the opportunity to participate fully. In US settings, attendance follows the care-planning participation rights of the setting rather than a conference-specific rule. Can a case conference happen by phone or video? Yes. Australian items allow each participant to join in person, by video, or by phone, and OHIP permits personal attendance, videoconference, or telephone. Record each participant's mode in the note, because a reviewer reconstructing the conference will look for it. Who can bill for a case conference? It depends on the scheme. In Australia, a medical practitioner organises and bills the organiser items (735, 739, 743 by time tier) while other providers bill participation items, including allied health items 82001 to 82003 for eligible clients. In Ontario, physicians bill the K-codes. In the US there is no separately paid Medicare conference fee for routine care; the work is documented inside care management services, the chain the care coordination note covers. Do all participants have to write their own note? No. In Australia, only the practitioner who bills an item must meet all of that item's record requirements; other participants document under their ordinary professional obligations. In Ontario, one shared chart note signed or initialled by the physician participants satisfies the record requirement. Non-billing clinicians should still note their participation and any tasks they accepted, because those tasks live in their own scope of practice. How many case conferences can one client have? No hard cap exists for most items, but Australia's consultant physician notes state that a patient would not normally require more than five case conferences in a 12-month period, and auditors treat that as a benchmark. If a client genuinely needs more, the records need to show why. Do team rounds or a journal club count as a case conference? No. OHIP expressly excludes educational meetings such as rounds, journal clubs, and continuing professional development sessions from case conference billing, and the Australian definition requires a conference about an individual patient's care. A recurring internal meeting that reviews several clients is a multidisciplinary team meeting, which has its own note. How long should a case conference note be, and how long is it kept? No jurisdiction reviewed mandates a template or a minimum length; most case conference notes run 150 to 400 words, and a billed conference needs the payer's enumerated elements more than it needs length. Retention follows the record the note sits in: state law in the US (Massachusetts, for example, requires at least 7 years from the last encounter), 10 years from last contact or the client's 18th birthday under Ontario college standards, and 7 years for adults or until age 25 for minors under the Australian framework. What exactly must an Australian case conference note contain? The regulations enumerate it: the client's recorded agreement, the day and the start and end times, the participants, and the discussion elements (history, multidisciplinary care needs, outcomes to be achieved, tasks allocated to team members, and review of previously identified outcomes), plus a copy in the medical record, a summary offered to the client and carer, and a summary to each team member. The conference must also be clinically separate from preparing or reviewing a care plan. How does BastionGPT help with case conference documentation? Bring the conference however you captured it: dictation on the walk back, the agenda with notes, or a transcript. BastionGPT drafts the record with participants, times, decisions, and allocated tasks structured for your payer's checklist, flags missing preconditions like the consent line or an unmet team minimum, and produces the client and team summaries the distribution rules expect. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Care Coordination Note Collaborative Care (BHI/CoCM) Note Discharge Summary Treatment Plan Review & Update Multidisciplinary Team Meeting Note Hospital Consult & Liaison Note Consultation Response & Report Chronic Care Management (CCM) Note Australia Telehealth & MBS Case Conference Note Case Management Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/consultation-liaison-note All note templates # Hospital Consult & Liaison Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A consultation-liaison (C-L) note records a psychiatric assessment of a hospital patient admitted under a non-psychiatric team, performed at that team's request. It answers the consult question, documents mental status, capacity, and risk, and returns numbered recommendations the requesting team can act on the same day. Most run 200 to 600 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Consulting psychiatrists, PMHNPs, and hospital psychologists on C-L services Audience The requesting medical or surgical team, the hospital record, payers, auditors Typical length 200 to 600 words · 15 to 25 minutes by hand (clinical team estimate) Format family Specialist consultation note (compare: psychiatric diagnostic evaluation, discharge summary) When it's used After a bedside psychiatric assessment requested by another hospital team Standards context No mandated format anywhere; hospital record rules and billing pathways shape the content On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a consultation-liaison note? A consultation-liaison note is the written record of a psychiatric assessment performed on a patient admitted under a medical, surgical, obstetric, or neurological team, at that team's request, in a general hospital. It works double duty: a permanent entry in the shared record and a communication the requesting team can act on. The recognizable format has a scholarly lineage, not a regulatory one: Garrick and Stotland proposed the schema in the American Journal of Psychiatry in 1982, and Alexander and Bloch refined it in 2002 around the referrer's needs. The subspecialty itself was renamed from Psychosomatic Medicine to Consultation-Liaison Psychiatry after a 2017 board vote, with the Academy of Consultation-Liaison Psychiatry adopting its current name in April 2018. You will also hear the note called a psychiatric consultation note, C-L consult, liaison psychiatry review, or, historically, a psychosomatic medicine consult. The legal reach is thinner than most clinicians assume. US hospital regulation touches the note only generically: 42 CFR 482.24 requires the record to contain "results of all consultative evaluations" and to be complete within 30 days of discharge, and The Joint Commission expressly leaves authentication timeframes to hospital policy. No US, Canadian, or Australian instrument prescribes a C-L format, requires a written consult request, or sets a turnaround clock. The nearest neighbor is the psychiatric diagnostic evaluation : a US initial C-L encounter is often billed as one (90792 with medical services), but that evaluation opens your own treatment episode, while the C-L note answers another team's question inside the shared hospital chart. The outbound letter that closes an outside referral loop is a consultation report; the C-L note never leaves the chart. ## Who uses consultation-liaison notes and when Consulting psychiatrists and psychiatric nurse practitioners write them on hospital C-L services, and hospital psychologists write the parallel health-psychology assessment when the primary diagnosis is medical. The trigger is a request from the admitting team: altered mental status, depression screening after a cardiac event, capacity questions around refusal of care, overdose follow-up, delirium versus depression, or agitation management. The initial assessment resembles a focused psychiatric diagnostic evaluation built around the consult question, with the mental status exam carrying the objective weight. When the patient leaves the hospital, findings worth keeping flow into the discharge summary . Two neighbors stay separate: psychiatric input delivered through a primary care program's registry review is documented in a collaborative care note , and the reply that leaves the hospital as a letter to an outside referrer is a consultation report, not a chart entry. ## Consultation-liaison note structure: what goes in each section Consult request. Who asked, the question as they asked it, and the urgency. The stated reason anchors medical necessity for the visit code and defines what the note must answer. Pitfall: no recorded consult question; a note that never says why psychiatry was called cannot show it answered anything. Medical context. Admission reason, relevant medications, labs, vitals, and the current medical course, briefly. The psychiatric assessment has to sit inside the medical picture, not beside it. Pitfall: a free-floating psychiatric note that ignores the anticholinergic burden, the sodium, or the post-operative day. Focused history. Psychiatric history, substance use, current psychotropics, and collateral, selected for the question. Alexander and Bloch's point still stands: the conventional full psychiatric history serves the psychiatrist, not the requesting team. Pitfall: a clinic-length history no hospitalist will read; length is not thoroughness here. Mental status examination. The objective core, with attention and orientation testing whenever delirium is in play. Pitfall: "alert and oriented" standing in for an examination; hypoactive delirium hides inside exactly that phrase. Capacity and risk. Address decisional capacity whenever the question touches consent or refusal, and risk in every note. A UK hospital audit found capacity documented in 19.8% of liaison reviews and risk in 27.9% at baseline, which is the gap to write against. Pitfall: silence; these two lines are the most-missed and the most-litigated. Impression and differential. Your clinical judgment against the medical differential: delirium versus depression, akathisia versus anxiety, intoxication versus psychosis. Pitfall: naming a psychiatric disorder without engaging the medical causes the team is worried about. Numbered recommendations. The section the requesting team actually reads. Cover safety, non-pharmacologic steps, medication with dose and route and interactions, monitoring, and disposition, one number each. Pitfall: recommendations buried in a paragraph; if the intern cannot turn them into orders in one pass, the consult has not landed. Communication and follow-up. Who on the team you spoke with and when, whether C-L will follow, and the plan if the picture changes. Pitfall: no documented closure; the conversation happened, the record cannot prove it. ### Blank template (copy and adapt) HOSPITAL CONSULT / CONSULTATION-LIAISON NOTE Patient: ______________ MRN: __________ Date/time of consult: __________ Requesting team/attending: _____________________________________________ Consult question (as asked): ___________________________________________ Urgency: [ ] routine [ ] urgent MEDICAL CONTEXT Admission reason/date: _________________________________________________ Relevant meds, labs, vitals: ___________________________________________ FOCUSED HISTORY (psychiatric, substance, medication, collateral) _______________________________________________________________________ _______________________________________________________________________ MENTAL STATUS EXAMINATION (include attention/orientation if delirium is in question) _______________________________________________________________________ CAPACITY (decision-specific, if relevant): _____________________________ RISK (self-harm, harm to others, vulnerability): _______________________ IMPRESSION / DIFFERENTIAL _______________________________________________________________________ RECOMMENDATIONS (numbered, actionable) 1. Safety: ____________________________________________________________ 2. Non-pharmacologic: _________________________________________________ 3. Medication (dose, route, interactions): ____________________________ 4. Monitoring / labs: _________________________________________________ 5. Disposition / follow-up: ___________________________________________ COMMUNICATION Discussed with (team member, time): ____________________________________ C-L follow-up plan: ____________________________________________________ Total time (if time-based billing): ________ Signature/credentials/date: ____________________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample consultation-liaison note Scenario: a hospital medicine team requests psychiatric consultation for a 68-year-old man on post-operative day 2 after hip fracture repair, asking about depression and refusal of physical therapy. All details are fictional. Consultation-Liaison Note. Patient: H.W., 68M · Date/time: 07/23/2026, 10:40 am · Requesting: hospital medicine (Dr. F. Agu) · Question: "depressed, refusing PT, please assess" · Urgency: routine, seen same day Medical context: Admitted 07/21 after fall at home; hip fracture repair 07/21, post-operative day 2. Oxycodone 5 mg q6h prn (last dose 6:00 am), home sertraline 50 mg held since admission, diphenhydramine given for sleep 07/22. Na 131, mild renal impairment, hemoglobin stable, afebrile with normal white count. Nursing reports daytime drowsiness, awake overnight. Focused history: Per chart, daughter, and patient: treated depression, stable on sertraline for 3 years, no prior psychiatric admissions, no substance use history. Daughter reports he was "sharp and cheerful" at baseline last week; decline began yesterday evening. MSE: Drowsy but rousable, inattentive; months of the year backward failed at October. Oriented to person and place, not date. Speech slowed; thought process wanders. Mood "tired"; affect blunted. Fluctuation documented by nursing overnight. No hallucinations elicited; no delusional content. Denies suicidal ideation; no self-harm history. Capacity and risk: Capacity for PT participation decisions currently impaired by inattention; expected to be restorable as the confusional state clears, reassess daily rather than escalate. Risk: no suicidal ideation, no agitation; fall risk elevated; 1:1 not indicated. Impression: Hypoactive delirium, multifactorial (post-operative state, opioid effect, hyponatremia, anticholinergic exposure, disrupted sleep), superimposed on treated depression. "Refusing PT" reflects inattention and somnolence, not a mood decision. Not a depressive relapse on current evidence. Recommendations: 1) Discontinue diphenhydramine; use non-pharmacologic sleep protocol. 2) Reduce opioid to lowest effective dose; scheduled acetaminophen. 3) Continue to hold sertraline today; SSRIs can contribute to hyponatremia. Resume 50 mg daily once sodium is trending up, and recheck sodium within 72 hours of restart. 4) Hyponatremia workup and correction per medicine team. 5) Reorientation measures, daytime mobilization, glasses and hearing aids at bedside; schedule PT late morning. 6) Delirium precautions; call C-L for agitation or no improvement in 48 hours. Communication and follow-up: Findings and recommendations discussed with Dr. Agu by phone 11:35 am; nursing informed of sleep and reorientation plan. C-L will re-see 07/24. Total time 45 minutes. R. Calloway, MD, consulting psychiatrist, 07/23/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works It answers the question that was asked. The team asked about depression and PT refusal; the note reframes both as hypoactive delirium with evidence, which is the consult actually doing its job. Objective findings carry the reasoning. Failed attention testing, fluctuation documented by nursing, and the medication and sodium contributors make the impression auditable rather than asserted. Capacity and risk are addressed explicitly. The two most-missed elements in published audits each get a decision-specific line, including what would change the answer. Recommendations convert to orders in one pass. Six numbered actions with drug names, doses, timing, and the interaction that matters; nothing requires the intern to interpret prose. The loop is closed on the record. Verbal handoff with name and time, a follow-up date, and total time documented for the billing pathway. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The binding US rules are generic, and the billing layer does the shaping. Medicare stopped recognizing the inpatient consultation codes on January 1, 2010, and the AMA deleted 99251 in 2023, so the initial C-L encounter is billed as initial hospital care (99221 to 99223) or as a psychiatric diagnostic evaluation (90792, or 90791 without medical services), with follow-ups as subsequent hospital care; some commercial payers still accept consult codes, so the answer is payer-specific. That makes time and medical decision making the load-bearing documentation: an OIG audit found 87% of inpatient psychiatric facility claims with outlier payments failed medical necessity or documentation requirements. Teaching settings add 42 CFR 415.172 (the attending's presence for the key portion, documented), with a psychiatry-specific exception at 415.184 allowing observation by one-way mirror or video. Records must be complete within 30 days of discharge; everything faster than that is hospital policy, not law. Outside the US, the written product itself is a payment condition. Australia's MBS consultant psychiatrist items treat the detailed report to the referrer as a fundamental component, and the Professional Services Review has recovered six-figure sums where records lacked a history, mental status examination, diagnosis, or management plan, including a $103,500 repayment noted in its July 2025 update. Ontario's OHIP consultation codes (A195, and A190 for 75 minutes or longer) are defined to include a written report to the referring physician. Retention: 5 years minimum under the US hospital conditions of participation with most states requiring more, 10 years under Ontario's college standard, and 7 years for adults or to age 25 for minors under the statutes in the ACT, NSW, and Victoria. Coordination work around the admission, calls to the outpatient prescriber, family meetings, and disposition legwork, belongs in a care coordination note ; the consultative assessment is this note, and at the end of the stay its findings feed the discharge summary . ↑ Back to the template and downloads ## Common consultation-liaison note errors auditors flag Two data points frame the risk. In the US, OIG estimated that 87% of inpatient psychiatric facility claims with outlier payments (FYs 2014 to 2015) did not meet Medicare necessity or documentation requirements, roughly $93 million overpaid. In the UK, the only national audit standard for liaison documentation found baseline rates of 19.8% for capacity and 27.9% for risk at a major teaching hospital. The BastionGPT Clinical Advisory Board sees the same errors most often in C-L note reviews: The consult question is never answered. The note documents an assessment but takes no position on the thing the team asked, leaving the referral loop open in the chart. Capacity and risk are missing. The two elements with the worst published documentation rates and the highest legal exposure, especially around refusal of care. Dead billing pathways. Consult codes billed to Medicare (not recognized since 2010), or an initial hospital care or 90792 claim with no time or medical decision making support behind it. Recommendations that read as an essay. Unnumbered advice inside paragraphs; requesting teams act on lists, and auditors look for actionable specificity. Copy-forward follow-ups. Daily notes cloned from the initial consult with no status change, no new examination, and no updated plan, which reviewers read as no service delivered. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on consultation-liaison notes. Draft the consult from your bedside dictation or bullets, structured around the question asked, with the medical context, examination, capacity, and risk in their places. Check the note before signing: consult question answered, capacity and risk addressed, recommendations numbered and dosed, time and decision making documented for the code you intend. Convert the consult into what travels: a one-paragraph verbal-handoff summary, a follow-up note scaffold for tomorrow, or a referrer letter when the reply leaves the hospital. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How do I bill an inpatient psychiatric consult now that Medicare does not recognize consult codes? Bill the initial encounter as initial hospital care (99221 to 99223, leveled by time or medical decision making) or, when the encounter was a psychiatric diagnostic evaluation that included medical services, as 90792. Follow-up visits are subsequent hospital care (99231 to 99233). The consultation codes remain valid CPT and some commercial payers still accept them, so check the payer before assuming either way. Do I need a written order or consult request before seeing the patient? No regulation requires one. The practical reasons to have it anyway: hospitals typically require an order for workflow, payers expect the requesting-team relationship to be visible for medical necessity, and in Australia and Ontario a valid referral is what makes the consultant item payable. Document who asked and what they asked in the note itself. Can a psychiatric nurse practitioner write and bill the initial consult note? Yes, within state scope of practice. PMHNPs can bill 90792 because it includes medical services. In facility settings, a physician and NP from the same group can share a visit, billed by whoever performs the substantive portion, defined in Medicare rules as more than half the total time. Incident-to billing does not apply to hospital inpatients. Where do psychologists fit on a C-L service? Hospital psychologists assess patients whose primary diagnosis is medical under the health behavior assessment and intervention codes (the 96156 family), or perform a psychiatric diagnostic evaluation without medical services under 90791; the HBAI codes cannot be reported by the same provider on the same day as a psychiatric service. The psychiatric C-L note as such is written by the psychiatrist or PMHNP, and the psychologist's parallel note follows the same discipline: answer the question, examine objectively, recommend actionably. How fast does the consult note have to be completed? There is no C-L-specific clock in any of the three countries. The enforceable US rule is that the record be complete within 30 days of discharge, and The Joint Commission leaves authentication timeframes to hospital policy. Same-day documentation is the professional convention because the requesting team is waiting on your recommendations, but that is your hospital's rule to set. Does my attending have to see the patient if I am a resident? For the key portion of the service, yes, under 42 CFR 415.172, and their participation must be documented. Psychiatry has its own exception: 42 CFR 415.184 allows the required presence to be met by observation through a one-way mirror or video equipment. Whoever bills should personally document the medical decision making. Is a consultation-liaison note the same as a psychiatric diagnostic evaluation? No. The psychiatric diagnostic evaluation is a billable service (90792 with medical services, 90791 without) with its own documentation shape, and a US initial C-L encounter is often coded as one when that fits better than initial hospital care. The C-L note is the chart document built around the requesting team's question; it can support either code. CPT allows 90791 or 90792 once per day, not on the same day as an E/M service by the same provider, and many payers limit it to once per episode of care. How long do consultation-liaison notes have to be kept? As long as the hospital record they sit in. The US hospital conditions of participation set a 5-year minimum and most states require more, commonly 7 to 10 years for adults and longer for minors. Ontario's college standard is 10 years from the last entry. Australia's statutory minimums, 7 years for adults or to age 25 for minors, apply in the ACT, NSW, and Victoria; South Australia has no legislated period, so hospital policy governs there. Keep the record indefinitely where a complaint or claim is foreseeable. How does BastionGPT help with consultation-liaison documentation? Dictate the consult on the walk back from the bedside, or hand it bullets. BastionGPT drafts the note around the consult question with the examination, capacity, and risk lines in place, numbers the recommendations with doses so the team can order from them, and checks the elements reviewers test first: the question answered, time and decision making supported, and the handoff documented. BastionGPT is HIPAA-compliant with a signed BAA, and data is never used to train models. ## Related templates Care Coordination Note Collaborative Care (BHI/CoCM) Note Discharge Summary Psychiatric Diagnostic Evaluation Mental Status Exam (MSE) Note Multidisciplinary Team Meeting Note Case Conference Note Consultation Response & Report Chronic Care Management (CCM) Note Capacity & Competency Evaluation Psychiatric Consultation Note STOP-Bang Score Documentation Epworth Sleepiness Scale Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/consultation-report All note templates # Consultation Response & Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A consultation response, or consultation report, is the written reply a consultant sends to the provider who requested the consultation: findings, diagnostic impression, recommendations, and who does what next. It closes the referral loop the referral letter opened. Psychiatrists, psychologists, and therapists send one after an assessment or a completed course of treatment. Most run 150 to 700 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Consultants replying to a referrer: psychiatrists, psychologists, therapists, PMHNPs Audience The referring provider, the client's chart; payers in Australia and Canada Typical length 150 to 700 words · 10 to 20 minutes by hand (clinical team estimate) Format family Inter-provider correspondence (compare: consultation-liaison note, discharge summary) When it's used After an assessment, and after each completed course of treatment under a referral Standards context A payment condition in Australia and for Canadian physicians; convention for US psychologists On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a consultation response and report? A consultation response and report is the written communication a consultant sends back to the provider who requested a consultation, conveying findings, opinion, and recommendations. It belongs to the family of inter-provider correspondence: the referral letter opens the loop, the consultation report closes it, and a discharge summary closes the whole episode of care. Its modern shape descends from the US Medicare consultation service and its three R's (request, render, report): section 15506 of the old Medicare Carriers Manual made a written report to the requesting physician a mandatory element of billing a consultation, so the document grew up as a billing-integrity artifact as much as clinical correspondence. Clinicians and payers also call it a consultation letter, reply letter, referral response, letter back, GP letter, or closing-the-loop letter. Its weight now depends on where you practice and what you bill. Medicare stopped paying US consultation codes on January 1, 2010, so for most US clinicians, and for independent-practice psychologists especially, the report back is professional ethics and continuity of care rather than statute; the exception is the interprofessional consult family (99446 to 99451, with behavioral health G codes added for 2025), where a written report is again a payment element. Australia is the strongest mandate: the detailed written report is a fundamental component of the psychiatrist assessment items, and Better Access requires even psychologists to report to the referring GP after each course of treatment. In Ontario the consultant physician must submit findings, opinions, and recommendations in writing to the referring physician, and an incomplete consultation chain pays as a lesser assessment. Keep it distinct from the consultation liaison note, a chart entry inside a shared hospital record rather than outbound correspondence, and never paste psychotherapy notes into it; they stay out of correspondence entirely. ## Who uses consultation reports and when Any clinician who accepts referrals owes the loop a reply: psychiatrists reporting back to GPs after an assessment, psychologists writing to the referring physician after intake or after a course of therapy ends, therapists closing out an EAP or primary-care referral, and specialists answering an interprofessional e-consult. The empirical case for the document is stark: in a large US health system study, only 34.8% of 103,737 referrals produced a documented completed appointment, and referrers and specialists disagree sharply about whether information ever arrives. Send one after the initial assessment, after each completed course of treatment (a Better Access requirement in Australia), and when treatment ends or transfers. A full psychological evaluation report answers a formal assessment referral; an episode that ends produces a discharge summary ; the day-to-day coordination in between lives in care coordination notes . ## Consultation report structure: what goes in each section Addressee and referral anchor. Who referred, when, and the question they asked, restated in your first lines. The reply only closes the loop if it visibly answers the request that opened it. Pitfall: a letter that never restates the question; the referrer cannot tell whether their concern was addressed. Basis and limits. What the opinion rests on: sessions and dates, measures administered, records reviewed, collateral. State what the consultation did not cover. Pitfall: an unbounded opinion; scope you never stated is scope you will be held to. Findings. Concise and referrer-oriented. Research on what GPs want from psychiatrist reply letters (Selzer and colleagues, 2009) lands on six things: diagnosis, management plan, medication, prognosis, risk, and follow-up timing. Pitfall: pasting the intake note; the referrer needs your synthesis, not your raw record. Diagnosis and impression. Name it, with confidence level and differentials where honest. Pitfall: hedged formulations with no usable diagnosis; the referrer has to code and act on something. Numbered recommendations. What you advise, split into what you will do and what you suggest the referrer do, with medication specifics where relevant. Pitfall: advice without an actor; "consider SSRI adjustment" assigned to no one changes nothing. Responsibility and follow-up. Who owns ongoing management from today: returned to referrer, shared, or retained by you. Name who prescribes, who monitors risk, when you will see the client again, and what should trigger re-referral. Pitfall: the ambiguous handback; unowned follow-up is where referral loops fail. Sending and copy status. Date sent, method, and whether the client was copied. Australian guidance expects the psychiatrist's report within 2 weeks; Ontario's rules make the written report part of the consultation service itself. Pitfall: a report that was written but cannot be shown to have been sent; auditors have penalized exactly this. ### Blank template (copy and adapt) CONSULTATION RESPONSE / REPORT TO REFERRER To (referring provider): _______________________________________________ From (consultant, credentials): ________________________________________ Re (client, DOB): ______________________ Date of report: __________ Referral received: __________ Question asked: _________________________ BASIS OF THIS OPINION (sessions/dates, measures, records, collateral) _______________________________________________________________________ Limits of this consultation: ___________________________________________ FINDINGS (concise, referrer-oriented) _______________________________________________________________________ _______________________________________________________________________ DIAGNOSIS / IMPRESSION _______________________________________________________________________ RECOMMENDATIONS (numbered) 1. ____________________________________________________________________ 2. ____________________________________________________________________ 3. ____________________________________________________________________ RESPONSIBILITY AND FOLLOW-UP Ongoing management: [ ] returned to referrer [ ] shared [ ] consultant Who prescribes/monitors medication: ____________________________________ Follow-up with consultant (if any): ____________________________________ Re-referral triggers: __________________________________________________ SENT Date sent: __________ Method: __________ Client copied: [ ] yes [ ] no Signature/credentials: _________________________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample consultation report Scenario: a psychologist replies to the referring family physician after a two-session assessment of a 41-year-old woman referred for panic symptoms. All details are fictional. Consultation Report. To: Dr. M. Osei, family medicine · From: K. Brennan, PhD, licensed psychologist · Re: T.V., 41F · Date: 07/24/2026 · Referral received 07/06/2026: "recurrent panic attacks, rule out cardiac anxiety, advise on treatment" Basis and limits: Two assessment sessions (07/14, 07/21), clinical interview, PHQ-9 and GAD-7, panic symptom log, and your referral summary including last month's normal cardiac workup. No records from prior therapy several years ago were available. This consultation addressed diagnosis and treatment planning; it did not assess fitness for work. Findings: Ms. V. reports six panic episodes over the past ten weeks with palpitations, dyspnea, and fear of dying, followed by persistent worry about further attacks and avoidance of driving and exercise. GAD-7 13, PHQ-9 7 (item 9 zero). She checks her pulse repeatedly and has attended the emergency department twice despite the normal workup. Sleep onset delayed by anticipatory worry. No agoraphobic housebound pattern; alcohol one to two drinks weekly; caffeine four coffees daily. Impression: Panic disorder, moderate, with health-focused anxiety features; no current depressive episode. Prognosis good with structured treatment; risk is low, with no suicidal ideation reported at either session. Recommendations: 1) I will provide CBT for panic disorder, 8 to 12 weekly sessions, including interoceptive exposure and reduction of pulse-checking and reassurance-seeking. 2) Suggest reviewing the prn lorazepam you mentioned: as-needed benzodiazepine use during exposure-based treatment tends to maintain avoidance; a taper conversation at her next visit with you would help. 3) Gradual caffeine reduction. 4) No antidepressant is required for the current presentation in my opinion; if you elect one for her preference or course, treatment here is compatible. Responsibility and follow-up: I will provide the therapy course and send you a further brief report at completion or by session 12, whichever comes first. Medication decisions and monitoring remain with you. Re-referral triggers: emerging depressive symptoms, suicidal ideation, or no reduction in panic frequency by session 6. Sent: 07/24/2026 by secure message to your practice; copy offered to and accepted by Ms. V. Signed: K. Brennan, PhD, licensed psychologist. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral question is restated and answered. The GP asked three things (panic, cardiac anxiety, treatment advice) and can find each answered in order. It contains the six things referrers say they want. Diagnosis, management plan, medication position, prognosis, risk, and follow-up timing are all present and findable in seconds. Scope and basis are explicit. Sessions, measures, missing records, and what the consultation did not address are stated, which bounds the opinion defensibly. Responsibility is unambiguous. Therapy sits with the psychologist, prescribing with the GP, and the re-referral triggers are concrete, so nothing falls between providers. The send is documented. Date, method, and client copy status appear on the face of the report, the element payer audits have specifically penalized when absent. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations For US clinicians the duty is softer than most assume, until a code makes it hard. No statute obliges an independent-practice psychologist to report back to a referrer at all: the APA record-keeping guidelines are aspirational, and the referral loop runs on ethics and continuity of care. The exception is billing an interprofessional consultation (99446 to 99451, or the behavioral health G code family effective 2025), where the written report to the treating provider is part of the service definition. Two privacy boundaries hold in every case: psychotherapy notes as defined in 45 CFR 164.501 are excluded from the report and from electronic health information under the information-blocking rule, and a report to a clinical referrer needs no separate authorization, while reports to attorneys, employers, or insurers follow different consent rules entirely. In Australia and Canada the report is enforceable. Australia treats the detailed written report as a fundamental component of the psychiatrist assessment items, expects it within 2 weeks, and requires Better Access providers, psychologists included, to send a written report to the referring practitioner after each completed course of treatment. The Professional Services Review has penalized missing reports directly: one 2025 determination cited a practitioner who did not provide a written report back to the referrer, with $325,000 repaid, and another recovered $59,000 where reports could not be located. In Ontario, the Schedule of Benefits requires the consultant physician's findings, opinions, and recommendations in writing, and pays a lesser assessment fee when the consultation chain is incomplete. Retention follows your college or state rules: 10 years in Ontario with 15 advised, effectively 16 years in British Columbia per CMPA guidance, and state by state in Australia with longer periods for records made while the client was a minor. Document the send itself, and keep the deeper clinical record in the chart: the report summarizes the psychiatric diagnostic evaluation or assessment behind it rather than reproducing it. ↑ Back to the template and downloads ## Common consultation report errors auditors flag The loop fails more often than clinicians believe: 69.3% of primary care physicians said they usually send referral information, while only 34.8% of specialists said they usually receive it, and the OIG's 2006 audit of the old US consultation benefit found that about 75% of consultations billed to Medicare in 2001, $1.1 billion in improper payments, failed at least one program requirement, with 9% not documented in any record. Australia's Professional Services Review still penalizes the missing reply directly. The BastionGPT Clinical Advisory Board sees the same errors most often in consultation report reviews: The report is never sent, or the send is never documented. The letter exists in the chart with no date, method, or recipient trail; in PSR cases, reports that could not be located cost real money. A record dump instead of a reply. The full intake or test protocol pasted into correspondence, burying the answer and overexposing the client's record. No usable diagnosis or position. Pages of formulation with nothing the referrer can code, prescribe against, or explain to the client. An ambiguous handback. No statement of who manages what going forward, who prescribes, or what triggers re-referral; the client falls between two providers who each assume the other is watching. Psychotherapy notes leaking into correspondence. Process material that belongs in separated notes appears in a letter that will be forwarded, quoted, and filed in another practice's record. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on consultation responses and reports. Draft the reply letter from your assessment note, dictation, or bullets, restating the referral question and organizing findings the way referrers read them. Check the report before it goes: question answered, diagnosis stated, the six referrer essentials present, responsibility and re-referral triggers explicit, and nothing from psychotherapy notes inside. Convert one assessment into every audience it owes: the referrer letter, a plain-language summary for the client, and the chart note that stays home. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is sending a consultation report legally required? Rarely by statute, often by payment rule. In Australia the written report is a condition of the Medicare benefit: a fundamental component of the psychiatrist assessment items, and required from Better Access providers, psychologists included, after each completed course of treatment. In Ontario the fee schedule requires the consultant physician to submit findings, opinions, and recommendations in writing. In the US, Medicare eliminated consultation codes in 2010, and no statute obliges an independent-practice psychologist to report back at all; the duty is ethical until you bill an interprofessional consult code, which builds the written report into the service itself. How much detail belongs in the report versus the clinical record? Less than most first drafts contain. The report is a synthesis for a busy referrer: presentation in a few lines, impression and diagnosis, recommendations, and the plan. Research on GP preferences lands on six essentials: diagnosis, management plan, medication, prognosis, risk, and follow-up timing. The full history, session content, and test detail stay in your chart, summarized in the report only as far as the referrer needs to act. Is the referrer entitled to my psychotherapy notes? No. Psychotherapy notes as defined in 45 CFR 164.501 are separately protected under HIPAA, require a specific authorization to release, and are excluded from electronic health information under the information-blocking rule regardless of which type of mental health professional wrote them. They are never part of a consultation report. The report carries your findings and recommendations, not process notes. Can the client refuse to have the report sent? Generally yes, for a competent adult, and the referral itself usually implies consent to the return communication. The consequence is jurisdictional: in Australia the report to the referrer is a condition of the Medicare rebate for the psychiatry items and Better Access courses, so refusal can affect the claim. Talk it through with the client, document the decision, and where a rebate depends on the report, make that consequence clear before treatment starts. The referral came from an attorney or employer, not a clinician. Does the same report go back? No. That is a third-party evaluation, not a clinical consultation: the report goes to the retaining party under the engagement's terms, confidentiality warnings replace the treatment frame, and material compiled in anticipation of legal proceedings sits outside the ordinary record rules. Do not reuse a clinical reply-letter template for forensic work; the audiences and duties are different. Can my intake assessment double as the consultation report? The content overlaps, the document should not. Send a tailored summary: the referrer wants the answer, not the raw intake, and legacy US consultation guidance always preferred a separately identifiable report. Practically, a one-page letter derived from the intake reads better, travels safer, and answers faster than six pages of assessment detail. Is there a deadline for sending the consultation report? Only where a payer sets one. Australian guidance expects the psychiatrist's report to the GP within 2 weeks, and hospital records in the US must be complete within 30 days of discharge, but no US or Canadian rule sets a general deadline for a psychologist's reply. Convention is prompt: after the assessment, after each completed course, and at treatment end. The longer the gap, the less the referrer can use it. What happens when the consultation turns into ongoing treatment? Say so in the report. In Australia the psychiatrist moves from the single-assessment item to the ongoing-management items, and a report to the referrer is still expected. In the US, a consultation that becomes ongoing care is a transfer of care, and the interprofessional consult codes cannot be used to arrange one. Either way, the loop-closing letter is where you state that responsibility has shifted and what the referrer should still watch. How does BastionGPT help with consultation reports? Give it the assessment note or dictate the outcome, and it drafts the reply letter referrer-first: question restated, findings condensed, diagnosis and recommendations numbered, responsibility and re-referral triggers explicit. It checks the send block, keeps psychotherapy-notes material out, and produces the client's plain-language copy alongside. BastionGPT is HIPAA-compliant with a signed BAA, and your data is never used to train models. ## Related templates Care Coordination Note Collaborative Care (BHI/CoCM) Note Discharge Summary Psychiatric Diagnostic Evaluation Psychological Evaluation Report Multidisciplinary Team Meeting Note Case Conference Note Hospital Consult & Liaison Note Chronic Care Management (CCM) Note Medication Management Follow-Up Note Return-to-Primary-Care Summary Fitness-for-Duty Evaluation NDIS Plan Reassessment & Progress Report Forensic Psychological Evaluation Report Psychiatric Consultation Note Referral Letter Apgar Score Documentation Bishop Score Documentation MEWS & NEWS2 Early Warning Score Documentation Pressure Injury Staging Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ccm-note All note templates # Chronic Care Management (CCM) Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A chronic care management (CCM) note is the monthly documentation behind Medicare's CCM billing: a consent record, a comprehensive electronic care plan, and a time log showing at least 20 minutes of clinical staff care coordination for a patient with two or more chronic conditions. Care managers in primary care and integrated behavioral health teams keep it every enrolled month. Most months the record runs 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it RN, LPN, and MA care managers and behavioral health staff, under the billing practitioner's general supervision Audience The billing practitioner, the care team, Medicare and its contractors, auditors Typical length 150 to 400 words across the month's entries · 5 to 12 minutes per month by hand (clinical team estimate) Format family Monthly time-and-care-plan record (compare: collaborative care note, care coordination note) When it's used Every calendar month a consented patient with two or more chronic conditions is enrolled in CCM Standards context Medicare payment policy sets the required elements; the monthly narrative note is a convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a chronic care management (CCM) note? A chronic care management note is the medical-record documentation that supports Medicare's monthly, non-face-to-face care management service for patients with two or more chronic conditions, billed under CPT 99490 and its add-on 99439. In practice it is not one document but three linked artifacts: a consent record, a comprehensive patient-centered electronic care plan, and a time log tying dated activities and minutes to named staff. CMS created the service, not a professional society: the CY2014 Physician Fee Schedule rule first finalized a payable code, and the CY2015 rule adopted CPT 99490 effective January 1, 2015. The family now includes complex CCM (99487, 99489), practitioner-performed CCM (99491), transitional care management, principal care management, behavioral health integration, and, since 2025, the non-time-based Advanced Primary Care Management bundle. The governing guidance is the CMS booklet MLN909188, June 2025 edition. You will also hear the record called a 99490 note, a care management time log, or a monthly CCM summary. The distinction that carries this page: almost nothing about the CCM note is statute. The required elements are Medicare payment policy, and three of the most repeated "requirements" are not requirements at all. No CMS rule demands a per-encounter narrative note, a timestamp on each time entry, or a signature on the care plan; all three are vendor conventions, worth keeping only as internal practice. What payer policy does require: documented consent, the electronic care plan, a core dataset of demographics, problems, medications, and allergies kept in an EHR that meets the federal CEHRT standard, and time reaching the monthly threshold. The nearest neighbors are the care coordination note , which records the same kind of work without the billing frame, and the collaborative care note , the registry-based record behind CoCM and BHI billing. Canada and Australia have no monthly time-tracked equivalent; their analogues are annual care-plan documents. ## Who uses chronic care management notes and when Primary care practices, internal medicine and geriatrics clinics, rural health clinics and FQHCs, and integrated behavioral health teams run CCM programs, and the note is kept by the clinical staff who deliver the service: RN and LPN care managers, medical assistants, and the behavioral health clinicians and social workers embedded in primary care. Only physicians, NPs, PAs, clinical nurse specialists, and nurse-midwives may bill it; everyone else's minutes accumulate under the billing practitioner through general supervision. Write a CCM note for ongoing multi-condition coordination. Switch to a collaborative care note when the month's work is a psychiatric registry workflow, to a TCM record for the 30 days after the kind of transition a discharge summary closes, and to a care coordination note for coordination outside any billing program. The care plan at the center of CCM is the medical cousin of the mental health treatment plan . ## Chronic care management (CCM) note structure: what goes in each section Patient and month header. Patient identifiers, the service month, the billing practitioner whose claim the record supports, and the care manager(s) logging time. Medicare pays one CCM claim per patient per month, so the header names whose claim this is. Pitfall: the OIG's follow-up audit traced $1,427,930 of its $1.9 million overpayment finding to CCM billed more than once for the same patient in the same period; a record that never names the billing practitioner cannot show the one-practitioner rule was checked. Eligibility and diagnoses. The two or more chronic conditions, stated with the framing Medicare requires: expected to last at least 12 months or until death, and placing the patient at significant risk of death, acute exacerbation, or functional decline. Behavioral health conditions count; depression plus diabetes is a common qualifying pair. Pitfall: a bare problem list; the duration-and-risk language is the eligibility test, and reviewers match the documented diagnoses against the claim. Consent record. When consent was obtained, verbal or written, by whom, and the four disclosures: CCM availability, cost sharing, the one-practitioner-per-month rule, and the right to stop. Verbal consent has sufficed since January 1, 2017, and consent is obtained once, again only if the billing practitioner changes. Pitfall: re-collecting consent every January while never documenting the disclosures; auditors read for the four elements, not for a fresh signature. Comprehensive care plan. The electronic care plan's status this month: problems, prognosis, measurable goals, planned interventions with responsible parties, medication management, community services, and the review schedule, plus the operational line vendors skip: shared with the care team, copy offered to the patient. Pitfall: a plan created at enrollment and never touched again. CMS expects a living document; a signature on it appears nowhere in the element list, while sharing it does. Time log. One line per contact or activity: date, staff name and credentials, what was done, minutes. The billing practitioner's own minutes may count toward 99490 and 99439 so long as they are not the basis for 99491, a point most vendor guides get wrong in the other direction. Pitfall: identical round-number entries across patients and months; time not demonstrably spent is the failure mode behind the first False Claims Act settlement naming 99490. Concurrent services check. The month's overlaps, ruled out explicitly: no minute counted toward both CCM and an E/M visit, TCM, RPM, or RTM; general BHI may run alongside, CoCM generally should not; no CCM during home health or hospice oversight billing (G0181, G0182) or certain ESRD bundles. Pitfall: the denial arrives as CARC CO-97 with remark M15, a contractual bundling write-off you cannot pass to the patient. Month-end summary and codes. Total minutes for the calendar month and the code the total supports: 20 minutes for 99490, 40 for the first 99439 add-on, 60 for the second, two add-on units the ceiling. Note the initiating visit for patients new to the practice or not seen within the prior year. Pitfall: 99439 without 99490 on the claim, a third add-on unit, or a claim filed before the month's minutes actually cross the threshold. ### Blank template (copy and adapt) CHRONIC CARE MANAGEMENT (CCM) NOTE Patient: ______________ DOB: __________ Service month: ______________ Billing practitioner (one per patient per month): ______________________ Care manager(s), credentials: __________________________________________ ELIGIBILITY (2+ chronic conditions, expected to last 12+ months, with significant risk of death, acute exacerbation, or functional decline) 1. _____________________________ 2. _____________________________ 3. _____________________________ Initiating visit (if required): ______ CONSENT (once; again only if the billing practitioner changes) Date: __________ [ ] verbal [ ] written Obtained by: ________________ Disclosed: [ ] CCM availability [ ] cost sharing [ ] one practitioner per month [ ] right to stop anytime CARE PLAN (electronic, shareable; copy offered to the patient) Updated this month: [ ] yes [ ] no Date: __________ Next review: ____ Changes: _______________________________________________________________ TIME LOG (date | staff, credentials | activity | minutes) ______ | ________________ | _________________________________ | ______ ______ | ________________ | _________________________________ | ______ ______ | ________________ | _________________________________ | ______ ______ | ________________ | _________________________________ | ______ CONCURRENT SERVICES CHECK [ ] No minutes shared with E/M, TCM, RPM/RTM, or another practitioner [ ] Not a CoCM month [ ] No hospice or home health oversight (G0181/G0182) MONTH-END SUMMARY Total minutes: ______ Supports: [ ] 99490 [ ] 99439 x ____ (max 2) Author signature/credentials/date: _____________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample chronic care management (CCM) note Scenario: month-end view of a CCM record kept by an RN care manager in a primary care practice, for a 68-year-old patient with type 2 diabetes and recurrent major depressive disorder. The billing practitioner is the patient's internist. All details are fictional. Chronic Care Management Note. Patient: E.M., 68 · Service month: July 2026 · Billing practitioner: L. Ferro, MD (internal medicine) · Care manager: J. Patel, RN Eligibility: Type 2 diabetes (E11.9) and major depressive disorder, recurrent, moderate (F33.1). Both expected to persist beyond 12 months with significant risk of exacerbation and functional decline; documented at the 03/12/2026 initiating visit (annual wellness visit). Consent: Verbal consent on file, obtained 03/12/2026 by J. Patel, RN. Reviewed: CCM availability, monthly cost sharing, one billing practitioner per month, right to stop at any time. No change in billing practitioner since. Care plan: Updated 07/08/2026 after 06/30 A1c of 8.1: glycemic goal reset to 7.5 by 12/2026, walking goal 20 minutes 4 days per week, monthly PHQ-9 added while sertraline dose is titrated, medication list reconciled (metformin, sertraline, lisinopril), community diabetes education referral active. Plan shared with the practice care team; updated copy sent to the patient portal 07/08/2026. Next full review 10/2026. Time log: 07/02 · J. Patel, RN · Phone review of home glucose readings; hypoglycemia symptoms screened; metformin timing reinforced · 11 min 07/08 · J. Patel, RN · Care plan updated after A1c result; portal copy to patient; dose question routed to Dr. Ferro · 9 min 07/14 · J. Patel, RN · PHQ-9 by phone: 9, down from 13 in April; sleep and sertraline adherence reviewed; walking goal encouraged · 12 min 07/21 · R. Voss, LCSW · Coordination with community diabetes education program; enrollment confirmed for the 08/04 class · 6 min 07/28 · J. Patel, RN · Pharmacy coordination; lisinopril refill gap flagged to Dr. Ferro; patient confirmed pickup · 5 min Concurrent services check: No minutes shared with any E/M visit, TCM, RPM, or BHI service this month; no other practitioner billing CCM; patient not under home health or hospice oversight. Month-end summary: 43 clinical staff minutes in July 2026 supports 99490 plus one unit of 99439 on Dr. Ferro's claim. Entry closed 07/31/2026 by J. Patel, RN. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Eligibility is argued, not assumed. Two named conditions carry the duration-and-risk language Medicare's eligibility test uses, tied to a dated initiating visit. Consent is complete and datable. Date, mode, who obtained it, and all four disclosures, plus the one trigger that would require redoing it: a change of billing practitioner. The care plan is visibly alive. A goal changed after a lab result, a symptom score is trending, and the patient received the updated copy, the element CMS actually lists. Every minute is attributed and the code follows the arithmetic. Five entries, each with a date, named staff with credentials, a concrete activity, and minutes; 43 minutes maps to 99490 plus one 99439, inside the calendar month. The overlap check is explicit. One practitioner, no shared minutes, no hospice or home health conflict: the exact failure modes the OIG audit and CO-97 denials target. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Treat the CCM record as three linked artifacts and know where each lives. The consent, the care plan, and the time log are ordinary medical-record content: part of the designated record set, releasable on request, and discoverable, with none of the special protections psychotherapy notes carry. The EHR-technology requirement is narrower than vendors imply: only the core dataset of demographics, problems, medications, and allergies must sit in an EHR meeting the federal CEHRT standard; the rest of the documentation can live where your workflow puts it. Retention has no CCM-specific rule. What applies is the general Medicare obligation to keep documentation supporting claims, 7 years under 42 CFR 424.516(f), plus state law and payer contract, whichever runs longest. And because CCM carries monthly coinsurance, the patient sees a bill every month: the OIG put $540,680 of improper beneficiary cost sharing on the 2017 to 2018 error tape, and a consent conversation that skipped cost sharing is where patient complaints start. The payer layer rewards a spare template. Build to the MLN909188 element list and treat everything else as internal practice: no CMS rule requires a monthly narrative, per-entry timestamps, or a signed care plan, and presenting those conventions to auditors as legal requirements only raises the bar you will be held to. Supervision is general, meaning the billing practitioner directs the service without needing to be in the building, and behavioral health clinicians and social workers usually deliver CCM as clinical staff under that arrangement rather than billing it themselves. Watch the program boundary in both directions: a month dominated by a psychiatric registry workflow belongs in a collaborative care note , coordination outside any billing program in a care coordination note , and if minute-tracking is the burden, the Advanced Primary Care Management bundle (G0556 to G0558) drops the time thresholds in exchange for heavier population-management duties, with behavioral health add-on codes arriving in 2026. ## Common chronic care management note errors auditors flag The enforcement record here is unusually specific. The HHS Office of Inspector General's audit of 2017 and 2018 claims (A-07-19-05122) found $1,918,278 in CCM overpayments across 50,192 claims, $1,427,930 of it from CCM billed more than once for the same patient in the same period, plus up to $540,680 owed back to patients in cost sharing. CMS's response noted that fewer than 1% of CCM claims were noncompliant, so the errors concentrate in a narrow set of practices. June 2024 added the first False Claims Act settlement naming CPT 99490: Bluestone Physician Services paid $14.9 million over claims, including CCM, that did not support the level of service billed. The BastionGPT Clinical Advisory Board sees the same errors most often in chronic care management note reviews: Two claims, one month. Duplicate or overlapping CCM for the same patient, usually two practitioners each assuming they are the one; the single largest component of the OIG finding. Minutes that cannot survive a look. Round-number logs, entries cloned across patients, or totals the listed activities cannot plausibly support; time not demonstrably spent is what False Claims Act cases are built on. Consent that is assumed. No documented disclosure of cost sharing, the one-practitioner rule, or the right to stop. Verbal consent is fine; undocumented consent is not. A care plan that never moves. Created at enrollment, untouched since, never shared with the patient; reviewers read the update dates and the patient-copy line before they read the goals. Threshold and code arithmetic. 99439 without 99490, a third add-on unit, minutes shared with an E/M or TCM service, or a claim filed before 20 minutes existed; the denial returns as CO-97 with M15 and cannot be billed to the patient. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on chronic care management notes. Draft the month's entries and the month-end summary from a dictated recap, a call log, or a portal message thread, each entry dated and attributed with minutes. Merge the scattered sources, the problem list, last month's care plan, and this month's contacts, into one coherent time log and an updated care plan draft. Check the record before the claim: total minutes against 99490 and 99439, consent elements on file, care plan currency, and overlap flags for TCM, RPM, and CoCM months. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a monthly narrative note required to bill CCM? No. Medicare's element list (MLN909188) requires documented consent, a comprehensive electronic care plan, the core structured dataset, and time reaching the monthly threshold. A per-encounter narrative, a timestamp on every entry, and a signature on the care plan are conventions many vendors present as rules; no CMS instrument requires them. Many programs keep a light monthly summary anyway because it makes the time log defensible, but that is a choice, not a mandate. Does CCM consent have to be in writing? Not since January 1, 2017, when the CY2017 fee schedule rule allowed verbal consent documented in the record. Consent is obtained once, and again only if the patient changes billing practitioners; it does not expire on a schedule. Document the four disclosures: that CCM is available, that cost sharing applies, that only one practitioner can bill it per month, and that the patient can stop at any time. Whose time counts toward the 20 minutes? Clinical staff time furnished incident-to the billing practitioner under general supervision, and the billing practitioner's own time, provided those minutes are not also the basis for 99491. The staff can be employees or contractors: RNs, LPNs, medical assistants, behavioral health clinicians, social workers, within state scope-of-practice rules. What never works is a nurse or therapist billing 99490 under their own name; the claim belongs to the practitioner. Can we bill CCM in the same month as TCM, RPM, or behavioral health codes? Often yes, with boundaries. CCM and TCM may share a month under policy in effect since 2020, but no minute counts toward both. CCM can run with RPM or RTM, one of the two, if each service independently meets its requirements. General BHI (99484) can coexist with CCM; psychiatric CoCM (99492, 99493, 99494) generally should not be reported in the same month. CCM is off the table entirely during home health or hospice oversight billing (G0181, G0182) and certain ESRD months. What is the difference between a CCM note and a care coordination note? A care coordination note records coordination work wherever it happens and carries no billing clock. The CCM note is that same work placed inside Medicare's payment frame: documented eligibility, consent, a care plan, and a monthly time threshold under one named practitioner. If the record cannot show all four, it is a coordination note, not a billable CCM record. The registry-driven behavioral health cousin is the collaborative care note . Can a psychologist or therapist bill CCM? No. CCM billing is limited to physicians, nurse practitioners, physician assistants, clinical nurse specialists, and nurse-midwives; clinical psychologists are among the practitioners excluded. Behavioral health clinicians still sit inside CCM constantly, as the clinical staff whose minutes accumulate under the billing practitioner. When the behavioral condition is the center of the work, the collaborative care code family (CoCM and BHI) is the one built for it. How long do we keep CCM documentation? There is no CCM-specific retention period. What applies is the general Medicare requirement to keep documentation supporting claims, 7 years from the date of service under 42 CFR 424.516(f), plus state medical-record law and payer contracts, and the longest applicable period wins. Keep the consent record and each care plan version as long as the time logs; an audit reads all three together. Is there a Canadian or Australian version of the CCM note? No. Neither country pays for monthly, non-face-to-face, time-tracked care management. Canada funds chronic disease care through provincial fee schedules built on annual flow sheets and incentives, such as Ontario's diabetes management codes and British Columbia's condition-based fees. Australia moved to the GP Chronic Condition Management Plan on July 1, 2025, replacing GP Management Plans and Team Care Arrangements: a care-plan document with consent, patient-copy, and record-keeping duties, not a time log. A US-style CCM log used in either country is an internal convention, not a payer requirement. How does BastionGPT help with chronic care management notes? Bring the month as it actually exists: call notes, portal threads, a dictated recap. BastionGPT drafts the dated, attributed time log and the month-end summary, updates the care plan draft from what changed, and checks the claim-facing details: minutes against 99490 and 99439, consent elements on file, and overlap flags for TCM, RPM, and CoCM months. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Care Coordination Note Collaborative Care (BHI/CoCM) Note Discharge Summary Treatment Plan Claim-Support Billing Note Multidisciplinary Team Meeting Note Case Conference Note Hospital Consult & Liaison Note Consultation Response & Report Medication Management Follow-Up Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Communication Log & Portal Message Note Case Management Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/medication-management-note All note templates # Medication Management Follow-Up Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A medication management follow-up note is a prescriber's record of an established-patient psychiatric visit centered on medications: response, adherence, side effects, monitoring, and plan changes. Psychiatrists, psychiatric nurse practitioners, and physician assistants write one at every med check, billed in the US as an E/M visit (99212 to 99215). Most run 150 to 500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychiatrists, PMHNPs, and PAs; prescribing psychologists in the few states that authorize them Audience The prescriber at the next visit, the care team, payers and their auditors Typical length 150 to 500 words · 5 to 10 minutes by hand (clinical team estimate) Format family Established-patient E/M progress note (compare: psychotherapy progress note, psychiatric diagnostic evaluation) When it's used At every follow-up visit where assessing and adjusting psychiatric medication is the main work Standards context E/M coding rules and payer policy shape the content; no statute in any of the three countries prescribes a format On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a medication management follow-up note? A medication management follow-up note records an established-patient psychiatric visit whose primary work is pharmacologic: how the medication is working, whether the patient is taking it, what side effects and monitoring results have appeared, and what the prescriber changed. Its structure merged two lineages, the problem-oriented medical record Lawrence Weed published in 1968, ancestor of the SOAP note , and the much older psychiatric interview and mental status tradition. No body ever issued a med-check format; what standardized the note was a coding change. The AMA deleted the standalone medication management code 90862 effective January 1, 2013, with no grace period, moving prescribers onto the general E/M codes, and the January 1, 2021 E/M overhaul then dropped scored history and exam requirements, so the code level now rides on medical decision making or total time alone. The note as practiced today is an E/M artifact governed by general E/M rules, not by anything psychiatry-specific. You will also hear med check, med management note, psych follow-up, med review, and, in Australia, review attendance. The distinction that carries this page: almost nothing about the note's content is statute. Record-keeping law in all three countries requires that a record exist, be legible, be authenticated, and be retained; the field-level expectations, interval history, medication list, mental status, risk, plan, come from payer policy and professional convention, and the E/M framework itself is AMA convention that becomes payer policy when CMS and commercial payers adopt it. The note differs from the psychiatric diagnostic evaluation , the comprehensive intake generally payable once per episode, and from the psychotherapy progress note a therapist writes for the same patient: this one is prescriber-authored, pharmacologically focused, and coded from the E/M family. ## Who uses medication management follow-up notes and when Psychiatrists, psychiatric nurse practitioners, and physician assistants write these notes in outpatient psychiatry practices, community mental health centers, integrated primary care, addiction medicine, and telepsychiatry; prescribing psychologists write them in the handful of states that authorize psychologist prescribing, with 90863 as their add-on code. In collaborative care programs the psychiatric consultant recommends rather than prescribes, so that advice lives in a collaborative care note while the treating prescriber's own visits produce medication management notes. Use this format for routine follow-up. Switch to the psychiatric diagnostic evaluation for a new patient, a new episode after a substantial gap, or a documented major change needing full re-evaluation, and to a consultation report when another clinician asked a question you are answering. The therapist seeing the same patient keeps a separate psychotherapy progress note . ## Medication management follow-up note structure: what goes in each section Reason for visit. The condition and today's purpose, stated so medical necessity is visible: 6-week follow-up of sertraline titration for major depressive disorder, not a bare follow-up. Medicare contractor guidance treats a one-word reason as insufficient to support the claim. Pitfall: medical-necessity denials (CO-50) start at the first line; follow-up names an interval, not a reason. Interval history. What changed since the last visit: symptom course, response to the current dose, adherence with specifics, side effects asked about and answered, sleep, substance use, and stressors. Pitfall: a history copied forward until it contradicts the plan; a JAMA Internal Medicine study of one center's inpatient notes found roughly 46 percent of text copied, and auditors read consecutive notes side by side. Current medications and monitoring. The full list with name, dose, and frequency, reconciled against what the patient actually takes, plus the monitoring the regimen requires: lithium or valproate levels, metabolic panels on antipsychotics, AIMS at guideline intervals, ECG where indicated. Pitfall: monitoring that lives only in the lab module; results count toward medical decision making only when the note shows they were reviewed. Examination, mental status, and risk. A medically appropriate exam. Since January 1, 2021 no payer scores its extent, so document what you examined and found, and assess risk, suicidality and, where relevant, violence, whenever symptoms, medications, or history make it pertinent. The full format lives on the mental status exam page. Pitfall: a 12-domain MSE template copied forward unchanged; a shorter exam written fresh is stronger in an audit than boilerplate. Assessment and medical decision making. Each problem addressed, with status and severity; the data reviewed; the rationale for the medication decision, including the risks weighed; the diagnosis coded to the highest specificity supported. Pitfall: stable, continue meds under a 99214; downcoding clusters on under-described severity and unstated rationale, not on note length. Plan and prescriptions. Every change paired with its why: dose and titration schedule, new and stopped scripts, labs ordered, the PDMP query where state law requires one, safety-netting instructions, and the follow-up interval. Pitfall: a PDMP check performed but never attested; a minority of states require the note to record the query, and the attestation takes one line. Billing basis and signature. Either total time on the date of the encounter with a line on what filled it, or medical decision making, plus signature, credentials, and date. When a psychotherapy add-on (90833, 90836, 90838) is billed, the psychotherapy time and content are documented separately and the E/M level is selected by MDM alone. Pitfall: forcing a time entry into an MDM-based note; a documented 15 minutes sitting beside a 99214 does the auditor's work. ### Blank template (copy and adapt) MEDICATION MANAGEMENT FOLLOW-UP NOTE Patient: ______________ DOB: __________ Visit date: __________ Prescriber, credentials: ______________________________________ REASON FOR VISIT (condition + purpose, never a bare "follow-up") ________________________________________________________________ INTERVAL HISTORY (course since last visit, response, side effects, sleep, substance use, stressors) ________________________________________________________________ ________________________________________________________________ Adherence: [ ] as prescribed [ ] partial [ ] stopped Missed doses / pattern: ________________________________________ CURRENT MEDICATIONS (name | dose | frequency | start/change date) ________________________________________________________________ ________________________________________________________________ Monitoring reviewed (levels, labs, AIMS, vitals, ECG): __________ EXAM / MENTAL STATUS (medically appropriate scope) ________________________________________________________________ Risk: SI [ ] denied [ ] passive [ ] active Plan/intent: _______ Violence risk (if pertinent): __________________________________ ASSESSMENT / MDM (each problem + status + severity; data reviewed; rationale and risk of the medication decision) 1. ___________________________ [ ] improved [ ] stable [ ] worse 2. ___________________________ [ ] improved [ ] stable [ ] worse PLAN (changes + rationale, scripts, labs ordered, safety-netting, follow-up interval) ________________________________________________________________ ________________________________________________________________ PDMP checked (controlled scripts): [ ] yes, date ______ [ ] n/a BILLING BASIS: [ ] MDM [ ] total time ______ min (encounter date) Psychotherapy add-on: [ ] none [ ] 90833/90836/90838, documented separately (E/M level by MDM when an add-on is billed) Signature / credentials / date: ________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample medication management follow-up note Scenario: a psychiatric nurse practitioner's six-week follow-up of an established patient on a sertraline titration for depression with co-occurring anxiety. All details are fictional. Medication Management Follow-Up Note. Patient: R.T., 41 · Visit: 07/16/2026 · Prescriber: M. Osei, PMHNP-BC Reason for visit: Six-week follow-up of sertraline titration for major depressive disorder, recurrent, moderate (F33.1), with generalized anxiety disorder (F41.1). Interval history: Six weeks at sertraline 100 mg. Mood and energy improved; PHQ-9 today 11, down from 18 in May; GAD-7 9, down from 14. Early nausea resolved by week two; no other side effects reported on direct questioning. Sleep improved, using trazodone 2 to 3 nights per week. No alcohol since April. No missed doses in the past month per pill organizer and refill history. Medications: Sertraline 100 mg daily (since 06/03/2026); trazodone 50 mg at bedtime as needed (since 05/2026). No supplements. List reconciled with patient. Monitoring: no labs due; weight stable; BP 118/76. Exam and mental status: Alert, engaged, casually dressed. Speech normal in rate and tone. Mood reported as better; affect brighter, congruent. Thought process linear. No delusional content elicited; denies hallucinations. Cognition grossly intact. Insight and judgment good. Risk: Denies suicidal ideation, intent, or plan since last visit; no self-harm. No change in access to means; safety plan from intake remains current. Violence risk not indicated. Assessment and MDM: 1. MDD, recurrent, moderate (F33.1): improved, partial response at 100 mg with residual anhedonia and morning fatigue at week six. 2. GAD (F41.1): improving in parallel. Data reviewed: PHQ-9 and GAD-7 administered today; pharmacy fill history. Options weighed: hold at 100 mg versus increase; residual symptoms at an adequate trial interval support an increase; dose-related side effect risks reviewed with patient, who agrees with the plan. Plan: Increase sertraline to 150 mg daily from 07/17/2026; counseled on transient GI effects and the warning signs that warrant a call. Continue trazodone as needed. No controlled substances prescribed; PDMP not applicable this visit. Repeat PHQ-9 and GAD-7 next visit. Return in 4 weeks, sooner if worse; after-hours crisis contacts reviewed. Billing basis: Level selected by medical decision making: two chronic illnesses, one incompletely controlled, with a medication adjustment; time not recorded. Signed M. Osei, PMHNP-BC, 07/16/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Medical necessity is visible in line one. The reason for visit names the condition and the titration, not just follow-up. The interval history carries measurement. PHQ-9 and GAD-7 with direction, adherence with a source, side effects asked about and resolved. The exam is honest scope, and risk is fresh. A visit-appropriate mental status exam, written new, with suicidality assessed, dated, and tied to the standing safety plan. The dose change carries its rationale. Residual symptoms at an adequate trial, options weighed, risks counseled, and the diagnosis kept at full specificity. The billing basis is stated once and cleanly. MDM, no stray time figure, and no add-on claimed without separate documentation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A medication management note is part of the designated record set: patients can read it, it is discoverable, and it carries none of the special protections psychotherapy notes have. Retention is jurisdictional, and the most repeated claim about it is wrong: HIPAA's 6-year rule (45 CFR 164.316(b)(2)) covers compliance documentation, not clinical records. US state boards commonly require 7 to 10 years from the last encounter, longer for minors, and Medicare Advantage risk-adjustment audits reach back 10 years. CPSO in Ontario requires 10 years and recommends 15 because Ontario's Limitations Act lets certain claims begin up to 15 years after the events; CMPA advises 16 in British Columbia. Australia's rule is 7 years from last service, or to age 25 for records collected from a child, split across two instruments (Victoria's HPP 4.2 and NSW's s.25(1)) rather than one national law. No federal statute sets a signature deadline either: CMS's manuals ask for completion during or as soon as practicable after the encounter, the 24-to-48-hour figures come from contractors and institutions, and late entries follow the amendment rules, dated, signed, and never disguised as contemporaneous. The payer layer is where the format-versus-content distinction earns its keep: no statute prescribes the sections, but the claim stands on their content. Bill by MDM or by total time on the encounter date; if MDM, leave time out entirely. When separately identifiable psychotherapy happens in the same visit, use the add-on codes, document the psychotherapy on its own, and select the E/M level by MDM alone. Supervision changed on January 1, 2026: direct supervision for incident-to services under 42 CFR 410.26 can now be met by real-time audio-video availability, audio-only excluded, and incident-to still never fits a new patient or a new problem. In teaching clinics, 42 CFR 415.184 lets one-way-mirror or video observation satisfy the psychiatric presence requirement. The DEA's telemedicine flexibility for controlled-substance prescribing without a prior in-person exam runs through December 31, 2026, with the permanent framework still pending, so date-stamp any workflow built on it. And keep the program boundaries straight: consultant recommendations in collaborative care belong in a collaborative care note , and chronic-condition coordination time in a chronic care management note , not here. ## Common medication management note errors auditors flag The audit record for this note type is unusually quantified. CMS's Medicare Learning Network compliance material on outpatient psychiatric care reports a 16.1% improper payment rate, a projected $254.5 million, with 78.3% of it attributed to insufficient documentation against 2.6% for incorrect coding: psychiatry loses these audits on notes, not code choice, by roughly thirty to one. An HHS OIG review of pandemic-era psychotherapy claims found $580 million of roughly $1 billion paid was improper, and Australia's Professional Services Review has ordered individual psychiatrists to repay $100,000 and $500,000 in cases where inadequate records alone were capable of constituting inappropriate practice. The BastionGPT Clinical Advisory Board sees the same errors most often in medication management note reviews: The cloned note. Interval history, mental status, and risk identical across months; copy-forward reads as no reassessment, and insufficient documentation drives 78.3% of the improper payments in this category. Stable, continue at a moderate level. A 99214 with no stated severity, no data reviewed, and no rationale for continuing the regimen; the level is earned by documented decision making, not by the visit having happened. Add-on psychotherapy without a separate story. Standalone codes 90832, 90834, or 90837 billed beside an E/M trigger the CO-97 bundling denial; add-ons without separately documented time and content, or psychotherapy minutes double-counted into the E/M, fail on review. Incident-to shortcuts. A new problem or new patient billed incident-to, or supervision conditions unmet; since January 1, 2026 direct supervision can be satisfied by real-time audio-video availability, but the categorical exclusions did not move. A dead code family. 90862 lingering in templates thirteen years after deletion, 90863 used by anyone other than a prescribing psychologist, or the intake code 90792 re-billed for routine reassessment inside a continuous episode. ### Medication management follow-up notes in the US, Canada, and Australia Aspect United States Canada Australia Status The E/M framework is AMA convention adopted as payer policy by CMS and commercial payers; statutes require that authenticated records exist (42 CFR 482.24 in hospitals, state board rules in offices), not what the note contains A record with basic identifiers is law (Ontario's O. Reg. 114/94, s.18(1)); content expectations are college standards (CPSO, CPSA, CMQ), enforceable through discipline Record-keeping duties sit in the Medical Board's Good Medical Practice code, section 10.5, a regulatory standard; MBS item rules are payer policy; retention statutes in Victoria and NSW are law Terminology Med check, med management note, psych follow-up, E/M note Follow-up visit note, or the provincial fee code itself (Ontario's time-based K-codes) Review attendance, or the item number (the 300 to 308 face-to-face family, 91827 to 91839 telehealth) What changes Code selection by MDM or total time; psychotherapy add-ons documented separately; incident-to and teaching-physician rules decide who can author and bill Time-based K-codes must be supported by recorded time, the closest Canadian analogue to a content mandate; no province specifies mental status exam scope Items 300 to 308 carry a 50-service annual threshold before lower-rebate items 310 to 318 apply, which shapes visit length; a valid referral is a payment condition Retention State boards commonly 7 to 10 years from the last encounter, longer for minors; Medicare Advantage audits reach 10 years; HIPAA's 6-year rule covers compliance papers, not charts CPSO: 10 years from the last entry, or 10 years past majority for minors, with 15 recommended; CMPA advises 16 in British Columbia 7 years from last service, or until age 25 for records collected from a child (Victoria HPP 4.2; NSW HRIP Act s.25(1)) The clinical skeleton travels unchanged: interval history, medication response, an appropriate exam, risk, and a reasoned plan serve a prescriber identically in Boston, Toronto, and Melbourne. What changes is the payment scaffolding around the visit and the retention clock behind it. Document to the clinical standard, then check the coding layer your jurisdiction actually applies, and label each local rule for what it is: law, payer policy, or convention. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on medication management follow-up notes. Draft the full note from a dictated recap or visit transcript: interval history, exam, MDM rationale, and plan, in your clinic's format. Pull the moving parts into structure: medication changes with dates, scores with direction, monitoring due, and the follow-up interval. Check the note before you sign: risk addressed, monitoring documented, a rationale on every change, no stray time figure beside an MDM claim, add-on documentation separated. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to document time, or can I bill a med check by MDM alone? Either works for 99212 through 99215, and has since January 1, 2021. MDM usually serves brief but complex visits better. If you select the level by MDM, leave time out of the note entirely; a documented short time beside a higher-level code invites the exact comparison you do not want. If you bill by time, record total time on the date of the encounter and what filled it. One rule is fixed: when a psychotherapy add-on is billed, the E/M level must come from MDM, because psychotherapy minutes cannot also support the E/M. Can I bill an E/M visit and psychotherapy on the same day? Yes, when the psychotherapy is significant, separately identifiable, and separately documented. Use the add-on codes 90833, 90836, or 90838 alongside the E/M, never the standalone codes 90832, 90834, or 90837, which trigger the CO-97 bundling denial. Keep the psychotherapy time exclusive: minutes spent on psychotherapy count toward the add-on only, and the E/M level is selected by MDM. Physically separating the psychotherapy paragraph is a convention, not a rule, but it is the convention reviewers reward. Do I need a full mental status exam at every medication visit? No rule in any of the three countries requires one. Since the 2021 E/M overhaul, US code selection needs only a medically appropriate history and exam; no Canadian college standard or Australian item note specifies MSE scope either. Document what you examined and what you found, at the depth today's visit justified. The full format, when you need it, is on the mental status exam page. A 12-domain MSE copied forward unchanged is a larger audit exposure than a shorter exam written fresh. Is 90862 still a valid medication management code? No. The AMA deleted it effective January 1, 2013, with no grace period; medication management moved to the E/M codes that day and never returned. Its cousin 90863 exists only for prescribing psychologists in the states that authorize psychologist prescribing, and is never correct for psychiatrists, PMHNPs, or PAs. A template or payer document still built around either code is more than a decade out of date. How long should a medication management note be? Most run 150 to 500 words. No regulator sets a length, and no published time-and-motion study isolates how long psychiatric prescribers spend writing one, so treat any per-note minute figure as an estimate. Completeness beats length: each problem's status and severity, the data reviewed, the rationale for changes, and the risk assessment do the work. A long note assembled by copy-forward is weaker under review than a short one written fresh. How is this different from a psychiatric diagnostic evaluation? The psychiatric diagnostic evaluation (90791, or 90792 when a prescriber performs it) is the comprehensive intake: full history, diagnosis-setting, generally payable once per episode of care, with Medicare typically allowing one per year. The medication management note is the routine follow-up that rides the E/M family. Re-billing the intake code for a routine reassessment inside a continuous episode is a leading denial, and 90792 cannot be reported with an office E/M by the same provider on the same day. Save the intake format for a new patient, a return after a substantial gap, or a documented major change. Do I have to document that I checked the PDMP? The duty to query is state law in most of the US: 36 states plus DC mandate a check in at least some circumstances. Whether the note must record the query is separately specified by only a minority of states and left unspecified in most, and no primary source anywhere prescribes what the attestation must look like. One dated line satisfies the strictest versions, so most compliance teams standardize it for every controlled prescription. Canada and Australia run prescription monitoring through provincial and state systems with their own documentation rules. How long do medication management notes have to be kept? In the US, state rules mostly run 7 to 10 years from the last visit, longer for minors, and HIPAA's 6-year figure applies to compliance documentation, not clinical records. In Canada, CPSO requires 10 years and recommends 15, and CMPA advises 16 in British Columbia. In Australia, keep records 7 years from last service, or until age 25 for records collected from a child. Retention attaches to the whole record, and the medication history is often the part a future prescriber needs most. How does BastionGPT help with medication management notes? Dictate the visit or paste your transcript, and BastionGPT drafts the interval history, exam, MDM rationale, and plan in your clinic's format. Before you sign, it flags what is missing: risk not addressed, monitoring due, a dose change with no rationale, a stray time figure beside an MDM claim, or add-on documentation that is not separated. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychiatric Diagnostic Evaluation Psychotherapy Progress Note Mental Status Exam (MSE) Note Consultation Response & Report Chronic Care Management (CCM) Note Collaborative Care (BHI/CoCM) Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Australia Telehealth & MBS Case Conference Note Psychiatric Consultation Note COWS Protocol Documentation ASRS v1.1 Score Documentation Vanderbilt ADHD Scale Documentation Y-BOCS Score Documentation STOP-Bang Score Documentation Epworth Sleepiness Scale Documentation Insomnia Severity Index Documentation MADRS Score Documentation Morse Fall Scale Documentation OASIS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/tcm-note All note templates # Transitional Care Management (TCM) Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A transitional care management (TCM) note is the medical-record documentation behind Medicare's 30-day post-discharge service, billed under CPT 99495 or 99496. It evidences the three billable components: interactive contact within two business days of discharge, a face-to-face visit within 14 days (99495) or 7 days (99496), and moderate or high complexity medical decision making. Practices keep one record per qualifying discharge; most run 250 to 600 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The billing physician, NP, PA, CNS, or nurse-midwife; clinical staff log non-face-to-face work under general supervision Audience The billing practitioner, the care team and billers, Medicare and its contractors, auditors Typical length 250 to 600 words across the 30-day record · 10 to 20 minutes by hand (clinical team estimate) Format family Discharge-triggered 30-day service record (compare: CCM note, care coordination note) When it's used During the 30 days after a qualifying discharge, from first contact through the face-to-face visit and claim Standards context Medicare policy names four required data elements; every longer TCM checklist is convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a transitional care management (TCM) note? A transitional care management note is the medical-record documentation behind Medicare's bundled 30-day post-discharge service, billed under CPT 99495 and 99496. The codes were created by the American Medical Association's CPT Editorial Panel and adopted for Medicare payment effective January 1, 2013, through the CY2013 Physician Fee Schedule final rule, which makes TCM a sibling of the chronic care management and principal care management families. The record evidences the service's three moving parts: reaching the patient or caregiver interactively inside two business days of discharge, seeing them face to face inside 14 calendar days (99495) or 7 (99496), and carrying moderate or high complexity decision making through the 30-day period. You will also hear it called a post-discharge transitional care note or simply a 99495 note. One naming trap matters for behavioral health readers: in Medicaid and community mental health settings, "TCM" almost always means Targeted Case Management, a different benefit under 42 CFR 440.169 that has existed since 1995 and has nothing to do with these codes. The load-bearing fact about this document: CMS does not publish a required TCM note template. The Medicare Learning Network booklet MLN908628 (August 2025 edition) names only four data elements the record should contain: the discharge date, the date of the first interactive contact, the face-to-face visit date, and the level of medical decision making. That short list is payer policy; every longer "required checklist" circulating in vendor guides is convention. The TCM note is written by the receiving practice, which separates it from the discharge summary the hospital produces under its own regulation (42 CFR 482.43) and the TCM clinician reads on arrival, and from the chronic care management note , the monthly record for longitudinal coordination with no discharge trigger. Where the coordination is not billable at all, the plain care coordination note carries it. Canada and Australia have no bundled equivalent: post-discharge follow-up there rides on provincial fee codes, care-planning items, and handover standards, not a 30-day TCM service. ## Who uses transitional care management notes and when Primary care, internal medicine, family medicine, and geriatrics practices write most TCM notes, and the codes sit squarely inside behavioral health too: discharges from inpatient psychiatric hospitals and community mental health center partial hospitalization programs are qualifying settings, so a psychiatrist or psychiatric NP who assumes post-discharge care can bill TCM after them. Only practitioners who can independently bill evaluation and management services may bill it: physicians, NPs, PAs, clinical nurse specialists, and nurse-midwives. Psychologists, LCSWs, and counselors cannot, but they appear inside TCM records constantly as the clinical staff furnishing the non-face-to-face work under general supervision. Reach for a TCM note in the 30 days after a qualifying discharge; a CCM note for ongoing monthly coordination, which since 2020 may share a month with TCM when no minute is counted twice; and a care coordination note when the coordination has no billing frame. The behavioral health registry cousin is the collaborative care note . ## Transitional care management (TCM) note structure: what goes in each section Discharge header. Patient identifiers, the discharge date, the discharge setting, the day-30 end of the service period, and the billing practitioner. Qualifying settings include inpatient hospitals, SNFs, inpatient psychiatric hospitals, and CMHC partial hospitalization programs, and the period runs from the discharge date through 29 days after. Pitfall: the discharge date is one of only four elements CMS names, and it anchors every deadline in the record; a note that never states it cannot prove any timeframe was met. Interactive contact. The date and mode of the first two-way contact with the patient or caregiver, who made it, and every attempt if the first try fails; clinical staff may make the contact under general supervision. Pitfall: counting the discharge day. Business days start the next business day, Monday to Friday excluding holidays, so a Monday discharge runs to Wednesday and a Friday discharge to Tuesday; the discharge-day conversation itself never counts as the contact. Face-to-face visit. The visit date, its day count from discharge, and where it happened: inside 14 calendar days for 99495, inside 7 for 99496. Medicare does not allow this visit on the same day as discharge-day management. Pitfall: billing the required visit as a separate E/M. It is bundled into the TCM payment; only additional medically necessary visits during the 30 days are separately billable. Medication reconciliation. A dated reconciliation of the discharge list against what the patient is actually taking, completed no later than the date of the face-to-face visit, with discrepancies and the actions taken. Pitfall: a reconciliation dated after the visit; the on-or-before rule is payer policy, and the discharge medication list is exactly where transitions fail. Non-face-to-face services. A dated log of the 30-day coordination work: discharge summary review, referrals and community services, patient and caregiver education, communication with other providers. Pitfall: this is where behavioral health clinicians belong in a TCM record, as clinical staff whose documented work supports the claim, never as the billing practitioner; psychologists and LCSWs cannot bill TCM. Medical decision making. The level, moderate or high, with its basis: the conditions managed, the data reviewed, the risks carried through the period. Pitfall: the code is a pair test, not an MDM test alone. 99496 needs both high complexity and a visit inside 7 days; high complexity with a day-10 visit is still 99495. Billing check. The claim goes out with the face-to-face visit date as the date of service, after three overlap checks: no other TCM claim for this patient's 30-day period by any practitioner, no telephone E/M (99441-99443) during the period, and no global surgery period touching the service window. Pitfall: both active Recovery Audit topics live in this section. RAC topic 0225 targets more than one TCM per period and 0224 targets telephone E/M unbundling; run the checks before the claim, not after the recoupment letter. ### Blank template (copy and adapt) TRANSITIONAL CARE MANAGEMENT (TCM) NOTE Patient: ______________ DOB: __________ Discharge date: ______________ Discharge setting: ____________________ 30-day period ends: ___________ Billing practitioner (physician / NP / PA / CNS / CNM): ________________ Discharge summary reviewed: [ ] yes Date: __________ By: ____________ INTERACTIVE CONTACT (within 2 business days of discharge) Attempts (date / time / mode): _________________________________________ Reached: date ________ mode: [ ] phone [ ] portal [ ] in person With: [ ] patient [ ] caregiver By: _________________________________ FACE-TO-FACE VISIT (within 7 days: 99496; within 14 days: 99495) Date: __________ Days since discharge: ______ Location: ______________ Bundled into TCM: do not bill this visit separately. MEDICATION RECONCILIATION (no later than the visit date) Date: ________ By: ________________ Sources: _________________________ Discrepancies and actions: _____________________________________________ NON-FACE-TO-FACE SERVICES (date | by | activity) ______ | ______________ | ______________________________________________ ______ | ______________ | ______________________________________________ ______ | ______________ | ______________________________________________ MEDICAL DECISION MAKING [ ] moderate (99495) [ ] high (99496) Basis: _________________________________________________________________ BILLING CHECK Date of service = face-to-face visit date [ ] No other TCM billed this 30-day period, any practitioner [ ] No 99441-99443 this period [ ] No global surgery period overlap Code: [ ] 99495 [ ] 99496 Signature / credentials / date: ___________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample transitional care management (TCM) note Scenario: 30-day TCM record kept by an outpatient psychiatry practice after a 58-year-old patient's discharge from an inpatient psychiatric unit. The psychiatrist bills 99495; the practice's LCSW care coordinator furnishes the non-face-to-face work as clinical staff. All details are fictional. Transitional Care Management Note. Patient: J.L., 58 · Discharge: 07/06/2026, inpatient psychiatric unit · 30-day period: 07/06 to 08/04/2026 · Billing practitioner: S. Whitfield, MD (psychiatry) Discharge and source records: Discharged 07/06/2026 after a 9-day admission for major depressive disorder, recurrent, severe (F33.2), with comorbid type 2 diabetes (E11.9). Discharge summary received and reviewed 07/07/2026 by Dr. Whitfield. Inpatient team stopped sertraline, started venlafaxine XR titrated to 150 mg daily, and continued metformin. Interactive contact (within 2 business days): 07/07/2026 10:20, call to patient, voicemail left (attempt 1, business day 1). 07/08/2026 09:40, patient reached by phone (business day 2) by M. Duran, LCSW: discharge instructions and the new venlafaxine regimen reviewed, the 07/15 office visit confirmed, medication pickup screened; pharmacy did not stock venlafaxine XR. Non-face-to-face services: 07/08 · M. Duran, LCSW · Pharmacy called; venlafaxine XR located at a nearby branch; patient confirmed same-day pickup 07/09 · M. Duran, LCSW · Referral coordination with the intensive outpatient program; intake scheduled 07/20 07/14 · S. Whitfield, MD · Discharge labs and A1c reviewed; primary care office notified of the admission and the medication change Face-to-face visit: 07/15/2026, day 9 of 14, office visit with Dr. Whitfield. Mood improving, sleep 6 hours, PHQ-9 of 14 from 21 at admission. Venlafaxine XR continued at 150 mg; no orthostatic symptoms. This visit is bundled into TCM and not billed separately. Medication reconciliation (07/15/2026, on the visit date): Discharge list reconciled against the patient's bottles and pharmacy fill records by Dr. Whitfield. Discrepancy: leftover sertraline at home; instructed to discard, patient verbalized understanding. Final list: venlafaxine XR 150 mg daily, metformin 1000 mg twice daily. Medical decision making: Moderate. New antidepressant regimen requiring titration and monitoring, two chronic illnesses with one recent exacerbation, moderate risk of rehospitalization. Billing: Interactive contact 07/08 (business day 2), face-to-face visit 07/15 (day 9), medical decision making moderate: supports 99495 with date of service 07/15/2026. No other TCM claim this period, no telephone E/M billed, no global surgery period. Entry closed 07/15/2026 by S. Whitfield, MD. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works All four MLN908628 elements are dated and findable. Discharge date, first-contact date, visit date, and the decision-making level sit in labeled lines; an auditor can rebuild the timeline in seconds. The contact trail shows the business-day math. A Monday discharge, a documented day-1 attempt, contact completed on day 2, with mode, party, and the staff member named. Visit timing and complexity point at the same code. Day 9 plus moderate decision making is 99495, billed on the visit date, and the note says outright that the bundled visit was not billed separately. Medication reconciliation lands on the visit date, not after it. Sources named, one discrepancy found and closed, final list stated. The overlap checks mirror the audit topics. One TCM per period, no telephone E/M, no global surgery conflict: the exact targets of RAC topics 0225 and 0224. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Treat the TCM note as ordinary medical-record content: it sits in the designated record set, is releasable on request, and, as electronic health information, falls under the information blocking rule (45 CFR Part 171) that has applied to providers since April 5, 2021, so a records request cannot be parked. Authentication is payer policy with teeth: Medicare's Program Integrity Manual requires the author to sign with a handwritten or electronic signature, stamps do not qualify, and an illegible signature can be supported by a signature log or attestation, but a signature added retroactively is not accepted. Retention layers on: 42 CFR 424.516(f) requires documentation supporting Medicare claims to be kept 7 years from the date of service, the overpayment rule reaches back 6 years, and state boards run from 6 years (New York) through 7 (Texas, Pennsylvania) to 10 (Rhode Island), longer for minors; the longest applicable rule wins. Timely filing is regulation, not contractor discretion: 42 CFR 424.44 allows one calendar year from the date of service. The payer layer is where TCM claims die, and two stale rules still circulate. Since the CY2016 rule the claim is billed on the date of the face-to-face visit, not held to day 30, and since the CY2020 rule TCM and CCM may share a month when both are earned and no minute is counted twice. The four minimum elements are the requirement; the format is a convention, so build the note around dated proof: business-day arithmetic on the contact, attempts documented when the patient is hard to reach, medication reconciliation on or before the visit date, and the overlap checks run before the claim. If the patient is readmitted inside the 30 days, TCM can still be reported when the components are furnished, or billed after the second discharge for a fresh period. FQHCs and RHCs are not paid separately for TCM under the fee schedule, though the visit may count as a billable visit. And the note's neighbors matter: the hospital's discharge summary is the source document the reconciliation runs against, and ongoing coordination after day 30 belongs in a CCM note . ## Common transitional care management note errors auditors flag The enforcement file on TCM is short and specific. The HHS Office of Inspector General's dedicated audit (A-07-17-05100, July 2021) reviewed almost 1.8 million claims worth $249.5 million from 2015 and 2016 and found about $1.7 million in overpayments across 13,577 claims, almost all from more than one practitioner billing the same 30-day period: 0.006 percent of the dollars, and the report concluded the program "generally complied" with federal requirements. The live exposure is newer: on February 1, 2025, CMS approved two Recovery Audit topics dedicated to TCM, 0225 for excessive units and 0224 for unbundling telephone E/M codes. Front-end denials tell the learning-curve story: 37.3 percent of TCM services were denied in 2013, falling to 4.9 percent by 2018 as practices absorbed the timing rules (JAMA Network Open, 2020). The BastionGPT Clinical Advisory Board sees the same errors most often in transitional care management note reviews: Two claims, one period. More than one practitioner billing TCM for the same discharge, or the same practitioner twice; Medicare pays once per patient per 30-day period, it was the largest slice of the OIG finding, and it is the exact target of RAC topic 0225. Business-day arithmetic on the contact. Counting the discharge day, or forgetting that business days exclude weekends and holidays: a Friday discharge runs to Tuesday. Undocumented attempts compound it; two or more timely attempts with continued effort keep the claim defensible, an untraceable phone call does not. The bundled visit billed twice. Reporting the required face-to-face visit as a separate E/M on top of the TCM payment, or scheduling it on the same day as discharge-day management, which Medicare does not allow. Telephone E/M inside the period. Codes 99441-99443 reported for the same patient during the 30 days; CMS calls this unbundling, RAC topic 0224 exists to recoup it, and the fix is documentation inside the TCM record rather than a second claim. The four elements, undated. Notes long on narrative and short on the only things MLN908628 names: discharge date, first-contact date, visit date, decision-making level. Auditors reconstruct the timeline from dates; a record that cannot date its own deadlines fails regardless of length. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on transitional care management notes. Draft the contact entry, the visit note, and the 30-day record from a dictated recap, a call log, or the care team's messages, every component dated. Pull the discharge medications, pending results, and follow-up items out of a pasted discharge summary, ready for the reconciliation and the visit agenda. Check the record before the claim: all four MLN908628 elements present and dated, business-day math on the contact, visit window and complexity pointing at the same code, and overlap flags for duplicate TCM, telephone E/M, and global surgery periods. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does CMS require a specific TCM note template? No. The Medicare Learning Network booklet MLN908628 names four data elements the medical record should document: the date of discharge, the date of the first interactive contact with the patient or caregiver, the date of the face-to-face visit, and the complexity of medical decision making. Everything beyond that list, structured medication grids, care-plan sections, per-call narratives, is convention or vendor design, useful but not mandated. Build the note so the four dated elements are impossible to miss, then shape the rest to your workflow. How are the two business days counted for the interactive contact? Start counting on the first business day after discharge; the discharge day itself does not count, and business days are Monday through Friday excluding holidays, regardless of your practice hours. A Monday discharge means contact by the end of Wednesday; a Friday discharge means Tuesday. If you cannot reach the patient, document two or more timely attempts and keep trying. CMS guidance leaves contact completed after day 2 genuinely ambiguous, so many practices bill a regular E/M instead when the contact never lands in the window. What date of service goes on the TCM claim? The date of the face-to-face visit. Since the CY2016 Physician Fee Schedule rule you neither wait for day 30 nor report the 30th day; the older bill-on-day-30 instruction still circulates and is out of date. The 30-day service period itself still runs from the day of discharge through 29 days after it, and the work you document can continue through the whole period. How long do we keep TCM documentation? Under 42 CFR 424.516(f), documentation supporting Medicare claims must be kept 7 years from the date of service, and the overpayment rule adds a 6-year lookback. State medical-board rules layer on top: 6 years in New York, 7 in Texas and Pennsylvania, 10 in Rhode Island, longer for minors, and the longest applicable rule wins. Keep the contact log, the visit note, and the medication reconciliation together; an audit reads the 30-day record as one timeline. Can psychologists, LCSWs, or counselors bill TCM? No. TCM is limited to practitioners who can independently bill evaluation and management services: physicians, NPs, PAs, clinical nurse specialists, and nurse-midwives. Behavioral health clinicians still carry much of the work as the clinical staff furnishing non-face-to-face services under general supervision, and the discharges they care most about qualify: inpatient psychiatric hospitals and CMHC partial hospitalization programs are on the list of TCM discharge settings, so a billing physician or NP can anchor TCM after a psychiatric discharge. What is the difference between a TCM note and a CCM note? TCM is discharge-triggered and finite: one 30-day period, an interactive contact, a face-to-face visit, and moderate or high complexity decision making. CCM is longitudinal: a monthly time-threshold service for patients with two or more chronic conditions, built around a care plan, with no discharge trigger. Since 2020 the two may be billed in the same month when both are earned and no minute counts toward both. The CCM note page covers the monthly record; coordination outside any billing program belongs in a care coordination note . Is transitional care management the same as targeted case management? No, and the shared abbreviation causes real confusion. Transitional care management is the Medicare post-discharge service billed under 99495 and 99496. Targeted case management is a Medicaid benefit under 42 CFR 440.169, covered since 1995, that funds case management for defined populations such as adults with serious mental illness. A behavioral health agency's "TCM program" is usually the Medicaid one. Check which service a payer, contract, or job description means before assuming either set of rules applies. What happens if the patient is readmitted during the 30 days? TCM can still be reported if the required components are furnished within the original period, or the practitioner may instead bill TCM after the second discharge, starting a fresh 30-day period, provided no other practitioner billed TCM for the first one. CMS addressed both paths in its TCM guidance. Either way, document which discharge anchors the period so the timeline reads cleanly. How does BastionGPT help with transitional care management notes? Bring the discharge summary, the call log, and the visit however you captured it. BastionGPT drafts the dated contact entry, the visit note, and the 30-day record, extracts the medication list for the reconciliation, and checks the claim-facing details: the four MLN908628 elements, the business-day window, the visit-and-complexity match to 99495 or 99496, and overlap flags for duplicate TCM, telephone E/M, and global surgery periods. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Chronic Care Management (CCM) Note Discharge Summary Care Coordination Note Collaborative Care (BHI/CoCM) Note Claim-Support Billing Note Medication Management Follow-Up Note Return-to-Primary-Care Summary Australia Telehealth & MBS Case Conference Note Transfer-of-Care Summary Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/return-to-primary-care-summary All note templates # Return-to-Primary-Care Summary: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A return-to-primary-care summary is the closing letter that transfers a stabilized behavioral health patient from specialty or integrated care back to their primary care clinician. It states diagnoses, treatment delivered, current outcome measures, medications and who prescribes them, a relapse prevention plan, and re-referral triggers. Psychologists, psychiatrists, and collaborative care teams send it at step-down; most run 300 to 600 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The discharging clinician: psychologist, psychiatrist, therapist, or prescriber; in CoCM the care manager co-writes the closing plan Audience The receiving primary care clinician or GP; the patient, payers, and auditors read it too Typical length 300 to 600 words · 10 to 20 minutes by hand (clinical team estimate) Format family Closing handback letter (compare: discharge summary, consultation report) When it's used At the end of a specialty or integrated behavioral health episode, when ongoing care returns to primary care Standards context Written into the MBS item descriptor in Australia; professional convention in the US and Canada On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a return-to-primary-care summary? A return-to-primary-care summary is the closing clinical letter that transfers responsibility for a stabilized behavioral health patient from a specialty or integrated care service back to the patient's primary care clinician. It answers the referral that opened the episode: what was found, what was done, how the patient measures now, who prescribes what from here, and exactly when to re-refer. The document has two lineages rather than one inventor. In the US it grew out of the Collaborative Care Model built by Wayne Katon and Jürgen Unützer's team at the University of Washington in the 1990s and proven in the IMPACT trial (JAMA, 2002), where the closing step is the relapse prevention plan the care manager co-creates with the patient before graduation. In Australia it has been part of Medicare's Better Access program since 2006 as the psychologist's end-of-course report to the referring GP. You will also see it called a step-down summary, graduation letter, handback summary, back-to-GP letter, or coordination-of-care letter. One fact separates this page from the template libraries: whether you must write the letter depends on where you practice. In Australia the report-back is written into the MBS item descriptors themselves, items 80000, 80005, 80010, and 80015 and their telehealth equivalents, so a course of treatment that ends without a written report to the referrer was not properly rendered as billed; Explanatory Note MN.6.2 confirms a report is due at the completion of each course. In the US, no statute or CMS rule names this letter. The nearest binding hook is the collaborative care codes' relapse prevention planning, a required service element of 99492, 99493, 99494, and G2214 that CMS never requires anyone to transmit. In Canada the duty comes from college continuity-of-care policy and provincial privacy law, not a fee schedule. And it is not a discharge summary , which closes an episode at any level of care: the handback letter specifically re-vests longitudinal responsibility in primary care, with re-referral triggers attached. ## Who uses return-to-primary-care summaries and when Australian psychologists write one at the end of every Better Access course, because the item descriptor requires it. US collaborative care teams write the graduation equivalent when a patient leaves the CoCM registry, and psychiatrists, psychiatric NPs, and psychologists in ordinary outpatient practice write one whenever a stabilized patient no longer needs specialty care. The volume keeps growing: commercial CoCM penetration rose roughly 27-fold from 2018 to 2023 (Milliman analysis for the Meadows Mental Health Policy Institute, 2025), and every graduated patient is a handback waiting to be written. Choose this letter when ongoing management returns to primary care with no specialty treater remaining. If another active specialty clinician takes over instead, that is a transfer of care, not a handback. If you are answering a referrer's question without ever assuming care, write a consultation report . The registry documentation that precedes a CoCM graduation lives in the collaborative care note , and coordination work outside any billing frame belongs in a care coordination note . ## Return-to-primary-care summary structure: what goes in each section Header and referral context. Patient identifiers, the receiving PCP or GP by name, episode start and end dates, the referral source and date, and the original referral question. Pitfall: this letter is the closing half of a referral loop. If it never names the referrer or the question the episode was meant to answer, the loop stays open on paper; Australian psychologists must also keep the referral itself for 24 months from the first service. Reason for the episode and diagnoses. The presenting problem at entry and the diagnostic picture at handback, coded where your payer expects it. Pitfall: restating the intake diagnosis untouched. The PCP needs to know what changed: resolved, in remission, revised, or unchanged with a reason. Treatment delivered and response. Modality, session count and date range, assessments used, and how the patient responded. Pitfall: a session count with no response line. Sixteen sessions of CBT is logistics; response to treatment is the clinical content the next decision rides on. Current status and outcome measures. Scores at entry and at handback with dates, PHQ-9, GAD-7, or whatever the episode tracked, plus a one-line functional status. Pitfall: "much improved" is an adjective, not a baseline. Primary care monitors against your closing numbers; without them the next screening has nothing to compare to. Medications and prescribing responsibility. The current list with doses, changes made during the episode, and one explicit line naming who prescribes from today. Pitfall: the most dangerous ambiguity in the letter. If the psychiatrist's last refill runs out in 30 days and the PCP believes psychiatry still prescribes, the patient stops medication cold. Relapse prevention plan. Early warning signs in the patient's own words, what helped this episode, first steps if signs return, and confirmation the patient holds a copy. Pitfall: stapling a generic handout. In collaborative care the plan is co-created with the patient in the months before graduation; a plan the patient has never seen fails at its only job. Re-referral triggers and follow-up. Concrete thresholds for coming back: score cutoffs, safety conditions, medication events, plus the route back and a suggested follow-up interval. Pitfall: "return as needed." A PCP managing thousands of patients acts on thresholds; name the number, the condition, and the direct line. ### Blank template (copy and adapt) RETURN-TO-PRIMARY-CARE SUMMARY Date: __________ From: _______________________________________________ To (PCP / GP): _________________________________________________________ Patient: ______________ DOB: __________ Episode: ________ to ________ Referral source / date: ________________________________________________ Original referral question: ____________________________________________ REASON FOR EPISODE AND DIAGNOSES At referral: ___________________________________________________________ At handback (resolved / remission / revised / unchanged): ______________ TREATMENT DELIVERED AND RESPONSE Modality and focus: ____________________________________________________ Sessions: ______ Dates: ________ to ________ Completed: [ ] yes [ ] no Response: ______________________________________________________________ CURRENT STATUS AND MEASURES Measure: _________ At entry: ______ At handback: ______ Date: ______ Measure: _________ At entry: ______ At handback: ______ Date: ______ Functional status: _____________________________________________________ MEDICATIONS AND PRESCRIBING Current medications and doses: _________________________________________ Changes during the episode: ____________________________________________ Prescribing from today: [ ] PCP / GP [ ] psychiatry until ____________ RELAPSE PREVENTION PLAN Early warning signs: ___________________________________________________ What helped this episode: ______________________________________________ Patient holds a copy: [ ] yes Built together on: _____________________ RE-REFERRAL TRIGGERS AND FOLLOW-UP Re-refer if: ___________________________________________________________ How to re-refer / direct contact: ______________________________________ Suggested follow-up interval: __________________________________________ Clinician signature / credentials: _________________ Date: ____________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample return-to-primary-care summary Scenario: an outpatient psychiatrist returns a patient with major depressive disorder, now in remission on sertraline, to the referring primary care physician, transferring prescribing back. All details are fictional. Return-to-Primary-Care Summary. From: R. Okafor, MD, Lakeview Psychiatry · To: A. Patel, MD, Fern Hill Primary Care · Date: 07/23/2026 · Re: T.K., 41 · Episode: 01/20/2026 to 07/21/2026 Dear Dr. Patel, Referral context: You referred T.K. on 01/06/2026 after two months of worsening depression despite sertraline 50 mg, asking us to take over medication management and assess for treatment resistance. This letter answers that referral and returns T.K. to your care as of today; we are not scheduling further visits. Diagnoses: At referral: major depressive disorder, recurrent, moderate (F33.1). At handback: in remission on measures since April. Structured screening found no bipolar spectrum features and no substance use disorder. Treatment delivered and response: Six visits from 01/20/2026 to 07/21/2026. Sertraline titrated to 150 mg daily by March after a partial response at 100 mg, with brief behavioral activation coaching alongside. Improvement was steady rather than dramatic: measurable gains by the second month, remission-range scores from April onward, held through today. Current status and measures: PHQ-9 4 on 07/21/2026, from 16 at intake (01/20/2026); GAD-7 3, from 9. Sleeping about 7 hours, exercising twice a week, work performance back to baseline by self-report. Medications and prescribing: Sertraline 150 mg daily, unchanged since March. Prescribing transfers to you with this letter; a final 90-day supply went to the pharmacy today to cover the transition. I recommend holding the current dose for at least 9 to 12 more months before any taper discussion, given two lifetime episodes. Relapse prevention plan: Built with T.K. on 07/21/2026; T.K. holds a copy. Early warning signs, in T.K.'s words: skipping the gym for two weeks, dreading Monday by Saturday night, going quiet in the family group chat. What helped: morning walks, activity scheduling, not skipping doses. First steps if signs return: restart the activity log and call your office. Re-referral triggers: Please send T.K. back if PHQ-9 reaches 15 at any visit or 10 to 14 on two consecutive screenings, if item 9 is positive at any score, if symptoms return during a future taper, or if you want a medication consult for any reason. Direct scheduling line: 555-0187; returning patients are typically seen within two weeks. Follow-up suggestion: PHQ-9 at T.K.'s next routine visit, then quarterly for a year. Records from the episode are available on request with T.K.'s authorization. Thank you for the referral. R. Okafor, MD · Lakeview Psychiatry · Signed 07/23/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral loop closes in writing. The letter names the referral date, restates the question it was sent to answer, answers it, and dates the transfer of responsibility to a specific day. Entry and exit numbers give primary care a baseline. PHQ-9 16 to 4 and GAD-7 9 to 3, all dated; the next screening in primary care has something to compare against. The prescribing handoff is one unambiguous sentence. Who prescribes from today, a bridge supply to cover the transition, and a duration recommendation before any taper. The relapse prevention plan is the patient's own. Warning signs in T.K.'s words, built together, copy in hand, first steps named. Re-referral triggers are thresholds a busy PCP can act on. Score cutoffs, an item 9 rule, a taper condition, a direct line, and a two-week return window. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The summary is part of the medical record and the designated record set: releasable to the patient on request, and disclosable to the PCP for treatment purposes under HIPAA without a separate authorization, while psychotherapy notes stay behind their own protection. One US exception has teeth now: if the episode involved substance use disorder treatment at a Part 2 program, the handback is a disclosure governed by 42 CFR Part 2, whose 2024 Final Rule passed its compliance date on February 16, 2026 and is enforced by OCR, so check that consent covers the disclosure before the letter leaves. Retention follows the longest applicable rule. US: APA guidance says 7 years after the last service for adults and 3 years past majority for minors, with states ranging from 5 years (Pennsylvania) to 7 (California, North Carolina, Tennessee, Connecticut). Canadian provinces commonly run 10 years. Australia expects 7 years, or until age 25 for minors, and separately requires the psychologist to keep the referral itself for 24 months from the first service. The payer layer splits by country, and the split is the point. In Australia the report to the referrer is a condition inside the Better Access item descriptors, due at the completion of each course of treatment under Note MN.6.2, and the obligation survived the November 2025 Better Access changes intact: skip it and the item was not properly rendered. In the US no rule requires sending this letter at all; relapse prevention planning is a service element of the CoCM codes 99492 to 99494 and G2214, and since G0512 was retired at the end of 2025, FQHCs and RHCs bill those individual codes too. The Joint Commission's hand-off standard PC.02.02.01 and CARF's transition standards bind accredited organizations to structured handovers, and in Canada college continuity-of-care policies carry the duty into enforceable territory. Everywhere, the letter format is a convention; the continuity content is the requirement. Keep the clinical detail in the discharge summary and the record; this letter is the transmission that makes the step-down real. ## Common return-to-primary-care summary errors auditors flag The research says the biggest failure is not writing quality but transmission. In the best-documented US study, only 34.8 percent of 103,737 referral scheduling attempts produced a documented completed appointment (Patel and colleagues, Journal of General Internal Medicine, 2018), and a classic communication study found 25 percent of PCPs still had no information from the specialist four weeks after a referral visit (Gandhi and colleagues, 2000); the return half of the loop is even less studied. Enforcement is indirect but real. Australia's Professional Services Review has penalized record-keeping too thin for another practitioner to safely take over care, including a January 2022 agreement in which a GP repaid $135,000 and accepted item disqualifications, and in December 2025 the College of Physicians and Surgeons of Alberta found a psychiatrist guilty of unprofessional conduct partly for an inadequate consultation report to the referring family physician. The BastionGPT Clinical Advisory Board sees the same errors most often in return-to-primary-care summary reviews: The letter is written but never lands. Filed in the chart, faxed to a dead number, or parked in an EHR queue; the loop closes only when the PCP receives it, and the closure numbers above are the base rate. Track the send and the confirmation, not just the authorship. Adjectives where the PCP needs numbers. "Significantly improved" without entry and exit scores leaves primary care monitoring against nothing; the closing PHQ-9 or GAD-7 with dates is the whole baseline. Nobody owns the prescription. The medication list is present but no line names who renews it; the gap surfaces 30 days later as an abrupt discontinuation. "Return as needed" instead of triggers. No score threshold, no safety condition, no route back; six months later the re-referral starts over in the general queue instead of on the direct line. Treating the Australian report as a courtesy. Under Better Access the end-of-course report to the referrer sits inside the item descriptor; a course billed without one was not properly rendered, and the referral itself must be kept for 24 months. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on return-to-primary-care summaries. Draft the full handback letter from a dictated recap, the episode's closing notes, or the measure scores, with the referral question answered up top. Convert a discharge summary or final progress note into a PCP-ready letter, keeping the scores, the medication list, and the relapse prevention plan intact. Check the letter before it goes: referrer named, entry and exit measures dated, a prescriber assigned, re-referral thresholds concrete, and the report content Australian Better Access courses require. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I legally have to send a letter to the PCP or GP when I discharge a client? It depends on the country, and the difference is the most important fact on this page. In Australia, yes: Medicare's Better Access item descriptors condition the benefit on a written report to the referring practitioner at the completion of each course of treatment, so the report is part of properly rendering the item, not a courtesy. In the US and Canada, no statute or fee schedule names this letter: the duty comes from professional convention, college continuity-of-care policy, and, in accredited organizations, hand-off standards from the Joint Commission or CARF. The Australian loop opens with the Better Access treatment plan and referral; this report is how it closes. What goes in a return-to-primary-care summary? Payers and auditors converge on the same core: patient identifiers and the referral source; the presenting problem and diagnoses at entry and at handback; treatment delivered with dates and the response to it; current status with validated measures such as the PHQ-9 or GAD-7; medications with an explicit line naming who prescribes from now on; the relapse prevention plan; any outside referrals; and recommendations with concrete re-referral triggers. Australia's MBS notes expect the end-of-course report to cover the assessments done, the treatment given, and what the psychologist recommends next. If a section is empty, say so in a phrase rather than deleting it; the PCP reads absence as information. How long should a return-to-primary-care summary be? Most run 300 to 600 words: one page. No regulation in any of the three countries sets a length, a format, or a signature deadline for this letter, so length is a convention shaped by what a primary care clinician will actually read. Front-load the four decisions the PCP needs on first scan: what the diagnosis is now, what the medications are and who prescribes them, what the closing scores are, and when to re-refer. Everything else belongs in the record, which the PCP can request. Is the CoCM relapse prevention plan the same as this letter? No, and the two are conflated constantly. In the Collaborative Care Model, the relapse prevention plan is a patient self-management tool the behavioral health care manager co-creates with the patient in the months before graduation; CMS set no minimum credential for the care manager who co-writes it (81 FR 80231). It is a required service element of the CoCM billing codes, but CMS never requires that it be transmitted to anyone, and the primary care clinician is already the billing provider, so inside CoCM there is often no letter at all. The handback letter earns its keep when a separate specialty service returns care to a PCP who was not running the episode. The monthly registry record is the collaborative care note , and the plan itself has its own relapse prevention plan template. How is a return-to-primary-care summary different from a discharge summary? A discharge summary closes an episode of care at any level and can go to the patient, the next facility, or simply the chart; it is about ending. A return-to-primary-care summary is about what starts next: it re-vests longitudinal responsibility in a named primary care clinician and arms them with a baseline and re-referral triggers. In Australia the two collapse into one document, because the end-of-course report to the GP is the discharge communication for a Better Access episode. In US and Canadian practice you will often write both: the summary for the record, the letter for the receiving clinician. Does US Medicare pay separately for writing this letter? No code pays for the letter itself. Inside collaborative care, relapse prevention planning is one of the service elements your monthly CoCM time already covers under 99492, 99493, 99494, and G2214; and since G0512 was retired at the end of 2025, FQHCs and RHCs bill those same individual codes from 2026. Outside CoCM, the writing time is part of the final visit's work. There is also no denial code specific to a missing handback letter: no CARC or RARC names it, so failures surface indirectly, as insufficient-documentation findings or continuity complaints rather than a line-item denial. How long do we keep a return-to-primary-care summary? Keep it with the episode's record under the longest applicable rule. In the US, APA guidance says 7 years after the last service for adults, or 3 years past the age of majority for minors, with state law layering on top: California, North Carolina, Tennessee, and Connecticut run 7 years, Pennsylvania 5. Canadian provinces commonly require 10 years, longer for minors. Australia expects 7 years, or until age 25 for minors, and adds one obligation specific to this document type: the psychologist must keep the referral itself for 24 months from the date of the first service. How do I make sure the letter actually reaches the PCP? Treat transmission as a workflow step with its own evidence, because this is where handbacks die: only 34.8 percent of 103,737 referral scheduling attempts in the best-documented US study produced a completed, documented appointment, and 25 percent of PCPs in a classic study still had no specialist information four weeks after a referral visit. Send through a channel that produces a receipt: direct secure messaging, an EHR referral module, or a fax with confirmation. Log the send date and the confirmation in the record, and task a follow-up if nothing confirms within a week. A perfect letter that stays in the chart closes nothing. How does BastionGPT help with return-to-primary-care summaries? Bring the closing session however you captured it: a dictated recap, the episode's final notes, or the measure scores. BastionGPT drafts the handback letter with the referral question answered first, converts a discharge summary into a PCP-ready letter, and checks the details that make one defensible: entry and exit measures with dates, an explicit prescriber line, concrete re-referral thresholds, and the report content Australian Better Access courses require. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Collaborative Care (BHI/CoCM) Note Discharge Summary Relapse Prevention Plan Care Coordination Note Chronic Care Management (CCM) Note Consultation Response & Report Medication Management Follow-Up Note Transitional Care Management (TCM) Note Australia Telehealth & MBS Case Conference Note 42 CFR Part 2 Consent Transfer-of-Care Summary Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mbs-case-conference-note All note templates # Australia Telehealth & MBS Case Conference Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An MBS case conference note is the clinical record entry an Australian practitioner makes to substantiate a multidisciplinary or mental health case conference item under the Medicare Benefits Schedule. The regulations legislate recorded facts, not a form: the patient's agreement, the day with start and end times, participant names, and the matters discussed. Telehealth attendance notes add a documented eligibility pathway. Most run 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The practitioner claiming the item: the organising medical practitioner, or each participating practitioner, including eligible allied mental health professionals Audience The patient's medical record, the care team, Services Australia, Professional Services Review Typical length 150 to 400 words · 10 to 15 minutes by hand (clinical team estimate) Format family Medicare substantiation record (compare: case conference note, team meeting note, Better Access plan) When it's used Whenever an MBS case conference or telehealth attendance item is claimed; written at the time or as soon as practicable Standards context The recorded facts are delegated legislation (GMST Regulations); the note layout itself is convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an MBS case conference note? An MBS case conference note is the contemporaneous record entry a practitioner makes to substantiate a claim for a multidisciplinary or mental health case conference item under Australia's Medicare Benefits Schedule, and its telehealth sibling is the attendance note that substantiates a video or phone item. Case conferencing entered the MBS with the Enhanced Primary Care package in 1999; the recording rules now sit in the Health Insurance (General Medical Services Table) Regulations 2021 , read with explanatory note AN.0.49 . The mental health case conferencing family arrived on 1 July 2023 out of the Better Access evaluation, and the telehealth items, created by emergency determination in March 2020, became permanent on 1 January 2022. Clinicians file the same content under several names: case conf note, MDT note, case conference record, telehealth attendance note, or loosely CDM documentation. The distinction this page exists to make: there is no legislated case conference template. What the regulations fix, in the organising and participating clauses (2.16.15 and 2.16.16), is the set of facts a biller must record: the patient's agreement, the day with start and end times, the participants, the matters the team worked through, a copy filed in the medical record, and, for the organiser, summaries offered and given and outcomes discussed. No participant signature, no written consent form, no minimum length, and no video-platform field appears anywhere in the instruments. The general-purpose meeting record, with its US and Canadian rules, lives on the case conference note page; this page covers the Australian claim-substantiation layer that turns that meeting record into evidence. ## Who uses MBS case conference notes and when GPs and prescribed medical practitioners organise most conferences; psychiatrists, paediatricians, and consultant physicians have their own item families; and psychologists, mental health nurses, accredited mental health social workers, and occupational therapists participate, with eligible allied mental health professionals claiming the participation items in the mental health family. The settings are the coordination-heavy ones: shared GP and psychologist care under Better Access, residential aged care, palliative teams, and community mental health. Uptake has trailed the need from the start: Health Insurance Commission data showed only 11,095 case conference claims in 2001 against 155,486 multidisciplinary care plans, and GPs named the recording rules among the barriers. Reach for this note whenever an item will be claimed; use the team meeting note for the recurring internal meeting that reviews several clients without billing, and the Better Access treatment plan when the work product is the plan itself rather than a conference. ## MBS case conference note structure: what goes in each section Consent, recorded before the conference. Explain the nature of the conference to the patient, ask whether they agree, and record the agreement. The patient or carer need not attend, and nothing requires the consent to be in writing or on a signed form: the recorded fact is the requirement. For the mental health items, record agreement to each participant as well. Pitfall: consent obtained in conversation and recorded nowhere; a missing agreement line can unwind an otherwise perfect claim. Day, start and end times. Record the day and the exact start and end times, written at the time. The items are tiered at 15 to 20, 20 to 40, and 40 or more minutes, so the times are not context, they select the item. AskMBS confirms no minute-by-minute template is needed, and only time actually spent in the conference counts. Pitfall: a tier the recorded times cannot prove; absent or reconstructed times are the most common failure in Professional Services Review outcomes. Participants and team mix. Name every participant with their profession and the kind of care they bring. A claimable conference needs at least three participants present at the same time, no more than two of them medical practitioners, each providing a different kind of care; the patient does not count toward the minimum. Pitfall: an email thread, a pair of clinicians, or three doctors written up as a conference; each fails the definition before the note is even read. Matters discussed: the enumerated set. Record what the team worked through: the patient's history and multidisciplinary care needs, the outcomes to be achieved, the tasks allocated with a named member for each, and a review of previously identified outcomes. Pitfall: "case discussed, plan agreed"; an empty discussion section reads to an auditor as a service that cannot be shown to have happened. Copy filed in the record. Place a copy of the conference record in the patient's medical record; a note that lives only in the organiser's meeting file fails the clause. This is also where the two-year retention clock for claim-condition documents starts. Pitfall: a compliant record filed in the wrong place; a records reviewer reads the patient's chart, not your meeting folder. Organiser extras: summaries and outcomes. The organiser must also offer a summary to the patient and carer, give a summary to each team member, and discuss the conference outcomes with the patient or carer. Participants have none of these duties, and the organiser cannot claim a participation item on top: organising includes participating. Pitfall: an organiser item claimed on a participant-grade record, with no distribution trail to show. Telehealth attendances: the eligibility line. For video and phone attendance items, add one line documenting the eligibility pathway: a face-to-face service with you or your practice in the preceding 12 months, or the exemption relied on. Modality is captured by the item number, and no instrument requires the platform to be named. Pitfall: a note that names the video platform but not the pathway; the first is optional, the second is what the November 2025 AskMBS advisory expects. Attribution and timing. Sign off with your name and credentials, and complete the entry at the time or as soon as practicable, as a separate dated entry for each service, comprehensible enough for another practitioner to continue the patient's care from it. Pitfall: batch-writing entries days later; a reconstructed entry is exactly what PSR findings describe as inadequate. ### Blank template (copy and adapt) MBS CASE CONFERENCE NOTE (Australia) Patient: ______________ DOB: __________ Conference date: __________ Role: [ ] organiser [ ] participant Item claimed / tier: __________ Patient agreement recorded (date, how): ________________________________ Patient/carer attending: [ ] yes [ ] no (recorded consent is required, attendance is not) Start time: ________ End time: ________ Total minutes: ________ Tier check: [ ] 15 to 20 min [ ] 20 to 40 min [ ] 40+ min PARTICIPANTS (name, profession, kind of care, mode: in person/video/phone) 1. _____________________________ 2. _____________________________ 3. _____________________________ 4. _____________________________ All present at the same time: [ ] Mix: 3+ members, no more than 2 medical practitioners, each providing a different kind of care: [ ] MATTERS DISCUSSED History and multidisciplinary care needs: ______________________________ Outcomes to be achieved: _______________________________________________ Tasks allocated (task, named member, timeframe): _______________________ Previously identified outcomes reviewed: _______________________________ ORGANISER ONLY Summary offered to patient/carer: [ ] Summary given to each member: [ ] Outcomes discussed with patient/carer: [ ] TELEHEALTH ATTENDANCES (if claiming video/phone items) Eligibility pathway: [ ] face-to-face with practitioner/practice in the last 12 months [ ] exemption relied on: ____________ (document it) Copy of this record filed in the patient's medical record: [ ] Author name/credentials/date: __________________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample MBS case conference note Scenario: a GP organises a mental health case conference by video for a 29-year-old with major depressive disorder and harmful alcohol use whose treatment has stalled; the treating psychologist and an accredited mental health social worker participate. All details are fictional. MBS Mental Health Case Conference Note. Patient: R.T., 29 · Conference date: 21/07/2026 · Role: organiser · Purpose: coordinate depression and alcohol care across GP, psychologist, and AMHSW after eight weeks without improvement Consent: Nature and purpose of the conference explained at the 14/07/2026 attendance; R.T. agreed to the conference and to each named participant; agreement recorded that day. R.T. elected not to attend. Day, times, mode: 21/07/2026, 14:05 to 14:37 (32 minutes). All three participants present at the same time by video. Participants: Dr H. Patel, GP (organiser, medical care); C. Bruns, clinical psychologist (fortnightly psychological therapy); M. Okafor, accredited mental health social worker (alcohol and housing supports). Three participants, one medical practitioner, each providing a different kind of care. Matters discussed: History and needs: PHQ-9 19 on 14/07/2026, minimal change over eight weeks; drinking 6 to 8 standard drinks on about three nights weekly; rent arrears adding stress; shift work disrupting appointments. Outcomes to be achieved: one shared behavioural activation plan, a coordinated drinking-reduction target, and rent assistance lodged. Tasks allocated: C. Bruns to align session goals with the shared activation plan and repeat the PHQ-9 by 18/08/2026; M. Okafor to lodge the rent assistance application by 28/07/2026 and begin weekly check-ins; Dr Patel to review sertraline dose and pathology at the 24/07/2026 attendance. Previously identified outcomes reviewed: the 04/2026 referral plan lapsed when R.T. stopped attending; barriers (shift work, cost) addressed by moving to evening telehealth sessions. Organiser items and filing: Summary offered to R.T. and accepted, sent 22/07/2026; summaries provided to both participants 22/07/2026; outcomes to be discussed with R.T. at the 24/07/2026 attendance. Copy of this record filed in R.T.'s medical record. Entry completed 21/07/2026 immediately after the conference. Dr H. Patel, MBBS FRACGP. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every legislated recording activity is on the face of the note. Agreement, day and times, participant names, the enumerated matters, the filed copy, and the organiser's summaries each get a line an auditor can point to. The tier is provable. 14:05 to 14:37, written at the time, supports the middle time tier without any reconstruction. The team mix is auditable. Three named participants, one medical practitioner, each providing a different kind of care, all present at the same time. Outcomes are reviewed, not just set. The note explains why the April plan stalled before allocating new tasks to named members with dates. The adequacy standard is met. A separate dated entry, completed immediately after the conference, detailed enough for another practitioner to continue care from it. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Every claimed conference and telehealth attendance lives or dies on the record. Section 81(1) of the Health Insurance Act 1973 sets the "adequate and contemporaneous records" standard, and regulation 6 of the PSR Scheme Regulations 2019 gives it four teeth: the record names the patient and carries a separate dated entry for each service; the entry holds enough clinical information to explain the service; it is completed at the time or as soon as practicable afterwards; and it is comprehensible enough that another practitioner could take over the patient's care in reliance on it. Two retention clocks then run: a copy of any document created as a condition of claiming an item must be kept for two years from the date of service, and the clinical record itself follows state and territory law, seven years for adults and until age 25 for minors in Victoria, NSW, and the ACT, no statutory period in South Australia, and longer wherever a complaint or litigation is foreseeable. The payer layer decides what else must appear. Case conference items are time-tiered, so the recorded duration selects the item; the conference must be clinically separate from preparing or reviewing a care plan, and the descriptors exclude same-occasion claims with a list of chronic condition management items, which since 1 July 2025 means the GP Chronic Condition Management Plan items 965 and 967 that replaced the GPMP and TCA architecture (the asynchronous coordination those plans generate belongs in a care coordination note ). For telehealth attendances, document the eligibility pathway: a face-to-face service with the practitioner or practice in the preceding 12 months, or the exemption relied on, which the November 2025 AskMBS advisory expects in the clinical notes. Scale explains the scrutiny: the Australian National Audit Office recorded more than 86.3 million COVID-era telehealth services to 16.1 million patients, worth $4.4 billion in benefits, and two statutory tripwires, the 80/20 rule and the 30/20 phone rule, refer prescribed patterns of service straight to PSR. The note format is a convention; the recorded facts are the requirement. ↑ Back to the template and downloads ## Common MBS case conference note errors auditors flag These claims are policed under Part VAA of the Health Insurance Act 1973 by the Professional Services Review, and the published outcomes are specific. The Director's update for June 2026 describes a practitioner who rendered the highest volume of item 758 nationally and acknowledged inappropriate practice across items including 743 and 758; the July 2025 update records a 12-month disqualification where the record did not always contain the documentation required for the billed service, across chronic disease and mental health items and their telehealth equivalents. The pattern is stable: PSR's 2021-22 annual report found most reviewed cases involved "inadequate clinical notes, failure to comply with the particular requirements of the MBS items", or prescribing concerns. The BastionGPT Clinical Advisory Board sees the same errors most often in MBS case conference note reviews: Times that cannot prove the tier. Start and end times missing, rounded, or reconstructed after the fact; the tier claimed becomes an assertion instead of a recorded fact. Consent explained but never recorded. The conversation happened, the agreement line does not exist, and the legislated recording activity is unmet. A team that fails the mix. Fewer than three synchronous participants, three medical practitioners, or members providing the same kind of care; the headcount alone proves nothing. Telehealth with no eligibility line. Video and phone attendances billed without the face-to-face pathway or the exemption documented, the gap the November 2025 AskMBS advisory closes. Conference work tangled with plan work. A case conference claimed as part of preparing or reviewing a care plan, which the descriptors expressly exclude; the two services must be clinically separate. ### MBS case conference notes in the US, Canada, and Australia Aspect United States Canada Australia Status No analogue: Medicare bundles the team conference codes 99366 to 99368 (status B), so no separate payment and no claim-substantiation regime Provincial payer policy: Ontario pays time-unit K-codes (K700 series, K121, K124) under OHIP with college and PHIPA record rules Delegated legislation: the GMST Regulations make the recording activities a condition of billing items 735 to 758 and the mental health case conferencing items Terminology Team conference, IDT or care planning conference Case conference (K-codes), care conference in long-term care Multidisciplinary case conference, MH case conference, telehealth attendance record What changes Conference work is documented inside care management and E/M services; the note supports medical necessity rather than a conference claim Time-unit billing with per-code caps, for example K701 (mental health) is restricted to psychiatry and capped at four per patient per physician per 12 months Recorded consent, day and start/end times, participant names, enumerated matters, a filed copy, organiser summaries, and a documented telehealth eligibility pathway Retention HIPAA keeps required documentation six years; patient charts follow state law, often longer College standards: commonly about ten years from last entry, or from the age of majority for minors Two years for MBS claim-condition documents; the chart runs seven years for adults and until age 25 for minors in Vic, NSW, and the ACT The meeting looks identical in all three countries; the paperwork does not. A US clinician writes the conference into care management documentation, an Ontario physician bills time units against college record rules, and an Australian biller works against an enumerated statutory list with an audit agency attached. This page is written to that strictest list. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on MBS case conference notes. Draft the conference record from a dictated recap, the agenda, or a transcript, with the legislated facts in their places: agreement, day and times, participants and kinds of care, and the enumerated matters. Check a note against the organiser or participant checklist and the telehealth eligibility line before the item is claimed, and flag the tier the recorded times actually support. Convert the record into the summaries the organiser must distribute: a plain-language version for the patient and carer, and a team summary ready to send to each participant. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there an official MBS case conference template? No. No regulation or MBS instrument prescribes a case conference form. What the law fixes is a list of facts that must be recorded when an item is billed: the patient's recorded agreement, the day with start and end times, the participant names, the matters discussed, a copy filed in the medical record, and the organiser's summaries. Any layout that captures those facts is compliant, including the general-purpose case conference note adapted with the MBS fields. Does the patient have to attend an MBS case conference? No. The patient or their carer must agree to the conference before it happens and the agreement must be recorded, but attendance is optional under both the primary care and the mental health case conferencing notes. Nothing requires the consent to be in writing either: the recorded agreement, not a signed form, is the legislated element. How do I prove the time tier for a case conference item? Record the day and the exact start and end times when they happen. The items are tiered at 15 to 20, 20 to 40, and 40 or more minutes, and the tier claimed must be provable from the record. AskMBS has confirmed no minute-by-minute template is required and that only time actually spent in the conference counts. Absent or reconstructed times are the most common failure point in Professional Services Review outcomes. How many participants does an MBS case conference need? At least three, present at the same time, each providing a different kind of care, and no more than two of them medical practitioners; the patient does not count toward the minimum. One more rule surprises organisers: organising includes participating, so the organiser cannot also claim a participation item for the same conference. Team members who bill nothing do not have to meet the item requirements at all. Does the telehealth 12-month rule apply to case conference items? No, it is an attendance rule. Video and phone attendance items generally require a face-to-face service with the practitioner or their practice in the preceding 12 months, or a listed exemption, and the November 2025 AskMBS advisory expects the exemption relied on to be documented in the clinical notes. Case conference items carry no such requirement, and the mental health family expressly waives any existing-relationship rule for providers other than the patient's usual practitioner. A video therapy session is a single-provider attendance with its own record; see the telehealth therapy note . Do I have to record which video platform I used? No MBS instrument requires the platform to be named in the note, and phone versus video is captured by the item number you claim rather than by a mandated modality field. What privacy law does require is that you are satisfied the software meets security and privacy requirements, a practice-level obligation rather than a per-note field. Many clinicians still note the modality and any technical interruptions; that is good practice, not law. Can a provisional psychologist claim the participation items? No. Provisional psychologists cannot obtain a Medicare provider number, so they cannot claim any MBS item, including the mental health case conference participation items 80176 to 80178. Those items belong to eligible allied mental health professionals, such as registered and clinical psychologists qualified to provide psychological therapy or focussed psychological strategies, and the wider family separates GP, psychiatrist, prescribed medical practitioner, and allied health roles. How long do MBS case conference records have to be kept? Two clocks run at once. A document created as a condition of claiming an MBS item must be kept for two years from the date of service. The clinical record it sits in follows state and territory law: seven years from last service for adults, and until age 25 for minors, in Victoria, NSW, and the ACT, while South Australia sets no statutory period. Where a complaint or litigation is foreseeable, keep the record until the matter is fully closed. How does BastionGPT help with MBS case conference documentation? Bring the conference however you captured it: a dictated recap, the agenda with notes, or a transcript. BastionGPT drafts the note with the legislated facts in their places, flags a missing agreement line, an unprovable time tier, or an unmet team mix before you claim, and produces the patient and team summaries the organiser must distribute. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Case Conference Note Multidisciplinary Team Meeting Note Care Coordination Note Collaborative Care (BHI/CoCM) Note Better Access Mental Health Treatment Plan Referral Telehealth Psychotherapy Note Medication Management Follow-Up Note Transitional Care Management (TCM) Note Return-to-Primary-Care Summary Canada Telepsychology & Virtual Care Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/good-faith-estimate All note templates # Good Faith Estimate (No Surprises Act): Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A good faith estimate (GFE) is the written notice of expected charges that the No Surprises Act requires US providers to give uninsured and self-pay clients, including therapy clients who pay out of pocket. It itemizes expected services, codes, and charges for up to 12 months of care. Required since January 1, 2022 under 45 CFR 149.610; most therapy GFEs run one to two pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The convening provider or their practice admin; therapists, psychologists, counselors, social workers, and MFTs are all covered Audience Uninsured and self-pay clients; a copy stays in the record for dispute reviewers and auditors Typical length 400 to 700 words · 10 to 20 minutes by hand (clinical team estimate) Format family Standardized financial disclosure (compare: fee agreement, superbill, ABN) When it's used At scheduling or on request for uninsured and self-pay clients; re-issued on fee or scope changes and at least every 12 months Standards context US federal law (45 CFR 149.610) since January 1, 2022; Canada and Australia rely on fee-disclosure conventions instead On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a good faith estimate? A good faith estimate (GFE) is the written notice of expected charges that United States federal law requires health care providers, including therapists, psychologists, counselors, social workers, and marriage and family therapists, to give clients who are uninsured or who choose not to bill their insurance. Unlike almost every other document in a therapy practice, this one was created by statute, not by a professional body: the No Surprises Act (Division BB of the Consolidated Appropriations Act, 2021) added section 2799B-6 to the Public Health Service Act, HHS implemented it in an interim final rule published October 7, 2021, and the requirement now lives at 45 CFR 149.610, applying to care scheduled or requested on or after January 1, 2022. You will also see it called a self-pay estimate, an NSA estimate, or a No Surprises Act estimate. It shares a name with, and has nothing to do with, the mortgage-industry good faith estimate that RESPA retired in 2015. The scope question trips more practices than the form does. Self-pay is broader than uninsured: an insured client who pays your full fee and files a superbill for their own out-of-network reimbursement is self-pay for GFE purposes and is entitled to an estimate, which makes the claim-support billing note the after-session partner of this before-care document. Enrollees in federal programs such as Medicare and Medicaid are excluded, and a Medicare beneficiary generally cannot be relabeled self-pay for a covered service unless the clinician has formally opted out with a private contract. The GFE is also not a fee agreement: your financial policy is a contract that governs the ongoing relationship, while the estimate is a forward-looking disclosure that says on its face it is not a contract. Keep both in the intake packet, and keep them separate. ## Who uses good faith estimates and when Every licensed mental health clinician who sees even one uninsured or self-pay client needs a GFE workflow: the rule reaches anyone practicing within the scope of a state license, solo or group, in office or telehealth-only, with no exemption for modality or caseload size. That is most of the profession. In the APA's 2024 Practitioner Pulse Survey, 34 percent of psychologists reported accepting no insurance at all, and the January 2022 request that HHS exempt routine behavioral health care, signed by 11 professional groups including the APA and NASW, was never granted. Issue the estimate when a self-pay client schedules an intake or a new course of care, whenever anyone asks what treatment will cost (the rule treats any cost discussion as a request), when fees or scope change, and again before a recurring estimate passes its 12-month limit. Coverage-facing paperwork runs in a different lane: requests to a plan go through a prior authorization letter , not an estimate. ## Good faith estimate structure: what goes in each section Client details and estimate date. Name and date of birth, the date you issued the estimate, and how it was delivered. The rule requires a written estimate the client can save and print, in the method the client chooses; oral explanation is allowed only on top of the written one. Pitfall: adding a client signature line as if the rule required it. No provision of 45 CFR 149.610 asks for a signature; log the delivery date and method instead, which is the receipt evidence a dispute reviewer actually uses. Primary service, in plain language. What the client is scheduling, described so a non-clinician can understand it, with the scheduled date if one exists. Pitfall: code-only descriptions. The regulation requires clear, understandable language; a bare 90837 with no words fails that requirement even though the code itself belongs elsewhere in the estimate. Itemized services, codes, and expected charges. Each item or service reasonably expected for the period of care, with expected service codes, applicable diagnosis codes, and the expected charge for each. Pitfall: listing rack rates when you have already agreed on a discount. The rule defines the expected charge as the cash rate reflecting any discounts; a sliding-scale practice lists the discounted rate where income is known, and CMS's documented fallback when it is not known yet is the undiscounted price. Provider identifiers. Name, NPI, and TIN of each provider expected to furnish care, plus the state and location where services will be delivered. Pitfall: omitting the NPI or TIN, one of the documented failure modes in practice. A telehealth-only clinician still lists the states involved; there is no telehealth exemption. Recurring-care scope statement. For ongoing therapy, one estimate can cover a course of care if it states the expected frequency, the expected number of sessions, and the per-session charge, for no more than 12 months. Pitfall: writing weekly, ongoing. An open-ended estimate is not compliant; state the scope, diarize the 12-month expiry, and issue a fresh estimate to continue. The required disclaimers. Printed text, not blanks: the estimate-only disclaimer, the separate-scheduling disclaimer, the additional-services disclaimer, the dispute-rights disclaimer with instructions, and the not-a-contract disclaimer. Pitfall: counting to twelve. Element (vii) of the regulation's content list is Reserved, so a complete GFE has eleven elements; templates promising all twelve are miscounting the rule they cite. Delivery and the paper trail. Who issued the estimate, when, and through what channel, with a copy kept in the record. Pitfall: treating the front-desk cost conversation as small talk. Any discussion of costs is a GFE request that starts a 3-business-day clock, and the answer must arrive in writing. ### Blank template (copy and adapt) GOOD FAITH ESTIMATE (uninsured / self-pay) This is an estimate of expected charges. It is not a bill and it is not a contract. Client name: ________________________________ DOB: _______________ Date issued: ___________ Delivered by: [ ] paper [ ] email [ ] portal Trigger: [ ] scheduling [ ] client request / cost inquiry (date: ____) PROVIDER Name and credentials: _____________________________________________ NPI: _______________ TIN: _______________ State(s) of care: _____ Practice and location: ____________________________________________ PRIMARY SERVICE (plain language) Description: ______________________________________________________ First / scheduled date: ___________ Setting: [ ] office [ ] video EXPECTED SERVICES AND CHARGES Period: _________ to _________ (a single recurring estimate may cover at most 12 months) Service (plain language) Code Diagnosis Charge _____________________________ _______ __________ $_________ _____________________________ _______ __________ $_________ _____________________________ _______ __________ $_________ Expected frequency: __________ Expected number of visits: _______ Total estimated charges for this period: $__________ SERVICES NEEDING SEPARATE SCHEDULING (estimated separately) ___________________________________________________________________ REQUIRED DISCLAIMERS (print with every estimate) This is only an estimate; actual items, services, and charges may differ. Additional recommended services are not reflected here. If you are billed $400 or more above this estimate, you may start the patient-provider dispute resolution process within 120 days; using it will not affect the quality of your care. This estimate is not a contract and does not require you to obtain services from the providers listed. Issued by: ______________________________ Date: _______________ Delivery logged (date / method): __________________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample good faith estimate Scenario: a solo therapist re-issues a recurring good faith estimate for an established self-pay client, because the prior estimate is approaching its 12-month limit and the practice fee rises in August. All details are fictional. Good Faith Estimate. Uninsured / self-pay · Client: R.S. · DOB: 03/14/1990 · Date issued: 07/06/2026 · Delivered: secure portal (client's chosen method), 07/06/2026 Provider: J. Alvarez, LCSW, Cedar Lane Counseling LLC. NPI 1234567893, TIN 87-6543210. Office: Portland, Oregon; video sessions delivered to the client in Oregon. Primary service: Weekly individual therapy sessions, 55 minutes, for generalized anxiety disorder (diagnosis code F41.1), in office or by video. Next scheduled session 07/13/2026. Expected services and charges, 07/13/2026 to 07/12/2027: Individual therapy session, service code 90837: 3 sessions in July 2026 at $170 each ($510), then up to 45 sessions at $180 each from 08/01/2026 ($8,100), when the practice's posted fee changes. Expected frequency: weekly. Expected total: up to 48 sessions. Total estimated charges for the period: $8,610. Services that may need separate scheduling: None anticipated. If any service is recommended later, it will be scheduled separately and receive its own estimate. Disclaimers: This is an estimate of reasonably expected charges. It is not a bill and not a contract, and it does not require you to obtain services from this provider; actual items, services, and charges may differ, and additional recommended services are not reflected here. If you are billed $400 or more above this estimate for this provider, you may start the patient-provider dispute resolution process within 120 calendar days of the bill; using it will not affect the quality of your care. Instructions: cms.gov/medical-bill-rights or ask this office for the dispute packet. Record note: Prior recurring estimate issued 08/04/2025 lapses 08/03/2026; this estimate replaces it ahead of the August fee change. Delivery logged: portal message, 07/06/2026. No client signature required by 45 CFR 149.610; receipt tracked by delivery log. Issued by J. Alvarez, LCSW, 07/06/2026. This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works All eleven content elements are present, and nothing extra pretends to be required. The five disclaimers are printed in full, and there is no signature line, because element (vii) of the rule's list is Reserved and no element asks for a signature. The recurring scope is explicit. Frequency, expected session count, per-session charge, and a period that stays inside the 12-month cap, so open-ended weekly therapy is covered without an open-ended estimate. The August fee change is inside the numbers. The new rate appears weeks before the first affected session, far ahead of the 1-business-day deadline, so no bill should ever run $400 past this page. Delivery is logged by date, method, and the client's chosen format. That is the receipt evidence the rule contemplates, with no signature chase. Itemized codes and charges produce a clean aggregate. A dispute reviewer can compare billed against estimated per provider, and the client sees the year's real cost, $8,610, not just a per-session number. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The estimate is part of the client's record. The rule's own retention hook is retrieval, not destruction: on request, you must be able to produce any GFE you issued in the last 6 years, a duty that parallels HIPAA's 6-year documentation rule and sits underneath longer state chart-retention laws, commonly 7 to 10 years and longer for minors. Client requests for copies run through your normal release of information workflow. Alongside the copies, keep the delivery evidence (date and method for each estimate) and maintain the availability notice the rule requires: a prominent, searchable notice on your website and in the office, plus an oral mention when scheduling or whenever costs come up. Federal-program enrollees are outside the GFE lane entirely: Medicare, Medicaid, TRICARE, VA, and FEHB clients do not receive one, and a Medicare beneficiary cannot simply be relabeled self-pay for a covered service; that generally takes a formal opt-out and private contract, a path psychologists have had for years and marriage and family therapists and mental health counselors gained with Part B recognition in 2024. On enforcement, hold both truths. No published civil monetary penalty against a therapist for a GFE violation has been identified; the co-provider element has sat under CMS enforcement discretion since December 2, 2022; and the insured-client version with its Advanced Explanation of Benefits remains unenforced pending rulemaking, with the most recent federal agenda pointing at a proposed rule in 2026. But the authority is real: up to $10,000 per violation, a consumer complaint process that has accepted GFE uploads since January 2024, and patient-provider dispute resolution for any bill that runs $400 or more above the estimate, aggregated per provider, inside a 120-day window for a $25 fee. The GFE is never adjudicated as a claim, so no CARC or RARC code will ever flag a bad one; failure surfaces as a dispute or a complaint, not a denial. A good-faith safe harbor covers errors corrected as soon as practicable. For most documents on this site the format is a convention and the content is the requirement; here the content list itself is the law, and only the wording is yours. ## Common good faith estimate errors auditors flag There is no public GFE audit dataset, and that absence is itself a finding: HHS projected more than 26,000 patient-provider disputes per year before launch, yet no PPDR case count or outcome has ever been published by CMS, GAO, or CRS, so the observed record for therapists is one of complaints and corrections rather than penalties. Do not borrow numbers from the insured side either: the nearly 490,000 disputes GAO counted in the No Surprises Act's first 15 months belong to the insurer-facing IDR process and say nothing about self-pay estimates. What is documented is how estimates fail: never issued, missing disclaimers, missing NPI or TIN, cost inquiries not treated as requests, and bills that outgrow a stale estimate. The BastionGPT Clinical Advisory Board sees the same errors most often in good faith estimate reviews: Waiting for the client to say the word estimate. Any discussion of costs is legally a request that starts the 3-business-day clock; a front-desk question about what weekly sessions cost is a trigger, not small talk. Running the GFE as an annual form. The rule has no calendar trigger. New scheduling, any change in fees or scope (new estimate at least 1 business day before the affected session), and the 12-month ceiling on a recurring estimate are the real triggers; annually merely approximates the last one. Skipping superbill clients. A client who pays in full and seeks their own reimbursement is self-pay under the rule and entitled to an estimate; equating insured with exempt is the quiet compliance gap in otherwise careful practices. Chasing signatures instead of logging delivery. Nothing in the rule requires a client signature, and a signed estimate that went out late is still late. The defensible artifact is the delivery log: date, method, and the client's chosen format. Letting the bill outgrow the estimate. Raise a fee, add a service, or extend care without re-issuing and every dollar past the old numbers walks toward the $400 aggregate dispute threshold; update the estimate before the change, not after the bill. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on good faith estimates. Draft a complete GFE from your fee schedule and the expected course of care, with the plain-language service description, the recurring scope statement, and all five disclaimers in place. Pull services, codes, and charges out of a treatment plan, superbill, or fee schedule and turn them into clean estimate line items with a period total. Check a draft against the eleven required elements before it goes out: identifiers, description, codes and charges, NPI and TIN, states of care, the 12-month scope, and the disclaimers. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does my client have to sign the good faith estimate? No. No provision of 45 CFR 149.610 requires a client signature or a provider signature; a signature line is a practice or software convention for documenting receipt, not a regulatory element. What protects you in a dispute is proof of timely delivery, so log the date, the method, and the client's chosen format for every estimate you issue, including re-issues after a fee change. If your intake software insists on a signature field, treat it as optional and never hold care to it. Do I have to give a new good faith estimate every year? The rule contains no annual or calendar-year trigger; the advice to re-issue annually is a simplification that happens to approximate one real rule. The actual triggers are: each new scheduling or request; any change in expected charges, services, frequency, or providers, which requires a new estimate no later than 1 business day before the affected service; and the 12-month maximum life of a recurring-services estimate, which must be replaced to continue care beyond it. Diarize each recurring estimate's expiry and your fee-change dates, and the annual myth takes care of itself. How do I write a good faith estimate for open-ended weekly therapy? Use the recurring-services provision: one estimate may cover a course of care for up to 12 months if it states the expected scope. Give the expected frequency, the expected number of sessions, and the charge per session, then total it. A statement like weekly 55-minute sessions from July 13, 2026 through July 12, 2027, up to 48 sessions at $180 per session, satisfies the scope requirement as CMS and the professional associations describe it. To continue past the period, issue a new estimate before the old one lapses; to change fees mid-period, issue the new estimate at least 1 business day before the first session at the new fee. My client has insurance but pays me directly and files superbills. Do they need a GFE? Yes. The rule defines self-pay to include insured clients who choose not to have claims submitted to their plan, which is exactly the superbill arrangement: the client pays your full fee and seeks their own out-of-network reimbursement. They are entitled to a good faith estimate before care, and the two documents then bracket each session: the GFE is pre-service and never goes to a payer, while the claim-support billing note and superbill are post-service records the client submits. Practices that equate having insurance with not needing a GFE miss this group, and it is often the largest self-pay group they have. Do insured or Medicare clients ever get a good faith estimate? Not the self-pay version this page covers. When a claim will go to a plan, the statute contemplates a plan-facing estimate feeding an Advanced Explanation of Benefits, but those provisions are unenforced pending rulemaking, with the most recent federal agenda pointing at a proposed rule in 2026. Enrollees in federal programs (Medicare, Medicaid, TRICARE, VA care, FEHB plans) are excluded from the GFE entirely; for Original Medicare the analogous form is the ABN, a different instrument that does require a signature and an option choice. A Medicare beneficiary also cannot simply be relabeled self-pay for a covered service: that takes enrollment or a formal opt-out with a private contract, a path open to psychologists for years and to marriage and family therapists and mental health counselors since they became Part B providers in 2024. Requests to a plan for coverage run through a prior authorization letter , a different lane entirely. What happens if my final bill is higher than the estimate? If billed charges for you as a provider run $400 or more above your estimated total, aggregated across the estimate rather than per line item, the client may start the patient-provider dispute resolution process within 120 calendar days of the bill for a $25 fee, and a federally contracted reviewer decides what is owed. Below $400 there is no penalty mechanism at all, and the rule includes a good-faith safe harbor for errors corrected as soon as practicable. The practical defense is boring: update the estimate whenever fees or scope change, at least 1 business day before the affected session, so the bill never outruns the paper. How does a good faith estimate work with sliding scale or free care? The rule defines the expected charge as the cash rate the client will actually be charged, reflecting any discounts, so a sliding-scale practice lists the discounted rate whenever income and family size are known. When they are not known yet, CMS's FAQ guidance from December 2022 gives the compliance path: list the undiscounted price, and attach your sliding-fee schedule so the client can see what may apply. Free and pro bono care still requires an estimate, because a $0 expectation is not a $0 guarantee; CMS created an abbreviated GFE format for providers who do not expect to bill at all. I practice in Canada or Australia. Do I need a good faith estimate? No. The GFE is a US federal statute with no equivalent in either country, so do not import the US form as if it were required. In Canada, fee disclosure is a professional-college obligation, convention backed by college discipline rather than statute: college standards and guidelines address financial arrangements and billing, and none requires an itemized, coded written estimate. In Australia, the analogue is informed financial consent, and the bar rose on 1 December 2025, when the Psychology Board of Australia's Code of Conduct became the enforceable standard and embedded financial clarity (fees, rebates, third-party involvement, cancellation terms) in its consent expectations. In both countries the discipline is the same even though the statute is absent: document fees and payment terms in writing before treatment starts. Can BastionGPT draft my good faith estimates? Yes. Give it your fee schedule, the expected frequency and duration of care, and the codes you bill, and it drafts the estimate with the plain-language service description, the recurring scope statement, and the five disclaimers in place, then checks the result against the eleven required elements before you send it. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Prior Authorization Support Letter Release of Information (ROI) Authorization Intake Note Psychotherapy Notes Authorization Financial Policy & Fee Agreement Appointment No-Show & Outreach Note Advance Beneficiary Notice (ABN) Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/communication-log All note templates # Communication Log & Portal Message Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A communication log or portal message note documents a client contact that happens outside a scheduled session: a patient portal message, phone call, text, or secure email. Therapists, psychologists, and practice teams use it to record what was communicated, their clinical read of it, and the follow-up. Most entries run 50 to 200 words; a billed e-visit entry runs longer. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, psychiatrists, and practice staff triaging messages under a supervising clinician Audience Treating clinician and practice team; payers and auditors when a contact is billed; clients can read messages through access rights Typical length 50 to 200 words · 2 to 5 minutes by hand (clinical team estimate) Format family Brief operational chart entry (compare: care coordination note, billing note, crisis note) When it's used After any material out-of-session contact: portal messages, phone calls, texts, secure email Standards context A professional convention, not a named legal form; messages sit in the designated record set, and billed digital services carry payer-specific rules On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a communication log? A communication log, also called a portal message note, phone note, or telephone encounter, is the chart entry for a client contact that happens outside a scheduled session: a patient portal message, a phone call, a text, or a secure email. It records that the contact happened, what was clinically material in it, and how you handled it. Unlike the SOAP note , which traces to Lawrence Weed's problem-oriented medical record, this document type has no inventor, no founding publication, and no standards body. It is the meeting point of two separate streams: billing documentation expectations that arrived when the CPT 2020 code set created e-visit codes for patient-initiated digital exchanges, and licensing-board record-keeping standards that require material between-session contacts to be documented. Most of its names come from EHR vendors: telephone encounter and in-basket message from Epic, MyChart message, secure message note, virtual check-in note. Two boundaries keep the log honest. It is not a progress note : the progress note documents the session, the communication log documents the space between sessions, and a billed e-visit sits in between, a communication entry that must also carry decision-making and time. And no law requires a separate note for every call or text; what United States law actually says is narrower and stronger. Portal messages and documented calls that inform decisions about a client sit inside the HIPAA designated record set (45 CFR 164.501), which makes them reachable by the client's right of access and, as electronic health information, subject to the information blocking rule. The famous exclusion is narrower than practitioners hope: psychotherapy notes means separately kept process notes, not routine messages about symptoms, scheduling, or medication, so the working assumption is that anything you type to or about a client can one day be read by them. ## Who uses communication logs and when Any practice with a portal, a phone line, or a texting policy produces these entries, and the volume keeps climbing: an Epic Cosmos analysis of roughly 8 billion encounters found patient-authored portal messages rose 153 percent from 2020 to 2025, to 2.5 messages per active patient per year, while office visits rose 17 percent, so messaging arrives on top of session work, not instead of it. Telehealth-forward and asynchronous practices lean on the log hardest, and front-desk or clinical staff often write the first entry when they triage the inbox under a supervising clinician. The boundaries matter more than the format: a provider-to-provider exchange belongs in a care coordination note , a contact that becomes active risk work belongs in a crisis note , and an exchange you intend to bill has to double as a claim-support record . ## Communication log structure: what goes in each section Date, time, mode, and initiator. When the contact happened, through what channel (portal, phone, text, secure email), and who started it. Patient initiation is a billing prerequisite for e-visit and virtual check-in codes, so record it even when you have no plan to bill. Pitfall: an entry with no initiator. A log that is silent on who started the exchange can never support a billed digital service later, and it hides the difference between outreach and response. Participants and identity. Who was on each end, and how you knew. A portal message arrives through an authenticated login; a phone call does not, so note how you confirmed you were talking to your client, and name any third party who relayed information. Pitfall: a relative's call logged as the client's own report. When a parent or partner calls about the client, the entry is a collateral contact and should say so, which also keeps later release decisions clean. Substance of the contact. The clinically material content: what was reported, asked, or disclosed. Quote sparingly and summarize the rest. If the message thread stays in the EHR, no regulation requires re-typing it; the thread can itself be the record, and the log entry adds the clinical handling. Pitfall: both extremes. Re-transcribing a thread that is already retained wastes hours, and the bare 'spoke with client' fails the adequate-records standard licensing boards apply. Clinical assessment and risk. What you made of it. If the contact conveyed symptoms, side effects, or risk, the entry should show your judgment: what changed, what did not, and whether screening or a safety response was indicated. Pitfall: advice with no reasoning. This section also decides billing: an exchange with no medical decision-making cannot be billed as an e-visit, however long it took. Action taken and follow-up. Advice given, changes made, scheduling, referral, escalation, and who does what by when. Close the loop the message opened, and note any resulting appointment. Pitfall: 'will follow up' with no owner or date. The resulting visit matters too: an exchange that leads to a visit within 24 hours, or follows a related one within 7 days, is bundled into that visit rather than separately billable. Billing status, time, and consent. Most entries end with 'not billed', and writing that is protective, not risky. When you do bill, the entry has to carry the claim: the code, time spent, cumulative time across the rolling 7-day window that selects among the e-visit codes, and consent for digital services, which Medicare lets you obtain verbally once a year. Pitfall: reaching for deleted codes. The physician telephone codes 99441 to 99443 ended January 1, 2025; the nonphysician telephone codes 98966 to 98968 remain active, and Medicare prices the nonphysician e-visit codes 98970 to 98972 at zero. Signature and authentication. Author name, credentials, and date, with cosignature where supervision rules require it. Staff can and should document their own triage under their own logins. Pitfall: a staff-handled exchange billed as a clinician service. Billed e-visits and virtual check-ins require the billing practitioner's own interaction with the client; a message answered entirely by staff is documentation, not a billable service. ### Blank template (copy and adapt) COMMUNICATION LOG / PORTAL MESSAGE NOTE (out-of-session contact; a session gets a progress note instead) Client: [initials] DOB: __________ Date: ____________ Time: ________ Mode: [ ] portal [ ] phone [ ] text [ ] email Initiated by: [ ] client [ ] clinician [ ] staff [ ] other: _____ Participants / identity confirmed by: _____________________________ SUBSTANCE OF CONTACT (summarize; the retained thread can be the record) ___________________________________________________________________ ___________________________________________________________________ CLINICAL ASSESSMENT / RISK (required if symptoms or risk were conveyed) ___________________________________________________________________ ___________________________________________________________________ ACTION TAKEN / FOLLOW-UP (advice, scheduling, referral; owner and date) ___________________________________________________________________ ___________________________________________________________________ BILLING STATUS [ ] Not billed (documentation only) [ ] Billed: code ________ Time this contact: ______ min Cumulative time, rolling 7 days (e-visits): ______ min Digital-service consent on file: [ ] yes Date: ________ Related visit within prior 7 days / next 24 hours: [ ] no [ ] yes Author (name / credentials): ____________________ Date: ________ Supervisor cosign (if required): ________________ Date: ________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample portal message note Scenario: an established psychotherapy client sends a patient portal message between weekly sessions; the psychologist replies the same day and logs the contact without billing it. All details are fictional. Communication log. Portal message · Client: D.K., 41 · Date: 07/21/2026 · Time: 2:15 PM · Initiated by: client (authenticated portal account) Substance: Client messaged at 11:52 AM: "Rough week at work and the breathing exercise isn't cutting it. Can we get to the sleep piece sooner?" Full thread retained in the EHR message center; no attachments. Assessment / risk: Message reports increased work stress and early-week insomnia, consistent with the existing generalized anxiety presentation; no new symptoms and no medication questions. No risk content in the message; client denied ideation at the 07/17/2026 session, and nothing here changes that screening. No urgent response indicated. Action taken / follow-up: Replied through the portal at 2:15 PM: validated the stress, coached one adjustment to the paced-breathing cadence, and moved the sleep module up to the next session. Offered an earlier appointment; client confirmed Thursday 07/23/2026 at 3:00 PM, replacing the standing Friday slot. Agenda flag added to the appointment. Billing status: Not billed; brief supportive reply inside the existing treatment plan (6 minutes; the practice does not bill portal messages). Digital-communication consent on file 01/12/2026. No related visit inside the billing windows either way; noted for completeness. Authentication: L. Moreno, PsyD. Entered 07/21/2026, 2:24 PM. No cosign required. This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works Date, time, mode, and initiator are all present. The entry could support a billed digital service if one ever applied, even though this contact was not billed. The thread is pointed to, not re-typed. The retained portal thread is the record of what was said; the log adds the clinical reading of it, which is the part no thread contains. Risk is addressed even though the message contained none. The record shows screening rather than silence, tied back to the last session's denial. The action line closes the loop with an owner, a date, and a confirmed appointment. A reviewer can see the message was handled, not just received. The billing decision is explicit either way. Not billed, with the reason and the time, costs nothing and defuses the myth that unbilled documentation creates exposure. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Portal messages and documented calls are not a private side channel. Whenever they are used to make decisions about a client, they sit inside the HIPAA designated record set (45 CFR 164.501), the client's right of access reaches them (45 CFR 164.524), and their content is electronic health information under the information blocking rule (45 CFR Part 171), where unreasonable withholding carries consequences. The psychotherapy-notes carve-out will not rescue a routine message: that exclusion covers separately kept process notes, not scheduling, symptom, or medication exchanges (the psychotherapy notes authorization page covers the narrow category that does get special handling). Texts that carry clinical content are in the record set regardless of whose device they sit on, which is a Security Rule problem worth solving with policy before it becomes one. On retention, HIPAA's famous 6-year rule applies to compliance documentation such as policies and risk analyses, not to clinical records; how long you keep messages and logs is set by state law in the US and by provincial, state, and territory law in Canada and Australia, where common answers run 7 to 10 years and longer for minors. When a between-visit exchange is billed, the log entry doubles as the claim-support record, and the payer rules are specific: a patient-initiated exchange with an established client, medical decision-making in your reply, time accumulated across a rolling 7-day window, no related visit in the 7 days before, none generated in the 24 hours after, and digital-service consent documented annually. Denials arrive as bundling (CO-97: the payment sits inside another service) or as non-coverage (CO-96 with remark N448 for codes a fee schedule never pays, which is where Medicare puts the nonphysician e-visit codes). The billed message is the exception, not the rule: UCSF billed under 2 percent of patient messages after turning message billing on, and other large systems report under 1 percent. For everything else the format is a convention and the content is the requirement: date, initiator, substance, your handling, and follow-up, at whatever length the contact deserves. The claim-support billing note covers the billed case in detail, and a scheduled video session is not a between-visit contact at all: document it as a telehealth session note . ## Common communication log errors auditors flag Between-visit services are audited as telehealth, and the findings are documentation findings. In one HHS Office of Inspector General national sample, 31 of 100 telehealth claims failed Medicare requirements, an estimated $3.7 million in avoidable spending for the audit period. The Massachusetts state auditor found MassHealth paid at least $91.8 million for telehealth behavioral health services that were not properly documented, with every one of the 47 sampled claims carrying at least one documentation defect. And a 2024 OIG review of $1.4 billion in telehealth E/M payments listed the failures reviewers always list: no record at all, illegible entries, and detail too thin to support the service billed. The BastionGPT Clinical Advisory Board sees the same errors most often in communication log reviews: Billing telephone codes that no longer exist. The physician telephone codes 99441 to 99443 were deleted January 1, 2025, and claims on them now reject; the nonphysician telephone codes 98966 to 98968 remain active, and of the 2025 telemedicine family 98000 to 98016, Medicare recognizes only 98016, the virtual check-in. Medicare e-visit claims from clinicians who cannot report E/M. The nonphysician codes 98970 to 98972 carry an invalid status on the Medicare fee schedule and pay zero, so counselor and social worker claims come back as CO-96 with remark N448, while some Medicaid programs and commercial payers do pay them. Check the payer before promising anyone reimbursement. Billing inside the bundling windows. An e-visit that follows a related visit within 7 days, or leads to one within 24 hours, is folded into that visit's payment; the CO-97 denial means another service already paid for this work. Entries with no initiator and no time. Patient initiation and cumulative 7-day time are the two facts that decide whether a digital exchange is billable at all, and they are the two most often missing from message documentation. All-or-nothing texting habits. Writing a note for every scheduling text wastes clinical hours, and skipping the message that carried symptoms or risk violates the adequate-records standards boards enforce. The standard is materiality: document the contacts that matter clinically, and let the appointment system hold the rest. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on communication logs and portal message notes. Draft a complete log entry from a pasted portal thread, a phone call summary, or a few quick bullets, with the date, mode, initiator, and follow-up in place. Pull the clinically material content out of a long message thread and condense it into a defensible entry that points back to the retained thread. Check an entry before you sign for the facts that decide billability and defensibility: initiator, time, decision-making, risk handling, and the follow-up owner. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to write a note every time a client texts or messages me? No law in the US, Canada, or Australia requires a note for every contact. What licensing boards require is adequate records, and adequacy runs on materiality: the College of Psychologists of Ontario expects the date of every material consultation to be recorded, and the Psychology Board of Australia's code of conduct, in force since December 2025, requires accurate, factual records that report relevant details. A scheduling text needs no clinical note. A message that carries symptoms, risk, or a treatment decision does. Write your messaging boundaries into the informed-consent paperwork and the judgment calls get easier. Can I bill for time spent answering patient portal messages? Sometimes, inside strict rules. Clinicians who can report E/M services, including clinical psychologists, can bill the e-visit codes 99421 to 99423 when an established client initiates the exchange, the reply involves medical decision-making, and time accumulates across a rolling 7-day window, with no related visit in the prior 7 days and none generated within 24 hours after. Purely administrative replies never qualify. The scale is worth knowing: after UCSF turned on message billing, billed e-visits rose nearly fivefold yet stayed under 2 percent of messages received. In Canada, psychologists have no public payer for this, and in Australia there is no MBS item for asynchronous psychology messaging. The claim-support billing note covers what a billed entry must carry. Do portal messages count as clinical records I have to release? In the US, yes. Messages and call records used to make decisions about a client are part of the HIPAA designated record set under 45 CFR 164.501, the right of access in 45 CFR 164.524 reaches them, and their content is electronic health information under the information blocking rule. The psychotherapy-notes exclusion is narrow: separately kept process notes analyzing a session, not routine messages about symptoms, scheduling, or medication. The psychotherapy notes authorization page covers the one category that does get special handling. A message you would not want read back is a message to handle in session instead. Do I have to copy the whole message thread into my note? No regulation in the US, Canada, or Australia says a thread must be re-transcribed. If the portal thread is retained in the EHR, the thread itself is part of the record, and the better log entry is short: point to the thread, summarize what mattered clinically, and record your assessment and action. Re-typing conversations adds hours and no defensibility. The exception is content that lives outside the chart, such as texts on a personal phone: get the clinical substance into the record, because it belongs to the designated record set wherever it sits. How do I document a crisis message that arrives between sessions? Document the material content, your risk assessment, and the action you took, at the time you took it: what the client reported, what you evaluated, who you contacted, and the plan you landed on. If the contact turned into active risk work, a crisis note carries it better than a brief log entry, and any safety plan you build or revise is its own document. Then close the loop in your communication policy: clients should know from intake that messages are not monitored around the clock and where to turn urgently instead. Which phone and e-visit codes are still billable in 2026? The physician telephone codes 99441 to 99443 were deleted effective January 1, 2025, and many pages still list them as current. Still active: the nonphysician telephone codes 98966 to 98968, the e-visit codes 99421 to 99423 for clinicians who can report E/M, and the nonphysician e-visit codes 98970 to 98972, which Medicare prices at zero but some Medicaid programs and commercial plans pay. Of the telemedicine code family 98000 to 98016 created for 2025, Medicare recognizes only 98016, the virtual check-in that replaced G2012. Codes move; verify against the current year's fee schedule before promising anyone reimbursement. When should I write a care coordination note instead of a communication log? Follow who is on the other end. Client or family on the other end: communication log. Another provider or agency on the other end: care coordination note , which documents referrals, handoffs, and what was shared under which consent. Ongoing between-visit care management with monthly time tracking belongs in a chronic care management note , and a contact that becomes risk work belongs in a crisis note. The communication log is the patient-facing member of the family. How long do I need to keep portal messages and communication logs? Longer than HIPAA says, because HIPAA does not say. Its 6-year rule covers compliance documentation such as policies and risk analyses, not clinical records or message content. Retention comes from state, provincial, and territory law: US states vary widely, Canadian provinces commonly expect 10 years from last contact or from a minor reaching the age of majority, and Victoria, New South Wales, and the ACT require 7 years from last service or until a client who was a minor turns 25. If a message informed care, it keeps for as long as the chart it belongs to. Can BastionGPT draft my communication log entries? Yes. Paste a portal thread, describe a phone call, or dictate what happened, and it drafts the entry with the date, mode, initiator, substance, assessment, and follow-up in place, then checks it for the details that decide billability when you plan to bill. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Prior Authorization Support Letter Care Coordination Note Telehealth Psychotherapy Note Crisis Psychotherapy Note Chronic Care Management (CCM) Note Appointment No-Show & Outreach Note Patient Instructions & After-Visit Summary Records Request & Chart Copy Packet Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/utilization-review-summary All note templates # Utilization Review Summary: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A utilization review summary is the clinical document a treating clinician or facility submits to a payer to justify continuing an authorized course of treatment or level of care. Payers call the process concurrent or continued-stay review. It compiles current symptoms, objective measures, risk status, treatment response, and the barriers to step-down. Most run 300 to 600 words or fill a payer portal form. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it UR staff or care coordinators compile it; a licensed treating clinician owns and signs the clinical assertions Audience Payer care managers and physician reviewers; the chart and auditors when a copy is filed Typical length 300 to 600 words · 15 to 25 minutes by hand (clinical team estimate) Format family Structured medical-necessity summary (compare: prior authorization request, appeal letter) When it's used During an authorized episode of care, at each concurrent or continued-stay review interval the payer sets Standards context Payer policy, not statute, defines the document; federal and state law set the payer's deadlines and denial duties On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a utilization review summary? A utilization review summary is the document a treating clinician or facility submits to a payer to support continuing an already-authorized course of treatment or level of care. Utilization review itself entered US law with the 1965 Medicare statute, which made a hospital utilization review plan a condition of participation (now 42 CFR 482.30), ran through the Professional Standards Review Organizations created in 1972 and the peer review organizations that replaced them under TEFRA in 1982, and moved into private managed care in the late 1980s and 1990s, with URAC incorporated in 1990 to write the first national utilization review standards. The clinician-facing document this page covers answers to many names: concurrent review, continued stay review, clinical review, level of care review, treatment report, and, in outpatient behavioral health, the outpatient treatment report or request (OTR). Two boundaries do the most work. First, the summary is the input, not the decision: the clinician supplies clinical information, and the determination belongs to the payer, whose adverse medical-necessity decisions generally must be made by a physician reviewer. Second, no US federal law requires an outpatient behavioral health clinician to produce a standalone utilization review summary at all. The utilization review regulations that exist (42 CFR 482.30 and 42 CFR Part 456) bind hospital and Medicaid facility review committees, not treating therapists, so in outpatient care the obligation to submit clinical information is contractual, defined by the payer's utilization management policy. That makes this document the mid-treatment sibling of the prior authorization letter : authorization gets care started, and the utilization review summary keeps it authorized. ## Who uses utilization review summaries and when Utilization review staff, care coordinators, and treating clinicians in inpatient psychiatric units, residential and substance use programs, partial hospitalization, and intensive outpatient programs produce them on a payer-set cadence, often every few days at the higher levels of care. Outpatient therapists meet the same document as the outpatient treatment report: routine weekly psychotherapy rarely needs prior authorization, but some plans begin concurrent review once a client passes a session threshold, commonly somewhere in the 10 to 20 visit range. Use this format whenever a payer asks you to justify continuing care that is already authorized. Before care starts, that request is a prior authorization; after a denial, it becomes a peer-to-peer conversation or a written appeal; and the internal document that tracks the same question for clinical rather than payer purposes is the treatment plan review . ## Utilization review summary structure: what goes in each section Authorization and review header. Member identifiers, plan, authorization number, current level of care, the authorization period, units used against units approved, and the review type. Pitfall: mismatches between this header and the authorization on file. Wrong dates, codes, or units produce administrative denials such as CO-197, and no amount of clinical writing further down the page fixes a clerical mismatch up here. Diagnosis and current presentation. Working diagnoses with ICD-10 codes, today's symptoms with severity and frequency, functional impairment, and objective measures with dates and direction of change. Pitfall: restating the admission picture. Reviewers read this section against your last submission, and payer guidance on outpatient treatment requests explicitly returns forms that change the dates without changing the clinical information. Current risk status. Suicidality, self-harm, danger to others, and withdrawal or medical risk, dated today and compared with the last review. Pitfall: a risk section that never changes. Static copy-forward risk language is an audit red flag, and it quietly argues the client is stable enough to step down. Treatment response since last review. Interventions delivered, attendance, measurable response or lack of response, medications and adherence, and what you changed when response stalled. Pitfall: narrating attendance instead of response. Nine of nine days attended proves the program happened; a falling symptom score and a widening window of function prove it is working and still needed. Continued-stay rationale tied to criteria. Name the criteria set the payer applies (ASAM for substance use, LOCUS or CALOCUS-CASII for mental health levels of care, InterQual or MCG on the general medical side), the specific criterion or dimension met, and why a lower level of care is not appropriate today. Pitfall: arguing that the client would benefit from more care. Benefit is the wrong test; the reviewer is deciding whether the member can be safely and adequately treated at a lower level right now, so answer exactly that question. Step-down plan and barriers. The discharge or step-down plan, the specific barriers preventing it today, and the measurable thresholds that will trigger it. Pitfall: omitting this section. With no step-down plan on the page, the summary reads as an open-ended stay, which is precisely what level-of-care criteria are built to end. Request and sign-off. The days, sessions, or dates requested, who compiled the summary, and the licensed treating clinician's attestation of the clinical content, with a supervisor co-signature where a pre-licensed clinician drafted it. Pitfall: a vague or overlapping request. Units without dates, or dates that overlap the current authorization, hand the plan an administrative reason to return the request unread. ### Blank template (copy and adapt) UTILIZATION REVIEW SUMMARY (concurrent / continued-stay review) The format is a convention; the payer's criteria set defines the content. Confirm the plan's form, portal fields, and review cadence. AUTHORIZATION AND REVIEW Member: ____________________ ID: ____________ DOB: _____________ Payer / plan: ______________________ Auth #: ____________________ Level of care: [ ] OP [ ] IOP [ ] PHP [ ] Residential [ ] Inpatient Auth period: _________ to _________ Units used / approved: __ / __ Review date: _________ Type: [ ] scheduled [ ] payer request DIAGNOSIS AND CURRENT PRESENTATION Diagnoses (ICD-10): _______________________________________________ Symptoms today, severity and frequency: ___________________________ ___________________________________________________________________ Objective measures (instrument, score, date, change): _____________ Functional impairment: ____________________________________________ CURRENT RISK STATUS (dated; compare with last review) Suicidality / self-harm: __________________________________________ Danger to others: _____________ Withdrawal / medical: ____________ Change since last review: _________________________________________ TREATMENT RESPONSE SINCE LAST REVIEW Interventions and attendance: _____________________________________ Response or lack of response: _____________________________________ Medications and adherence: ________________________________________ CONTINUED-STAY RATIONALE Criteria set and edition: _________________________________________ Criterion / dimension met: ________________________________________ Why a lower level is not appropriate today: _______________________ ___________________________________________________________________ STEP-DOWN PLAN AND BARRIERS Plan and target level: ____________________________________________ Barriers today: ___________________________________________________ Thresholds that trigger step-down: ________________________________ REQUEST AND SIGN-OFF Days / sessions requested: ___________ Dates: ____________________ Compiled by: ______________________ Date: ________________________ Licensed clinician (clinical content): ____________________________ Credential: ____________ Signature / date: _______________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample utilization review summary Scenario: the clinical team at an intensive outpatient program submits a scheduled concurrent review to continue an adult client in IOP care beyond the initial authorization. All details are fictional. Utilization Review Summary (concurrent review). Member: M.T. · DOB: 04/12/1988 · Plan: Harborline Health (commercial) · Auth #: HL-26-40917 · Level of care: IOP · Auth period: 07/01/2026 to 07/24/2026, 9 of 12 days used · Review date: 07/21/2026, scheduled concurrent review Diagnosis and current presentation: Major depressive disorder, recurrent, moderate (F33.1); generalized anxiety disorder (F41.1). Depressed mood most of the day, early-morning waking 4 to 5 nights per week, impaired concentration, and low motivation. PHQ-9 14 today (07/21) from 19 at admission (07/01); GAD-7 11 from 15. Working half days only; missed 3 of the last 10 scheduled workdays. LOCUS composite 17 today from 19 at admission, scored 07/21. Current risk status (07/21): Passive thoughts of death without plan, intent, or preparatory behavior, unchanged from the 07/14 review. No self-harm, no aggression, no withdrawal or acute medical risk. Safety plan reviewed 07/16; client named two reachable supports and used one this week. Treatment response since last review: Attended 9 of 9 IOP days (3 days per week: group CBT, behavioral activation, weekly individual session, psychiatric consultation). Behavioral activation homework completed 5 of 6 assigned days this week, up from 2 of 6 at the last review. Sertraline increased from 50 mg to 100 mg on 07/08 after partial response; adherent by report and refill record, initial nausea resolved. Sleep extended from about 4 to about 5.5 hours. Continued-stay rationale: Criteria set per the plan's provider manual: LOCUS. Composite 17 remains in the Level 3 range (high intensity community based services), driven by functional status and recovery environment dimensions. A step to weekly outpatient therapy is not appropriate today: symptom severity still prevents full-day work function, the 07/08 medication increase needs monitored titration through week 4, and gains have not yet held on non-program days. Step-down plan and barriers: Target: weekly individual therapy with the current clinician plus psychiatry every 4 weeks. Thresholds: PHQ-9 below 10 sustained for 2 weeks, full-day work tolerance, stable sleep. Barriers today: residual early-morning waking, impaired concentration, and the active titration. Request and sign-off: 10 additional IOP days, 3 days per week, 07/27/2026 to 08/14/2026. Compiled by R. Okafor, care coordinator, 07/21/2026. Clinical content reviewed and attested by D. Whitfield, LCSW, treating clinician, 07/21/2026. This sample is fictional and for educational purposes. It does not describe a real client, clinician, or health plan. ↑ Back to the template and downloads ### Why this sample works Every clinical claim carries a number and a date. PHQ-9, GAD-7, and LOCUS scores appear with admission and current values, so partial response and remaining severity are both measurable instead of asserted. Risk is current and compared, not copied. The risk section is dated, references the prior review, and records what changed, which defeats the copy-forward pattern payers and auditors flag. The rationale answers the reviewer's actual test. It names the criteria set and the dimensions driving the score, then states why a lower level is not safe and adequate today, preempting the standard can-be-treated-at-a-lower-level denial line. The step-down plan makes the stay finite. Concrete thresholds (PHQ-9 below 10 for 2 weeks, full-day function) show the program is a bridge with an exit, not an open-ended placement. The request matches the authorization mechanics. Specific days and dates that start after the current period ends, compiled by the coordinator, attested by the licensed treating clinician. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations No regulation requires you to file the utilization review summary in the clinical record; filing it is a convention, and many programs keep concurrent review paperwork in a utilization review or billing file instead. Once filed in the record, it inherits the record's retention clock, commonly 5 to 10 years in the US depending on program and state, with minors' records held past majority. Wherever it lives, keep its voice distinct from the treatment record: the summary and the psychotherapy progress note serve different audiences, and copying the same narrative across dates or documents is a documented audit red flag. Feed the summary from measurement instead, with current dated scores from an outcome measure note workflow, medication changes, and functional specifics. Substance use programs get one genuine easing: under the 2024 revision of 42 CFR Part 2 (compliance date February 16, 2026), a single properly drafted consent covering treatment, payment, and health care operations now reaches utilization review disclosures, replacing the per-disclosure consents that used to complicate concurrent review. Clinicians in Canada and Australia rarely face US-style concurrent review at all; the closest analogues are workers' compensation and disability insurer reports, with retention set by provincial or state health-records law. The payer-facing mechanics run on two clocks, and only one of them is yours. No federal rule sets a deadline for when the treating clinician must submit a concurrent review; that timeline is contractual. The payer's clock is law: urgent concurrent decisions run in as little as 72 hours under ERISA and Medicaid managed care rules, CMS's interoperability rule holds standard prior authorization decisions for impacted plans to 7 calendar days from January 1, 2026, and a Medicaid managed care plan that misses its deadline has, by regulation, issued a denial. The law also protects a client mid-course: under the ERISA concurrent-care rule, reducing or ending an approved course of treatment early is an adverse benefit determination requiring advance notice and appeal rights, and most state utilization review statutes require a physician, often of the same or similar specialty, to make that determination after offering a peer-to-peer. Parity law reaches the criteria themselves: concurrent review is a classic nonquantitative treatment limitation, and plans still owe a comparative analysis of those criteria on request even while the 2024 MHPAEA rule sits under a non-enforcement posture. When a denial does land, read the remittance next to the billing note : an authorization mismatch surfaces as CO-197, an administrative code, while a true medical-necessity denial arrives as code 50, and the two have entirely different fixes. ## Common utilization review summary errors auditors flag When the HHS Office of Inspector General audited denials issued by 15 of the largest Medicare Advantage organizations (report OEI-09-18-00260, April 2022), it found that 13 percent of denied prior authorization requests actually met Medicare coverage rules, and 18 percent of denied payment requests met both coverage and billing rules, with plans applying clinical criteria not found in Medicare coverage rules and requesting documentation they did not need. KFF's analysis of HealthCare.gov insurers adds the consumer side: nearly one in five in-network claims was denied in 2023, fewer than 1 percent of denials were appealed, and insurers upheld most of the appeals they did receive. The uncomfortable lesson is that criteria mismatch, not documentation quality, drives many denials, which makes the summaries that survive review the ones engineered against the named criteria. The BastionGPT Clinical Advisory Board sees the same errors most often in utilization review summaries: Resubmitting last week's review with new dates. Payer guidance on outpatient treatment requests explicitly returns forms that update the dates without updating the clinical information, and cloned text across submissions is the copy-forward pattern auditors are trained to catch. Arguing benefit instead of necessity. Would benefit from continued care concedes the reviewer's case. The test is whether the member can be safely and adequately treated at a lower level today; name the criterion and answer that question directly. Freezing risk and measures at admission. An intake score and an unchanged risk paragraph tell the reviewer nothing is being measured, and a summary without current data gets decided on the plan's criteria alone. Leaving the step-down plan out. Reviewers read a missing discharge plan as an open-ended stay. State the target level, the thresholds that will trigger it, and the specific barriers today, even when the honest answer is not yet. Fighting an administrative denial with clinical prose. CO-197 means the authorization is absent, expired, or mismatched, a clerical problem fixed by aligning the claim with the authorization on file. Code 50 is the medical-necessity denial that calls for a peer-to-peer or appeal. Misreading the code wastes the appeal window. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on utilization review summaries. Draft the continued-stay rationale from your recent notes and the payer's criteria set, with the current presentation, response since last review, and the why-not-a-lower-level argument in the payer's structure. Pull the reviewer-facing numbers out of the chart: attendance, objective scores with dates and direction of change, medication changes, and units used against the authorization. Check a draft before it goes out: current dated risk status, a named criterion, a step-down plan with thresholds, a specific request, and nothing cloned from the last submission. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a utilization review summary required by law? For outpatient behavioral health, no. No US federal statute requires a treating clinician to produce a standalone utilization review summary; the duty to submit clinical information for continued care comes from the payer contract and its utilization management policy. The utilization review regulations that do exist (42 CFR 482.30 for hospitals, 42 CFR Part 456 for Medicaid facilities) bind facility review committees, and the ERISA claims rule regulates the payer's deadlines and notices, not your prose. Treat the document as payer policy with legal consequences: contractual in origin, but decisive for whether authorized care continues. What should a utilization review summary include? Across ASAM, LOCUS and CALOCUS-CASII, InterQual, and MCG, reviewers look for the same core: current symptoms with severity and objective measures, functional impairment, current risk status, response or lack of response to treatment, medications and adherence, the step-down plan with the barriers preventing it today (the same planning that later feeds the discharge summary ), an explicit statement of why a lower level of care is not appropriate right now, the specific criterion or dimension met, and the exact days or sessions requested with dates. The strongest single upgrade is objective measurement with dates: a score that moved is evidence of response, and a score that has not yet reached target is evidence of continued need. How long should a utilization review summary be? Most run 300 to 600 words, and many payers replace free text entirely with portal fields or a fixed outpatient treatment report form, in which case length is decided for you. No published study isolates how long clinicians spend drafting one, a gap worth remembering when a plan calls the burden minimal; the AMA's 2024 physician survey found practices completing an average of 39 prior authorization requests per physician per week and spending about 13 hours a week on them. Write to the criteria, not to a word count: a short summary that names the criterion beats a long one that narrates the stay. How is a utilization review summary different from a prior authorization request? Timing. A prior authorization request is submitted before care starts to obtain initial approval; the utilization review summary is submitted during care to continue an approval that already exists, which is why payers call the process concurrent or continued-stay review. The distinction carries legal weight: under the ERISA concurrent-care rule, cutting an approved course of treatment mid-stream is an adverse benefit determination with advance-notice and appeal rights, protections a fresh authorization request does not enjoy. Downstream of both sit the peer-to-peer conversation and, after a denial, the written appeal. The payer cut my client's authorization mid-course. What can we do? Treat it as a denial, not a lapse. If the plan approved a course of treatment and is reducing or ending it early, ERISA's concurrent-care rule makes that an adverse benefit determination: the plan must give advance notice and an appeal opportunity before the benefit stops, and urgent concurrent appeals run on an expedited clock measured in hours, not weeks. The fastest lever is usually a peer-to-peer review with the plan's physician reviewer, which many state utilization review laws require the plan to offer, often with a physician of the same or similar specialty, before the determination becomes final. If the peer-to-peer is refused or too slow, file the expedited internal appeal and preserve external review rights. Which criteria sets will the payer apply to my summary? General medical reviews run on InterQual or MCG. In behavioral health, the ASAM Criteria dominates substance use placement (its Fourth Edition, released October 2023, is rolling into payer use through 2026), and mental health level-of-care decisions increasingly use LOCUS for adults and CALOCUS-CASII for ages 6 to 18. You are entitled to know which one applies: Medicaid managed care enrollees may obtain the criteria relied on free of charge, ERISA and ACA appeal rules give claimants the rule behind a denial, and several state laws require disclosure on request. Ask in writing, then write the summary to the named criterion. Can a pre-licensed clinician or UR staff member write the utilization review summary? They can compile and transmit it, and in facility settings a utilization review coordinator usually does. The clinical assertions, though, must come from the licensed treating clinician, and pre-licensed drafters typically need a supervisor's co-signature under state licensure and payer credentialing rules. The asymmetry is worth remembering: on the payer's side, most state utilization review laws require that an adverse medical-necessity determination be made by a physician, so the exchange is licensed clinician to licensed reviewer even when other staff carry the paperwork. Time spent on the review itself is generally administrative and not separately billable. Does the utilization review summary go in the client's record, and how long is it kept? No regulation requires it to be filed in the clinical record; filing is a convention, and many programs keep concurrent review documents in a utilization review or billing file instead. Once filed in the record, it inherits the record's retention clock: commonly 5 to 10 years for US federal programs and state rules, longer for minors. Clinicians in Canada and Australia rarely face US-style concurrent review; the closest analogues are workers' compensation and disability insurer reports, and retention follows provincial or state law (Ontario physicians keep records at least 10 years from the last entry, and several Australian states require 7 years for adults or until a child patient turns 25). Can BastionGPT draft my utilization review summaries? Yes. Give it your recent notes, the current scores, and the criteria set the plan applies, and it drafts the summary with the current presentation, response since last review, the named criterion, the step-down barriers, and a specific request, then checks that nothing was cloned from the prior submission. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Prior Authorization Support Letter Claim-Support Billing Note Treatment Plan Review & Update Outcome Measure Review Note Discharge Summary 42 CFR Part 2 Consent Peer-to-Peer Review Note Prior Authorization Request Insurance Appeal Letter GAF Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/after-visit-summary All note templates # Patient Instructions & After-Visit Summary: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An after-visit summary (AVS), also called a clinical summary or patient instructions, is a plain-language handout a client takes home after an appointment. It covers what happened in the visit, medications, the between-session plan, follow-up, and who to call. No US, Canadian, or Australian law requires one for outpatient therapy. Most run 150 to 400 words, about half a page. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any treating clinician; in EHR practices the system generates it from the encounter record and staff hand it over Audience The client and, where involved, family or caregivers; a copy in the chart shows what instructions the client received Typical length 150 to 400 words · 5 to 10 minutes by hand (clinical team estimate) Format family Plain-language client handout (compare: progress note, discharge summary, safety plan) When it's used At or right after an appointment, printed or sent by portal; re-issued whenever instructions, medications, or follow-up plans change Standards context Convention in outpatient behavioral health; the CMS field list and delivery deadlines are retired incentive-program policy, not law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an after-visit summary? An after-visit summary (AVS) is a client-facing document produced at or shortly after an appointment that tells the client, in plain language, what happened and what to do next. CMS described its ancestor, the clinical summary, as "relevant and actionable information and instructions" (Stage 2 measure specification), and that remains the best one-line test of a good one. The name has an unusual pedigree: "After Visit Summary" began as an Epic Systems trademark, filed with the USPTO on April 3, 1998, more than a decade before the 2010 Meaningful Use rules (75 FR 44314) turned a generic clinical summary into a federal incentive measure and EHRs learned to print one after every encounter. You will also see it called a visit summary or patient instructions; its inpatient cousins are discharge instructions and the shared-record summaries Australia uploads to My Health Record. Two boundaries carry most of this page. First, the AVS is not the progress note : the note is your signed clinical record, written for clinicians, payers, and auditors, while the AVS is a derivative written for the client, and no rule requires the two to match. Second, the AVS is never a psychotherapy note. The HIPAA definition at 45 CFR 164.501 protects only process notes kept separate from the record, and the information-blocking rule excludes those from electronic health information; an AVS is generated from the encounter record, built to be handed over, and ordinary PHI from the moment it exists. Nothing you place on it can claim psychotherapy-note protection, which is exactly why the format works: it forces a clean split between what the client takes home and what stays in the chart. ## Who uses after-visit summaries and when Health systems and integrated-care clinics generate an AVS automatically at checkout, so psychiatrists, psychiatric NPs, and therapists embedded in those settings already produce one whether they think about it or not. In private practice it is a choice, and the best evidence for making it is old and sturdy: Kessels (2003) found that 40 to 80 percent of medical information is forgotten immediately, and in a randomized primary-care trial clients recalled about half of their medication information two days after the visit regardless of how detailed the summary was. A written summary earns its keep after an intake , after any medication change, and in any week where the homework matters more than the hour. Use it for what it is, a per-visit client handout: a discharge summary closes a whole episode for the next clinician, and a safety plan is a clinical intervention in its own right that an AVS never substitutes for. One more honest note: every published AVS study is primary care; there is no published study of after-visit summaries in adult psychotherapy, so behavioral-health practice here runs on judgment, not trial data. ## After-visit summary structure: what goes in each section Visit identification and contact details. Client name, visit date and modality, clinician name, practice phone, and the portal address. The retired CMS field list opened with the provider's name and office contact information, and that instinct was right: the handout exists so the client can act on it later. Pitfall: a summary with no phone number or portal path. A client holding instructions with no way to reach you has a souvenir, not a plan. What we covered, in plain language. Three moves, borrowed from the 2012 ONC technical guide on clinical summaries: here is what is going on, here is what it means, here is what to do next. Write at a reading level the client can use on a bad day. Pitfall: pasting sentences from the progress note. The note is written for clinicians and auditors; an AVS that reads like a chart entry fails the only audience it has. Diagnosis and problem list, when appropriate. The CMS list contemplated encounter diagnoses, and many clients want the name in writing for their own reading or paperwork. It stays a clinical decision. Pitfall: treating the diagnosis as mandatory. Nothing compels a mental-health diagnosis on a client handout, and the Stage 2 specification expressly allowed withholding information whose disclosure could cause substantial harm. Medications and today's changes. The current list with doses, anything that changed today clearly marked, and who prescribes each one, especially where medication is managed outside your practice. Pitfall: the stale list. Medication details are what clients misremember most, about 53 percent recall in the JABFM randomized trial, and a wrong list on an official-looking handout actively misleads. Your plan between sessions. The homework or skill, how often, and in what situation, matched to whatever tracking tool the client is using. This is the section that makes an AVS clinically worth the paper. Pitfall: writing "practice coping skills." Name the skill, the frequency, and the trigger; vague homework produces vague weeks. Follow-up and referrals. The next appointment with date and modality, any referral or test still pending, and exactly how to reschedule or ask a question before then. Pitfall: open loops without owners. "Call to schedule" with no number, or a referral with no name attached, quietly becomes no follow-up at all. Crisis and after-hours contacts. The after-hours line, a crisis line, and when to use the emergency number, stated plainly even in routine care. Pitfall: skipping this because today's session was unremarkable. The AVS is often the only piece of paper the client keeps between sessions; it should work on the worst night of the month, not just the afternoon it was printed. ### Blank template (copy and adapt) AFTER-VISIT SUMMARY / PATIENT INSTRUCTIONS (plain-language client copy) Client: ______________________ Visit date: ____________ Time: _______ Visit type: [ ] office [ ] video [ ] phone Clinician: ___________________________ Practice: _____________________ Phone: ________________ Portal: ______________________________________ REASON FOR TODAY'S VISIT _______________________________________________________________________ WHAT WE COVERED TODAY What's going on: ______________________________________________________ What it means: ________________________________________________________ What to do next: ______________________________________________________ DIAGNOSIS / PROBLEM LIST (include when clinically appropriate) _______________________________________________________________________ MEDICATIONS (current list; mark changes made today; note who prescribes) _______________________________________________________________________ _______________________________________________________________________ YOUR PLAN BETWEEN SESSIONS (skill or homework, how often, when) _______________________________________________________________________ _______________________________________________________________________ FOLLOW-UP Next appointment: _______________ With: ______________ How: _________ Referrals or tests pending: ___________________________________________ To reschedule or ask a question: ______________________________________ IF YOU NEED HELP BEFORE YOUR NEXT VISIT After-hours line: _____________________________________________________ Crisis support: call or text 988 (US) or your local crisis line Emergency: 911 or your local emergency number RESOURCES SHARED TODAY _______________________________________________________________________ No signature is needed. This summary is written for you; it is not your full record. Ask the practice if you would like a copy of your records. Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample after-visit summary Scenario: a counselor in an outpatient practice sends a plain-language after-visit summary through the portal after a weekly CBT session for panic disorder. All details are fictional. After-Visit Summary. Client: M.T. · Visit: 07/21/2026, 3:00 pm, video · Clinician: A. Rivera, LPC · Lakeview Counseling, (555) 014-2266 · Portal: Lakeview client portal Reason for today's visit: Weekly therapy session for panic disorder. What we covered today: What is going on: your panic attacks are down from about four a week to two, and you handled Tuesday's attack at work without leaving the building. What it means: the practice is working; panic spikes are your body's alarm firing early, not a sign of danger, and each time you ride one out the alarm resets a little lower. What to do next: keep the daily breathing practice and this week's exposure exercise, and let the sensations come without fighting them. Diagnosis we discussed: Panic disorder. We talked about what this label does and does not mean; bring questions any time. Medications, for your reference: Sertraline 50 mg, one tablet each morning, prescribed by Dr. Osei (family medicine). No changes were made today. Questions about this medication go to Dr. Osei's office, (555) 014-8890. Your plan between sessions: 1) Paced breathing, 5 minutes, twice a day (after breakfast and after lunch). 2) One stair-climb practice on three different days, using the sensation log we built together; rate the peak (0 to 10) and how long it took to pass. 3) Note any panic attack in the tracking sheet the same day. Follow-up: Next appointment Tuesday 07/28/2026, 3:00 pm, video. To reschedule or ask a question, send a portal message or call (555) 014-2266; portal messages are answered within 1 business day. If you need help before then: After hours, call (555) 014-2267. If you are in crisis, call or text 988, or go to the nearest emergency department. These numbers are on the back of the card we gave you at intake. This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works It is written for the client, not the chart. The what-is-going-on, what-it-means, what-to-do-next structure comes straight from the ONC technical guide's framing, and nothing in it needs a glossary or a clinician to translate. The medication line does real safety work. It names the prescriber, states that nothing changed, and routes medication questions to the right office, exactly the details clients misremember most in the recall research. The homework is specific enough to do. Skill, frequency, and situation, tied to the log the client already has, so the week between sessions has a plan rather than a sentiment. Every open loop has a path. Reschedule route, portal response time, after-hours line, and crisis contacts are all present even though the visit was routine. It contains nothing the clinician would mind disclosing. The clinical reasoning stays in the progress note; the handout is ordinary PHI written in the knowledge that the client, and anyone the client shows it to, will read it. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The AVS sits in an odd legal spot: no law makes you create one, but once created it is ordinary PHI. If its content is used to make decisions about the client it belongs to the designated record set, which makes it accessible to the client and, in the US, part of electronic health information under the information-blocking rule (45 CFR Part 171). The psychotherapy-notes exclusion never covers it: 45 CFR 164.501 protects only process notes kept separate from the record, a boundary the psychotherapy notes authorization page walks through, so write every summary knowing it will be read. Retention attaches to the encounter documentation rather than the handout: US state rules commonly run 6 to 10 years for adult records (HIPAA separately requires 6 years for its required documentation), Canadian colleges commonly expect about 10 years after last contact, and Australian state law runs 7 years from the last entry, or until age 25 for minors. No rule anywhere requires the AVS itself to be kept as a separate document; a copy in the chart, or an EHR that can regenerate it, simply documents what instructions the client received. The payer story explains the document's shape without governing it today. Stage 1 of Meaningful Use (2010) paid eligible professionals to provide a clinical summary within 3 business days for more than half of office visits; Stage 2 tightened the deadline to 1 business day; the MACRA era then folded the measure into MIPS Promoting Interoperability's electronic-access measure, so no current federal measure counts paper summaries at all. Two details from the retired specification are still worth honoring: providers could not charge a fee for the summary, and a client who was offered one and declined still counted, so document the offer rather than pressing paper on anyone. The AVS is not a billable service and has no CPT code; the reimbursement weight stays on the encounter note and the claim-support billing note . The one place a legal duty attaches is hospital discharge instructions under 42 CFR 482.43, a Medicare Condition of Participation for hospitals and post-acute transitions, not outpatient psychotherapy. For most documents on this site the format is a convention and the content is the requirement; here even the content list is optional, which makes the AVS the rare document whose entire value is clinical rather than regulatory. ## Common after-visit summary errors auditors flag The audit record around this document is an incentive-program story, not a chart-review story. When OIG audited Meaningful Use payments to eligible professionals (report A-05-14-00047, 2017), 14 of 100 sampled clinicians could not support their attestations, worth $291,222 in the sample and an estimated $729.4 million extrapolated across the program; a 2019 OIG report put hospital-side incorrect payments at $93.6 million, under 1 percent of $10.8 billion paid. Both findings turned on missing attestation documentation, never on a missing summary in a chart, and no claim denial for an absent AVS has been documented; there is no denial code for one. The errors that matter are clinical and practical. The BastionGPT Clinical Advisory Board sees the same errors most often in after-visit summary reviews: Handing the client the progress note. The note is the record; the AVS is the translation. Chart shorthand, rule-out language, and clinical reasoning confuse the reader the handout exists for, and audience adaptation is the entire point of the format. Writing it as if it were private. An AVS is ordinary PHI with no psychotherapy-note protection, and the client is its addressee. Anything you would not want read aloud at the kitchen table belongs in the note, not the handout. Leaving the client without a path. No practice phone, no portal instructions, no next-appointment details, no crisis line. The retired CMS field list led with contact information because a summary the client cannot act on is decoration. A medication list nobody checked. Clients recalled about 53 percent of medication information in the JABFM trial even with a summary in hand; an outdated list on an official handout converts a memory problem into a safety problem. Importing requirements that never existed. A signature block, a mandatory diagnosis, a verbatim match to the note, a standalone AVS retention rule: none of these is real. The Stage 2 specification even allowed withholding information whose disclosure could cause substantial harm, a clinical-judgment exception most templates never mention. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on patient instructions and after-visit summaries. Turn a finished progress note or your session bullets into a plain-language client summary: what we covered, what it means, and what to do next, at a reading level the client can use. Draft the between-session plan, follow-up details, and crisis-contact block from your instructions, ready to print or send through the portal. Check the draft before it goes out: it flags jargon, clinical reasoning that belongs in the note rather than the handout, and missing follow-up or contact details. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is an after-visit summary legally required? No. No US, Canadian, or Australian law requires an outpatient therapist, psychologist, or counselor to give clients an after-visit summary. In the US it entered practice as payer policy: the Meaningful Use program paid eligible professionals to deliver a clinical summary, first within 3 business days (Stage 1) and then within 1 business day (Stage 2), and that measure has since been folded into MIPS Promoting Interoperability's electronic-access measure. The nearest true legal duties sit elsewhere: hospital discharge instructions under 42 CFR 482.43 and electronic record access under the information-blocking rule. For outpatient therapy the AVS is professional convention, which leaves the decision, and the format, to you. What goes in an after-visit summary? The retired CMS clinical-summary specification listed 20 fields, from provider contact information and visit details through problem list, medications, allergies, instructions, and future appointments. For outpatient behavioral health the working core is shorter: visit identification with practice contact details, a plain-language account of the session (here is what is going on, what it means, what to do next), medications when relevant, the between-session plan, follow-up details, crisis contacts, and any resources you shared. Most therapy versions run 150 to 400 words, about half a page; where a field has nothing in it, saying so was enough even under the old measure. Does the after-visit summary have to match my progress note? No rule requires the two documents to match, and they should not read the same. The progress note is your signed clinical record, written for clinicians, payers, and auditors; the AVS is a derivative written for the client. The Stage 2 specification even authorized withholding particular information from the summary where disclosure could cause substantial harm, so the AVS may deliberately contain less than the note. What the handout can never do is replace the note: the encounter documentation remains the record that billing and licensure depend on. Do I have to put the diagnosis on it? No. The CMS field list contemplated problem lists and encounter diagnoses, but nothing compels a mental-health diagnosis on a client handout, and the substantial-harm provision expressly permitted leaving particular information off. Include the diagnosis when it serves the client; many people want the name in writing for their own reading or their paperwork. Leave it off when a printed label would outrun the clinical conversation. Either way, the diagnosis lives in the encounter note regardless of what the handout says, and a line in the note about your reasoning closes the loop. What if the client does not want one? Offer it, document the decline, and move on. The Stage 2 measure explicitly counted a client who was offered and declined a clinical summary, so refusal was never a compliance failure even while the measure was live. Two other details from the retired specification are worth keeping: providers could not charge a fee for the summary, and clinicians could withhold particular information where disclosure risked substantial harm. A one-line entry in the note, offered visit summary, client declined, records the practice without pressing paper on anyone. Is an after-visit summary protected like a psychotherapy note? No, and the distinction matters more here than almost anywhere. Psychotherapy notes under 45 CFR 164.501 are process notes a therapist keeps separate from the record, and the information-blocking rule excludes them from electronic health information; releasing them takes the dedicated psychotherapy notes authorization . An AVS fits none of that definition: it is generated from the encounter record, designed to be handed over, and ordinary PHI. Never put anything on an after-visit summary you would fight to keep out of the client's hands; the client is the addressee. How is an AVS different from a discharge summary or a safety plan? Scope and audience. The AVS is per-visit and client-directed. A discharge summary documents an entire episode of care at its close and speaks mainly to the next clinician; ending a course of therapy calls for one, and an AVS does not substitute. A safety plan is a clinical intervention in its own right, built collaboratively for crisis moments; giving the client a copy does not turn it into an AVS, and pasting safety-plan content into a summary does not satisfy any safety-planning standard. The AVS can point to both, which is its actual job: carrying the next steps home. Do I need to keep a copy, and for how long? No rule requires the AVS itself to be retained as a separate document; retention duties attach to the underlying encounter record. In the US, state rules commonly require 6 to 10 years for adult records, longer for minors, and HIPAA separately requires 6-year retention of its required documentation. Canadian colleges commonly expect about 10 years after last contact. Australian state law runs 7 years from the last entry, and until age 25 for records of children. If your EHR can regenerate the summary from the encounter, that covers the practical need; if you hand-build one, a copy in the chart, or a note of what was given, documents the instructions the client received. Can BastionGPT write my after-visit summaries? Yes. Give it the session's key points or the finished note, and it drafts the client-facing summary in plain language: what we covered, what it means, the between-session plan, follow-up details, and crisis contacts, with the clinical reasoning kept back in the note where it belongs. It also checks the draft for jargon and missing contact or follow-up details before you send it. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Prior Authorization Support Letter Psychotherapy Progress Note Discharge Summary Safety Plan Psychotherapy Notes Authorization Communication Log & Portal Message Note Bishop Score Documentation Wong-Baker FACES Pain Scale Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/no-show-note All note templates # Appointment No-Show & Outreach Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An appointment no-show note is a brief clinical-record entry documenting that a client missed a scheduled session or cancelled late, along with any outreach attempts and a risk-informed next step. Front desk staff record the missed appointment; the clinician documents outreach and any risk decision. Most run 75 to 200 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Front desk staff record the miss; therapists, psychologists, and prescribers sign the risk review and outreach Audience The chart, the care team, auditors, and reviewers if a fee, termination, or adverse event is later questioned Typical length 75 to 200 words · 3 to 10 minutes by hand (clinical team estimate) Format family Operational chart entry (compare: progress note, communication log, care coordination note) When it's used Same day a client misses or late-cancels a session, and after each outreach attempt Standards context No US, Canadian, or Australian statute requires the note itself; no-show billing and record retention are the regulated parts On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an appointment no-show note? An appointment no-show note is the clinical-record entry a practice makes when a scheduled client does not attend or cancels inside the practice's notice window. It records the missed appointment, any outreach that followed, and a risk-informed decision about what happens next. You will also see it called a missed appointment note, a failure-to-attend (FTA) or did-not-attend (DNA) note, a non-attendance note, or an outreach or attempted-contact note; in Australian NDIS settings the same entry doubles as the short-notice cancellation record that supports a claim. Unlike the SOAP note , which traces to Lawrence Weed's problem-oriented medical record, the no-show note has no founding author or publication: it is a convention that grew out of the general duty to keep adequate records, which is why its format varies so much between practices and EHRs. The load-bearing distinction: a no-show note documents a non-service. A progress note records a session that happened; this note records that one did not, which is why it is never billed to a payer as a session. And almost nothing about the note itself is law. No US, Canadian, or Australian statute requires a no-show to be documented, sets a deadline or format for the entry, or fixes how many missed appointments justify discharge; all of that is professional convention. What law and payer policy do regulate is the money and the chart: whether you may charge for the missed time (a billing question that pairs this page with the billing note ) and how long the record is kept. Keeping those three layers separate, law, payer policy, and convention, is most of what makes the note defensible. ## Who uses appointment no-show notes and when Front desk staff, therapists, psychologists, psychiatrists, and case managers all write them, from solo practices to hospital outpatient clinics: the front desk logs the factual miss, and a clinician adds the outreach and any risk judgment. Volume alone makes the habit worth building. A 2021 systematic review across 19 general-practice studies found a mean missed-appointment rate of 15.2 percent, a safety-net outpatient psychiatry study measured about 13 percent of psychotherapy appointments missed, and the missed-appointment rate for mental health services in England has been estimated at 19.1 percent, nearly one in five. The note matters most when a higher-risk client misses a session, when repeated no-shows push a practice toward termination, and when a fee or an NDIS cancellation claim needs evidence behind it. If outreach turns into contact with a GP, family member, or crisis service, document that conversation in a care coordination note and keep this entry focused on the missed appointment itself. ## Appointment no-show note structure: what goes in each section Appointment details. The scheduled date and time, the service or appointment type, the clinician, and the setting (office or telehealth). This anchors everything else in the entry. Pitfall: a note that says "missed appointment" with no scheduled date or time; months later, in a fee dispute or an audit, nobody can reconstruct which slot the entry covers. Attendance status. Say precisely which event occurred: no-show, late cancellation inside your notice window (record when notice arrived), or clinician cancellation. Pitfall: treating the three as interchangeable. Your fee policy, payer rules, and, in Australia, the NDIS short-notice cancellation windows all turn on which one it was. Outreach attempts. One line per attempt: date, time, method (phone, text, portal, letter), and outcome, with the underlying call or message log kept as the evidence trail. Pitfall: "attempted contact" with no date, method, or outcome. In an abandonment or adverse-event review that line proves nothing; dated attempts prove diligence. Risk review. A short statement of whether the chart carries risk flags and what that means for outreach. For flagged clients, same-day outreach is the defensible benchmark, and the entry should show it happened. Pitfall: silence on risk for a client with a documented history. A blank reads as absence of monitoring, not absence of risk. Billing and fee decision. Record that no claim goes to any payer (a missed appointment is a non-service with no procedure code) and whether a client fee applies under your written policy. Pitfall: letting the missed slot reach a payer anyway. Medicare denies it with denial code CARC 204, and billing a session that did not occur is the fraud pattern behind real settlements. Next step. What happens now, with an owner and a date: a held or rescheduled slot, an attendance letter, a welfare check, or the start of a documented termination process. Pitfall: a bare "will follow up." The next reviewer, and the next clinician, need something they can verify happened. Signature and credentials. Who recorded the factual miss and who made the clinical call, each signed and dated. Administrative staff can record non-attendance, but a clinician signs the risk judgment and the outreach decision. Pitfall: a front-desk-only entry that quietly carries a clinical judgment, such as "no outreach needed," that no clinician ever signed. ### Blank template (copy and adapt) APPOINTMENT NO-SHOW & OUTREACH NOTE (documents a non-service: never bill a payer for a missed session) Client: [initials] Chart #: Date of entry: Scheduled appt: [date] [time] Service type: Clinician: Setting: [ ] office [ ] telehealth ATTENDANCE STATUS [ ] No-show [ ] Late cancellation (notice received: ____________) [ ] Clinician cancellation Fee policy applies: [ ] yes [ ] no OUTREACH ATTEMPTS (date / time / method / outcome, one line each) 1. ________________________________________________________________ 2. ________________________________________________________________ 3. ________________________________________________________________ RISK REVIEW (required when the chart carries risk flags) Risk flags: [ ] none [ ] yes: _____________________________________ Same-day outreach: [ ] done [ ] n/a Safety plan activated: [ ] n/a Clinical concern / rationale: ______________________________________ BILLING / FEE DECISION Claim to payer: [ ] none (non-service) Client fee per policy: $___ NEXT STEP (owner and date) [ ] Reschedule: __________ [ ] Letter: __________ [ ] Welfare check / other: _________________________________________ [ ] Termination process started (notice and referrals documented) Recorded by (front desk / clinician): Date / time: Clinician signature / credentials: Date signed: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample appointment no-show and outreach note Scenario: an established adult client in weekly CBT for major depressive disorder, with a safety plan on file from intake, misses a telehealth session without notice. The therapist documents same-day outreach. All details are fictional. Appointment No-Show & Outreach Note. Client: D.K., 41 · Date of entry: 07/21/2026 · Chart: flagged (safety plan on file) Scheduled appointment: 07/21/2026, 2:00 PM, individual psychotherapy, telehealth, R. Osei, LMFT. Client did not join the video session; the waiting-room log shows no connection attempt. No notice received before or after the start time. Status: No-show. The practice notice window is 24 hours and no cancellation arrived, so late cancellation does not apply. Outreach: (1) 07/21/2026, 2:12 PM, phone: no answer; voicemail asked D.K. to call the office and confirmed the held slot for next week. (2) 07/21/2026, 4:40 PM, portal message: reschedule link, a copy of the attendance policy, and a reminder that R. Osei can be reached through the office line. Full text retained in the portal log. Risk review: Chart is flagged: safety plan on file from the 03/2026 intake (history of passive suicidal ideation, no attempts). Per practice policy, a missed appointment for a flagged client triggers same-day outreach, completed above. No acute-risk indicators at the last session on 07/14/2026: client denied suicidal ideation and PHQ-9 was 9, down from 16 at intake. If no response by end of day 07/23/2026, R. Osei will call the emergency contact consented to on file. Billing: No claim submitted; a missed appointment is a non-service. The $75 no-show fee is waived as a first occurrence under the written attendance policy, which was re-sent with the portal message. Next step: Hold the 07/28/2026, 2:00 PM slot. If no response by 07/23/2026: second call, then an attendance letter. Any move toward termination goes through supervision first and would follow the practice's written termination process. Signatures: Missed appointment logged by front desk (J. Tran), 07/21/2026, 2:05 PM. Outreach, risk review, and plan documented and signed by R. Osei, LMFT, 07/21/2026, 5:10 PM. This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works The status is precise and evidenced. No-show is distinguished from late cancellation, the notice window is stated, and the waiting-room log is cited as the evidence trail. Every outreach attempt carries a date, time, method, and outcome. The record proves diligence instead of asserting it, which is exactly what an abandonment or adverse-event review checks. Risk is addressed because the chart is flagged. Same-day outreach is documented, last-session risk data is cited, and an escalation trigger with a date and a named contact ties back to the safety plan and consent on file. The billing decision is explicit and separate. Nothing goes to a payer, and the fee decision follows the written policy, applied the same way for every client. The next step has an owner and dates. A held slot, a response deadline, and a supervision gate before any termination talk: the paper trail that separates discharge from abandonment. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The no-show note is part of the clinical record: it sits in the chart, clients can request it, and it is discoverable and producible under a records request or subpoena like any other entry. Retention therefore follows chart rules, and the widely repeated "HIPAA six-year rule" is not one of them: that rule governs compliance documentation such as policies and procedures, not patient charts. Chart retention is set by state law in the US, by provincial law and college standards in Canada (Ontario expects at least 10 years from the last interaction, or 10 years past the client's 18th birthday), and by state law in Australia (commonly 7 years from last service, or until age 25 for minors). No statute anywhere sets a completion deadline for the entry itself; same-day documentation is what auditors and courts credit, and it is the practical standard too, because outreach details fade fast. Front desk staff may record the factual miss, but the risk judgment and the outreach decision belong under a clinician's signature. The money is where the real rules live. A missed appointment is a non-service: no procedure code exists for it, Medicare returns denial code CARC 204 if one is billed, and in Australia no MBS rebate applies because a benefit attaches only to a service actually rendered. Charging the client directly is a separate question with jurisdiction-specific answers: Medicare permits a uniformly applied fee billed to the patient (its claims manual calls the charge "a missed business opportunity"), Medicaid programs generally prohibit it, Canadian colleges such as Ontario's permit it with advance notice and a working cancellation system, and the NDIS lets providers claim up to 100 percent of the agreed fee for a short-notice cancellation, with a 2-clear-business-day window for therapy since July 2024. With repeated no-shows the exposure flips from billing to abandonment: ending care over attendance is legitimate, but only on top of documented outreach, written notice, referrals, and a transition period, the record that feeds a proper discharge summary . And when a flagged client goes quiet, outreach is a clinical event: document it here, and if it activates the client's safety plan or an emergency contact, record that too. For most documents on this site the format is a convention and the content is the requirement; here even the note is a convention, and the rules worth memorizing are the money and retention rules around it. ## Common appointment no-show note errors auditors flag Enforcement in this area is lopsided: no regulator has ever fined a clinician for failing to write a no-show note, but billing for sessions that did not occur is a standing false-claims pattern. A Maryland psychiatrist paid $400,000 in 2018 to resolve allegations of billing Medicare and Medicaid for psychotherapy time not actually delivered, a Connecticut psychologist repaid $126,760 for psychotherapy billed to Medicaid but not provided, and a federal OIG review of Maine's 2023 fee-for-service Medicaid behavioral claims estimated $45.6 million in improper payments, with session notes failing documentation requirements among the top issues. On the clinical side, an AHRQ patient-safety commentary cites an estimate that up to 70 percent of patients evaluated in an emergency department for suicidal ideation never attend their first follow-up appointment, which is why outreach after a miss is treated as a safety task, not a courtesy. The BastionGPT Clinical Advisory Board sees the same errors most often in no-show and outreach note reviews: The missed session reaches a payer. There is no procedure code for a missed appointment and no payer that covers one; Medicare denies the claim with CARC 204, and a billed session that did not occur is a false claim, not a paperwork slip. Outreach without evidence. "Attempted to contact client" with no date, method, or outcome. When a termination or an adverse outcome is later questioned, undated outreach is weightless; the dated attempt log is the defense. Risk silence for flagged clients. The chart carries a safety plan or documented ideation, the client goes quiet, and the note says nothing about risk. Same-day outreach, recorded with outcomes, is the standard reviewers apply. Three strikes, then a termination letter, with nothing in between. No rule sets a number of misses; what boards and courts examine is the middle: documented attendance conversations, outreach, written notice, referrals, and a transition period. Fees applied inconsistently or to the wrong programs. Medicare's fee permission requires the same policy for every patient and bars billing Medicare itself; Medicaid programs generally prohibit no-show fees; and NDIS practices still applying a 7-day notice window to therapy are using the wrong rule, 2 clear business days since July 2024. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on appointment no-show and outreach notes. Draft a complete no-show and outreach note from a sentence about what happened plus your call or portal log. Turn a string of contact attempts into a clean, dated outreach record with methods, outcomes, and a next step. Check the entry before you sign it for the pieces reviewers hunt for: attendance status, dated attempts, a risk statement when the chart is flagged, and a next step with an owner. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to write a note when a client no-shows? No statute in the US, Canada, or Australia requires it; documenting a missed appointment is professional convention flowing from the general duty to keep adequate records. Treat it as non-optional anyway. Colorado Medicaid guidance, for example, calls a missed appointment a non-service that is not reimbursable but "should be documented in the clinical record," and every risk-management source treats the note plus dated outreach as expected practice. It is also the record you will want if attendance, termination, or an adverse outcome is ever examined. Can I bill insurance for a missed appointment? No. A missed appointment is a non-service: no procedure code exists for it, and no payer covers it. A claim that reaches Medicare anyway is denied with code CARC 204 (not covered under the patient's benefit plan), and in Australia no MBS rebate applies because a Medicare benefit is payable only for a service actually rendered. Billing a session that did not occur is treated as a false claim, the pattern behind the settlements described on this page. Can I charge Medicare or Medicaid clients a no-show fee? Medicare: yes, under conditions in its claims manual (Section 30.3.13, effective October 2007): the charge is for a missed business opportunity, it is billed to the patient and never to Medicare, and the same policy must apply to Medicare and non-Medicare patients alike. Medicaid: generally no; participating providers accept the state's payment as payment in full, and some states, California among them, expressly bar billing behavioral health beneficiaries for missed appointments. Check your state manual before charging anyone in a public program, and put the policy in writing for everyone else. How many no-shows before I can discharge a client? No law or board rule sets a number; the common three-strikes threshold is practice convention. Repeated no-shows are a recognized, valid ground for ending care, but the burden is on you to show the process was not abandonment: documented outreach, an attendance conversation where possible, written notice with a reasonable transition period, and referrals. The discharge summary then closes the record properly. Do I have to call a client who no-shows if they are at risk? Not by universal statute. The closest formal requirement, the Joint Commission's suicide-prevention goal (NPSG.15.01.01), ties written follow-up policy to discharge, not to missed appointments, and applies only in accredited organizations. Clinical duty of care fills the gap: for a client with documented risk, same-day outreach recorded with date, method, and outcome is the defensible standard, and your own written policy becomes the yardstick a reviewer measures you against. If outreach activates the client's safety plan or an emergency contact, record that in the note too. Can front desk staff write the no-show note? Yes, for the factual part: administrative staff routinely record that the client did not attend, when, and for what appointment type. A clinician should document and sign anything that involves clinical judgment: the risk review, the outreach decision, and the next step. Pre-licensed clinicians follow the same supervisory co-signature rules that apply to their other notes; no jurisdiction has a special signature rule for no-show entries. What is the NDIS short-notice cancellation rule for therapy? Since 1 July 2024, the short-notice window for therapy and psychology services is 2 clear business days; the 7-day figure many practices still use now applies to disability-support-worker services, not therapy. Under the current NDIS pricing arrangements a provider may claim up to 100 percent of the agreed fee for a short-notice cancellation or no-show if the cancellation terms are in the service agreement, the worker could not be redeployed to other billable work, and the worker is paid for the scheduled time. Record the non-redeployment in the note, and re-check the pricing arrangements each July, when a new version takes effect. How long do I keep no-show notes, and can they be subpoenaed? They are part of the clinical record, so they are discoverable and must be produced in a records request or subpoena like any other chart entry. Retention follows chart rules, not HIPAA's six-year rule, which covers compliance documentation such as policies: in the US the chart is governed by state law, Ontario's colleges expect at least 10 years from the last interaction (or 10 years past the client's 18th birthday), and Australian state law commonly requires 7 years from last service, or until age 25 for minors. Can BastionGPT write my no-show and outreach notes? Yes. Tell it what happened in a line or two, or paste the call log, and it drafts a chart-ready entry with attendance status, dated outreach attempts, a risk statement when the chart is flagged, and a next step with an owner. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Prior Authorization Support Letter Psychotherapy Progress Note Care Coordination Note Safety Plan Discharge Summary Financial Policy & Fee Agreement Good Faith Estimate (No Surprises Act) Communication Log & Portal Message Note Records Request & Chart Copy Packet Advance Beneficiary Notice (ABN) Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/financial-policy All note templates # Financial Policy & Fee Agreement: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A financial policy and fee agreement is the signed contract between a therapy practice and a client that sets fees, billing and insurance terms, cancellation and no-show charges, and collection rules before treatment starts. Practices present it at intake alongside informed consent. No US statute requires one; ethics codes and payer contracts define what it must say. Most run one to three pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Practice owners, practice managers, and solo clinicians; pre-licensed clinicians add their supervisor's name and license Audience Clients, who sign it; front-desk and billing staff, who apply it; the payer contracts and licensing boards it must not contradict Typical length 600 to 1,200 words · 45 to 90 minutes to draft (clinical team estimate) Format family Signed intake-packet contract (compare: informed consent, good faith estimate, assignment of benefits) When it's used Signed at or before the first session; re-signed whenever fees or terms change Standards context No statute mandates it; ethics codes (APA 6.04, ACA A.10, NASW 1.13) and payer contracts control what it must say On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a financial policy and fee agreement? A financial policy and fee agreement is the client-facing contract, signed at or before the first session, that discloses a mental health practice's fees, billing methods, insurance and assignment-of-benefits terms, cancellation and no-show charges, and what happens when a balance goes unpaid. No government body created or standardized it. Its closest ancestor sits on the ethics shelf: APA Standard 6.04, in place since the 1992 ethics code, expects psychologists and clients to reach an agreement on compensation and billing "as early as is feasible," and the ACA code (Section A.10) and NASW code (Standard 1.13) carry parallel duties for counselors and social workers. You will also see it called a fee agreement, payment agreement, financial responsibility form, fee disclosure, office policies document, or, for insured clients, an assignment-of-benefits form. The document's legal status confuses more practices than its content does. No statute in the US, Canada, or Australia requires a signed financial policy; the signature is a convention backed by contract law and professional ethics. What surrounds the signature is law: the No Surprises Act's good faith estimate (45 CFR 149.610, in force since January 1, 2022) is a separate, one-way cost disclosure for uninsured and self-pay clients that states on its face it is not a contract, Medicaid regulations bar balance billing no matter what a client signed, and payer contracts decide which balances are ever the client's to pay. The agreement is also not informed consent: consent covers the nature and course of therapy, and ethics codes fold fees into it, but a consent form is not a fee contract. Keep both in the intake packet, and keep them distinct. ## Who uses financial policies and fee agreements and when Every setting that bills clients directly needs one: solo and group private practices, training clinics whose pre-licensed clinicians see clients under supervision, and outpatient programs where a billing office applies the terms the clinician promised. The agreement is drafted once, presented with the intake paperwork before the first session (the clinical side of that visit lands in the intake note ), and re-signed when fees or terms change. It earns its keep at the edges of care rather than the middle: the missed third appointment, the deductible that resets in January, the balance nobody mentioned for four months. Insurance-based practices pair it with claim documentation such as the claim-support billing note ; cash-pay practices lean on it even harder, because it is the only money document the client ever signs. ## Financial policy and fee agreement structure: what goes in each section Practice and clinician identification. The practice's legal name, the treating clinician's name, credentials, and license number, and, for pre-licensed clinicians, the supervisor's name and license. This is who the client is contracting with. Pitfall: leaving the supervisor off a supervisee's agreement. Payers credential the supervisor and boards treat an unnamed supervisory relationship as a disclosure failure, so the money paperwork must match the supervision paperwork. Fee schedule. Each service in plain language with its bare code number and dollar fee: intake assessment (90791), 55-minute individual session (90837), 45-minute session (90834), and whatever else you actually deliver, plus how much notice clients get before fees change. Pitfall: promising one blanket rate. For in-network care your contracted rate controls, not your posted fee, so state that contracted rates apply and quote your full fee for self-pay and out-of-network work. Insurance, assignment of benefits, and client responsibility. Whether you bill plans directly, what the client owes at time of service (copays, coinsurance, unmet deductible), and what happens for non-covered services. Pitfall: the phrase "you owe whatever insurance does not pay." When a remittance labels a balance CO, for contractual obligation, that amount is your write-off under the payer contract; only balances coded PR, patient responsibility, can ever move to the client, and no signed policy overrides the group code. Cancellation and no-show terms. The notice window, the fee, and the two rules that keep it enforceable: the charge is billed to the client directly, never to insurance, and the same policy applies to every client. Pitfall: a Medicare carve-out. Medicare permits missed-appointment charges only when the policy applies equally to all patients, so one commercial contract that bans no-show fees breaks the uniformity Medicare requires, and Medicaid members generally cannot be charged at all. Payment methods, card on file, and payment plans. When payment is due, which methods you accept, the card-on-file authorization, and the returned-payment fee. Pitfall: a generous installment plan that quietly makes you a creditor. A written plan of more than four installments, or any plan with a finance charge, triggers federal Truth in Lending disclosures, so cap plans at four payments with no added charge unless you want Regulation Z paperwork. Overdue balances: interest, hardship, and collections. Any interest rate and when it starts, the sliding-scale or hardship process, and what happens before an account leaves the practice. Pitfall: surprise enforcement. Interest is lawful only if disclosed here first and kept within state limits, hardship discounts must be individualized and documented rather than advertised, and ethics codes require notice and a chance to pay before any collector hears a client's name. Signature and date, on the current terms. Client and clinician sign and date at or before the first session, and again whenever fees or material terms change; every signed version stays in the record. Pitfall: the stale signature. Raise your fee without a re-signed agreement and the version the client signed, at the old number, is the version a board or small-claims judge reads. ### Blank template (copy and adapt) FINANCIAL POLICY AND FEE AGREEMENT Signed at or before the first session. This agreement sits alongside your informed consent; it does not replace it. PRACTICE AND CLINICIAN Practice legal name: ____________________________________________ Clinician and credentials: ____________________ License #: ______ Supervisor and license # (if pre-licensed): ______________________ FEES (plain-language service, code number, fee) Intake assessment (90791): ....................... $______ Individual session, 55 min (90837): .............. $______ Individual session, 45 min (90834): .............. $______ Other service: ................................... $______ Fees reviewed annually; ____ days written notice before changes. INSURANCE AND YOUR RESPONSIBILITY [ ] In network: we bill your plan. You owe copays, coinsurance, and unmet deductible at time of service. Contractual write-offs are never billed to you. [ ] Out of network / self-pay: payment in full at time of service; superbill provided on request. CANCELLATION AND MISSED APPOINTMENTS Notice required: ____ hours Late-cancel / no-show fee: $______ Billed to you directly, never to insurance. Applies equally to every client; not charged to Medicaid members. Emergencies considered case by case. PAYMENT, BALANCES, AND COLLECTIONS Due at time of service by: [ ] card [ ] cash [ ] check [ ] Card-on-file authorization signed (PCI-compliant processor) Returned-payment fee: $______ Interest on balances over ____ days: ____% per month, disclosed here. Payment plans: up to four installments, no finance charge. Before collections, we send written notice and a chance to pay or arrange a plan. HARDSHIP AND SLIDING SCALE Reduced fees by documented, individualized financial-need review: [ ] available on request [ ] not offered AGREEMENT Client signature: ______________________ Date: _______________ Clinician signature: ___________________ Date: _______________ Re-signed when fees or terms change; every version stays in your record. Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample financial policy and fee agreement Scenario: a two-clinician private practice, in network with two commercial plans, employs one pre-licensed supervisee and re-issued its agreement for a July 2026 fee change. All details are fictional. Financial Policy and Fee Agreement. Harbor Lane Psychology PLLC · Effective: 07/01/2026 · Client: A.B. · Signed: 07/09/2026 Practice and clinicians: Harbor Lane Psychology PLLC, Columbus, Ohio. L. Whitman, PhD, licensed psychologist, Ohio license 7712. J. Osei, MA, psychology trainee, practicing under the supervision of L. Whitman, PhD; supervised services are billed under Dr. Whitman according to plan rules. Fees: Intake assessment (90791) $210. Individual session, 55 minutes (90837) $180. Individual session, 45 minutes (90834) $150. Telehealth is billed at the same rates. Fees are reviewed each July; you receive 30 days written notice and a new agreement to sign before any change applies. Insurance and your responsibility: We are in network with two commercial plans and bill them directly. You owe your copay, coinsurance, and any unmet deductible at time of service. Amounts our network contracts require us to write off are never billed to you; only balances your plan assigns as your responsibility are. Out-of-network clients pay in full at time of service and receive a superbill on request. Cancellation and missed appointments: We ask for 48 hours of notice. Late cancellations and missed appointments carry a $90 fee, billed to you directly and never to your insurance. The policy applies equally to every client of the practice; it is not applied to Medicaid members, and genuine emergencies are considered case by case. Payment, balances, and collections: Payment is due at time of service. With your written authorization we keep a card on file, stored by our PCI-compliant payment processor; the practice never keeps card numbers in your chart. Returned payments carry a $25 fee. Balances more than 60 days old accrue simple interest at 0.5% per month, as disclosed here. Payment plans are available: up to four installments with no added charge. Before any account is referred to collections, we send written notice and give you 30 days to pay or arrange a plan. Hardship: A reduced fee is available through a documented, individualized financial-need review; ask either clinician. We do not advertise discounts or routinely waive plan cost-sharing. Agreement: Signed A.B. (client), 07/09/2026. Signed L. Whitman, PhD (for the practice), 07/09/2026. This version replaces the agreement signed 01/15/2026; both copies remain in the record. This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works Only patient-responsibility balances ever reach the client. The agreement promises contractual write-offs are never billed, which is exactly what remittance group codes will later enforce, so the policy and the payer contract cannot collide. The no-show clause has the shape Medicare permits. Disclosed in advance, a set fee, billed directly to the client and never to a plan, applied equally to every client, and switched off for Medicaid members. The payment plan stops at four installments with no added charge. That keeps a courtesy inside the practice instead of turning it into consumer credit with federal Truth in Lending disclosure duties. Interest and collections are disclosed up front, with notice and a chance to pay before enforcement. The sequence tracks what ethics codes require, so the practice can actually use its own terms. The supervisee's supervisor is named, licensed, and tied to billing. The money paperwork matches the supervision and credentialing paperwork, which is where training practices usually get burned. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The signed agreement is part of the client's record: keep every signed version, not just the current one, because a fee dispute is judged against the version in force when the charge arose, and produce copies through your normal release of information workflow. On retention, do not let the HIPAA six-year rule mislead you: 45 CFR 164.530(j) covers HIPAA compliance documentation, while the clinical and billing record, this agreement included, is governed by state law, commonly 5 to 10 years and longer for minors. A few states also put fee disclosure into licensing law itself, California among them for psychologists and master's-level clinicians, so read your board's language rather than assuming the ethics code is the ceiling. Boards discipline around this document in predictable ways: fees not disclosed in advance, missed-session charges the client never agreed to, and collections launched without the notice the ethics codes promise. For insured care, the payer contract outranks anything the client signs. A remittance that labels a balance CO makes it a contractual write-off no financial policy can convert into client debt, Medicaid providers must accept the plan payment plus allowed cost-sharing as payment in full, and routine waivers of federal-program copays sit inside a standing OIG fraud alert unless they follow a documented, individualized hardship determination. If you see uninsured or self-pay clients, federal law adds a separate document to the packet: a good faith estimate of expected charges, a one-way disclosure rather than a contract, which does not replace this agreement. The signed form is the convention; the money rules around it are the law. When coverage itself is in question, that argument lives in a prior authorization letter , and the session-level record behind each billed charge lives in a claim-support billing note . ↑ Back to the template and downloads ## Common financial policy and fee agreement errors auditors flag The stakes are not hypothetical. Missed appointments ran 18.0% to 21.9% of mental health visits in a VA pragmatic trial covering more than 38,000 appointments, roughly double the primary care rate in the same system (Journal of General Internal Medicine, 2023), and studies of community mental health intakes report 25% to 40% of first appointments missed (American Journal of Managed Care, 2009). The unpaid-balance side is just as heavy: the CFPB's March 2022 analysis found medical debt made up 58% of all collection tradelines on credit reports, roughly $88 billion, and the 2022 KFF Health News investigation counted health care debt on more than 100 million people in America, 41% of adults. An unclear or unenforceable fee agreement converts those percentages straight into write-offs and board complaints. The BastionGPT Clinical Advisory Board sees the same errors most often in financial policy and fee agreement reviews: Balance-billing language that ignores group codes. "You owe whatever insurance does not pay" reads naturally and is unenforceable for in-network care: CO-coded balances are contractual write-offs, and billing them breaches the payer contract you signed. No-show fees pointed at the wrong payer. A missed appointment is not a covered service; a claim for one comes back denied (Medicare uses code 204), and the enforceable version bills the client directly, at a disclosed fee, under a policy applied to everyone. Advertised copay waivers. "We waive all copays" in marketing invites federal fraud scrutiny the moment it touches Medicare or Medicaid cost-sharing; the OIG's standing guidance protects only unadvertised, documented, individualized hardship decisions. Collections and interest the client never agreed to. Interest that was never disclosed in the agreement, or a collections referral with no notice and no chance to pay, is board-complaint material as much as a contract problem. A stale signature under new numbers. The fee went up in July, the agreement on file is from January, and the client legitimately owes the January price; without a re-signing routine, every fee change quietly rewrites your receivables. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on financial policies and fee agreements. Draft a complete financial policy from your fee schedule, cancellation window, and payer mix, in plain language clients can actually read. Check an existing policy for the traps boards and payer contracts punish: balance-billing language, a no-show clause that is not uniform, undisclosed interest, and installment plans that cross the four-payment line. Rewrite dense policy language at a client-friendly reading level, and produce a short intake summary of what clients will owe and when. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a signed financial policy legally required? No statute in the US, Canada, or Australia requires a signed financial policy. The obligation comes from ethics codes and contract law: APA Standard 6.04 expects an agreement on compensation "as early as is feasible," the ACA and NASW codes carry parallel duties, and a handful of state licensing laws, California's among them, require written fee disclosure. Treat the signature as your primary defensible instrument anyway: it is what makes cancellation fees, interest, and collection terms enforceable as contract terms rather than surprises. Can my policy make clients pay whatever insurance does not cover? Not for in-network care. When a payer processes a claim, each unpaid amount carries a group code: CO, contractual obligation, means the balance is your write-off under the network contract and cannot be billed to the client no matter what they signed, while PR, patient responsibility, marks the copays, coinsurance, and deductible amounts that can. A financial policy that promises otherwise is unenforceable on that point and can put you in breach of the payer contract. Medicaid is stricter still: providers must accept the plan payment plus allowed cost-sharing as payment in full. Can I charge no-show fees to insured, Medicare, or Medicaid clients? Usually, with three conditions. First, the fee must be disclosed and agreed in advance, which is this document's job. Second, bill the client directly, never the plan: a missed appointment is not a covered service, and a claim for one is denied (Medicare uses code 204). Third, Medicare permits missed-appointment charges only when the same policy applies equally to all clients, so a single commercial contract that prohibits no-show fees breaks the uniformity Medicare requires; check your contracts before charging Medicare beneficiaries. Medicaid members generally cannot be charged no-show fees at all. The session-level documentation behind billed services lives in a claim-support billing note ; the no-show fee never appears on a claim. Can I charge interest or offer payment plans? Yes to both, with disclosure doing the heavy lifting. Interest on overdue balances is lawful when the agreement discloses the rate in advance and the rate respects state usury limits; ethics codes add that clients get notice before enforcement begins. Payment plans have a federal wrinkle: a written plan payable in more than four installments, or any plan with a finance charge, makes the practice a creditor under the Truth in Lending Act, with Regulation Z disclosure duties. The safe pattern is up to four installments with no added charge, in writing, dated. What are the rules for collections and credit reporting? Ethics first: give written notice and a genuine chance to pay or arrange a plan before any account is referred, and use a collector bound by the FDCPA. Credit reporting has shifted under practices' feet: the national bureaus stopped reporting paid medical collections and those under $500 across 2022 and 2023, and the CFPB rule that would have removed most medical debt from credit reports, finalized in January 2025, was vacated by a federal court on July 11, 2025, in a ruling that also read the FCRA as preempting similar state bans. The practical takeaway for your policy: promise notice and process, not specific credit consequences, because that landscape keeps moving. Is a sliding scale legal, and will it trigger an audit? For self-pay clients with no federal payer in the picture, a sliding scale is private pricing and entirely yours to set. The risk zone is federal cost-sharing: routinely or promotionally waiving Medicare or Medicaid copays and deductibles can implicate the Anti-Kickback Statute and civil monetary penalty rules under a fraud alert the OIG has kept in force since 1991. The protected path is narrow and practical: no advertising, no routine waiver, and a documented, individualized determination of financial need behind each discount. Put that process in the agreement and then follow it. How is the financial policy different from a good faith estimate or informed consent? Three documents, three jobs, one intake packet. The financial policy is a two-way signed contract covering all money terms of the relationship. The good faith estimate is a one-way federal disclosure for uninsured and self-pay clients, required since January 2022, that itemizes expected charges and states on its face it is not a contract; it triggers dispute rights if billing runs $400 or more past it. Informed consent covers the nature, risks, and course of therapy; ethics codes fold fees into consent, but a consent form is not a fee contract. Present all three at the first visit, then document the clinical side in the intake note . What changes if I practice in Canada or Australia? The signature stays conventional in both countries, and the rules around it differ sharply from the US. In Canada, college standards require agreement on fees and billing before services start, and Ontario's standards have long allowed interest on overdue accounts only if the client was told at the outset; since June 20, 2024, qualifying psychotherapy and counselling therapy services are exempt from GST/HST, which changes what a compliant receipt says, and records are commonly kept 10 years, longer for minors. In Australia, bulk billing means accepting the Medicare benefit as full payment, so no gap can be charged on a bulk-billed session; standard-form cancellation clauses have been exposed to unfair-contract-term penalties since November 2023, even where the clause is transparent; and NDIS providers can claim short-notice cancellations only where the service agreement allows it, inside the scheme's notice windows. Records commonly run 7 years, or to age 25 for minors. Can BastionGPT draft my financial policy and fee agreement? Yes. Give it your fee schedule, cancellation window, payer mix, and state, and it drafts the agreement in plain language, then checks the result for the classic traps: balance-billing language a payer contract forbids, a no-show clause that is not uniform, interest that was never disclosed, and installment plans that cross the four-payment line. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Prior Authorization Support Letter Release of Information (ROI) Authorization Intake Note Psychotherapy Notes Authorization Good Faith Estimate (No Surprises Act) Appointment No-Show & Outreach Note Advance Beneficiary Notice (ABN) Informed Consent for Treatment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/incident-report All note templates # Incident & Adverse Event Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An incident or adverse event report is an internal risk-management record of an unexpected event affecting a client, visitor, or staff member: a fall, medication error, elopement, assault, or suicide attempt. Staff file it with the practice or facility, separate from the clinical record. Facility and program rules often mandate it; for solo private practice it is convention. Most run 300 to 600 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The staff member who observed or discovered the event: clinicians, front-office and support staff, program managers; a supervisor or risk lead reviews Audience Practice leadership and risk management, quality and safety committees, malpractice carriers, and, where mandated, state agencies, accreditors, and oversight commissions Typical length 300 to 600 words · 15 to 30 minutes by hand (clinical team estimate) Format family Risk-management / quality-assurance record, kept outside the chart (compare: crisis note, mandated report, progress note) When it's used As soon as practicable after a fall, medication error, elopement, assault, restraint or seclusion event, self-harm event, death, privacy event, or near miss Standards context Mandated for facilities and funded programs (CMS, state, NDIS, aged-care rules); for solo private practice a risk-management convention, not law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an incident and adverse event report? An incident and adverse event report is the internal record a practice or facility creates when something unexpected happens to a client, visitor, or staff member: a fall, a medication error, an elopement, an assault, a restraint or seclusion event, a suicide attempt or death, a privacy event, or a near miss. You will also see it called an occurrence report, variance report, unusual occurrence report, critical incident report, serious incident report, or safety event report. It descends from the hospital occurrence report of the mid-century risk-management movement and was standardized by three milestones: the Joint Commission's Sentinel Event Policy (1996), the Institute of Medicine's To Err Is Human report (1999), which estimated up to 98,000 hospital deaths a year from medical error, and the Patient Safety and Quality Improvement Act of 2005, which created Patient Safety Organizations and a conditional federal privilege for patient safety work product. What defines the document is where it lives: outside the clinical record. New York's mental health incident-management regulation states the split in black letter: incident-related documents are quality assurance records "maintained separately from the patient's clinical record," while the clinical impact of the event must be recorded in the chart ( 14 NYCRR 524.9 ). The report is an organizational account written for risk management and quality improvement; the chart keeps the client's condition, assessment, and care. That is also what separates it from its neighbors: a crisis note documents a clinical crisis contact and belongs in the chart, and mandated reporting documentation records a legal duty discharged to an outside authority, also in the chart. One event can generate several documents; this is the only one the client's record should never mention. ## Who uses incident and adverse event reports and when Facility-based and publicly funded settings use them because they must: hospitals and inpatient units under Medicare conditions of participation and state rules, community programs licensed by state behavioral health agencies, organizations accredited under Joint Commission or CARF incident-management standards, and NDIS and aged-care providers in Australia under their reportable-incident regimes. Group practices and solo clinicians use them because their malpractice carriers ask them to: a short internal form, completed the same day, is the version of events written while memory is fresh rather than two years into a claim. The trigger is any unexpected safety-relevant event, and near misses count. When the event is a clinical crisis, the chart documentation comes first (see the crisis note ); the incident report is the organizational record filed alongside, never instead. ## Incident report structure: what goes in each section Event basics. Date, exact time, precise location, and the event type: fall, medication error, elopement, aggression, restraint or seclusion, self-harm event, death, privacy event, or near miss. Pitfall: "during the afternoon shift" instead of a clock time. Regulatory clocks run from when the organization knew, and a report that cannot establish when that was cannot prove a deadline was met. Persons involved and witnesses. Who was affected (client, visitor, or staff member) and every person who saw any part of the event, with role and contact details. Pitfall: the unlisted witness. By the time anyone investigates, the visitor who saw the fall is unreachable and the report is the only account left. Factual narrative. First person, in sequence, only what the reporter saw and heard, with the words people used. Pitfall: conclusions and blame. "Client was being difficult" or a guess at the cause converts a safety record into an admission drafted for the opposing attorney; describe the observable sequence instead. Immediate actions and care provided. The assessment offered, first aid given, the EMS decision, who examined the person, and what the person chose. Pitfall: an event with no documented response. Care that was given but never recorded reads as neglect when the record is the only witness. Injury and outcome. Apparent injury or "none observed," plus a follow-up window and who owns the reassessment. Pitfall: letting "no injury" close the file at the scene. Some injuries surface days later, and the strongest reports anticipate that with a planned check. Notifications made. Each person and agency notified, with date, time, and by whom: supervisor or practice owner, family or guardian where appropriate, and any regulator, agency, or carrier your setting requires. Pitfall: the missed clock. CMS expects a restraint or seclusion death reported by close of the next business day, the NDIS and Australian aged-care regimes run 24-hour deadlines for the most serious categories, and Saskatchewan gives three business days. Reporter identity and review. The person who observed the event writes and signs the report, with role and date; a supervisor reviews and dates the review. Pitfall: secondhand authorship. A report written by someone who was not there is hearsay layered on hearsay, and its evidentiary value drops with each retelling. Routing and the chart cross-walk. Where the report is filed (risk management or practice QA files) and a reminder of what goes in the chart instead: the client's condition, assessment, and care. Pitfall: writing "incident report filed" in a progress note. The chart reference ties the protected record to the discoverable one and can waive the separation that QA privilege depends on. ### Blank template (copy and adapt) INCIDENT / ADVERSE EVENT REPORT (internal QA record; file apart from the clinical record) Practice/program: ____________________ Report #: ______________ EVENT Date: __________ Time: ________ Exact location: ______________ Type: [ ] fall [ ] medication error [ ] elopement [ ] aggression or assault [ ] restraint or seclusion [ ] self-harm or suicide attempt [ ] death [ ] privacy event [ ] near miss [ ] other: _____________________________________ PERSONS INVOLVED Person affected (initials): __________ [ ] client [ ] visitor [ ] staff Witnesses (name, role, contact): ___________________ ________________________________________________________________ FACTUAL NARRATIVE (first person, in sequence; only what was seen and heard; no conclusions, blame, or speculation) ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ IMMEDIATE ACTIONS AND CARE PROVIDED Assessment offered / first aid: ________________________________ EMS called: [ ] yes [ ] no Examined by: ______________________ INJURY / OUTCOME Apparent injury: [ ] none observed [ ] described here: ________ ________________________________________________________________ Follow-up window and who reassesses: ___________________________ NOTIFICATIONS MADE (person or agency, date, time, by whom; note any mandatory deadline for your setting) Supervisor / practice owner: ___________________________________ Family or guardian (if applicable): ____________________________ Regulator, agency, or carrier (if required): ___________________ REPORTER Name and role (the person who observed): _______________________ Signature: ____________________ Date/time completed: __________ Supervisor review: ____________________ Date: _________________ ROUTING: file with risk management or practice QA files, apart from the clinical record. Chart the clinical impact (condition, assessment, care) in the clinical record; do not reference this report there. Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample incident and adverse event report Scenario: an adult client trips on an entry mat in a group-practice waiting room; the front-office coordinator who saw it happen writes the report the same afternoon. All details are fictional. Incident / Adverse Event Report. Cedar Grove Counseling Associates · Report #: 2026-014 · Internal QA record, filed apart from the clinical record Event: 07/16/2026, 2:47 PM, front waiting room near the check-in window. Type: client fall. Persons involved: Client R.L., 58 (initials per practice policy on this form). Witness and reporter: D. Marsh, front-office coordinator. Responder: T. Okafor, LPC. Factual narrative (D. Marsh): At approximately 2:47 PM I was at the front desk and saw R.L. walking from the check-in window toward the seating area. R.L.'s left shoe caught the corner of the entry mat. R.L. fell forward, landing on both knees and both hands. I did not see R.L.'s head contact the floor or any furniture. R.L. said "I'm fine, my shoe caught the mat." I stayed with R.L. and asked T. Okafor to come to the waiting room at 2:48 PM. Immediate actions and care: T. Okafor helped R.L. to a chair at 2:49 PM and offered an assessment. R.L. reported mild soreness in both knees, rated 2 of 10, and no other pain. Skin intact, no swelling observed. R.L. declined emergency services when offered. Ice pack provided and accepted. R.L. chose to stay for the scheduled 3:00 PM appointment and walked to the office without apparent difficulty. Injury / outcome: No injury observed beyond reported knee soreness. R.L. given written guidance to monitor and to contact the practice or a physician if pain increases. Front office to check in with R.L. at the 07/23/2026 visit. Notifications: Practice owner C. Hale, PhD, notified in person by D. Marsh on 07/16/2026 at 3:35 PM. Malpractice carrier notified administratively per practice policy on 07/17/2026; no claim asserted. C. Hale confirmed no state or program reporting requirement applies to this outpatient setting for this event. Chart cross-walk: T. Okafor documented the reported soreness, the brief assessment, and the session's clinical content in the clinical record. This report is filed with practice QA records and is not referenced in the chart. Reporter: D. Marsh, front-office coordinator (observer). Signed 07/16/2026, 4:10 PM. Supervisor review: C. Hale, PhD, 07/17/2026. This sample is fictional and for educational purposes. It does not describe a real client, practice, or event. ↑ Back to the template and downloads ### Why this sample works The person who saw it wrote it. First person, in sequence, with the client's own words; no cause assigned, no blame, nothing the reporter did not observe. Exact times anchor every step. Event, response, and each notification carry clock times, which is how every regulator's deadline is measured. The response is documented as fully as the event. Assessment offered, care given, and the client's declinations recorded, which is what defends the practice later. "No injury" does not close the file. The outcome carries a follow-up window and an owner, so an injury that surfaces later meets a record that anticipated it. The split is visible on the form. The clinical impact went to the chart, the report went to QA files, and the chart does not mention the report, so the separation that protects the record is part of the record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Write the report as if people you did not write it for will read it, because they may. Its protections are real but conditional: the federal Patient Safety Act privilege attaches only to material created inside a patient safety evaluation system for reporting to a listed Patient Safety Organization ( 42 CFR Part 3 ), and state quality-assurance privileges vary widely. The Kentucky Supreme Court held in Tibbs v. Bunnell (2014) that information normally contained in an incident report was not privileged under the federal act and could be discovered, while an Illinois appellate court upheld the privilege for properly channeled PSO material in 2018; Florida's constitution goes further and gives patients a right to obtain records of adverse medical incidents. Access runs differently than the chart, too: an incident report ordinarily sits outside the HIPAA designated record set ( 45 CFR 164.501 ), so the client right of access usually does not reach it, Florida again excepted. The working rules that follow: objective facts only, completed promptly, kept out of the chart, routed per policy. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The reporting duties around the form are a separate layer, and for facilities and funded programs they are law with unforgiving clocks. Medicare hospitals must report a death associated with restraint or seclusion to CMS by close of business the next business day after learning of it, and must note the date and time of that report in the medical record ( 42 CFR 482.13(g) ). NDIS providers in Australia notify the Commission within 24 hours for the most serious categories; aged-care providers report Priority 1 incidents within 24 hours under rules that commenced 1 November 2025; Saskatchewan requires written notice within three business days and a final investigation report within 60 days; Quebec makes declaring every incident or accident mandatory on the AH-223 form and requires disclosure to the user after any accident with consequences. None of this touches a claim: an incident report is not billed, has no procedure code, and has no denial codes, which is why billing-oriented template sites rarely cover it well. For a solo outpatient practice, no statute reviewed for this page requires an internal incident report at all; it is risk-management convention, and your malpractice carrier's guidance is the document to follow. When an incident involves a death, a report to an authority, or a likely claim, consult your attorney or board; state rules vary. The clinical side of the same event belongs in the chart: a crisis contact in a crisis note , a duty-to-warn action in mandated reporting documentation , and the client's condition and care in the progress note . ↑ Back to the template and downloads ## Common incident report errors auditors flag The dominant failure is silence. HHS Office of Inspector General work found hospital incident reporting systems captured an estimated 14 percent of patient-harm events experienced by Medicare beneficiaries (January 2012), and its follow-up study found 49 percent of sampled 2023 harm events entirely absent from hospital incident-reporting and surveillance systems, with another 16 percent indeterminate (OEI-06-18-00401). Timeliness fails almost as often: Saskatchewan's Provincial Auditor reported in 2024 that 62 percent of initial critical-incident notices reached the Ministry later than the three-business-day requirement, and 90 percent of final reports missed the 60-day deadline. The Joint Commission's 2024 sentinel event data shows behavioral health settings carry a distinct profile: suicide and self-inflicted injury led at 41 percent of reported behavioral health events, ahead of falls at 21 percent and treatment delays at 19 percent. The BastionGPT Clinical Advisory Board sees the same errors most often in incident report reviews: The event never becomes a report. In the OIG's 2012 study, 62 percent of unreported harm events went unreported because staff did not perceive them as reportable. The fix is a practice-specific list of reportable events, with near misses on it, reviewed with every new hire. Conclusions where facts belong. "Client was being difficult," a guess at the cause, or an assignment of fault converts a quality-assurance record into an admission drafted for the opposing attorney. The narrative is what was seen and heard, in sequence, by the person who saw it. The chart references the report. "Incident report filed" in a progress note ties the protected record to the discoverable one. Risk-management carriers and professional associations state the rule plainly; the APTA's version reads "Do not enter your incident report into the patient's or client's chart." The convention is not statute, but its logic is sound: the chart gets the clinical impact, the report stays with QA. The clock runs out. Deadlines are setting-specific and short: close of the next business day for a CMS restraint or seclusion death report, 24 hours for the most serious NDIS and Australian aged-care categories, three business days in Saskatchewan. Late filing is a finding on its own, as the Saskatchewan audit's 62 and 90 percent figures show. "No injury" closes the file. No follow-up window, no supervisor review, no trend analysis. Medicare's hospital QAPI condition (42 CFR 482.21) and accreditor standards expect adverse events to be tracked and analyzed, not archived; even a two-clinician practice can review its handful of reports once a year. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on incident and adverse event reports. Draft a complete, objective incident report from your bullets or dictation, with the facts in sequence and the conclusions left out. Check a draft before filing: blame language, speculation, missing times, missing notifications, and an outcome with no follow-up plan. Draft the separate progress note that records the event's clinical impact without referencing the report, so the two records stay properly apart. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does an incident report go in the client's chart? No. It is filed with risk management or the practice's QA files, separate from the clinical record; New York's mental health regulation codifies exactly that split, and risk-management guidance across professions repeats it. The chart gets the clinical impact of the event: the client's condition, your assessment, and the care provided, documented in the progress note as usual. Do not write "incident report filed" in the chart; the clinical note should stand on its clinical content alone. Is an incident report legally required? It depends on the setting, and the difference is the most misunderstood thing about these reports. Facility and program rules are law: Medicare hospitals must report restraint and seclusion deaths to CMS, New York licensed programs report through the OMH and Justice Center regimes, NDIS and aged-care providers in Australia carry 24-hour clocks for the most serious categories, and several Canadian provinces mandate critical-incident reporting. For a solo outpatient private practice, no statute reviewed for this page requires an internal incident report; it is risk-management convention backed by your malpractice carrier's guidance. What counts as a reportable incident in a mental health practice? No universal list exists; each regime defines its own, and that absence is worth knowing in itself. Common categories: falls, medication errors, elopement from a program, aggression or assault, restraint or seclusion, suicide attempts and deaths, abuse or neglect allegations, privacy events, and near misses. The Joint Commission defines a sentinel event as a patient safety event resulting in death, permanent harm, or severe temporary harm, and expects accredited organizations to define and review their own list. A practical private-practice threshold: any client death, suicide attempt, serious injury, restraint use, or event likely to generate a complaint or claim. Can a client get a copy of an incident report? Generally no. An incident report is an administrative quality-assurance record, ordinarily outside the HIPAA designated record set, so the right of access that covers the chart does not usually reach it. Florida is the standing exception: its constitutional Amendment 7 gives patients a right to records of adverse medical incidents. Disclosure is a separate duty from access: Quebec and Ontario law require telling the client what happened after a harmful incident, and Australia's open disclosure framework sets the same expectation, so keeping the report out of the record never means keeping the client in the dark. Is an incident report privileged, or can it be discovered in a lawsuit? Treat it as potentially discoverable. The federal Patient Safety Act privilege applies only to material created within a patient safety evaluation system for a listed Patient Safety Organization, and it expressly does not protect the medical record or information kept outside that system. Courts split: Kentucky's Supreme Court allowed discovery of incident-report information in Tibbs v. Bunnell (2014), while an Illinois appellate court upheld the privilege for properly channeled PSO material in 2018, and state QA privileges vary as much. The defensible posture: objective facts only, no conclusions, filed per policy, never referenced in the chart, and questions to your carrier or attorney before anything is produced. What deadlines apply after a serious incident? Whatever your tightest applicable regime says, and they are short. Medicare hospitals: a restraint or seclusion death reported to CMS by close of the next business day. Joint Commission accredited organizations: a root cause analysis within 45 business days of becoming aware of a sentinel event. NDIS providers: 24 hours for the five most serious categories. Australian aged care: Priority 1 incidents within 24 hours. Saskatchewan: written notice in three business days and a final report in 60 days. Solo practices with no mandated regime should still complete the internal report the same day; contemporaneous is the standard that holds up. A client died by suicide. What do I document, and where? Two records, two purposes. In the clinical record, document the facts contemporaneously: what you knew and when, your risk assessments and the reasoning behind the level of care, contacts made, and consultation sought. Never alter earlier notes; anything added later is dated as an addendum. Complete your organization's critical-incident report separately, per policy and any state program rule (some states require licensed programs to report a client death within 24 hours). If a safety plan existed, it stays in the chart as part of the record of care. Involve your malpractice carrier and attorney early, and use postvention support for yourself and your team; losing a client is one of the hardest events in clinical practice. How is an incident report different from a crisis note or a mandated report? By where it lives and who it serves. A crisis note is chart documentation of a clinical crisis contact and the intervention delivered. Mandated reporting documentation is chart evidence that a legal reporting duty was met. The incident report is the organizational record of the event itself, kept outside the chart for risk management and quality assurance. One event can generate all three: a client's threat during a session might produce a crisis note, a duty-to-warn record, and an agency incident report, each in its own place. Can BastionGPT write an incident report? Yes. Give it the facts as bullets: what you saw, times, who was present, what you did, and who you notified, and it drafts an objective, first-person report with the conclusions left out, then checks it for blame language and missing elements. It can also draft the separate progress note that records the clinical impact without referencing the report. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Crisis Psychotherapy Note Mandated Reporting & Duty-to-Protect Documentation Safety Plan Suicide Risk Assessment Claim-Support Billing Note Psychotherapy Progress Note Records Request & Chart Copy Packet Workers' Compensation Initial Report Glasgow Coma Scale (GCS) Documentation Braden Scale Documentation RASS Score Documentation Morse Fall Scale Documentation MEWS & NEWS2 Early Warning Score Documentation Pressure Injury Staging Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/records-request All note templates # Records Request & Chart Copy Packet: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A records request and chart copy packet is the set of documents a mental health practice produces when a client asks to inspect or copy their record: a cover letter, an itemized fee notice, a denial letter when one applies, and an internal log. US practices must act within 30 calendar days under 45 CFR 164.524. A typical packet runs 300 to 700 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Privacy officers, records and HIM staff, practice managers; a licensed clinician only for an endangerment-based denial Audience The client or their personal representative, any third party they direct records to, OCR investigators, your own compliance log Typical length 300 to 700 words across the packet · 20 to 40 minutes by hand (clinical team estimate) Format family Administrative response packet (compare: release of information authorization, record amendment response, subpoena response) When it's used Whenever a client asks to inspect or copy their record; the US response clock is 30 calendar days from receipt Standards context Timeline, fee cap, denial mechanics, and documentation are law (45 CFR 164.524); the cover letter format and clinician signature are convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a records request and chart copy packet? A records request and chart copy packet is the set of documents a practice generates when a client asks to inspect or obtain a copy of their own health record: a cover or response letter, a cost-based fee notice, a denial or partial-denial letter where one applies, and an internal log entry proving what happened and when. No standard-setting body invented it. The packet crystallized around the HIPAA Privacy Rule, in effect for most covered entities since April 14, 2003, and it is the operational output of the right of access at 45 CFR 164.524, which reaches everything in the "designated record set" defined at 45 CFR 164.501. Clinicians, HIM staff, and payers also call it an ROI response packet, chart copy, records release packet, access request response, or right-of-access response. The load-bearing distinction is between an access request and an authorization-based disclosure, because confusing the two is the most common and most expensive error in this corner of practice administration. When a client exercises their own right, or directs a copy to a third party in a signed writing, the 30-calendar-day clock and the reasonable, cost-based fee limit apply. When a third party wants records for its own purposes under a signed release of information authorization , neither the access clock nor the fee cap governs, and since the January 23, 2020 Ciox Health v. Azar ruling the fee limit also no longer extends to third-party directives in non-electronic formats. One more boundary saves grief: psychotherapy notes sit outside the access right only when they are kept separate from the chart. Commingle them with progress notes and they are part of the accessible record, with their own authorization rules for anything beyond. ## Who uses records request packets and when Any practice that holds records receives access requests: solo therapists, group practices, community clinics, hospital HIM departments, and training clinics where requests route through a supervisor. Processing is administrative by design: HIPAA lets records staff or a privacy officer handle a routine response without a clinician signature, and a licensed clinician enters only when a denial rests on professional judgment, such as likely endangerment. Reach for this packet when the client themselves is asking, when a lawyer holds a client-signed directive naming where the copy goes, and for the portal-era request that arrives as an email. When a payer, employer, or another provider wants records for its own use, that is a disclosure under a release of information , not an access response, and different fee and timing rules apply. ## Records request packet structure: what goes in each section Request intake and the log entry. Date-stamp the request the moment it arrives, whoever receives it: the 30 calendar days run from receipt, even when the chart lives with a business associate and even while a permitted fee is unpaid. Verify identity by a reasonable method matched to how the request arrived, then open the log entry: date received, requester, records sought, format requested, due date. Pitfall: parking the clock behind a ritual. Requiring your own form, an office visit, or notarized ID before you count day one is exactly the barrier OCR's access cases punish; you may require a written request, but the received date never moves. Scope: the designated record set. Everything used to make decisions about the client is requestable: progress notes , medication records, test results, treatment plans, discharge summaries , and the billing record . Two carve-outs only: psychotherapy notes kept separate from the chart, and information compiled in anticipation of litigation. Pitfall: treating "therapy notes" as a magic phrase. The exclusion follows storage, not labels; process notes filed in the progress-note stream are part of the accessible record. The cover or response letter. State the date the request arrived, the date of this response, what is enclosed, and the format provided. Provide the copy in the form and format requested whenever it is readily producible; if you used the one 30-day extension, the written notice with reasons and an expected date must have gone out inside the first 30 days. Pitfall: substituting the format you prefer. Refusing to send records by email when that is what the client asked for was the most common failure in the Ciitizen access reviews, and a summary may replace a copy only when the client agreed in advance, fee included. The fee notice. For a client's own access, a reasonable, cost-based fee only, itemized as labor for copying, supplies, postage when mailed, and a summary the client agreed to, with an estimate up front; a flat $6.50 is a recognized option for electronic copies. Prepayment may be requested, and the clock still runs. Pitfall: importing state per-page schedules. Those govern other disclosures, not access, per-page math has no place in an electronic copy pulled from an EHR, and retrieval, search, and system costs are never chargeable. The denial or partial-denial letter. Deny narrowly, in plain language, and name the ground. An endangerment denial must rest on a licensed professional's judgment that access is reasonably likely to endanger someone's life or physical safety, and the letter must offer review by a different licensed professional who took no part in the original decision, plus how to complain to the practice and to OCR. Pitfall: the blanket "mental health records" denial. No such ground exists; expected distress does not qualify, and the unreviewable grounds are a short, specific list. The outcome entry and retained documentation. Close the loop in the log: action date, format provided, fee charged, outcome, any extension. Keep the packet documentation six years under 45 CFR 164.530(j), alongside the standing record of your designated record sets and the titles responsible for requests that 164.524(e) requires. Pitfall: a response with no log. In an OCR complaint the log entry is the only evidence that the action date beat day 30; without it a timely response is unprovable. ### Blank template (copy and adapt) RECORDS REQUEST AND CHART COPY RESPONSE PACKET Four parts: log entry, cover letter, fee notice, denial letter if a ground applies. US clock: 30 calendar days from receipt. REQUEST LOG ENTRY Date received: ____________ Received by: ______________________ Requester: [ ] client [ ] personal representative [ ] third party directed in a signed writing (attach it) Identity verified by (reasonable, no barriers): ________________ Records requested / date range: ________________________________ Format requested: [ ] electronic [ ] paper [ ] inspection Response due (30 calendar days): _______________________________ Extension: [ ] none [ ] notice sent ________ new due _________ COVER LETTER (send with the copies) Request received on: ____________ This response sent: _________ What is enclosed and the format provided: ______________________ ________________________________________________________________ Items withheld and the plain-language ground, if any: __________ Psychotherapy notes are kept separate from the chart and are not part of the record set released. Contact for questions or complaints (practice and OCR): ________ FEE NOTICE (client access: cost-based only) Labor to copy: $_____ Supplies: $_____ Postage: $_____ Summary, agreed in advance (optional): $_____ Total: $_____ [ ] Estimate given in advance [ ] Prepayment requested Never charge retrieval, search, or system costs for access. DENIAL / PARTIAL DENIAL (only when a ground applies) Ground, in plain language: _____________________________________ [ ] Unreviewable ground [ ] Reviewable: endangerment judgment by a licensed professional: ____________________________________ Review offered by a different licensed professional: ___________ How to complain to the practice and to OCR: ____________________ OUTCOME Completed on: ___________ Format: ___________ Fee: $_________ Logged by: _____________________________________________________ Keep this documentation 6 years (45 CFR 164.530(j)). Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample records request and chart copy packet Scenario: an adult client asks by portal message for a complete electronic copy of her chart; a two-clinician private practice responds in 11 days at no charge. All details are fictional. Records request and chart copy packet. Alder Point Counseling LLC · Client: R.V. · Request received: 07/06/2026 · Response sent: 07/17/2026 Log entry (opened 07/06/2026): Portal message from R.V. requesting a complete copy of her record, January 2024 to present, electronic format. Received by C. Marsh, office manager. Requester: the client herself. Identity verified by portal credentials and date of birth on file; no additional ID required. Response due 08/05/2026, 30 calendar days from receipt. No extension used. Routed to the privacy officer the same day. Cover letter (sent 07/17/2026 by secure portal): Dear R.V.: We received your records request on July 6, 2026. Enclosed is a complete electronic copy of your designated record set from January 5, 2024 through July 15, 2026, provided as a searchable PDF through the secure portal, the format you requested. It includes progress notes, your treatment plan and reviews, medication and care coordination entries, test and screening results, and billing records. Psychotherapy notes are kept separate from your chart and are not part of the record set; nothing else has been withheld. If a file does not open, or you would also like a paper copy, contact our privacy officer, D. Okafor, at the practice number. You may raise concerns with us or with the HHS Office for Civil Rights; asking never affects your care. Fee notice: Electronic copy exported from the EHR: no copying labor beyond the export, no supplies, no postage. Fee charged: $0.00. Practice policy: cost-based fees only, itemized and estimated in advance; retrieval and search are never charged. Outcome (logged 07/17/2026): Completed on day 11 of 30. Format: searchable PDF by portal, as requested. Fee: $0.00. Items withheld: none beyond separately kept psychotherapy notes. Logged by C. Marsh; documentation retained six years per 45 CFR 164.530(j). This sample is fictional and for educational purposes. It does not describe a real client or practice. ↑ Back to the template and downloads ### Why this sample works The action date is provable. Receipt 07/06, response 07/17: the log shows day 11 of 30, which is the exact fact OCR right-of-access cases turn on. Format follows the request. Electronic asked, searchable PDF delivered through the portal; form-and-format refusals were the leading remaining failure in the Ciitizen access reviews. The fee is cost-based, itemized, and lands at zero. No retrieval or search charges and no per-page math on an EHR export, which is where fee findings concentrate. The psychotherapy-notes carve-out is earned. The letter states the notes are kept separate, so the withholding rests on storage, as the rule requires, not on a label. Verification stayed reasonable. Portal credentials plus date of birth on file, with no form ritual or ID visit standing between receipt and day one of the clock. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations This packet is a live enforcement target, not paperwork theater. OCR's Right of Access Initiative, running since 2019, reached its 54th enforcement action in December 2025, and small behavioral health practices sit on the settlement list by name: Patricia King MD & Associates at $3,500, Wise Psychiatry at $10,000, Riverside Psychiatric Medical Group at $25,000, Arbour Hospital at $65,000, Beth Israel Lahey Health Behavioral Services at $70,000. Right of access was the second most common HIPAA complaint category in 2024, with 541 complaints. On outsourced records handling, OCR Acting Director Anthony Archeval put it plainly: "A covered entity's responsibility to provide timely access continues." Your defense is the log: receipt date, action date, format, fee, outcome. Retention then splits in a way that trips practices: the access documentation is HIPAA compliance documentation kept six years under 45 CFR 164.530(j), while the chart itself follows state retention law, commonly 5 to 10 years and longer for minors. Where state access law is more protective, with a shorter deadline or a lower fee, the state rule governs; read your own statute and board rules rather than assuming the federal floor is the ceiling. Two more realities shape the frame. First, nothing here is payer-adjudicated: no CARC or RARC code exists for an access response because it is not a claim, and the permitted fee is paid by the individual, so no billing office will catch a fee mistake later; the discipline lives in this packet or nowhere. Electronic records add the information-blocking overlay at 45 CFR Part 171, under which practices that interfere with access to electronic health information face Medicare disincentives. Second, the ground is moving, so date your policies. As of July 2026, the 2021 federal proposal to shorten the response window from 30 days to 15 remains unfinalized, with a new access-timing rule projected for late 2026, and the 2024 reproductive-health privacy rule was vacated nationwide on June 18, 2025, returning practices to pre-December-2024 handling. The cover letter format is a convention; the timeline, the cost-based fee, the denial mechanics, and the documentation are the requirement. When a third party seeks records for its own purposes, step out of this packet and into the release of information workflow, and route requests aimed at separately kept process notes through a psychotherapy notes authorization . ↑ Back to the template and downloads ## Common records request errors auditors flag The compliance record on this document is unusually well measured. A 2018 JAMA Network Open study of 83 top US hospitals found request handling inconsistent between authorization forms and phone staff, with some hospitals quoting more than $500 for a 200-page record against the $6.50 federal flat option for electronic copies, and 8% out of step with state processing-time limits. The Ciitizen patient-record scorecards trace the same problem over time: more than half of 210 providers scored in 2019 were initially out of compliance, 56% of telephone responses indicated noncompliance, and by the third scorecard of 820 providers the noncompliant share had fallen to 27%, with form-and-format failures the leading remaining error at 65% of the noncompliant group and only 6% still charging fees. OCR's 2024 numbers keep right of access the second most common HIPAA complaint category at 541. The BastionGPT Clinical Advisory Board sees the same errors most often in records request response reviews: Billing a client's own request like a third-party disclosure. State per-page schedules and contract rates belong to authorization-based disclosures; a client's access request is capped at cost. The confusion cuts both ways: since the 2020 Ciox ruling, the cap no longer extends to third-party directives in non-electronic formats, so quoting the $6.50 figure to a records vendor is the mirror-image mistake. Charging for retrieval, search, or per-page math on an EHR export. Labor to copy, supplies, postage, and an agreed summary are the entire permitted list; per-page fees on electronic copies are the error OCR guidance singles out. Letting the clock wait. Day one is receipt, not fee payment, not the completed form, not the ID visit. Prepayment may be requested, but the action date must still land inside 30 calendar days, extension included only if the written notice went out in time. Blanket withholding of anything labeled mental health. The access right covers the chart. Only separately kept psychotherapy notes and litigation-compilation material fall outside it, and commingled process notes travel with the record. A denial letter with no review path. An endangerment denial needs a licensed professional's judgment, a named route to review by a different licensed professional, and complaint instructions for both the practice and OCR; a denial missing them converts one violation into several. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on records request and chart copy packets. Draft the full packet from the request details: cover letter, itemized cost-based fee notice, and, where a ground genuinely applies, a denial letter carrying the required review language. Check a drafted response before it goes out: receipt and action dates inside 30 days, format matching the request, fee basis, psychotherapy-notes handling, and complaint paths. Convert each completed request into a dated log entry that satisfies the documentation duties at 45 CFR 164.524(e) and 164.530(j). See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long do we have to respond to a records request? US practices must act within 30 calendar days of receiving the request: provide the records, deny with a proper letter, or send a written extension notice stating the reasons and the expected date, which buys one additional 30-day period at most. The clock runs from receipt even when a business associate holds the chart and even while a permitted fee is unpaid. State law can set a shorter deadline, and where it does, the shorter rule governs. A 2021 federal proposal to cut the window to 15 days had still not been finalized as of July 2026. How much can we charge for copies of mental health records? Only a reasonable, cost-based fee for the client's own access: labor for copying, supplies, postage when the copy is mailed, and preparation of a summary the client agreed to in advance. Retrieval, search, and system costs are never chargeable, per-page fees do not fit an electronic copy from an EHR, and a flat $6.50 is a recognized option for electronic copies. Many practices simply charge nothing; by the third Ciitizen scorecard only 6% of providers charged any fee. State per-page schedules apply to other disclosures, not to an access request. Is a records request the same as a release of information authorization? No, and the difference sets both the fee and the deadline. A records request is the client exercising their own right of access, or directing a copy to a third party in a signed writing: the 30-day clock and the cost-based fee cap apply. A release of information authorization permits the practice to disclose records to a third party for that party's own purposes: no access clock, and fees set by state schedules or agreement. Since the 2020 Ciox ruling, the access fee cap also does not extend to third-party directives in non-electronic formats. Do we have to release psychotherapy notes? Only if they are not truly psychotherapy notes. Process notes kept separate from the rest of the chart are excluded from the access right, and disclosing them generally takes a specific psychotherapy notes authorization . Everything else must be released: medication management, session start and stop times, test results, diagnosis, functional status, the treatment plan, symptoms, prognosis, and progress to date. Notes commingled with the progress-note stream are part of the accessible record no matter what they are called. Can we require our own request form or photo ID? You may require that requests be in writing if you tell people so, and you may offer a form, but you cannot make your form a precondition, and identity verification must be reasonable rather than a barrier: matching a portal login or date of birth on file is usually enough, while an in-person visit or notarization requirement is the kind of obstacle regulators flag. The same pattern holds elsewhere: Australia's OAIC says an entity may ask a person to use a form but cannot require it, and Ontario's PHIPA lets custodians honor oral requests. Can we deny access because release might harm the client? Only on narrow grounds, and almost never as a blanket matter. The main reviewable ground is a licensed professional's judgment that access is reasonably likely to endanger the client's or another person's life or physical safety; expected distress does not qualify. The denial letter must state the ground in plain language, offer review by a different licensed professional who took no part in the original decision, and explain how to complain to the practice and to OCR. The unreviewable grounds, such as separately kept psychotherapy notes and litigation-compilation material, are a short and specific list. What are the timelines and fees in Canada and Australia? Canada's PIPEDA gives private-sector organizations 30 days, extendable by 30 with written notice, at minimal or no cost; Ontario's PHIPA also runs on 30 days, and its regulator reads reasonable cost recovery as $30 for the first 20 pages plus 25 cents per page after that. Australia's federal APP 12 requires access within a reasonable period, which the OAIC reads as not more than 30 days, with any charge not excessive; the Victorian and NSW health-records Acts set a harder 45-day deadline, Victoria with a regulated fee schedule updated each July and NSW with no prescribed fee scale at all. How long do we keep the request documentation? Two different clocks. The access-request documentation itself, meaning the log, the letters, and the record-set designations, is HIPAA compliance documentation kept at least six years under 45 CFR 164.530(j). The clinical chart follows state, provincial, and professional rules instead: commonly 5 to 10 years for adults in the US and longer for minors, at least 10 years after last contact for Ontario psychologists, and 7 years from last service or until age 25 in Australia under the Victorian and NSW Acts and the APS code. Can BastionGPT draft our records request responses? Yes. Give it the request details, meaning what was asked for, when it arrived, the format requested, and your fee inputs, and it drafts the cover letter, the itemized fee notice, and, where a ground genuinely applies, a denial letter with the review language 45 CFR 164.524 requires, plus a log entry for your compliance file. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Psychotherapy Notes Authorization Discharge Summary Claim-Support Billing Note SOAP Note Consent for Treatment of a Minor Record Amendment & Correction Response 42 CFR Part 2 Consent Communication Log & Portal Message Note Appointment No-Show & Outreach Note Incident & Adverse Event Report Advance Beneficiary Notice (ABN) Subpoena Response & Records Affidavit HIPAA Notice of Privacy Practices Acknowledgment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/minor-consent All note templates # Consent for Treatment of a Minor: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A consent for treatment of a minor is the record showing who authorized a child's mental health care and on what legal basis: parent, guardian, custody order, or the minor under a state statute. Therapists complete it before the first session with any client under the age of majority. No federal law prescribes the form; state law governs who may consent. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Child and adolescent therapists, counselors, psychologists, intake staff; the clinician personally for a minor self-consent determination Audience The treating clinician and practice, licensing boards, custody attorneys and courts, auditors Typical length 300 to 600 words · 10 to 20 minutes by hand (clinical team estimate) Format family Consent and intake documentation (compare: informed consent, release of information, psychotherapy notes authorization) When it's used Before the first session with any client under the age of majority; revisited when custody or the minor's legal status changes Standards context The form is a convention; who may consent is state law, and HIPAA's personal-representative rule follows it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a consent for treatment of a minor? A consent for treatment of a minor is the record showing who authorized a child's or adolescent's mental health treatment, on what legal basis, and what the clinician verified before relying on it. Unlike the SOAP note , which traces to a named author and year, this document has no origin story and no standard-setting body: every practice assembles its own from general informed-consent doctrine, its state's minor-consent statutes, and privacy rules, and that absence of a canonical form is itself worth knowing. Practices call it a consent to treat a minor, an authorization to treat, a parent or guardian consent, or a custody attestation, and in most EHRs it lives inside the intake packet. Two boundaries do the heavy lifting. First, no federal law requires this form: HIPAA's personal-representative rule at 45 CFR 164.502(g)(3) decides who stands in the minor's shoes and generally hands that role to a parent or guardian, with exceptions that track state minor-consent law, but it never demands a signed document. Second, this is not the same thing as informed consent for treatment. Informed consent records that the nature, risks, and benefits of treatment were explained and accepted; the minor-consent layer records that the person accepting had legal authority to accept. And in several states the legally operative element is not a signature at all but a clinician determination: New York Mental Hygiene Law 33.21(c) and California Family Code 6924(d) both turn on reasoning the practitioner writes into the record. ## Who uses minor consent forms and when Anyone treating clients under the age of majority: child and adolescent therapists, school-linked and community clinics, group practices, and training clinics where supervisors review intake packets. The trigger is age plus session one: complete it before the first appointment, alongside the intake note and the informed consent it sits on top of. It earns its keep in three situations: separated or divorced parents, where authority follows the custody order rather than the parent in the waiting room; a teen seeking services alone under a state statute; and third-party care, where a grandparent, stepparent, or foster caregiver presents without obvious authority. Family work adds one wrinkle worth planning for: each minor in family therapy needs the authority question answered individually, not one signature stretched across siblings. ## Minor consent form structure: what goes in each section Minor and clinician identification. Open with the facts that anchor everything else: the minor's full name, date of birth, and age at intake, the clinician and credentials, and the date treatment begins. Age decides which consent pathways exist at all, so record it precisely. Pitfall: a consent dated after care has begun. Reviewers read a late signature as no consent for the sessions already delivered, and consent obtained after the fact was a counted finding in a 2022 Texas OIG behavioral hospital audit. Consenting adult and legal authority. Name the adult, the relationship, and the legal basis: parent in an intact family, joint or sole legal custody, guardianship, a court order, a caregiver authorization affidavit, or the minor's own consent under a state statute. This is the load-bearing section. The Ohio counselor and social work board, which reports it "continues to receive many complaints involving custody matters," frames discipline around whether established legal authority existed. Pitfall: treating relationship as authority. A stepparent, a grandparent who drives the child to sessions, or a parent with physical custody only may hold no legal decision-making authority at all. Custody documentation. For separated or divorced families, record the custody status, whether the order requires both parents' consent for mental health care, and whether a copy of the current order or parenting plan is on file. No statute identified requires you to hold the order; requesting it whenever there is ambiguity, and documenting that you asked, is the convention that wins disputes. Pitfall: taking one parent's account of custody. The order's language decides; "she lost custody" from the waiting room is not a legal basis. Scope of this consent. State what is being consented to: individual counseling, family sessions, telehealth, parent check-ins, and what is excluded without new consent, such as psychological testing or a records release, which runs through a separate release of information . Pitfall: a signature attached to nothing. A consent that never says what was authorized proves only that someone signed something. Minor self-consent determination (where a statute applies). When a minor consents alone, the operative record in several states is yours, not theirs: California Family Code 6924(d) requires the record to state whether and when you attempted parental contact and why involvement would be inappropriate, and New York Mental Hygiene Law 33.21(c) directs the practitioner to "fully document the reasons" for the determination. Pitfall: collecting the teen's signature and skipping the written determination. The signature is not the legally operative element; the documented clinician reasoning is. Confidentiality plan and signatures. Set expectations in writing before session one: what the consenting adult will be told, what stays between the minor and clinician, and the safety limits on both. HIPAA defers parental access to state law and professional judgment, so the plan must fit your statute rather than your preference. Close with signature and date for the consenting adult, an age-appropriate minor assent, and the clinician. Pitfall: promising the minor blanket secrecy. Parental access rights vary by state, and a promise the law cannot keep damages the treatment it was meant to protect. ### Blank template (copy and adapt) CONSENT FOR TREATMENT OF A MINOR Documents who authorized the minor's care and on what legal basis. Complete before the first session; verify authority against documents, not accounts. Adapt to your state. MINOR AND CLINICIAN Minor's name: ____________________ DOB: ________ Age: ____ Clinician and credentials: ________________________________ First session date: ______________ Service/location: ______ CONSENTING ADULT AND LEGAL AUTHORITY Name: ____________________ Relationship to minor: _________ Basis of authority (check one; attach proof where held): [ ] Parent, intact family [ ] Joint legal custody [ ] Sole legal custody [ ] Legal guardian [ ] Court order [ ] Caregiver affidavit [ ] Minor consents under state statute: __________________ Documents reviewed, with dates: ___________________________ CUSTODY STATUS (separated, divorced, or third-party care) Order requires both parents' consent: [ ] yes [ ] no Copy of current order or parenting plan on file: [ ] yes Other parent notified or consent obtained: ________________ SCOPE OF THIS CONSENT Services consented to (modality, frequency, telehealth): ___________________________________________________________ Not covered without new consent: __________________________ MINOR SELF-CONSENT DETERMINATION (where a statute applies) Statutory basis and criteria met: _________________________ Parental contact attempted, or why inappropriate: _________ ___________________________________________________________ Clinician reasoning documented in the record: [ ] yes CONFIDENTIALITY PLAN What the consenting adult will be told; what stays between minor and clinician; safety limits explained: _____________ ___________________________________________________________ SIGNATURES Consenting adult: ________________________ Date: __________ Minor assent (age-appropriate): __________ Date: __________ Clinician: _______________________________ Date: __________ Retention: keep to the age of majority plus your state's added years (example: California psychologists, 18 + 7). Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample consent for treatment of a minor Scenario: a mother brings her 14-year-old to weekly counseling for anxiety; the parents are divorced and share joint legal custody. All details are fictional. Consent for treatment of a minor. Maple Reach Counseling LLC · Minor: J.R., 14 · Completed: 07/17/2026 · First session: 07/21/2026 Minor and clinician: J.R., date of birth 03/09/2012, age 14 at intake. Clinician: T. Nguyen, LPC. First session scheduled 07/21/2026: weekly individual counseling for anxiety, in office, telehealth as needed. Consenting adult and legal authority: L.R., mother. Basis: joint legal custody under the parenting plan entered 06/2023. Plan reviewed by the clinician 07/17/2026; copy on file. The plan grants both parents joint decision-making for health care and contains no clause requiring both signatures for routine outpatient mental health treatment. Custody documentation: Order requires both parents' consent: no. Father, D.R., holds joint legal custody. Notified of the referral by phone message and follow-up email 07/18/2026 with an invitation to contact the practice; no objection received as of 07/21/2026. The practice will provide him treatment information on request, consistent with state law. Scope of this consent: Weekly individual counseling with T. Nguyen, LPC; telehealth sessions when school or transportation interferes; monthly parent check-ins limited to goals, attendance, and safety. Not covered without new consent: psychological testing, group therapy, release of records to any third party. Minor self-consent determination: Not applicable; consent provided by a parent with legal authority. Confidentiality plan: Reviewed with J.R. and L.R. together on 07/17/2026. Session content stays between J.R. and the clinician, except safety concerns the clinician will raise with a parent: risk of harm to self or others, abuse, or a court order. Parent check-ins follow the scope above. J.R. voiced understanding and agreed to begin. Signatures: L.R., mother and joint legal custodian, signed 07/17/2026. J.R. signed an age-appropriate assent 07/17/2026. T. Nguyen, LPC, attested and filed the form with the parenting plan 07/17/2026. Retention: to age 18 plus the state's added period. This sample is fictional and for educational purposes. It does not describe a real client, family, or practice. ↑ Back to the template and downloads ### Why this sample works Authority is verified, not assumed. The record names the legal basis, joint legal custody, and shows the clinician read the parenting plan on a stated date, with a copy filed. The both-parents question is answered from the order's language. The one clause that could invalidate a single signature is checked and ruled out, and the second custodian is notified, with dates. Scope runs in both directions. What was consented to and what needs new consent are both stated, so later services cannot lean on a signature that never covered them. The confidentiality plan predates session one. What parents hear, what stays private, and the safety limits are in writing before any disclosure question can arise. Every signature is dated before care begins. Consent 07/17, first session 07/21: the sequence a reviewer checks first is provable from the form itself. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations This record is read most often not by payers but by licensing boards and custody attorneys. Boards rarely publish discipline indexed to this exact violation; they fold it into unprofessional conduct, and custody-related complaints are a steady stream, by the Ohio counselor and social work board's own account. The one on-point appellate case cuts the other way: in 2024 a California appeals court reversed a psychologist's license revocation that rested partly on evaluating two minors without their father's consent, holding that a genuine emergency made that consent unnecessary, the territory the crisis note covers. Statutes vary sharply by state. Florida's Parents' Bill of Rights makes treating a minor without written parental consent a first-degree misdemeanor and a discipline ground, while California and New York authorize minors to consent alone once the clinician documents the statutory determination. Substance use treatment adds a federal layer: under 42 CFR Part 2, where state law lets a minor obtain SUD treatment alone, only the minor may consent to disclosures of those records, a rule fully enforceable since the 2024 final rule's compliance date passed in February 2026. The payer picture is quieter than most pages suggest: no Medicare, Medicaid, or commercial payer manual identified requires a who-consented form as a condition of payment, and no claim adjustment code is tied to missing minor consent. The pressure arrives through audits instead, where HHS OIG Medicaid reviews of ABA services list missing parental signatures among their dominant documentation findings. The form is a convention; the authority behind it is the law. Retention is the other place hard law bites: minors' records must generally be kept well past the age of majority, seven years past age 18 for California psychologists and until age 25 in New South Wales, Victoria, and the ACT for records made while the client was under 18. No regulation identified requires re-obtaining consent when custody changes or the minor reaches the age of consent; revisiting it then is prudence, not mandate. Route disclosures to schools, pediatricians, or the other parent through a release of information , and remember that psychotherapy notes keep their own authorization rules , even from parents. ↑ Back to the template and downloads ## Common minor consent errors boards and auditors flag Consent documentation is an audited element in behavioral health, and the numbers are specific. A 2022 Texas Health and Human Services OIG audit of a psychiatric hospital (Report AUD-22-025) found 18 of 137 required medication-consent forms were never obtained, 13 of the 119 on file were signed only after the medication had been given, and 46 of 296 psychoactive-medication orders in the sample, 15.5 percent, were not signed as required. In HHS OIG's Medicaid reviews of applied behavior analysis services, missing parental signatures sat among the dominant findings behind at least $56 million in improper payments identified in Indiana and $18.5 million in Wisconsin. The BastionGPT Clinical Advisory Board sees the same errors most often in minor consent reviews: Consent from an adult without legal authority. A stepparent, a grandparent, or a parent with physical custody only signs the form. Legal custody governs consent; residence and biology do not, and boards frame the question as whether established legal authority existed. The custody order nobody requested. The file says "parents divorced" over one signature, with no record of whether the order vests sole authority or requires both parents. When any ambiguity exists, request the order and document that you did. A teen signature standing in for the statutory determination. In minor self-consent states the record must carry the clinician's written reasoning, and in California whether and when parental contact was attempted. A signature line alone fails the statute the treatment relies on. Consent that postdates the care. The first session comes before the signature. Reviewers read a late consent as no consent for the services already delivered, the exact pattern the Texas audit counted. A confidentiality promise state law cannot keep. The minor is promised blanket secrecy, then a parent exercises access rights HIPAA defers to state law. Put the plan in writing, not the promise. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on minor consent forms. Draft the complete record from intake details: the consenting adult, the claimed basis of authority, custody documents reviewed, scope, and the confidentiality plan. Check a completed form before session one: authority named and matched to a document, the both-parents question answered, scope stated, and every date in the right order. Turn a teen's self-consent session into the written determination your statute expects, with parental-contact attempts and clinician reasoning stated the way California and New York require. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I need a signed consent form to treat a minor? The signature is convention; the authority is law. No federal rule, HIPAA included, requires a signed minor-consent form, and no payer manual identified makes one a condition of payment. What the law does demand is that the person consenting held legal authority, set by your state's statutes and any custody order, and some states go further: Florida's Parents' Bill of Rights makes treating a minor without written parental consent a first-degree misdemeanor. A signed, dated form remains the cleanest proof of authority a chart can hold. Which parent can consent when parents are divorced? Legal custody decides, not physical custody. Either parent with joint legal custody may consent alone unless the order requires both parents' agreement or grants one parent sole authority over health decisions. A parent who only sees the children on weekends may hold full legal authority, and the custodial home may hold none. When anything is ambiguous, request the order or parenting plan, read the health-decision clause, and document what you reviewed and when. Can a teenager consent to counseling without a parent? In some US states, yes, with conditions. California's overlapping statutes let a minor 12 or older consent to outpatient mental health treatment when the professional finds them mature enough to participate intelligently, one statute adding danger or abuse criteria, and the record must state whether and when parental contact was attempted. New York permits treatment on the practitioner's documented determination, with the minor signing a voluntary-services statement. In these states the legally operative element is the clinician's written reasoning, not the teen's signature, so the determination belongs in the record before treatment proceeds. Does HIPAA require a parent's signature before treating a minor? No. HIPAA decides who may exercise the minor's privacy rights, not whether a form exists. Under 45 CFR 164.502(g)(3), a parent or guardian is generally the minor's personal representative, except when the minor may lawfully consent alone and does, when a court or another authorized person consents, or when a parent agrees to a confidential relationship between the minor and the clinician. Parental access to the record then defers to state law and, where state law is silent, to a licensed professional's judgment. Nothing in HIPAA requires a signed minor-consent form. What if one parent withdraws consent after treatment starts? Unilateral withdrawal does not automatically end treatment. Where both parents hold joint legal custody, professional-liability guidance reads the decision to end treatment as belonging to both, just like the decision to start it, unless the order says otherwise. Pause, reread the order's health-decision language, and document the withdrawal conversation with dates. Where the parents deadlock, the order or a court resolves it; what ends treatment is the order's allocation of authority, not the louder phone call. Is this the same as informed consent for treatment? No. Informed consent documents that the nature, risks, and benefits of treatment were explained and accepted. The minor-consent layer documents that the person accepting had legal authority to accept, a question that only exists, and always exists, when the client is under the age of majority. Practices bundle both into the intake packet with the privacy-notice acknowledgment. Disclosures to schools, pediatricians, or the other parent are a third thing again, running through a release of information . How long do minor consent records have to be kept? Longer than adult records, by statute. The common US pattern is the age of majority plus a period: California requires psychologists, with parallel rules for MFTs and clinical social workers, to keep a minor's records at least seven years after the client turns 18. In Australia, New South Wales, Victoria, and the ACT require records made while the client was under 18 to be kept until age 25. Vendor retention tables disagree with one another, so set your schedule from your own statute or board rule and default to the longest applicable period. Who can consent to a minor's therapy in Canada and Australia? Capacity, more than age. Most Canadian provinces follow the mature-minor doctrine the Supreme Court entrenched in A.C. v. Manitoba: a minor with capacity decides. Quebec fixes the line at 14, and Ontario presumes capacity without a minimum age, adding an age-16 presumption for information decisions. Australia applies Gillick competence, and one recent change matters for separated families: since May 2024, family law no longer presumes equal shared parental responsibility, so who may consent must be checked against current orders rather than assumed. Can BastionGPT draft minor consent documentation? Yes. Give it the intake facts: who presented with the minor, the claimed relationship and custody situation, the documents you reviewed, and the planned services, and it drafts the consent record, the custody-documentation entry, and, where a statute applies, the clinician determination language with parental-contact attempts stated. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Psychotherapy Notes Authorization Intake Note Family & Couples Therapy Note Crisis Psychotherapy Note Records Request & Chart Copy Packet Record Amendment & Correction Response 42 CFR Part 2 Consent Return-to-School Letter Informed Consent for Treatment Telehealth Consent Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/record-amendment All note templates # Record Amendment & Correction Response: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A record amendment and correction response is the set of documents a practice creates when a client asks to change something in their health record: the decision letter, the appended amendment or the denial with its required elements, and the request log. Under HIPAA, US practices must act within 60 days of receipt (45 CFR 164.526). A typical decision letter runs 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Privacy officers and practice owners process and log; the clinician who authored the disputed entry judges accuracy Audience The requesting client or their personal representative, downstream providers and business associates holding the disputed information, privacy regulators, your compliance log Typical length 150 to 400 words per letter · 15 to 30 minutes by hand (clinical team estimate) Format family Administrative response letter set (compare: release of information authorization, psychotherapy notes authorization, progress note addendum) When it's used Whenever a client asks to amend or correct their record; the US decision clock is 60 days from receipt Standards context The 60-day clock, the four denial grounds, and append-only mechanics are law (45 CFR 164.526); the letter format and any request form are convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a record amendment and correction response? A record amendment and correction response is the set of documents a practice generates when a client asks to change information in their health record: the acceptance or denial decision, the letter notifying the client, the amendment appended or linked to the disputed entry, the client's statement of disagreement with any rebuttal, and the log entry proving who decided what and when. The right behind it was created by the HIPAA Privacy Rule at 45 CFR 164.526 , published in December 2000, and professional handling was standardized by AHIMA's amendment practice briefs. You will also see the pieces called an amendment request response, record correction letter, addendum, late entry, or statement of disagreement. The load-bearing distinction, muddled almost everywhere online, is between what the client requests and what the clinician fixes. A patient-initiated amendment runs on the privacy rule's 60-day clock and is policed by privacy regulators; a provider-initiated correction, addendum, or late entry is the clinician repairing their own documentation under CMS program integrity standards, judged by payers and auditors. The two share one mechanic: nothing is deleted. HIPAA's method of amending is to append or link, and integrity guidance expects original content preserved, with a single-line strike-through on paper. The edges matter too: psychotherapy notes kept separate from the chart sit outside the amendment right, with their own authorization rules , while everything in the designated record set is amendable, billing records included. ## Who uses record amendment responses and when Any practice that keeps records eventually receives an amendment request, and portals made the trigger routine: clients now read their notes, and some of what they read they dispute. Solo therapists handle the request themselves; group practices route it to the privacy officer HIPAA requires them to designate; clinics and hospital HIM departments run formal amendment queues. The processing is split by role. The privacy officer owns the clock, the log, and the letters; the clinician who authored the disputed entry judges accuracy, because only the author can say whether the note records what happened. Use this workflow when the client disputes content in the record. When they want a copy, that is an access request on a 30-day clock with different mechanics, and when a third party wants records for its own purposes, use a release of information instead. ## Record amendment response structure: what goes in each section Request intake and the log entry. Date-stamp the request on arrival: the 60 days run from receipt, not from a completed form. Log who is asking (the client or a personal representative with authority), the exact entry disputed, the change requested, and the stated reason, then calendar the due date. A practice may require writing and a reason only if its Notice of Privacy Practices announced that in advance. Pitfall: importing the access rule's 30-day clock, or the folklore six-month window. Amendment runs on 60 days with one 30-day extension, and the received date never moves. The accuracy review. Route the request to the clinician who wrote the entry, and test it against the only four grounds that permit denial: the information was not created by the practice (unless the originator is no longer available), is not part of the designated record set, would not be available for inspection under the access rule, or is accurate and complete. Pitfall: treating disagreement as a ground. The client disputing an entry is not on the list; either the record is inaccurate or incomplete and you amend, or you stand behind it and deny. The acceptance path: append or link. Identify every affected record, append the amendment or link it to the disputed entry, tell the client it is done, and obtain their agreement on who else should hear about it. Then make reasonable efforts to notify the people the client names plus any business associates who hold the information and could rely on the old version to the client's detriment. Pitfall: overwriting. The original entry stays; an amendment that erases what it corrects is an alteration, and alterations are what integrity reviewers escalate. The denial letter: four required elements. Plain language, and all four: the basis for denial; the right to file a written statement of disagreement and how; the client's option, if they file none, to have the request and denial accompany future disclosures; and how to complain to the practice and to HHS, naming the contact. Pitfall: the one-element letter. Most free samples state the basis and stop, and every dropped element is a separate process failure waiting for a complaint. Statement of disagreement and rebuttal handling. Accept the statement, within a reasonable length limit you set in advance; write a rebuttal only when it earns its place, and give the client a copy if you do. Append the request, the denial, the statement, and the rebuttal to the record set, and include the statement or an accurate summary with every future disclosure of the disputed information. Pitfall: the orphaned statement. Filed in a side folder and missing from the next disclosure, it silently unravels the compromise the rule is built on. The provider-correction crosswalk. When the review shows the error is the practice's own, repair it as a correction, addendum, or late entry under integrity standards: the change clearly and permanently denoted, its date and author identifiable, and the original content preserved, struck through with a single line on paper rather than obliterated. Pitfall: correcting after the audit. Changes made after a payment determination based on medical review are disregarded, and the contractor reviews only the original record. The outcome entry and retention. Close the log: decision, date, who decided, notifications completed, and where the amendment now lives. Keep the documentation, including the titles of the persons responsible for processing, for six years under 45 CFR 164.530(j). Pitfall: an undated decision. The 60-day question is answered by this entry or not at all, and an undecided request quietly aging past its deadline is the most common way this process fails. ### Blank template (copy and adapt) RECORD AMENDMENT AND CORRECTION RESPONSE Patient-initiated amendment under 45 CFR 164.526. Decision due 60 days from receipt; one 30-day extension, written reasons only. Never delete the original entry: append or link. REQUEST LOG ENTRY Date received: __________ Received by: ________________________ Requester: [ ] client [ ] personal representative (basis): ____ Entry in dispute (note, date, author): _________________________ Change requested and reason given: _____________________________ Decision due (60 days): ___________ Extension notice: _________ ACCURACY REVIEW (by the entry's author where possible) Reviewed by: ______________________ Date: _____________________ Finding: [ ] inaccurate or incomplete: amend [ ] accurate and complete: deny [ ] not created here: deny [ ] outside the record set: deny [ ] not inspectable: deny IF ACCEPTED Amendment appended or linked at: _______________________________ Original preserved: [ ] yes Client informed on: ______________ Client agreement to notify others: _____________________________ Persons / business associates notified: ________________________ IF DENIED (the letter must contain all four, plainly) 1 Basis for denial: ____________________________________________ 2 Statement-of-disagreement right and how to file: _____________ 3 Option: request + denial accompany future disclosures: _______ 4 Complaint paths (practice contact and HHS OCR): ______________ STATEMENT OF DISAGREEMENT / REBUTTAL Statement received: _________ Rebuttal (copy to client): ______ Appended to record set: [ ] request [ ] denial [ ] statement [ ] rebuttal Travels with future disclosures: [ ] set up OUTCOME Decision: [ ] amended [ ] denied Date: _________ By: _______ Keep this documentation 6 years (45 CFR 164.530(j)). Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample record amendment and correction response Scenario: an adult client reads her notes through the portal and asks for two changes: a wrongly recorded medication dose, and removal of a diagnosis she feels no longer fits. The practice amends one and denies the other. All details are fictional. Record Amendment and Correction Response. Larkfield Counseling Group · Client: J.M., 41 · Request received: 07/06/2026 · Decision sent: 07/24/2026 (day 18 of 60) Log entry (opened 07/06/2026): Portal message from J.M. requesting amendment of the 07/01/2026 session note. Request 1: the medication-coordination line reads sertraline 50 mg daily; J.M. states the prescriber's instruction discussed in session was 100 mg. Request 2: remove the adjustment disorder diagnosis, which J.M. feels no longer describes her. Received by front office and routed the same day to D. Nwosu, privacy officer. Decision due 09/04/2026; no extension anticipated. Accuracy review (completed 07/15/2026): S. Whitaker, LMFT, author of the disputed note, compared the entry against the prescriber's visit summary on file. Finding 1: the note is inaccurate; the documented instruction was 100 mg daily. Amend. Finding 2: the diagnosis was supported at intake, is linked to the active treatment plan, and remains the clinician's judgment; the record is accurate and complete as written. Deny, offering the statement-of-disagreement path. Decision letter (sent 07/24/2026 by portal): Dear J.M.: Thank you for reviewing your record and writing to us. Your first request is accepted. An amendment has been appended to the July 1 note stating the correct dose; the original entry remains visible and is marked as amended, as the rule requires. With your agreement, we will send the amendment to your prescriber's office so their file matches ours. Your second request is denied, because the diagnosis is accurate and complete as a record of your assessment and care to date. You may send us a written statement of disagreement of up to 500 words, addressed to our privacy officer; it will be attached to your record and included with any future disclosure of the diagnosis. If you choose not to, you may instead ask that your request and this letter accompany those disclosures. You may also complain to us (D. Nwosu, privacy officer, at the practice number) or to the HHS Office for Civil Rights; neither affects your care here. Your current progress will be documented at your next treatment plan review. Outcome log (07/29/2026): Amendment appended and linked in the EHR 07/24/2026; original preserved. J.M. agreed 07/27/2026 to notification of the prescriber's office; copy of the amendment sent 07/28/2026. No statement of disagreement received to date; the request, the decision letter, and this log are appended to the designated record set. Documentation retained six years per 45 CFR 164.530(j). Logged by D. Nwosu. This sample is fictional and for educational purposes. It does not describe a real client, clinician, or practice. ↑ Back to the template and downloads ### Why this sample works The clock is provable. Received 07/06, decided 07/24, due date calendared: the log answers the 60-day question with dates, not memory. Each disputed item got its own finding. The factual dose error was amended; the clinical opinion was retained: exactly the split the rule draws between accuracy and disagreement. Nothing was deleted. The original entry stays visible and marked as amended, so the record shows its own history instead of hiding it. The denial carries all four required elements. Basis, statement-of-disagreement right, the accompany-future-disclosures option, and both complaint paths, in plain language. Downstream holders were closed out. The prescriber's office got the amendment with the client's agreement, which is the notification duty most responses forget. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The amendment right reaches everything in the designated record set, billing records included, for as long as the practice holds them. It does not reach psychotherapy notes kept separate from the chart, because whatever is unavailable for inspection under the access rule is also outside the amendment right; requests aimed at those notes route through their own authorization rules instead. Enforcement runs asymmetrically. OCR has brought more than fifty right-of-access enforcement actions since 2019, including a $200,000 civil monetary penalty against Oregon Health & Science University in March 2025, while no publicized settlement rests on the amendment rule, and amendment is not among OCR's top five complaint categories. Read that as under-enforced, not optional: an amendment dispute that slows the client's copy of the chart can implicate the information-blocking rules at 45 CFR Part 171, so provide the copy promptly even while the decision is pending. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The payer layer judges the other workflow: your own corrections. CMS's Medicare Program Integrity Manual expects the date and author of any amendment, correction, or delayed entry to be identifiable, the change permanently denoted, and the original content preserved; reviewers are told to treat obliterated sections, white-out, and excessive late entries as possible falsification and refer suspected cases to program integrity contractors. Timing is the sharp edge: corrections completed before a claim or review are considered, while Medicare contractor guidance is explicit that after a payment determination based on medical review, only the original record is reviewed. An addendum written because an audit letter arrived is documentation that arrives too late to count. The letter format is a convention; the clock, the exhaustive grounds, the four denial elements, and the append-only mechanics are the requirement. State law can add rights on top: California caps a patient addendum at 250 words per disputed item, and New York runs record challenges through its public health law, so read your own statute before templating your letters, and when a request arrives wrapped in litigation or a licensing complaint, consult your attorney or board; state rules vary. The disputed entry itself is usually a progress note , and a well-documented note is the best defense against ever writing a denial letter. ↑ Back to the template and downloads ## Common record amendment errors auditors flag The measured problem is large, and the enforcement lands where clinicians least expect it. In a 2020 JAMA Network Open study of patients reading their own ambulatory notes, 21 percent of 22,889 readers reported finding a mistake, 42.3 percent of those rated the mistake serious, and readers 65 and older were about twice as likely to report a serious one. Yet OCR enforcement concentrates on the neighboring right of access, with no publicized settlement resting on the amendment rule, while on the payer side integrity reviewers disregard corrections made after a payment determination and review only the original record. The BastionGPT Clinical Advisory Board sees the same errors most often in record amendment reviews: One workflow where two belong. The patient amendment (60-day clock, privacy regulators) and the provider correction (integrity manual, payer auditors) get run through the same template, so client requests are charted like corrections and the practice's own errors wait weeks for no reason. Build both paths and label them. Deleting or overwriting the disputed entry. The rule's own method is appending or linking, and on paper an accepted correction is a single line struck through with the original legible. Obliterated sections, white-out, and missing pages sit on the integrity reviewers' falsification indicator list. Clock errors imported from the access rule. Sixty days from receipt, not thirty, and not from a completed form; one 30-day extension, available only if written reasons and a completion date go out inside the original window. A second extension does not exist. The one-element denial letter. The basis alone is not compliant: the letter must also explain the statement-of-disagreement right and how to file it, the option to have the request and denial accompany future disclosures, and the complaint paths to the practice and to OCR. A statement of disagreement that never travels. Filed in a side folder, missing from the record set, absent from later disclosures of the disputed information: each is its own violation, and forgetting to give the client a copy of the practice's rebuttal completes the set. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on record amendment and correction responses. Draft the decision letter from your finding: an acceptance with the appended amendment language, or a denial carrying all four required elements in plain language. Check a drafted response before it goes out: the 60-day arithmetic, the ground cited against the exhaustive list, statement-of-disagreement handling, and who still needs notice. Keep the workflows apart: when the error is the practice's own, draft a dated, attributed provider correction or late entry that preserves the original entry. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long do we have to act on a record amendment request? Sixty days from receipt under HIPAA, and receipt means the day the request arrives, not the day your form is completed. One extension of up to 30 days is available, and only if the client gets a written statement of the reasons and a completion date inside the original 60; a second extension is prohibited. The commonly cited 30-day deadline belongs to the access rule, not amendment. Ontario custodians run on 30 days with a possible 30-day extension under PHIPA, and Australian agencies must respond within 30 days under APP 13, with private organizations held to a reasonable period. Do we have to remove a diagnosis the client disagrees with? No. A practice may deny an amendment when the record is accurate and complete, and a diagnosis you continue to stand behind is exactly that; the client's remedy is a statement of disagreement that travels with the record, which keeps their voice in the chart without rewriting your clinical judgment. Ontario's PHIPA makes the same point structurally: a professional opinion or observation made in good faith is exempt from the duty to correct. Quietly deleting a diagnosis you believe is correct can itself raise record-integrity concerns. Can we just delete the entry the client disputes? No. HIPAA's method of amendment is to append the change or provide a link to it; nothing in the rule authorizes deletion, and the original entry stays in the record. Provider-side correction standards point the same direction: on paper, a single-line strike-through that leaves the original legible, and in an EHR, a correction that preserves the prior version with date and author identifiable. Deletion is not a correction; to an integrity reviewer it is an alteration, and obliterated or missing content is a falsification indicator. What must an amendment denial letter contain? Four things, in plain language: the basis for the denial; the client's right to submit a written statement of disagreement and how to file it; a statement that, if they do not, they may ask that their request and your denial accompany any future disclosure of the disputed information; and how to complain to the practice and to the HHS Office for Civil Rights, including the practice contact. Most free sample letters include the basis and stop, which turns one defensible decision into several process violations. Is a patient amendment the same as a provider correction or addendum? No, and mixing them up is the most consequential error in this corner of documentation. A patient amendment is the client's request to change the record, governed by the privacy rule's 60-day process and judged by privacy regulators. A correction, addendum, or late entry is the clinician fixing their own progress note , governed by CMS integrity standards and judged by payer auditors, who expect the change permanently denoted, dated, attributed, and the original preserved. One event can involve both: a client points out a genuine charting error, you accept the amendment, and the fix itself follows correction mechanics. Can we require a request form or charge a fee for amendments? No regulation prescribes a form. A practice may require that amendment requests be in writing and include a reason, but only if its Notice of Privacy Practices tells people so in advance. There is no fee: the amendment rule has no fee provision, unlike access, which permits a cost-based charge, and Australia's APP 13 prohibits charging for correction requests outright. Making your own form a precondition is the kind of barrier privacy regulators read as obstruction; the 60 days run from receipt regardless. Do amendment rights reach psychotherapy notes or records from another provider? Psychotherapy notes kept separate from the chart sit outside the amendment right, because whatever is unavailable for inspection under the access rule cannot be amended either; disclosure of those notes runs on its own authorization . Records another provider created may be denied on the not-created-here ground, unless the originator is no longer available to act, in which case the practice holding the record handles the request. Denying is not forwarding: tell the client who the originator is so they can request amendment at the source. How long do we keep amendment documentation? Six years under HIPAA's documentation rule, measured from creation or from the date the document was last in effect: that covers the request, the decision, the denial letter, the statement of disagreement, the rebuttal, and the titles of the people responsible for processing. The amended record itself follows clinical retention rules instead, which usually run longer: commonly 5 to 10 years for adults under US state law, about 10 years for Canadian colleges, and 7 years from last service or until age 25 in Victoria and New South Wales. Can BastionGPT draft our amendment and correction responses? Yes. Give it the request, your finding, and the ground, and it drafts the acceptance or denial letter with the required elements, the appended amendment language, and a dated log entry for your compliance file; it can also draft a provider correction or late entry that keeps the original visible. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Psychotherapy Notes Authorization Psychotherapy Progress Note Claim-Support Billing Note SOAP Note Records Request & Chart Copy Packet Consent for Treatment of a Minor 42 CFR Part 2 Consent Subpoena Response & Records Affidavit HIPAA Notice of Privacy Practices Acknowledgment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/part-2-consent All note templates # 42 CFR Part 2 Consent: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A 42 CFR Part 2 consent is the written patient authorization, required by federal regulation, that lets a substance use disorder treatment program share a patient's SUD records. Clinicians and records staff in federally assisted SUD programs complete it before any consent-based disclosure. Since February 16, 2026, one consent can cover all treatment, payment, and operations uses until the patient revokes it in writing. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it SUD program clinicians, counselors, and records staff prepare it; the patient, or a person authorized under 2.14 or 2.15, signs Audience The disclosing program and its records office, recipients including providers, health plans, and HIEs, OCR reviewers, auditors Typical length 300 to 700 words · 10 to 20 minutes by hand (clinical team estimate) Format family Consent and privacy documentation (compare: release of information, psychotherapy notes authorization) When it's used Before any consent-based use or disclosure of SUD records; a TPO consent may be signed once at intake and lasts until revoked in writing Standards context The eight consent elements are federal law (42 CFR 2.31); the form itself is not mandated, and OCR civil enforcement began February 16, 2026 On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a 42 CFR Part 2 consent? A 42 CFR Part 2 consent is the written authorization, on paper or electronic, that permits a federally assisted substance use disorder program, or another lawful holder of its records, to use or disclose those records. Its required contents are enumerated at 42 CFR 2.31 . The confidentiality scheme behind it dates to federal statutes of 1970 and 1972 and to regulations first issued on July 1, 1975, now codified at 42 U.S.C. 290dd-2; the CARES Act of 2020 ordered the regime aligned with HIPAA, and the 2024 final rule (89 FR 12472) carried that out, with a compliance date of February 16, 2026 that has now passed. Clinicians and payers call the document a Part 2 consent, a SAMHSA consent, a 42 CFR release, or an SUD release of information under Part 2. The load-bearing distinction is with the ordinary HIPAA authorization behind a general release of information . HIPAA lets treatment, payment, and operations disclosures proceed with no authorization at all; Part 2 requires signed consent before those same disclosures whenever the records come from a Part 2 program, and a generic medical-records release does not satisfy 2.31. The two instruments also part ways on conditioning: a HIPAA authorization generally cannot be made a condition of treatment, while a Part 2 consent for treatment, payment, and operations can be. One newer neighbor matters too: the 2024 rule created SUD counseling notes, a Part 2 twin of the psychotherapy notes HIPAA already walls off behind a separate psychotherapy notes authorization . Both kinds of notes need their own standalone consent that cannot be combined with any other. ## Who uses Part 2 consents and when Programs that meet the regulation's two-part test: federally assisted, a status most providers hold through Medicare or Medicaid participation, federal tax exemption, DEA registration, or grant funding, and holding themselves out as providing SUD diagnosis, treatment, or referral. That reaches opioid treatment programs regulated under 42 CFR Part 8, outpatient and residential SUD programs, withdrawal management services, and the SUD arms of integrated care practices. The trigger is any consent-based use or disclosure: records to a primary care physician, a health plan, an HIE, a family member, or a court. A general therapy practice that neither takes federal assistance nor advertises SUD care is usually outside Part 2 and documents releases with a standard release of information ; a substance use assessment written inside a Part 2 program, by contrast, is a Part 2 record from the moment it exists. The pressure point is care coordination : a 2017 qualitative study across 76 Oregon stakeholders found the pre-reform consent rules "perceived as a barrier to care coordination and integration," the burden the one-time TPO consent now exists to remove. ## 42 CFR Part 2 consent structure: what goes in each section Patient and disclosing program. Name the patient, then the program or person authorized to make the disclosure. Both are enumerated elements, and the discloser line is what distinguishes this from a form the recipient drafted for its own purposes. Pitfall: assuming another agency's form covers you. The SSA-827 the Social Security Administration sends does not satisfy 2.31; a Part 2 program responding to SSA still needs its own compliant consent. Information to be used or disclosed. The regulation requires a description that identifies the information in a specific and meaningful fashion: record types, the treatment episode, date ranges. Pitfall: any-and-all boilerplate. A description that would fit every chart in the building identifies nothing in a specific and meaningful way, and it invites the recipient to take more than the patient understood. Recipients. Name the person or class receiving the records. Since the 2024 rule, a consent for treatment, payment, and operations may use a general designation, in the regulation's example wording, "my treating providers, health plans, third-party payers, and people helping to operate this program." When records flow through an intermediary such as an HIE, ACO, or care management organization, the consent must name the intermediary and either name the member participants or limit them to those with a treating relationship. Pitfall: naming the HIE alone. An intermediary designation without the participant limitation fails the recipient element even when everything else is right. Purpose, and the two TPO statements. State why the disclosure is being made. A TPO consent must also carry the two statements at 2.31(a)(10): that records may be redisclosed by the recipient and may then lose Part 2 protection, and what the consequences of declining to sign would be. Pitfall: a form last revised before 2024. Pre-final-rule consents almost never carry either statement, and the compliance date has passed. Expiration and revocation. Give an expiration date or an expiration event, plus the patient's right to revoke in writing, except to the extent the program has already acted in reliance, and how to revoke. For a TPO consent the regulation expressly accepts "none" or "end of the treatment," so no fixed date is required. Pitfall: importing an annual-renewal rule that is not in the regulation. A TPO consent lasts until written revocation; letting it lapse on an invented schedule interrupts lawful data flow for no compliance gain. Signature and date. The patient signs and dates the consent; for minors, a person authorized under 2.14, and for patients adjudicated to lack capacity or deceased, a person authorized under 2.15. Paper or electronic consent is valid, and electronic signatures are acceptable where other law does not prohibit them. Pitfall: holding disclosures for a wet-ink original. SAMHSA guidance accepts a photocopy, fax, or valid e-signature; no witness or notary appears anywhere in the rule. The 2.32 notice that travels with every disclosure. Each consent-based disclosure must be accompanied by one of two written statements, long form or short, and, since the 2024 rule, by a copy of the consent or a clear explanation of its scope ( 42 CFR 2.32 ). The section was renamed from "Prohibition on redisclosure" to reflect the addition. Pitfall: stamping the old redisclosure paragraph alone. Without the consent copy or scope explanation the disclosure fails the amended rule, HIEs included. ### Blank template (copy and adapt) 42 CFR PART 2 CONSENT FOR RELEASE OF SUD RECORDS Written consent (paper or electronic) is required before a federally assisted SUD program uses or discloses patient records (42 CFR 2.31). No federal form is mandated; adapt this to your program, your EHR, and any stricter state law. Verbal consent is never sufficient. PATIENT AND DISCLOSING PROGRAM Patient name: ______________________ DOB: _________________ Program or person authorized to disclose: _________________ ___________________________________________________________ INFORMATION TO BE USED OR DISCLOSED Describe in a specific and meaningful way (types, dates): ___________________________________________________________ ___________________________________________________________ SUD counseling notes are NOT covered; they require their own separate consent. RECIPIENT(S) Name or class of recipients: ______________________________ For a TPO consent, a general designation is permitted: [ ] My treating providers, health plans, third-party payers, and people helping to operate this program Intermediary (HIE, ACO), if any, plus named participants or a treating-relationship limit: _________________________ PURPOSE [ ] Treatment, payment, and health care operations (TPO) [ ] Other (state it): _____________________________________ TPO CONSENT STATEMENTS (2.31(a)(10)) [ ] Records disclosed under this consent may be redisclosed by the recipient and may no longer be protected by 42 CFR Part 2. [ ] Consequences of declining to sign were explained: ___________________________________________________________ EXPIRATION AND REVOCATION Expires on (a date or an event; "none" or "end of the treatment" is sufficient for a TPO consent): ______________ I may revoke this consent in writing at any time, except to the extent the program has already acted in reliance on it. How to revoke: ____________________________________________ SIGNATURE Patient (or person authorized under 2.14 or 2.15): ________ Signature: ____________________ Date signed: ______________ Attach the 2.32 notice plus a copy of this consent, or a clear explanation of its scope, to every disclosure made under it. Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample 42 CFR Part 2 consent Scenario: an adult in outpatient SUD treatment authorizes his program to share records with his primary care physician so ongoing care can be coordinated. All details are fictional. 42 CFR Part 2 consent. Cedarline Recovery Services · Patient: M.T. · Signed: 07/14/2026 · Disclosure logged: 07/16/2026 Patient and disclosing program: M.T., date of birth 05/22/1988. Discloser: Cedarline Recovery Services, an outpatient substance use disorder treatment program and Part 2 program. Information to be used or disclosed: Attendance summary, current medication list, treatment plan status, and discharge recommendations from the outpatient treatment episode that began 03/09/2026. SUD counseling notes are excluded; disclosing them would require a separate consent. Recipient: Dr. A. Patel, Fairhaven Family Medicine, the patient's primary care physician. Purpose: Treatment: coordination of ongoing medical care, including prescribing decisions that depend on current treatment status. TPO statements: M.T. acknowledged in writing that records disclosed under this consent may be redisclosed by the recipient and may then no longer be protected by Part 2, and that declining to sign would not affect his treatment at Cedarline. Expiration and revocation: Expires at the end of the current treatment episode. M.T. may revoke in writing at any time through the records office, except to the extent Cedarline has already acted in reliance on the consent. Signature: Signed electronically by M.T. on 07/14/2026; identity verified at check-in. Electronic signature accepted under program policy and applicable law. Disclosure processing: Records sent to Dr. Patel on 07/16/2026 with the 2.32 notice and a copy of this consent attached; the disclosure was entered in the program's log. This sample is fictional and for educational purposes. It does not describe a real patient, program, or practice. ↑ Back to the template and downloads ### Why this sample works Every 2.31 element is present and findable. Patient, discloser, description, recipient, purpose, expiration, revocation terms, and a dated signature can each be pointed to on one page. The description is specific and meaningful. Named record types and a dated treatment episode, not any-and-all boilerplate that would fit every chart. Counseling notes are expressly excluded. The standalone-consent rule for SUD counseling notes stays intact, and the exclusion is stated where the recipient will see it. Both TPO statements appear. Redisclosure risk and the consequences of declining are acknowledged in writing, the two elements pre-2024 forms miss. The expiration is an event the rule permits. End of the treatment episode is expressly sufficient, so the consent cannot silently lapse mid-care. The 2.32 attachment step is documented. Notice plus consent copy went with the records, and the disclosure was logged, closing the loop the 2024 rule added. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The enforcement ground shifted while many consent forms stood still. For five decades Part 2 violations were criminal matters, rarely prosecuted, with fines capped at $500 for a first offense and $5,000 for repeats under the pre-2024 statute. The CARES Act imported HIPAA's civil machinery instead: the HHS Secretary delegated Part 2 enforcement to the Office for Civil Rights on August 25, 2025, OCR announced a Part 2 Civil Enforcement Program on February 13, 2026, and complaints and breach reports have been accepted since February 16, 2026, under the tiered civil money penalty schedule, a range 2026 law firm alerts summarize as roughly $141 to $2.1 million per violation depending on culpability. Two facts keep the picture in proportion: OCR has said it will not investigate conduct predating February 16, 2026, and as of mid-2026 no Part 2 resolution agreement or civil penalty had been announced, so the first action will set the practical tone. What belongs in the record now: the signed consent itself, the disclosures made under it with the notice attached, and care coordination entries that state the consent's scope rather than assuming it. Around the consent sit layers of different weight, and the differences are the compliance program. The eight consent elements, the two TPO statements, and the 2.32 notice are law. The form is not: HHS and SAMHSA publish only voluntary model documents, refreshed most recently on February 13, 2026, and no regulation prescribes a form, a font, a witness signature, a wet-ink original, or a retention period for the consent itself. Retention arrives from elsewhere: HIPAA's six-year documentation rule at 45 CFR 164.530(j) reaches consents held by covered entities, and state licensing and payer rules commonly run five to ten years. State confidentiality statutes can be stricter than Part 2 and control when they are; New York's Mental Hygiene Law 33.13 is the standing example. One interaction catches even careful programs: under 42 CFR 2.36, reporting a patient's SUD medications to a state prescription drug monitoring program requires patient consent under 2.31, even where state law mandates PDMP reporting. Canada and Australia have no Part 2 equivalent; SUD records there fall under PIPEDA and provincial acts such as Ontario's PHIPA, or the Privacy Act 1988 and its Australian Privacy Principles, but a clinician in either country who receives US Part 2 records takes them subject to the redisclosure restriction carried by the accompanying notice. These questions always sit on top of state law and program policy, so consult your attorney or board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ## Common Part 2 consent errors auditors flag Part 2 consent defects rarely surface as claim denials; no claim adjustment or remark code specific to a Part 2 consent exists, and no published study measures consent error rates. The pressure arrives as enforcement exposure and blocked coordination instead, and the scale is documented: SAMHSA's 2023 national survey collected data from 20,681 substance use and mental health treatment facilities (84.9 percent response rate), all of them now inside the civil enforcement perimeter; an OCR paperwork notice from September 2025 estimated 1,864,367 patients in SUD treatment annually and projected roughly 1,864 Part 2 complaints a year at 45 minutes each; and a 2017 Psychiatric Services study built on 114 interviews with 76 Oregon stakeholders documented the consent regime itself as a coordination barrier, the finding that drove the one-time TPO consent. The BastionGPT Clinical Advisory Board sees the same errors most often in Part 2 consent reviews: A general release standing in for the Part 2 consent. A HIPAA authorization or standard ROI does not satisfy 2.31: it lacks the redisclosure statement, the consequences statement, and the intermediary designation. The disclosure it produces is unauthorized even though the patient signed willingly. An any-and-all description of the records. The element requires identifying the information in a specific and meaningful fashion. Blanket phrases fail it; name the record types and the date range of the episode. A pre-2024 form missing the two TPO statements. Consents covering treatment, payment, and operations must say that records may be redisclosed and may lose Part 2 protection, and what declining to sign would mean. Forms last revised before the final rule usually carry neither. The notice without the consent. Since February 16, 2026, every consent-based disclosure must carry the 2.32 statement plus a copy of the consent or a clear explanation of its scope. Stamping the old prohibition-on-redisclosure paragraph alone is the most common holdover, and HIE traffic is not exempt. Combined consents the rule requires to stand alone. SUD counseling notes take a separate consent that cannot ride with a TPO consent, and consents for civil, criminal, administrative, or legislative proceedings cannot be combined with any other. One signature line covering all three is three defects. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on 42 CFR Part 2 consents. Draft a complete consent from the disclosure facts: patient, discloser, records described in a specific and meaningful way, recipient, purpose, expiration or event, and both TPO statements. Check a form before signature: every 2.31 element present, the recipient designation valid for an HIE or ACO, revocation language included, and no combined consent where a standalone one is required. Assemble the outgoing packet: the 2.32 notice with the consent copy or a clear scope explanation, ready to travel with the records. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does a 42 CFR Part 2 consent need an expiration date? Not a fixed one. The regulation asks for an expiration date or an expiration event, and for a consent covering treatment, payment, and operations it expressly accepts "none" or "end of the treatment." An open-ended TPO consent stays valid until the patient revokes it in writing. Many pages ranking for this form still call a fixed date mandatory; that requirement is not in the rule. If your program prefers dated consents for workflow reasons, that is policy, not law. Is a regular HIPAA release enough to share SUD treatment records? No, not when the records come from a Part 2 program. A general release of information rides on HIPAA, which permits treatment, payment, and operations disclosures without any authorization; Part 2 requires signed consent even for those, and its consent must carry elements a generic form lacks, including the redisclosure statement and, since 2024, the consent copy that travels with each disclosure. Use the Part 2 consent whenever a Part 2 program's records are involved, and the ordinary release for everything else. What changed for Part 2 consents on February 16, 2026? That date ended the transition to the 2024 final rule and switched civil enforcement on. Programs must now use consents that track HIPAA's authorization elements, may rely on a single TPO consent signed once, must include the two 2.31(a)(10) statements, and must attach the renamed 2.32 notice plus a copy of the consent, or a clear explanation of its scope, to every consent-based disclosure. The same day, the Office for Civil Rights began accepting Part 2 complaints and breach reports under HIPAA-style penalties, while stating it will not investigate conduct from before that date. Can a program require patients to sign a TPO consent to get treatment? Unlike a HIPAA authorization, a Part 2 TPO consent may lawfully be made a condition of treatment. HHS added a caution in the rule's preamble: programs should not condition treatment on the consent unless they have some capacity to honor patients' requests for restrictions. So the practice is permitted by the regulatory text and discouraged in some circumstances by the commentary. Document your program's choice either way, and remember the consent must state the consequences of declining to sign. Is verbal, faxed, or electronically signed consent valid under Part 2? Verbal consent is never sufficient; Part 2 consent must be in writing. Beyond that, the rule is flexible: consent may be paper or electronic, electronic signatures are acceptable wherever other law does not prohibit them, and SAMHSA guidance confirms a photocopy or fax works. No witness, no notary, and no wet-ink original appear anywhere in the regulation. Do SUD counseling notes travel under a TPO consent? No. The 2024 rule created SUD counseling notes, a category HHS said almost exactly matches HIPAA's psychotherapy notes: a clinician's private session analysis kept separate from the rest of the record. They need their own standalone consent that cannot be combined with a TPO consent or any other, just as psychotherapy notes need a separate authorization . Consents for use in civil, criminal, administrative, or legislative proceedings must stand alone as well. What has to accompany a disclosure made with consent? Two things, under the 2.32 rule the 2024 final rule renamed and expanded. First, one of two written statements; the short form reads "42 CFR part 2 prohibits unauthorized use or disclosure of these records." Second, a copy of the consent itself or a clear explanation of its scope. The pre-2024 practice of stamping the old prohibition-on-redisclosure paragraph alone no longer satisfies the rule, and disclosures through an HIE are not exempt. Does anything like Part 2 apply in Canada or Australia? No equivalent exists in either country. Canadian SUD records fall under general health privacy law: PIPEDA federally and provincial statutes such as Ontario's PHIPA, which rely on implied consent inside the circle of care and express consent beyond it. Australia treats them as sensitive information under the Privacy Act 1988 and its Australian Privacy Principles, with state health records acts layered on. The catch is imported: a clinician in either country who receives records from a US Part 2 program takes them subject to the redisclosure restriction stated in the accompanying notice. Can BastionGPT draft 42 CFR Part 2 consent documentation? Yes. Give it the disclosure facts: who is disclosing, to whom, which records, and why, and it drafts a consent carrying all eight 2.31 elements, flags a missing TPO statement or an any-and-all description before signature, and produces the 2.32 notice packet for the outgoing disclosure. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Psychotherapy Notes Authorization Substance Use Assessment Care Coordination Note Collaborative Care (BHI/CoCM) Note Relapse Prevention Plan Records Request & Chart Copy Packet Consent for Treatment of a Minor Record Amendment & Correction Response Return-to-Primary-Care Summary Utilization Review Summary HIPAA Notice of Privacy Practices Acknowledgment CAGE Questionnaire Documentation AUDIT-C Score Documentation DAST-10 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/testing-administration-note All note templates # Psychological Testing Administration Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychological testing administration note documents a test administration and scoring encounter: which standardized instruments were given, who administered each one (psychologist or technician), and the total face-to-face minutes supporting each billed unit of CPT 96136 to 96139. Psychologists, neuropsychologists, and testing technicians write one for every administration session. Most notes run 150 to 350 words plus the time log. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The administering psychologist, or a trained testing technician (psychometrist) with the supervising psychologist accountable Audience The supervising and interpreting psychologist, Medicare and commercial payers, auditors, state boards Typical length 150 to 350 words plus the time log · 4 to 10 minutes by hand (clinical team estimate) Format family Structured diagnostic-test encounter record (compare: psychological evaluation report, outcome measure note) When it's used Every test administration and scoring session billed under 96136 to 96139, including each day of a multi-day battery Standards context The ordering and supervision framework is law; the contents are payer policy; a discrete administration note is convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychological testing administration note? A psychological testing administration note documents the administration and scoring leg of psychological or neuropsychological testing: the standardized instruments given, each named; who administered each one; the total face-to-face minutes behind every billed unit; and the supervision under which a technician worked. No professional body ever published an official template, and no author or year can be cited. The document is an emergent byproduct of the 2019 CPT restructuring, effective January 1, 2019, which split administration and scoring into a provider pair (96136, 96137) and a technician pair (96138, 96139), replacing legacy codes such as 96102, layered on Medicare's longstanding diagnostic-test rules. Clinicians and payers also call it a test administration and scoring note, psychometrist note, technician note, test administration log, time log, or start-stop time note; none of those names is official either. The load-bearing boundary is with the psychological evaluation report . Administration and scoring time lives here; test selection, interpretation, integration, and feedback are the professional work of the evaluation codes 96130 to 96133 and belong in that report. Just as load-bearing: no statute or regulation requires an administration note as a stand-alone document. What the law compels, as a condition of Medicare payment, is the diagnostic-test framework at 42 CFR 410.32 : testing ordered by the treating practitioner, furnished under the right level of supervision, with documentation that is legible, signed, and kept in the record. The note is simply the one artifact that proves all of it at once. ## Who uses testing administration notes and when Assessment-heavy settings write these daily: neuropsychology and psychological assessment practices, hospital and academic medical centers, rehabilitation units, and any clinic running a technician model. That model is the norm, not the exception: a 2002 practice survey found 51.2 percent of more than 1,300 respondents employed testing technicians, and the field's 2020 practice and salary survey of 1,677 doctoral-level practitioners ( Sweet et al., 2020 ) found reliance on technicians remains popular, especially in institutional and specialty neuropsychology settings, while observing that the 2019 code changes appear to have had "more negative than positive effects" on practice. Write one for every administration and scoring session, whoever administers: the psychologist's own session supports 96136 and 96137, the technician's supports 96138 and 96139. For a single brief symptom questionnaire scored in session, an outcome measure note is the right artifact instead; a full battery earns this note, and the write-up conventions for individual instruments live in guides like our WAIS report guide . ## Testing administration note structure: what goes in each section Encounter header and the order. Patient identifiers, date of service, the treating practitioner who ordered testing, and the referral question. Medicare treats psychological and neuropsychological testing as a diagnostic test, so the order and the clinical rationale must exist in the record; this header ties the session to them. Pitfall: an administration note floating free of any order. When the order and medical-necessity rationale are nowhere in the chart, every downstream unit is deniable no matter how clean the time log looks. Tests administered, by name. List every instrument given that day by full name or standard abbreviation. Payer policy expects the specific tests to be identifiable, and full batteries typically run 4 to 8 hours of testing under the national LCD's own description. Pitfall: "comprehensive battery administered per referral." An auditor cannot match unnamed tests to billed time, and unnamed tests read as tests that may not have happened. Administrator and role. Say who gave each test and in what role: the psychologist personally, or a trained technician. The entire split between the provider codes and the technician codes turns on this fact, and under Medicare the technician needs no license or certification: "trained technician" is the Benefit Policy Manual's whole requirement, with anything stricter coming from state boards or commercial payers. Name and credential them anyway. Pitfall: a note that never says who ran the tests. With the administrator unstated, neither code pair is supported and the claim rests on nothing. Supervision statement. Name the supervising psychologist or physician and record the supervision actually delivered. The controlling regulation, 42 CFR 410.32(b)(2)(iii)(B), requires general supervision for these tests, meaning available rather than physically present, while the national LCD says the technician "must be directly supervised." Document availability, and any on-site time, and the conflict resolves in your favor either way. Pitfall: boilerplate claiming on-site direct supervision that the appointment book contradicts. Auditors read this line against the supervisor's schedule. The time log. Record total face-to-face administration and scoring minutes for each administrator. No CMS authority requires literal start and stop clock times; what is required is total time supporting each 30-minute unit, with an added unit earned 16 minutes past the prior increment. Start-stop notation is defensive practice worth keeping for long batteries. For multi-day testing, combine the time and report it on the last date of service. Pitfall: unit-shaped time. Minutes recorded as exactly 30 per unit, every session, is the first pattern integrity reviewers flag. Behavioral observations. Effort, engagement, breaks, accommodations, interruptions, and anything bearing on validity. CMS never separately requires them for this note; they are clinical convention, and their real job is protecting the interpretation that will later be built on these scores. Pitfall: the same "cooperative, adequate effort" line copied forward across sessions. Observations that never vary support nothing when validity is later questioned. Scoring and disposition. Scoring by the person who administered belongs inside these codes, so document it within the total minutes, then route raw results to the interpreting clinician and sign the note; payer policy expects documentation legible and signed. Pitfall: interpretation smuggled into the administration note. Integration and report writing are the evaluation codes' work, and billing that thinking here double-counts it. ### Blank template (copy and adapt) PSYCHOLOGICAL TESTING ADMINISTRATION NOTE Administration and scoring encounter (CPT 96136 to 96139). Name every test, say who gave it, and let the minutes support the units. Interpretation belongs in the evaluation report. ENCOUNTER Patient: __________________ DOB: __________ DOS: __________ Ordered by (treating practitioner): _________________________ Order date: __________ Referral question: __________________ TESTS ADMINISTERED TODAY (each instrument by name) 1 ___________________________ 3 ___________________________ 2 ___________________________ 4 ___________________________ Administered by: [ ] psychologist (96136/96137) [ ] trained technician (96138/96139) Administrator name / credentials: ___________________________ SUPERVISION (technician-administered testing) Supervising psychologist or physician: ______________________ Level delivered: [ ] general (available) [ ] on site: ______ Availability documented (how): ______________________________ TIME LOG (face-to-face administration + scoring minutes) Today: start ______ stop ______ breaks ____ total ________ Prior day(s) of this battery: _______________________________ Combined total, reported on last DOS: _______________________ Units supported (30 min, then 16+ past each increment): _____ BEHAVIORAL OBSERVATIONS (effort, breaks, validity-relevant) _____________________________________________________________ _____________________________________________________________ SCORING AND DISPOSITION Scored by: _______________ Scoring minutes (in total): _____ Raw results routed to (interpreting clinician): _____________ Signature / credentials: __________________ Date: __________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychological testing administration note Scenario: day two of a two-day adult attention and memory battery in an outpatient neuropsychology practice. A trained testing technician administers under the neuropsychologist's general supervision. All details are fictional. Psychological Testing Administration Note. Cedarbrook Neuropsychology Group · Patient: R.T., 34 · DOS: 07/28/2026 (day 2 of 2) · Technician: M. Alvarez, BA · Supervisor: L. Chen, PhD Order and referral question: Testing ordered 07/10/2026 by P. Rowan, MD (psychiatry) to clarify adult ADHD versus anxiety-related attention complaints; order and rationale filed with the 07/10/2026 referral. First administration day 07/21/2026. Tests administered today (by M. Alvarez, trained testing technician): CVLT-3; CPT-3; CAARS 2 self-report and observer ratings; WAIS-5 processing speed subtests carried over from day one. All administered per manual under standard conditions, marking the battery complete. Supervision: L. Chen, PhD, clinical neuropsychologist, provided general supervision as permitted for diagnostic psychological and neuropsychological tests: available by phone and secure message throughout testing, on site 1:00 to 3:30 pm, and reviewed the session with the technician at 3:15 pm. Time: Today 1:05 to 3:12 pm with one 10-minute break: 117 face-to-face technician minutes, administration and scoring. Combined with day one (96 minutes on 07/21/2026), total technician time for the battery is 213 minutes, reported today as the last date of service: 96138 for the initial 30-minute unit and 96139 for six added units; a seventh add-on would have needed 226 minutes. Behavioral observations: Arrived on time wearing corrective lenses. Effort sustained across tasks with mild fatigue late in CPT-3; one scheduled break at 2:10 pm; no interruptions. Responses complete; engagement adequate throughout and consistent with interpretable results. Scoring and disposition: All day-two instruments scored by M. Alvarez per test manuals within the minutes above; raw and scaled scores entered in the record. Results routed to Dr. Chen for interpretation and integration under the evaluation codes; feedback visit scheduled 08/04/2026. Signed: M. Alvarez, testing technician, 07/28/2026. Supervision acknowledged: L. Chen, PhD, 07/28/2026. This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or practice. ↑ Back to the template and downloads ### Why this sample works Every instrument is named. The test list is the first thing an auditor matches against billed units, and this note answers it without a chart hunt. The code family is provable. Administrator and role are explicit for every test, which is the entire difference between the provider codes and the technician codes. The minutes do the billing math. Daily totals, the multi-day combination on the last date of service, and the 16-minute midpoint arithmetic are all visible, so seven units defend themselves. Supervision is recorded as delivered. The supervisor is named, the level is stated, and availability plus on-site time are documented: exactly where the direct-versus-general conflict gets decided. Interpretation stayed out. Scoring sits inside the administration total and integration is routed to the evaluation report, so the two code families never blur. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Sort the obligations by strength and this note gets much easier. LAW: the diagnostic-test framework at 42 CFR 410.32, ordering by the treating practitioner and supervision by a physician or clinical psychologist. PAYER POLICY: the content, per the national LCD L34646 and its billing article: tests by name, total face-to-face minutes, who administered, medical necessity, documentation legible and signed and available on request. CONVENTION: the discrete, titled administration note itself. Inside that stack sits the field's strangest contradiction: the LCD says a technician "must be directly supervised," while the regulation and CMS's own testing FAQ require only general supervision, which does not demand physical presence. The risk-minimizing posture is to document availability and any on-site time as delivered, and to appeal a supervision denial on the regulation's text. On credentials, Medicare asks only for a trained technician and bills the work under the supervising professional's NPI; state boards regulate delegation separately, and commercial payers write their own rules, some keeping the supervising neuropsychologist much closer to the testing room. Keep the record at least seven years for Medicare; some states and payers run longer. Time is the audit surface. Total administration and scoring minutes support each 30-minute unit, an added unit is earned 16 minutes past the prior increment, and literal start-stop times are defensive convention rather than mandate. The oft-quoted 31-minute minimum belongs to the per-hour evaluation codes, not these 30-minute codes. Multi-day batteries combine their time onto the last date of service; same-day provider and technician administration is billable together with the right modifier under NCCI edits; and full batteries typically run 4 to 8 hours, with some payers capping combined testing time at 8 hours, as Washington's workers' compensation program does. If you administer by telehealth, watch the calendar: 96136 and 96138 sit on the Medicare telehealth list through December 31, 2026, and continuation depends on the CY2027 fee schedule final rule. The format is a convention; the content is the requirement. Elsewhere the note changes character: psychological testing is largely not publicly insured in Canada or Australia, so the record answers college and privacy-law standards instead. In Ontario, a written supervisory agreement with a psychometrist and supervisor co-signature are college requirements, and Australia has no technician administration item at all. The claim side of the same encounter is its own document: see the claim-support billing note , and for the interpretive product, the neuropsychological report . ↑ Back to the template and downloads ## Common testing administration note errors auditors flag The enforcement history here is concrete. In a Justice Department case charged in 2015, two companies billed Medicare more than $25.2 million for psychological testing in Gulf Coast nursing facilities, Medicare paid more than $13.5 million on the claims, and the owners were convicted in January 2017. In Michigan, a psychologist submitted roughly 1,700 false claims for neuropsychological testing and 140 more for psychological testing over three years, including one patient billed for neuropsychological testing 21 times in a single year at 24 hours each; the insurer lost about $5 million, and the January 2020 sentence was 51 months with a forfeiture judgment of $3,163,172.92. Note what the record does not contain: the $580 million improper-payment figure often waved at testing is a psychotherapy number from a May 2023 federal audit, and no published OIG or CERT improper-payment rate isolates the testing code family at all. The BastionGPT Clinical Advisory Board sees the same errors most often in testing administration note reviews: Unnamed tests. "Battery administered per referral" with no instrument list. Billed time cannot be matched to tests, and the units fail on documentation before medical necessity is even reached. The missing administrator. The note never states who gave the tests or in what role, and the choice between the provider codes and the technician codes becomes unsupportable. The technician's name and the supervisor's identity belong in the record even though Medicare puts only the supervisor's NPI on the claim. Unit-shaped time. Exactly 30 minutes per unit, no daily totals, or multi-day batteries billed day by day instead of combined on the last date of service. Totals that always land on unit boundaries read as billed-to-the-code rather than documented-as-delivered. Supervision claimed but not delivered, or delivered but not documented. On-site language contradicted by the supervisor's schedule is an integrity problem; real availability left undocumented is a denial waiting for an appeal. Record the level actually provided and know that the binding regulation requires general, not direct, supervision. Interpretation billed as administration. Integration, report writing, and feedback logged under the administration codes double-count the evaluation codes' work; scoring by the administrator is the only cognitive task that belongs in this note's minutes. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological testing administration notes. Draft the complete note from a shorthand log or dictation: tests, administrators, minutes, and observations become a clean, signature-ready administration note. Check a note before you sign: every instrument named, minutes supporting the billed units under the midpoint rule, administrator role matching the code family, supervision stated as delivered. Combine a multi-day battery's time into the last-date-of-service entry with the unit arithmetic shown, ready for the claim. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does the supervising psychologist have to be in the building while a technician tests? Under Medicare, no. The controlling regulation, 42 CFR 410.32(b)(2)(iii)(B), and CMS's own testing FAQ require general supervision for diagnostic psychological and neuropsychological tests: the supervising physician or clinical psychologist must be available, not physically present. The complication is that the national LCD says the technician must be directly supervised, so two CMS-associated texts point in opposite directions. The defensible posture is to document availability and any on-site time, and to appeal a supervision denial by citing the regulation. State boards and commercial payers can be stricter, so check both before staffing an off-site day. Do we have to record start and stop times for testing? No CMS authority requires literal clock times. What is required is total face-to-face administration and scoring time supporting each billed 30-minute unit, with an added unit earned once you pass 16 minutes beyond the prior increment. Start-stop notation is defensive documentation worth keeping for long batteries, but its absence violates no rule. A related confusion: the 31-minute minimum quoted in payer guidance applies to the per-hour evaluation codes, not to the 30-minute administration codes. Can the psychologist and a technician both bill testing time on the same day? Yes. When the psychologist personally administers some instruments and a technician administers others, both code pairs can be reported for the same day, provided each person's face-to-face time is separately documented in the record. Claim edits require an appropriate modifier, such as 59 or XE, on the same-day combination. The one thing that never works is billing a technician's administration under the provider codes: 96136 and 96137 are for the professional's own time. How do we bill a battery spread across several days? Combine the face-to-face time and report it on the last date of service; that is the Medicare convention for multi-day testing, and the sample above shows the arithmetic. Full batteries typically run 4 to 8 hours of testing, so long technician days with multiple add-on units are normal, though some payers cap combined provider-plus-technician testing time at 8 hours. Keep each day's minutes in the record even though the claim carries the combined total on the final date. What credentials does a testing technician need? Under Medicare, none: the Benefit Policy Manual asks only for a trained technician, and the work is billed under the supervising professional's NPI. Anything stricter comes from elsewhere: state psychology boards regulate delegation to unlicensed technicians as a scope-of-practice matter, and commercial payers can impose their own requirements. In Ontario, a written supervisory agreement with the psychometrist is mandatory and the supervisor must co-sign the work, even though psychometrist is not a protected title there. Does the supervisor have to co-sign the technician's administration note? No US Medicare authority requires a co-signature on the technician's note. Billing already runs under the supervising professional's NPI, and the supervisor's own cognitive work is separately documented under the evaluation codes. Co-signature is a real requirement in Ontario, where the college obliges supervisors to co-sign reports prepared by non-regulated providers. Many US practices co-sign anyway as internal policy, which is sensible risk management rather than federal law, and it is what the sample note models. How long is a testing administration note, and how long do we keep it? Most run 150 to 350 words plus the time log: a header, the test list, the administrator, the supervision line, minutes, brief observations, and disposition. Keep it at least seven years for Medicare, with some states and payers requiring longer. In Australia, psychology records are kept a minimum of seven years from last contact, and for minors until age 25; Canadian colleges set their own retention periods and expect a record of every material service activity, including who administered testing under supervision. What goes in the administration note versus the evaluation report? The administration note holds the mechanics: tests by name, who administered each, total minutes, observations, and scoring. The psychological evaluation report or neuropsychological report holds the professional work: test selection, interpretation, integration with history, diagnosis, and recommendations. The regulatory reason to keep them clean is that each defends a different code family in an audit: the note substantiates the administration units, the report substantiates the evaluation codes and medical necessity. Can BastionGPT draft psychological testing administration notes? Yes. Give it the day's shorthand: instruments, who administered, minutes, and anything notable about effort or breaks, and it drafts the full note with the supervision statement, the combined multi-day time entry, and the unit arithmetic shown for your review. It can also check a drafted note for the gaps auditors flag: unnamed tests, a missing administrator, or minutes that fail the midpoint math. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Psychoeducational Evaluation Report Outcome Measure Review Note Claim-Support Billing Note WAIS-5 / WAIS-IV Report Write-Up Psychological Testing Scoring Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/testing-scoring-note All note templates # Psychological Testing Scoring Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A psychological testing scoring note records the professional evaluation side of psychological or neuropsychological testing: verifying and summarizing scores, integrating them with clinical data, and logging the dated minutes that support CPT 96130 to 96133. Psychologists and neuropsychologists write one per assessment episode, alongside the administration record and the evaluation report. Most run 200 to 400 words plus the score summary and time log. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The interpreting psychologist or neuropsychologist; evaluation work cannot be delegated to a technician or trainee Audience Medicare and commercial payers, auditors, referring practitioners, regulatory colleges and boards Typical length 200 to 400 words plus the score summary and time log · 5 to 12 minutes by hand (clinical team estimate) Format family Structured worksheet: score summary plus a dated evaluation-services time log (compare: psychological evaluation report) When it's used Every assessment episode billed under 96130 to 96133, from first score review through report and feedback Standards context No rule names the note; medical-necessity law and MAC payer policy compel its contents, and the discrete document is convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychological testing scoring note? A psychological testing scoring note records the professional evaluation side of psychological and neuropsychological testing: the interpreting clinician's verification of scores, the pulling together of test data with history and collateral, interpretation, planning, report generation, and feedback, together with the dated minutes that make CPT 96130 to 96133 billable. No statute, payer manual, or standards body names the document; the label is functional shorthand, and clinicians also call it a scoring and interpretation note, an evaluation services note, a testing time log, or simply the 96130 note. The code family it serves is young: the AMA CPT Editorial Panel created the evaluation-services codes effective January 1, 2019, retiring the legacy testing codes 96101 to 96103 and 96119 to 96120 and splitting the old per-hour 96101 into evaluation services on one side and administration and scoring on the other ( APA Services ). Two boundaries carry the weight. The first is with the administration record: giving tests and mechanical scoring are separately coded work (96136 to 96139) that a trained technician can perform, while everything in this note is the interpreting clinician's own and cannot be delegated. The second is with the psychological evaluation report : the report is the clinical deliverable, and MAC billing articles expect a written report to be generated, but the report rarely proves the minutes. This note is the time-and-activity record that does. One structural rule from APA Services guidance surprises many practices: test selection is bundled into evaluation services, so the evaluation codes are billed together with administration services from the same episode, never on their own. Nothing in federal law names or formats any of it; the obligations flow from medical-necessity documentation rules and each MAC's coverage article, which is why the same note travels under so many names. ## Who uses testing scoring notes and when Psychologists and neuropsychologists write one for every assessment episode billed under the evaluation-services codes: private assessment practices, hospital and academic neuropsychology services, psychoeducational and disability evaluators, and forensic practices whose work must survive hostile reading. The stakes scale with the hours: in the field's professional practice survey of 1,658 neuropsychologists (Sweet and colleagues, 2015, The Clinical Neuropsychologist), a typical evaluation took about six hours of professional time, ranging from half an hour to 25 for complex forensic and diagnostic questions. Use this note whenever those hours span days or the interpretation happens after the patient leaves, which is most of the time. For a single screening questionnaire scored and discussed in session, an outcome measure note is the right record instead; instrument-specific write-up conventions live in guides like our MMPI-3 report guide , and school-context batteries feed the psychoeducational report . ## Testing scoring note structure: what goes in each section Episode header and referral. Patient identifiers, the referring or treating practitioner, the referral question, the ICD-10 linkage, and the episode span from first administration to expected feedback. The evaluation-services claim is an episode claim, so this header is what makes multi-day work cohere. Pitfall: treating each sitting as its own encounter. The base code is reported once per episode with combined time; a header that never defines the episode invites day-by-day billing that auditors unwind. Instruments and data integrated. Every test interpreted, by exact name, plus the non-test data pulled in: interview, records, informant ratings, prior evaluations. The list must reconcile with the administration record and the report, because reviewers read all three side by side. Pitfall: lists that drift. An instrument that appears in the report but not here, or here but not in the administration record, reads as work that may not have happened. Score summary and validity. Exact raw and standard scores, percentiles, and validity indices, with anomalies and corrections noted. Preserve exact values and exact test names; this is the section that lets anyone reconcile platform output, note, and report, and it is where verification of computer scoring is proven. Ontario's college makes the stake explicit by barring computer-generated reports as a substitute for the psychologist's professional opinion. Pitfall: pasting the scoring platform's narrative as the summary. An unverified auto-score with a miskeyed age band or wrong norm group corrupts everything downstream. Evaluation time log. Dated entries naming the activity and the minutes: score verification, integrating data, interpreting results against clinical data, planning, report writing, feedback. Keep evaluation minutes strictly separate from administration and scoring time, which belongs to the technician-side codes. The base code needs 31 documented minutes; add-on hours follow past each full hour. Pitfall: vague time. 'Approximately 1 to 2 hours' is a documented denial risk, and undifferentiated block time cannot be matched to units. Interpretation and clinical decision-making. The synthesis in your own words: what the verified scores mean against history, presentation, and collateral, the diagnostic conclusion forming, and the treatment planning that follows. This work is the interpreting clinician's alone; Medicare does not pay for evaluation services performed by students or trainees, and none of it can be handed to a technician. Pitfall: a signature on someone else's synthesis. A trainee-drafted interpretation the psychologist merely signs fails the performance requirement, not just style. Report, feedback, and sign-off. The written report's status and date, the feedback date, the date of service reported for the episode, and signature with credentials. MAC articles expect a written report, and the claim is reported on the date the service concluded, typically feedback, per CMS date-of-service guidance. Referral sources notice speed too: 73 percent in one survey said slow report turnaround hurt patient care (Postal and colleagues, 2017). Pitfall: an open-ended episode. No report status, no feedback date, and no stated conclusion date leave the claim's date of service unsupported and the episode impossible to close. ### Blank template (copy and adapt) PSYCHOLOGICAL TESTING SCORING NOTE Evaluation services record (CPT 96130 to 96133): score verification, integration, interpretation, report, and feedback minutes. Administration time is logged separately. EPISODE Patient: __________________ DOB: __________ Referring practitioner: _____________ ICD-10: ______________ Referral question: __________________________________________ Episode span (first administration to feedback): ____________ INSTRUMENTS INTERPRETED AND DATA INTEGRATED 1 ___________________________ 3 ___________________________ 2 ___________________________ 4 ___________________________ Collateral integrated (records, interview, informants): _____ _____________________________________________________________ SCORE SUMMARY AND VALIDITY (exact values, exact test names) Test / index: _____________ Raw: _____ Std: _____ %ile: _____ Test / index: _____________ Raw: _____ Std: _____ %ile: _____ Validity indices and response style: ________________________ Anomalies, corrections, rescoring: __________________________ EVALUATION TIME LOG (interpreting clinician only) Date ______ Activity _______________________ Minutes ______ Date ______ Activity _______________________ Minutes ______ Date ______ Activity _______________________ Minutes ______ Total: ______ Units (base code once per episode, 31+ min): _ INTERPRETATION AND CLINICAL DECISION-MAKING Findings against history and presentation: __________________ _____________________________________________________________ Diagnostic impression and plan: _____________________________ REPORT, FEEDBACK, AND SIGN-OFF Report completed: __________ Feedback given: __________ Date of service reported (episode conclusion): ______________ Signature / credentials: __________________ Date: __________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychological testing scoring note Scenario: conclusion of a two-day adult attention and memory battery in an outpatient neuropsychology practice. The neuropsychologist documents her own evaluation services for the episode, from score verification through feedback. All details are fictional. Psychological Testing Scoring Note. Cedarbrook Neuropsychology Group · Patient: R.T., 34 · Episode: 07/21 to 08/04/2026 · Clinician: L. Chen, PhD, licensed psychologist (neuropsychology) Episode and referral: Testing ordered 07/10/2026 by P. Rowan, MD (psychiatry) to clarify adult ADHD versus anxiety-related attention complaints; provisional F41.1. Battery administered by a trained technician on 07/21 and 07/28/2026; technician administration and scoring time (213 minutes) is documented and billed separately under the technician codes. This note records my evaluation services for the episode (96132, 96133). Instruments and data integrated: WAIS-5 selected indices; CVLT-3; CPT-3; CAARS 2 self-report and observer forms. Non-test data: 07/21/2026 clinical interview, psychiatric and pharmacy records from Dr. Rowan, partner-completed observer ratings, and sleep screening responses. Score summary and validity: All platform scoring verified against raw responses before interpretation. WAIS-5 Processing Speed Index 92; CVLT-3 Trials 1 to 5 standard score 102, delayed recall within one standard deviation of the mean; CPT-3 detectability T 58. CAARS 2 self-report Inattention T 68; observer form initially T 71, rescored to T 63 after I corrected a miskeyed age band in the scoring platform. Embedded validity indicators within expected ranges on all instruments; response style consistent across sessions. Evaluation time log (my time only): 07/28/2026, 35 minutes: verified scoring, reviewed validity indicators, began integrating scores with interview and records. 07/30/2026, 58 minutes: interpreted results against history and collateral; drafted report. 08/04/2026, 55 minutes: finalized report (25) and feedback session with R.T. (30). Total evaluation time 148 minutes. Units: 96132 reported once for the episode plus one unit of 96133; a second add-on would have required 151 minutes. Interpretation and clinical decision-making: Performance-based attention and memory measures are broadly average with low-average processing speed and no decline pattern. Elevated self-reported inattention contrasts with rescored observer ratings in the expected range. Integrated with interview, records, and symptom course, the picture is more consistent with anxiety-related attentional complaints than with a developmental attention disorder. Impression: generalized anxiety disorder (F41.1); adult ADHD not supported. Plan: CBT referral, sleep evaluation, no stimulant trial; full recommendations in the report. Report, feedback, and sign-off: Comprehensive written report completed and filed 08/04/2026; copy to Dr. Rowan with R.T.'s authorization. Feedback provided 08/04/2026. Date of service for evaluation services reported as 08/04/2026, the date the episode concluded, with work on the three dates above reflected in this record. Signed: L. Chen, PhD, licensed psychologist, 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or practice. ↑ Back to the template and downloads ### Why this sample works The scores are verified, not transcribed. The rescored observer form shows the clinician checked the platform, the exact correction is recorded, and every value matches the output it came from. Time is dated, named, and separated. Three dated entries each name the activity, technician administration time is walled off, and 148 minutes is checkable arithmetic rather than an estimate. The unit math shows its work. The base code appears once for the episode, one add-on is earned, and the second add-on is explicitly not claimed, with its 151-minute threshold stated. The date of service is defensible. The episode concludes at feedback, the claim carries that date, and the day-span sits in the record exactly as CMS date-of-service guidance expects. The interpretation is the clinician's own. The synthesis ties verified scores to history and collateral and ends in a signed impression, with nothing delegated and no pasted computer narrative. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Sort the obligations by strength before you template anything. LAW: coverage of clinical psychologist services rests on the Social Security Act and 42 CFR 410.71, plus general medical-necessity documentation rules; no federal text names a scoring or interpretation note. State law can go further: Massachusetts Medicaid regulation 130 CMR 411.413 requires, for psychological assessment records, behavioral observations, original responses, a summary of scores, and a comprehensive written report, a rare case of a statute defining the note's contents. PAYER POLICY: MAC coverage articles such as LCD L34646 with Article A57481, and L34520 with A57780, expect the record to support medical necessity, expect a written report to be generated, and set a 31-minute minimum before the base code; Nevada Medicaid adds prior authorization for 96130 to 96133 under announcements effective January 12, 2026. CONVENTION: the titled note itself, start and stop times, and any fixed report length; median reports ran 5, 6, and 8 pages across medical, rehabilitation, and private-practice settings in one study (Donders, 2001), and no regulation sets a length. The work itself is the interpreting clinician's own: Medicare does not pay for evaluation services performed by students or trainees, technicians cannot perform them, and master's-level clinicians are recognized unevenly by payer, so verify before the episode starts. Ontario adds a rule automated scoring makes urgent: the college standard carried forward as 14.8 bars substituting a computer-generated report for the psychologist's professional opinion. Time and dates are the audit surface. Evaluation work done on a day the patient is not present is billable, and the episode is reported on the date the service concluded, per CMS MLN Matters SE17023 , with the day-span reflected in the record; the base code is reported once per episode, and the claim waits until the evaluation is complete, typically the feedback date. Total dated minutes are the requirement; start and stop times are defensive convention, and entries like 'approximately 1 to 2 hours' are a documented denial risk. Expect scrutiny of long episodes: one commercial policy treats 2 to 8 hours as the usual complete evaluation and requires a battery list and rationale beyond 8. A cluster of CO-50 with N115, CO-16, or CO-151 denials on testing claims is the signal to pull your MAC's article and rebut it section by section. The format is a convention; the content is the requirement. Elsewhere the frame changes: Australia has no MBS item comparable to 96130, so psychological services bill as time-tiered attendance items and the record answers the Health Insurance Act's adequate and contemporaneous standard, where Reg 6 prescribes each entry's contents and notes added after the attendance do not count toward consultation time. Canadian obligations run through provincial colleges and privacy statutes rather than a testing code. Keep retention to the longest applicable rule: about 7 years for US adults under the APA guideline with a 6-year federal lookback in practice, Ontario 10 years, British Columbia 16 years from April 1, 2026, Alberta 10 to 11. The claim-side reconciliation is its own artifact, the claim-support billing note , and the clinical deliverable is the psychological evaluation report . ↑ Back to the template and downloads ## Common testing scoring note errors auditors flag The enforcement record around time-based mental health codes is concrete. In a May 2023 federal audit of pandemic-era psychotherapy services (HHS OIG report A-09-21-03021), reviewers examined 216 sampled enrollee-days and found time improperly documented or required information missing for 128 of them, with provider signatures absent for 54; the office estimated $580 million of roughly $1 billion in Medicare psychotherapy payments were improper. That audit is psychotherapy, not testing, and no published improper-payment rate isolates 96130 to 96133; what transfers is the failure modes, time and signatures, which are exactly what this note exists to prove. Australia shows the ceiling: a Professional Services Review determination effective 27 June 2025 ordered a practitioner to repay $300,000 and disqualified her from MBS billing for three years, with failure to keep adequate and contemporaneous records among the central findings on a psychological-strategies item, though the practitioner was a general practitioner rather than a psychologist. The BastionGPT Clinical Advisory Board sees the same errors most often in testing scoring note reviews: Vague or undated time. 'Approximately 1 to 2 hours' and undifferentiated block entries cannot be matched to units and draw CO-151 denials. Each entry needs a date, a named activity, and minutes. A base code every day. Reporting 96130 or 96132 for each sitting instead of once per episode, with combined time on the conclusion date, is the multi-day error auditors unwind first and payers recoup fastest. Delegated interpretation. A technician or trainee drafts the synthesis and the psychologist signs it. Medicare does not pay for evaluation services performed by students or trainees; the performance requirement fails even when the signature is real. Evaluation billed without administration. Test selection is bundled into evaluation services, so the evaluation codes ride with administration services from the same episode. A scoring note with no administration record behind it fails the pairing before medical necessity is even reached. Scores that do not match the platform. Transcription drift and uncorrected auto-scoring errors, or a computer narrative pasted as the interpretation. Preserve exact values and test names, record every correction, and keep the synthesis in your own words. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological testing scoring notes. Extract every score, index, and validity flag from scoring-platform output into a clean summary that preserves exact values and exact test names. Check the note before you sign: dated minutes supporting the units, the base code once per episode, the date of service matching the episode's conclusion, and report and feedback status closed out. Produce a de-identified score summary and time log for consultation, supervision, or teaching, with identifiers stripped. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Can I bill for scoring and interpretation done on a day the patient was not there? Yes. Integrating data, interpreting results, and report writing count toward the evaluation-services codes whether or not the patient is present, and no regulation requires the patient in the room for that work. When a service begins on one day and concludes on another, CMS MLN Matters SE17023 directs that the claim carry the date the service concluded, with the record reflecting the span. In practice: log each day's evaluation minutes, report the combined episode on the conclusion date, usually feedback, and use the base code once. Top-ranking billing pages routinely omit this rule, which makes it the most valuable one to get right. Do I need start and stop times in a testing scoring note? No CMS rule mandates literal start and stop clock times for 96130 to 96133. The codes are time-based, so the record must support the total time behind the units, and dated entries with a named activity and minutes satisfy that. Start-stop notation is a defensive convention that makes audits smoother, while vague entries such as 'approximately 1 to 2 hours' are a documented denial risk. If your MAC's billing article ever adds an explicit start-stop requirement, upgrade the log then; until that day, precise dated totals win. When is the base code billable, and how do the add-on units work? MAC billing articles set a 31-minute minimum before the base code is assigned, and the base (96130 or 96132) is reported once per assessment episode even when the work spans several days. Add-on hours (96131, 96133) accrue past each full hour of combined evaluation time, so 148 minutes supports the base plus one add-on, and a second add-on would need 151. Expect scrutiny of long episodes: one commercial policy treats 2 to 8 hours as the range of a complete evaluation and asks for a battery list and rationale past 8 hours. Can a technician, trainee, or master's-level clinician do the interpretation? Not for Medicare. Administration and scoring can be delegated to a trained technician under the technician codes, but evaluation services are the interpreting clinician's own work, and Medicare does not pay 96130 or 96132 performed by students or trainees. A technician's role in administration never disqualifies the professional's evaluation claim. Master's-level clinicians such as LPCs and LCSWs are recognized unevenly: some commercial payers allow these codes and Medicare generally does not, so verify each payer in writing before the episode starts. How is the scoring note different from the administration note and the evaluation report? Three artifacts, three jobs. The administration record proves the giving and mechanical scoring of tests, work a technician can perform. The scoring note proves the interpreting clinician's own evaluation work: score verification, integration, dated minutes, and decisions. The psychological evaluation report or neuropsychological report is the clinical deliverable those minutes produce, and MAC articles expect it to exist. In a time-based-code audit, this note is what survives; the report alone rarely proves the minutes. Do actual scores belong in the scoring note? Yes, in exact form. Massachusetts Medicaid regulation 130 CMR 411.413 makes behavioral observations, original responses, and a summary of scores part of the required assessment record at the state-law level, and preserving exact values and exact test names is what lets a reviewer reconcile the note, the scoring-platform output, and the report. Verify computer scoring before interpreting anything: Ontario's college standard bars substituting a computer-generated report for the psychologist's professional opinion, a rule written for precisely the moment an auto-scored profile is tempting to paste. How long must psychological testing records be kept? Set retention to the longest rule that touches you. In the US the APA guideline is seven years after last service for adults and three years after a minor reaches majority, with state law able to override and federal auditors using a six-year lookback. Ontario requires ten years, or ten past age eighteen, whichever is later; British Columbia moves to sixteen years from April 1, 2026; Alberta works out to ten or eleven. In Australia, follow state rules and note that the Better Access referral itself must be kept for two years. Is there an Australian MBS equivalent of 96130? No. Australia has no separate scoring-and-interpretation item; psychological services bill as time-tiered attendance items, so US-style evaluation-services claims do not transfer. What Australia regulates hard is the record: the Health Insurance Act requires adequate and contemporaneous records, Reg 6 of the PSR Scheme Regulations prescribes each entry's contents, and MBS notes count only details recorded at the time of attendance toward consultation time, so material added later does not help. A 2025 Professional Services Review determination ordering a $300,000 repayment with a three-year billing disqualification shows the enforcement ceiling. Can BastionGPT draft psychological testing scoring notes? Yes. Give it the day's work, scores verified, anomalies found, minutes per activity, and it drafts the dated log, the score summary with every value preserved, and the sign-off block for your review. It can also extract exact scores from platform output into a clean summary, check a finished note for the gaps auditors flag (vague time, a missing report status, a base code repeated across days), and produce a de-identified version for consultation or teaching. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Psychoeducational Evaluation Report Claim-Support Billing Note Outcome Measure Review Note MMPI-3 Report Write-Up Psychological Testing Administration Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/fitness-for-duty-evaluation All note templates # Fitness-for-Duty Evaluation: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A fitness-for-duty evaluation (FFDE) is a formal psychological examination of an employee, ordered by the employer when objective evidence suggests a mental health condition may be impairing safe or effective job performance. Psychologists and psychiatrists retained by the employer conduct them; the employer, not the examinee, is the client. Most reports run 2 to 8 pages and answer a specific referral question. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Licensed psychologists and psychiatrists retained by the employer; police and occupational specialists Audience The employer or agency: HR, legal counsel, command staff; sometimes the examinee Typical length 1,000 to 4,000 words (2 to 8 pages) · 4 to 8 hours of interview, testing, collateral review, and writing (clinical team estimate) Format family Third-party occupational evaluation report (compare: psychological evaluation report, IME) When it's used When objective evidence ties a possible psychological condition to unsafe or ineffective performance of a specific job Standards context No law prescribes the report format; the ADA governs when an employer may compel one, and IACP guidelines shape the convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a fitness-for-duty evaluation? A fitness-for-duty evaluation is a specialized psychological examination of a current employee, triggered when two things line up: objective evidence that the employee may be unable to perform a defined job safely or effectively, and a reasonable basis to believe a psychological condition may be the cause. That two-part trigger comes from the IACP Police Psychological Services Section, the body that standardized the format for public safety: its FFDE guidelines were first adopted in 1998 and have been revised and re-ratified since, with the 2013 edition the most widely circulated. Clinicians and employers also call it an FFD exam, a psychological fitness-for-duty evaluation, or a return-to-duty evaluation, and under the Americans with Disabilities Act the operative term is simply a medical examination. In Canada and Australia the same evaluation usually travels as an independent medical examination (IME). Two boundaries define the document. First, the employer or agency, not the examinee, is the client: the referral question comes from the employer, the employer pays, and the report goes to the employer, which reverses most of the assumptions a treatment record runs on. An FFDE is not a psychological evaluation report written for diagnosis and treatment planning, and it is not the FMLA fitness-for-duty certification under 29 CFR 825.312, a short return-from-leave form with its own rules. Second, no US, Canadian, or Australian statute prescribes the report's sections: the format is convention, chiefly the IACP guidelines, which state expressly that they are aspirational rather than a standard of care. What the law does govern is the surrounding transaction: under the ADA, 42 U.S.C. 12112(d)(4), an employer may require the exam only where it is "job-related and consistent with business necessity", and the resulting medical information must be kept as a confidential medical record, separate from the personnel file. ## Who uses fitness-for-duty evaluations and when Employers request FFDEs wherever impaired performance carries safety consequences: police and corrections agencies, fire and EMS, aviation, rail, nuclear and utility operators, hospitals, and general employers acting on credible reports of concerning workplace behavior. The examiner is usually a licensed psychologist or psychiatrist with occupational and forensic competence, and sector rules can make qualifications hard requirements: California's POST Regulation 1955, for example, requires five years of experience diagnosing and treating mental disorders, three of them post-doctoral, before a psychologist may evaluate peace officers. The FFDE answers one narrow question, current capability to perform a specific job. When the concern is dangerousness itself, the violence risk assessment framework applies; when the referral turns on substance use, it pairs with a substance use assessment ; and a treating clinician asked about their own patient's work status should write a brief return-to-work letter instead, because the IACP guidelines treat evaluating your own therapy client as a conflict to decline. ## Fitness-for-duty evaluation structure: what goes in each section Referral question and business-necessity basis. Who ordered the exam, the objective evidence behind it (dated incidents, direct observations, credible reports), and the specific questions the employer wants answered, restated in your opening lines. Courts test FFDEs at this gate, not on report format. Pitfall: accepting a vague or diagnosis-hunting referral; if the employer cannot state objective evidence, the exam itself may violate the ADA. Notification and limits of confidentiality. What the examinee was told before questioning began: that the employer is the client, who will receive the report, that the exam is not treatment, and the possible employment consequences of declining. Record the explanation and the examinee's response. Pitfall: importing treatment-style consent; a compelled exam calls for documented notification of limits, and a signature alone proves little if the explanation was never recorded. Sources of information. Every basis for the opinion: interview dates and duration, records reviewed, collateral contacts, and each test administered. Pitfall: building the opinion on the employer's account alone; a one-sided foundation collapses under challenge. Job analysis and essential functions. The job description, the essential functions in play, and any list the employer supplied. This is what converts clinical findings into an occupational opinion. Pitfall: opinions floating free of the job; a limitation means nothing until it is tied to a specific essential function. Testing and validity. Instruments chosen with a stated reason each one fits this referral question and this job, plus response-style measures. Pitfall: a battery with no validity testing; in an adversarial exam, unmeasured over-reporting or under-reporting undermines every score. Functional opinion. Fit, unfit, or fit with restrictions, stated against the named essential functions, with the clinical detail retained in the examiner's file. Pitfall: sending the employer a diagnosis instead of functional language; several states restrict what an examiner may disclose to job-relevant limitations. Restrictions, prognosis, reevaluation, and distribution. Restrictions or accommodations considered, prognosis, a dated reevaluation window, the named recipients, and the instruction to file the report separately from the personnel file. Pitfall: an unfit opinion with no path back and no distribution limits; both invite the grievance the report could have prevented. ### Blank template (copy and adapt) PSYCHOLOGICAL FITNESS-FOR-DUTY EVALUATION (CONFIDENTIAL) To (employer contact, title): __________________________________________ From (examiner, credentials): __________________________________________ Re (employee, position): ______________________ Exam date(s): _________ Referral received: __________ Report date: __________ REFERRAL QUESTION AND OBJECTIVE EVIDENCE (as provided by the employer) _______________________________________________________________________ _______________________________________________________________________ NOTIFICATION GIVEN (client, report recipients, not treatment, consequences of declining, examinee response) _______________________________________________________________________ SOURCES (interviews with dates, records reviewed, collateral, tests administered with validity measures) _______________________________________________________________________ _______________________________________________________________________ ESSENTIAL JOB FUNCTIONS CONSIDERED (from the job description) _______________________________________________________________________ FUNCTIONAL FINDINGS (job-relevant; diagnosis only where authorized) _______________________________________________________________________ _______________________________________________________________________ OPINION: [ ] Fit for duty [ ] Fit with restrictions [ ] Unfit Restrictions or accommodations: ________________________________________ Prognosis and reevaluation date: _______________________________________ DISTRIBUTION (named recipients only; file separate from personnel file) _______________________________________________________________________ Signature and credentials: ____________________ Date: __________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample fitness-for-duty evaluation Scenario: a municipal water utility refers a 46-year-old control room operator for evaluation after documented episodes of disorientation and an unlogged alarm. A licensed psychologist evaluates for the employer. The sample is condensed but structurally complete; a full report runs 2 to 8 pages. All details are fictional. Psychological Fitness-for-Duty Evaluation (Confidential). To: R. Alvarez, HR Director, Calder Ridge Municipal Water Utility · From: L. Whitfield, PhD, licensed psychologist · Re: D.M., 46, Control Room Operator II · Exam dates: 07/16/2026 and 07/20/2026 · Report date: 07/24/2026 Referral question and basis: The utility requests this evaluation under its fitness-for-duty policy following documented objective evidence: supervisor memos of 06/26/2026 and 07/07/2026 describing episodes in which Mr. M. appeared disoriented at the operations console, and a low-pressure alarm on 07/07/2026 that was neither logged nor escalated and was discovered at shift handover. The written referral asks: (1) whether Mr. M. can currently perform the essential functions of Control Room Operator II safely and effectively; (2) if not, what restrictions or conditions would permit safe performance; (3) prognosis and a reevaluation timeline. No diagnosis was requested. Notification: Before the first interview I explained, orally and in writing, that the utility is the client for this evaluation, that a report limited to functional fitness will go to the HR director and the utility's counsel only, that the evaluation is not treatment and creates no treatment relationship, and that declining to participate could carry employment consequences determined by the employer. Mr. M. signed the notification and chose to proceed. Sources: Clinical interviews 07/16/2026 and 07/20/2026 (3.5 hours total); the utility's job description for Control Room Operator II (essential functions include continuous alarm monitoring, SCADA operation, emergency response initiation, and 12-hour rotating shifts); the two supervisor memos; attendance summary; psychological testing with embedded and stand-alone validity measures (instruments and scores retained in the examiner's file); collateral interview with the shift supervisor on 07/21/2026, conducted with Mr. M.'s written permission. Functional findings: Validity indicators supported an interpretable, candid presentation. Mr. M. described roughly four months of severely fragmented sleep following his move to a rotating 12-hour schedule, with involuntary dozing during night shifts, and testing showed sustained-attention and vigilance performance well below the level his role requires. There was no indication of substance involvement and no indication that he minimized the alarm event, which he reported himself in interview before being shown the memo. A specific, treatable clinical explanation was identified and discussed with Mr. M. with referral options; consistent with the notification and with confidentiality law, this report is limited to functional findings. Opinion: Mr. M. is currently unfit to perform the essential functions of Control Room Operator II, specifically continuous alarm monitoring and emergency response initiation on rotating shifts. The impairment appears treatable and time-limited rather than fixed. Restrictions and reevaluation: If the utility elects modified duty, Mr. M. could perform non-safety-critical daytime tasks that do not require sustained solitary monitoring. Prognosis for return to full duty is good, contingent on treatment engagement. Reevaluation is recommended in 60 to 90 days, or sooner if the utility receives documentation of treatment response. Distribution: Provided to R. Alvarez (HR Director) and utility counsel under the fitness-for-duty policy; maintain as a confidential medical record separate from the personnel file. Not authorized for further release. Signed: L. Whitfield, PhD, licensed psychologist, 07/24/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The business-necessity basis sits on the face of the report. Dated incidents and the employer's three specific questions are restated up front, the element courts examine first when an FFDE is challenged. Notification is documented as an event, not a form. The report records what was explained: who the client is, who receives the report, that the exam is not treatment, and the consequence of declining. The opinion is functional, not diagnostic. Capability is stated against named essential functions while clinical detail stays in the examiner's file, the posture state confidentiality law rewards. Every finding ties to a specific job function. Sustained attention maps to alarm monitoring and emergency initiation, so the unfit opinion reads as an occupational conclusion, not a judgment of the person. The report closes with a path back. Modified-duty options, prognosis, a dated reevaluation window, and a named, closed distribution list filed apart from the personnel record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations An FFDE report lives under different privacy law than a treatment chart. When an examiner evaluates for an employer, bills no health plan, and forms no treatment relationship, the report is generally not protected health information under HIPAA: the definition at 45 CFR 160.103 excludes employment records, and the examiner is typically not acting as a covered entity for that service. What governs instead is the ADA confidentiality rule, 29 CFR 1630.14(c)(1), which requires medical information to be collected on separate forms and kept as a confidential medical record apart from the personnel file, plus state law that can be stricter: California's Confidentiality of Medical Information Act, as applied in Pettus v. Cole, limits what an examiner may tell the employer to functional information rather than diagnosis. Access differs by country too: a US examinee has no general federal right to the report, while Canadian health-privacy statutes and Australia's APP 12 create access rights, and an Australian examiner should assume full Privacy Act obligations at collection because the employee-records exemption does not attach until the employer holds the record. Retention rules can conflict (EEOC recordkeeping, OSHA's employment-plus-30-years rule where exposure records are involved, and state board minimums such as California's seven years); apply the longest period that plausibly reaches the report. The payer rule is absolute: bill the employer under your engagement agreement, never a health plan. An employer-ordered FFDE is not for the examinee's diagnosis or treatment, so it sits outside Medicare's benefit under 42 U.S.C. 1395y(a)(1)(A) and outside private coverage, and submitting it to a health plan can constitute insurance fraud; EEOC guidance adds that when the employer chooses the examiner, the employer pays all costs. Sector rules then layer on top: state POST regulations for peace officers (California's Regulation 1955 sets examiner-qualification floors and 12 hours of biennial POST-approved continuing education), federal aviation and motor-carrier medical tracks, and the NRC fitness-for-duty program under 10 CFR Part 26. Keep the FMLA lane separate: a fitness-for-duty certification under 29 CFR 825.312 is a different document, limited to the condition that caused the leave, with the employer required to supply the essential-functions list. FFDE work is legal-adjacent by nature: when a referral, a records demand, or a disclosure question feels contested, involve your attorney or licensing board before acting, because state rules vary. ↑ Back to the template and downloads ## Common fitness-for-duty evaluation errors auditors flag FFDEs rarely fail on format; they fail at the gate and at the boundaries. In EEOC v. Dillard's, a policy requiring employees to disclose the specific nature of their medical conditions ended in a 2 million dollar consent-decree settlement in 2012, and the EEOC's resolved-case list for mental health conditions under the ADA includes a 75,000 dollar settlement where an employee cleared to return by the FFDE psychologist was forced out anyway. In a 2020 federal-sector appeal, Linn A. v. Department of the Army, the physical exam survived review but the psychological FFDE was found not job-related and consistent with business necessity, the two judged separately. The outcome literature, meanwhile, contains no peer-reviewed fit-versus-unfit base rates: the widely quoted four outcomes are a taxonomy from Rostow and Davis, not statistics. The BastionGPT Clinical Advisory Board sees the same errors most often in fitness-for-duty evaluation reviews: A referral with no objective evidence. The exam proceeds on rumor, an anonymous letter, or generalized worry. Courts have called employer discretion to order exams "hardly unbounded" (Sullivan v. River Valley School District), and a New Jersey court rejected an FFDE ordered on an anonymous letter alone. A diagnosis delivered to the employer. The report names a disorder where functional limitations would answer the question; in California and states like it, that disclosure can itself be unlawful. The exam billed to a health plan. A third-party evaluation submitted as if it were treatment: a non-covered service at best, insurance fraud at worst. Treatment habits imported into a third-party frame. HIPAA assumed to govern, treatment-style consent standing in for documented notification, or an evaluator examining their own therapy client despite the conflict. No path back from unfit. No restrictions considered, no reevaluation window, no distribution limits. The opinion reads as termination cover rather than evaluation, which is what draws grievances and litigation. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on fitness-for-duty evaluations. Draft the report from your interview notes, test summaries, and the employer's referral: basis restated, notification documented, findings organized by essential function. Check the draft before release: functional language only, no diagnosis where law restricts it, every limitation tied to a named job function, reevaluation and distribution stated. Convert one evaluation into each audience's document: the employer report, a plain-language explanation of the process for the examinee, and your own file memo. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Who is the client in a fitness-for-duty evaluation? The employer or agency, not the person being examined. The employer defines the referral question, pays for the evaluation, and receives the report, and no treatment relationship forms. That single fact reorganizes everything: consent becomes notification of limits, the record usually sits outside HIPAA, and the deliverable is an opinion about job capability rather than care. A psychological evaluation report written for diagnosis and treatment runs on the opposite assumptions. Is a specific FFDE report format required by law? No. No US, Canadian, or Australian statute prescribes the report's sections. The closest thing to a standard is the IACP Police Psychological Services Section guidelines, and they state expressly that they are aspirational rather than a standard of care. What the law does control is the trigger: under the ADA an employer may compel the exam only where it is job-related and consistent with business necessity. Most reports run 2 to 8 pages and follow the conventional sequence on this page. Can I bill an FFDE to the employee's health insurance? No. The evaluation is not for the examinee's diagnosis or treatment, so it falls outside Medicare's benefit under 42 U.S.C. 1395y(a)(1)(A) and outside private health coverage, and submitting it to a health plan can constitute insurance fraud. The employer pays under your engagement agreement, and EEOC guidance holds that when the employer chooses the examiner, the employer bears all costs. The same logic applies to provincial plans in Canada and to Medicare in Australia. Should the report give the employer a diagnosis? Generally no, and in some states you must not. Best practice, EEOC guidance, and state confidentiality statutes such as California's CMIA, applied in Pettus v. Cole, point the same direction: report fitness and job-relevant functional limitations, and keep diagnosis, test data, and clinical detail in the examiner's file unless law and a signed authorization clearly permit more. Does HIPAA protect a fitness-for-duty report? Usually not. HIPAA's definition of protected health information excludes employment records, and an examiner who has no treatment relationship and bills no health plan is generally not acting as a covered entity for that service. The operative protections are the ADA confidentiality rule, which requires the employer to keep medical information in a separate confidential file, and state law, which can be stricter. Assume the report will be read by HR, counsel, and possibly an arbitrator, and write accordingly. Can I do an FFDE on my own therapy client? It is strongly discouraged. The IACP guidelines treat evaluating a current or former therapy client as a conflict to decline, and multiple-relationship rules in professional ethics codes point the same way. No general statute prohibits it, but the dual role compromises both the therapy and the credibility of the opinion. If a patient's employer needs a fitness opinion, refer them to an independent evaluator and confine your own role to a treating clinician's return-to-work letter. Does the employee get a copy of the FFDE report? It depends on where you practice. In the US there is no general federal right for the examinee to obtain the report, though notification should say where it goes. In Canada, health-privacy statutes such as Ontario's PHIPA create access rights, and in Australia APP 12 does the same; Australian case law adds that the examining psychologist must meet full Privacy Act obligations at collection, because the employee-records exemption only attaches once the employer holds the record. How long should fitness-for-duty records be kept? Apply the longest rule that plausibly reaches the report, because the frameworks conflict. EEOC recordkeeping regulations set a floor of one year for employment records, OSHA requires employment plus 30 years for exposure-related medical records where that rule applies, and licensing boards set their own minimums, such as seven years from the last professional contact for California psychologists. Store the report as a confidential medical record separate from the personnel file in every case. Can BastionGPT help write a fitness-for-duty evaluation? Yes. Give it the referral, your interview notes, and test summaries, and it drafts the report with the basis restated, the notification documented, and findings organized by essential function, then checks that the language stays functional and the distribution stays closed. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Consultation Response & Report Violence & Homicide Risk Assessment Substance Use Assessment Forensic Psychological Evaluation Report Capacity & Competency Evaluation Workers' Compensation Initial Report Workers' Compensation Progress Report Return-to-Work & Accommodation Letter Disability Evaluation & Medical Source Statement FMLA & Leave Certification Support Note STOP-Bang Score Documentation Epworth Sleepiness Scale Documentation WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ndis-progress-report All note templates # NDIS Plan Reassessment & Progress Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 An NDIS progress report, called a plan reassessment report in current NDIA language, is the document a treating allied health provider writes toward the end of a participant's plan period. It summarises the supports delivered, evidences progress against a functional baseline using named outcome measures, and justifies recommended supports for the next plan against the NDIS reasonable and necessary criteria. Most run 3 to 6 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, occupational therapists, speech pathologists, physiotherapists and other treating allied health providers; provisional psychologists under supervision Audience The participant and their supports, the support coordinator, and the NDIA delegate deciding the next plan Typical length 1,200 to 3,000 words (3 to 6 pages) · 2 to 4 hours of scoring, collateral review and writing (clinical team estimate) Format family NDIS evidence report (compare: functional capacity assessment, support coordination report) When it's used In the weeks before the plan reassessment date, or when a participant seeks a s 48 reassessment after a change in circumstances Standards context No law or NDIA rule mandates the report or a template; NDIA guidance shapes the content, and the s 34 criteria decide the funding On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an NDIS plan reassessment report? An NDIS plan reassessment report, still called an NDIS progress report by most clinicians, is the document a treating allied health provider writes toward the end of a participant's plan period. It does three jobs: summarises the NDIS supports delivered under the current plan, evidences the participant's progress toward funded goals against a functional baseline, and recommends supports for the next plan period with justification the National Disability Insurance Agency can act on. The NDIA's guidance for allied health providers frames the report's purpose as helping the Agency decide whether the supports you recommend "meet our reasonable and necessary criteria to fund". The document is a creature of the NDIS Act 2013 rather than of any professional college: s 48 governs reassessment, s 50 lets the CEO call for information and reports, and s 34's reasonable and necessary criteria are the test every recommendation must survive. The renaming matters, too. The NDIA now says plan reassessment (s 48) to keep it distinct from internal review of a decision (s 100) and plan variation (s 47A), three mechanisms older template pages run together. The same document also answers to progress report, plan review report, therapy report, and S48 report. Two facts reorganise how the report is written, and both are missing from most template pages. First, no statute, NDIS rule, or NDIA instrument compels a treating provider to write one, and none prescribes a template: the NDIA standardises adjacent documents, the Evidence of Psychosocial Disability form, assistive technology and home modification templates, the early childhood provider report, but deliberately not this one. The format is a convention; the persuasive content is the requirement. Second, the report is participant-controlled: NDIA guidance directs the provider to give the finished report to the participant, who then shares it with the Agency, so the participant, not the clinician, decides release. That also draws the boundary with its heavier neighbour. An NDIS functional capacity assessment is a standalone, in-depth assessment used when a new baseline or a major funding question is at stake; the reassessment report is lighter, longitudinal and progress-focused, and usually cites the most recent FCA rather than repeating it. A treatment plan review lives in the clinical record and serves continuity of care; it is not written to the s 34 framework and does not substitute. ## Who uses NDIS plan reassessment reports and when Any treating allied health provider whose supports are funded in the plan writes one: psychologists delivering capacity-building supports for psychosocial disability, occupational therapists, speech pathologists, physiotherapists and behaviour support practitioners. Provisional psychologists and students on placement can deliver supports and contribute to the report under supervision, with the participant's consent documented in the service agreement; NDIA-managed plans require a registered provider, and the supervising psychologist remains responsible for the interpretation and the reporting. The trigger is the plan cycle: a report is commissioned in the weeks before the plan's reassessment date, or earlier when circumstances change and the participant asks the CEO to reassess under s 48, a request the Agency must answer within 21 days. Choose the document to match the question. Progress against existing goals belongs here; a new baseline or a high-stakes funding question belongs in a functional capacity assessment ; a diagnostic question belongs in a psychological evaluation report ; and Medicare's Better Access pathway is a different payer with a different statutory test, so a report written for one never serves the other. The audience is about to shift at the margins as well: from 1 October 2026 the Thriving Kids foundational-supports transition begins moving children aged 8 and under with lower-level developmental needs out of the scheme, reshaping who needs these reports at all. ## NDIS plan reassessment report structure: what goes in each section Participant, provider and service details. Participant name, NDIS number, date of birth and plan period; provider and author details with credentials and registration; and the service pattern: date of first service, most recent service, and total sessions delivered. The contact pattern is evidence in its own right: in Ray [2020] AATA 3452 the Tribunal preferred the treating psychologist who had seen the applicant on "approximately 50 to 60 occasions" over the NDIA's one-off independent assessor. Pitfall: omitting frequency and duration of contact, the very fact that gives a treating provider's observations their weight. Supports delivered this plan period. What was funded and what was actually delivered: support type, frequency, setting, attendance, and the plan goals each support served. Pitfall: a therapy diary instead of a delivery account; the delegate needs delivered-versus-funded and goal linkage, not session narratives. Functional baseline and outcome measures. The participant's functional capacity at the start of the plan period and the instruments that quantified it, repeated at reassessment. The NDIA's own examples are physical measures such as the Timed Up and Go and the Berg Balance Scale; psychologists more often reach for the WHODAS 2.0, goal attainment scaling, or a symptom measure tied to function (see the outcome measure note ). No instrument is mandated; a named one beats an adjective every time. Pitfall: progress claims with no baseline; change cannot be shown without a starting point. Progress toward each goal. Goal by goal: the measured change, what it means functionally, and an explanation wherever a goal was not achieved. Pitfall: documenting diagnosis and treatment instead of functional impact; evidence that explains the condition rather than how it affects daily activities is the dominant documented reason NDIS evidence falls short. Barriers, risks and informal supports. Barriers encountered and how they were addressed, risks in delivering supports and how they are managed, and the informal, community and mainstream supports in the participant's life. Pitfall: treating informal supports as none of the report's business; s 34 makes them part of the funding test, and silence reads as an unexamined recommendation. Recommendations for the next plan period. Each recommended support with frequency, intensity and duration, argued against the reasonable and necessary criteria: related to the disability, effective and beneficial per current good practice, value for money, informal supports taken into account, and not more appropriately funded by the health or education system. State expected outcomes and how they will be measured. Pitfall: recommendations stated as preferences; the delegate funds arguments, not requests. Author details, declaration and release. Name, credentials, registration details, signature and date; a supervisor countersignature where a provisional psychologist contributed; and a record that the report was provided to the participant. Pitfall: importing vendor rules as law; no instrument requires an AHPRA number on the report, per-page signatures or a fixed section count, and time spent on invented formalities is better spent on the s 34 argument. ### Blank template (copy and adapt) NDIS PLAN REASSESSMENT / PROGRESS REPORT (CONFIDENTIAL) Participant: ____________________ NDIS number: _______________ Date of birth: __________ Plan period: __________ to __________ Provider (registered / unregistered): __________________________ Author (name, credentials, registration): ______________________ First service: __________ Most recent: __________ Total: ____ SUPPORTS DELIVERED THIS PLAN PERIOD (type, frequency, setting, attendance, linked plan goals) _______________________________________________________________ _______________________________________________________________ FUNCTIONAL BASELINE AND OUTCOME MEASURES (instrument, date and score at plan start; repeat measure at reassessment) _______________________________________________________________ PROGRESS TOWARD EACH GOAL (measured change; explain any goal not achieved) _______________________________________________________________ _______________________________________________________________ BARRIERS AND RISKS; INFORMAL, COMMUNITY, MAINSTREAM SUPPORTS _______________________________________________________________ RECOMMENDATIONS FOR THE NEXT PLAN (support, frequency, duration; reasonable and necessary justification; expected outcomes) _______________________________________________________________ _______________________________________________________________ Provided to the participant on: __________ Signature, credentials and date: _______________________________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample NDIS plan reassessment report Scenario: a registered psychologist writes the plan reassessment report for an adult participant with psychosocial disability, ahead of the plan's August 2026 reassessment date. The sample is condensed but structurally complete; a full report runs 3 to 6 pages. All details are fictional. NDIS Plan Reassessment Report (Progress Report). Participant: J.L., 34 · NDIS number: 430 187 265 · Plan period: 26 August 2025 to 25 August 2026 · Author: R. Nguyen, MPsych, registered psychologist, Harbourline Psychology (registered NDIS provider) · Report date: 24 July 2026 Supports delivered: Under the plan's capacity-building funding (improved daily living), Ms L. received 31 of 36 scheduled individual psychology sessions between 4 September 2025 and 21 July 2026, weekly to fortnightly, in clinic and community settings. Work targeted the plan's three goals: independent travel, maintaining her tenancy and daily living routine, and weekly community participation. Non-face-to-face time for this report was agreed in the service agreement of 28 August 2025. Functional baseline and measures: At plan start (September 2025), WHODAS 2.0 (12-item) total 32, with getting-around and life-activities domains most affected; goal attainment scaling (GAS) baselines were set for each goal. At reassessment (July 2026), WHODAS 2.0 total 21, improvement concentrated in the same domains. Progress toward goals: Goal 1, travel independently to regular commitments: partly achieved. Ms L. now travels by bus to two standing commitments without support (baseline: none); GAS +1 against a +2 target, with anticipatory anxiety the remaining limiter. Goal 2, maintain tenancy and daily living routine: achieved. No missed rent or utility actions since November 2025; support worker collateral confirms independent completion of the weekly routine built in sessions; GAS +2. Goal 3, join one community activity weekly: not achieved. A two-week psychiatric admission in February 2026 and the recovery period that followed interrupted graded community work; attendance at her chosen group resumed in May 2026 at fortnightly; GAS 0. The goal remains appropriate with the staged approach recommended below. Barriers, risks and informal supports: The February admission was managed by the community mental health team, which continues to hold clinical treatment of Ms L.'s condition; no NDIS funding is sought for treatment itself. Informal support is one weekly shopping trip with her mother, who reports capacity to continue. No other mainstream services are engaged. Recommendations for the next plan (12 months): (1) Continue individual psychology fortnightly, 26 hours: related to the psychosocial disability, consistent with current good practice for functional recovery, and value for money in that it is designed to reduce the support-worker hours below. (2) Maintain support worker assistance at 2 hours per week, stepping down to 1 hour from month 6, contingent on Goal 3 progress. (3) No change to transport funding. Expected outcomes by month 9: independent travel to all regular commitments and weekly community participation, measured on the same instruments. Informal supports are unchanged; clinical treatment remains with the health system. Release: Provided to Ms L. on 24 July 2026 to share with the NDIA as she chooses; a copy is retained in the clinical record. Signed: R. Nguyen, MPsych, registered psychologist (AHPRA reg. PSY0009134712), 24 July 2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Function leads and diagnosis stays in the background. Every progress claim is a measured functional change; the condition is named once and never argued, avoiding the diagnosis-heavy, function-light evidence that most often sinks NDIS reports. Baseline, endpoint and instrument are named for every claim. WHODAS 2.0 and GAS scores at both ends of the plan period make progress auditable rather than asserted. The unmet goal is explained, not hidden. The admission, its effect on the graded community work, and the staged path back tell the delegate why continuing the support is reasonable rather than reading as evidence it failed. The contact pattern sits on the face of the report. Thirty-one sessions across eleven months is the sustained-treating-contact evidence tribunals have preferred over one-off assessments. Each recommendation argues the s 34 criteria and the release path is documented. Disability relation, current good practice, value for money, informal supports and the health-system boundary each get a sentence, and the report goes to the participant, who controls sharing. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The report has two lives: it is part of your clinical record under your profession's rules, and it becomes evidence in the participant's NDIS record the moment they share it, so write it for both readers. Retention follows health-records law and board expectations rather than any NDIS rule: for psychologists, keep records at least 7 years after the last occasion of service for adults, and for minors until age 25, with some state regimes and organisational policies running to 10 years; the Psychology Board's code of conduct in force since 1 December 2025 makes record-keeping an enforceable conduct standard. Release runs through the participant: NDIA guidance directs you to give the report to the participant, who shares it with the Agency, so record the date you provided it and leave distribution with them; only support coordination reporting routes through the provider portal. Accuracy is an integrity matter as much as a clinical one, because progress documentation supports payment claims and false claims have been prosecuted. Billing for the report is payer policy, not entitlement. Under the NDIS Pricing Arrangements and Price Limits , report writing is non-face-to-face service delivery, claimable against the plan only where the service agreement records the participant's agreement and the report relates directly to the funded support, and the hours draw on the participant's own therapy budget, which makes transparency about writing time part of the clinical relationship. The deeper point most template pages miss: the format is a convention; the content is the requirement. The NDIA lists what a persuasive report covers, baseline, measures, progress, barriers, informal supports, justified recommendations, and prescribes none of it as form, so aim the content at the s 34 criteria and the format takes care of itself. Date-stamp your practice, too, because the evidence model is moving: the 2024 amending Act created a statutory needs assessment report (s 32L) for new-framework plans, new variation and reassessment rules commenced in 2025, the pricing arrangements turn over annually, and from 1 October 2026 the Thriving Kids transition begins moving younger children with lower-level needs toward foundational supports outside the scheme. Keep Medicare separate: an NDIS report and a Better Access treatment plan answer different payers and different statutory tests, and neither substitutes for the other. Track outcomes on the instruments you will report with (see the outcome measure note ) and the reassessment report largely writes itself. ↑ Back to the template and downloads ## Common NDIS progress report errors auditors flag The system-level numbers explain why these reports are read sceptically. The Auditor-General found the NDIA's decision controls for reasonable and necessary supports wanting, with internal quality reviews of funding decisions "continually below target" (ANAO Report No. 14 of 2020-21), and in 2024-25 about 45% of substantive Administrative Review Tribunal decisions changed the NDIA position: 55% affirmed, 11% varied, 34% set aside ( NDIS Quarterly Report , Q4 2024-25). Integrity pressure runs the other way: the Fraud Fusion Taskforce, 23 agencies, has launched more than 630 investigations since 2022 and had disrupted over 1,900 problematic providers by 30 June 2025, and in 2025 two former Sydney providers were jailed over $1.2 million in fraudulent claims, so a progress report is read both as clinical evidence and as a claims-integrity document. The BastionGPT Clinical Advisory Board sees the same errors most often in NDIS progress report reviews: A diagnosis report in a progress report's clothing. The report proves the disability exists and describes treatment, but never shows how the disability affects daily activities or what changed. Evidence that focuses on diagnosis rather than functional impact is the dominant documented reason NDIS evidence falls short. No baseline, no named measure, no explanation. Progress is asserted without the start-of-plan measurement that would make it verifiable, and unmet goals pass in silence when NDIA guidance expects an explanation for every goal not achieved. Recommendations without the s 34 argument. Supports are listed with hours attached and no case made: nothing on disability relation, current good practice, value for money, informal supports, or why the health or education system is not the more appropriate funder. Submitting over the participant's head. The report goes straight to the NDIA when guidance directs the provider to give it to the participant, who controls release. The confusion imports the support coordinator's portal pathway into treating-provider work. Report-writing time handled badly in both directions. Hours claimed as non-face-to-face without the service-agreement basis the pricing arrangements require, or claimable writing time absorbed unpaid; either way the participant was never told the report draws on their own budget. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on NDIS plan reassessment reports. Draft the full report from the plan goals, your session notes and outcome measure scores: baseline restated, progress measured goal by goal, and every recommendation argued against the reasonable and necessary criteria. Check a finished draft before you sign: function leading diagnosis, an explanation for every unmet goal, informal and mainstream supports addressed, and no recommendation left as a bare request. Convert one plan period's session data into each audience's document: the reassessment report, a plain-language progress summary for the participant, and a handover note for the support coordinator. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to use an official NDIA template for a plan reassessment report? No. The NDIA publishes content guidance for allied health plan reassessment reports, not a mandatory template, and no statute or NDIS rule prescribes a format. The prescribed forms exist for adjacent documents instead: the Evidence of Psychosocial Disability form, assistive technology and home modification templates, and the early childhood provider report. Vendor pages selling a required multi-section structure are describing their own synthesis, not law. Cover the NDIA's expected content in a consistent house structure and spend the saved effort on the s 34 argument. Can I bill for the time it takes to write an NDIS progress report? Yes, as non-face-to-face service delivery under the NDIS Pricing Arrangements and Price Limits, provided the service agreement records the participant's agreement and the report relates directly to the funded support. It is not an automatic or unlimited entitlement, and the hours draw on the participant's own therapy budget, so quote the writing time up front and record the agreement. Support coordinators claim their own reporting separately under different line items. Do I send the report to the NDIA or give it to the participant? Give it to the participant. NDIA guidance for allied health providers is that the clinician shares the finished report with the participant, who then shares it with the Agency; the participant controls release and consent. Support coordination reports travel differently, increasingly through the provider portal, which is why so many template pages get this wrong. Record the date you provided the report in your clinical record. How is a plan reassessment report different from a functional capacity assessment? Depth and job. An NDIS functional capacity assessment is a standalone, in-depth assessment across functional domains, commissioned when a new baseline or a major funding question such as supported independent living or home modifications is at stake. The plan reassessment report is lighter and longitudinal: it tracks progress against the existing baseline over one plan period and argues the next plan's supports. A good reassessment report cites the most recent FCA rather than repeating it. Can a provisional psychologist write an NDIS progress report? A provisional psychologist or student on placement can deliver supports and contribute to the report under the supervision of a qualified practitioner, with the participant's consent documented in the service agreement. The supervising psychologist remains professionally responsible for the interpretation and reporting, so the supervisor reviews and countersigns as a matter of professional regulation, not NDIA rule. NDIA-managed plans require a registered provider; self-managed and plan-managed participants can engage provisional psychologists more freely. How recent does the evidence in a reassessment report need to be? No statute or rule sets a currency limit or a submission deadline. In practice the NDIA gives most weight to recent functional evidence, practice guidance commonly treats reports older than about 12 months as significantly weaker, and once a reassessment request is accepted the Agency's reported average completion time has been around 33 days, so a report finished close to the reassessment window lands while it is still current. Sustained treating contact adds weight of its own: tribunals have preferred a treating psychologist with dozens of contacts across settings over a one-off independent assessor. How long do I need to keep NDIS progress reports? Follow your profession's record rules, because no NDIS-specific retention rule replaces them. For psychologists that means keeping records at least 7 years after the last occasion of service for adults, and for minors until age 25 or 7 years after last service, whichever is longer, under Psychology Board expectations, APS guidance and state health records law; some state regimes and policies run to 10 years. The Psychology Board's code of conduct in force since 1 December 2025 makes record-keeping an enforceable conduct standard. Will a strong report guarantee the recommended supports are funded? No. The report is evidence; the funding decision belongs to the NDIA delegate under s 34, and the reasonable and necessary criteria decide it. Strong functional evidence still moves outcomes: in McCutcheon [2015] AATA 624 the Tribunal rejected the NDIA's insufficient-evidence position because the reports showed the support was maintaining function, and in 2024-25 about 45% of substantive tribunal decisions varied or set aside the NDIA position. Write the report so a delegate can say yes and a tribunal can see why. Can BastionGPT write an NDIS progress report from my session notes? Yes. Give it the plan goals, your session notes and outcome measure scores, and it drafts the report with the baseline restated, progress measured goal by goal, unmet goals explained, and recommendations argued against the reasonable and necessary criteria, ready for your clinical review. It can also convert the same material into a plain-language summary for the participant. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates NDIS Functional Capacity Assessment Report Psychological Evaluation Report Consultation Response & Report Better Access Mental Health Treatment Plan Referral Treatment Plan Review & Update Outcome Measure Review Note Aboriginal & Torres Strait Islander Culturally Safe Assessment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/capacity-evaluation All note templates # Capacity & Competency Evaluation: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A capacity evaluation is a clinician's structured assessment of whether a person can make a specific decision at a specific time, such as consenting to treatment, managing finances, or making a will. Psychologists, psychiatrists, and physicians write them for treating teams, courts, and tribunals. Capacity is the clinical judgment; competency is the court's legal determination. Most reports run 2 to 8 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, neuropsychologists, psychiatrists, and physicians; some jurisdictions require designated capacity assessors Audience Treating teams, courts and tribunals, attorneys, guardians, families Typical length 1,000 to 4,000 words (2 to 8 pages) · 2 to 6 hours of interview, testing, collateral, and writing; a MacCAT-T interview alone runs about 30 minutes (clinical team estimate) Format family Decision-specific evaluation report (compare: psychological evaluation report, neuropsychological report) When it's used When a specific decision is in doubt: consenting to treatment, managing money or property, making a will, or a guardianship petition Standards context The four-abilities model is convention, partly written into state law; where a court form applies (GC-335, Form 4), content is mandated On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a capacity evaluation? A capacity evaluation is a structured clinical assessment of whether a person can make a particular decision at a particular time: consent to a proposed treatment, manage money and property, make a will, or decide where to live. The dominant framework is the four-abilities model set out by Paul Appelbaum and Thomas Grisso in the New England Journal of Medicine in 1988: to hold capacity for a decision, the person must be able to communicate a choice, understand the relevant information, appreciate the situation and its consequences, and reason about the options. The assessment runs decision by decision, so the same person can hold capacity for one choice and lack it for another on the same afternoon, and neither a diagnosis nor an age makes anyone incapable by itself. Clinicians and courts also call the document a capacity assessment, a decision-making capacity (DMC) evaluation, a competency evaluation, a Capacity Declaration (California), or a Capacity Assessment Report (Ontario and Alberta). Two boundary lines organize the field. First, capacity is clinical and competency is legal: a clinician assesses and documents decision-specific abilities, while only a court or tribunal can change a person's legal status, a distinction the VA writes directly into its informed-consent directive. Second, the clinical model has been partly written into law. Many US states have incorporated one or more of the four abilities into their statutory definitions of decisional capacity, with the consequence, as a 2023 review in the Journal of the American Academy of Psychiatry and the Law put it, that "physicians in many jurisdictions are unable to modify these criteria on their own." Where a court form applies, content is mandated outright: California's GC-335 Capacity Declaration is a Judicial Council form signed under penalty of perjury, and Alberta's Form 4 must accompany guardianship and trusteeship applications. A capacity evaluation is narrower than a psychological evaluation report , which characterizes diagnosis and treatment needs broadly; the capacity report answers one legal-functional question. ## Who uses capacity evaluations and when Hospital consultation-liaison services carry the highest volume: a systematic review in JAMA found that 26% of medicine inpatients lack capacity for treatment decisions, against 2.8% of healthy older adults, and that clinicians had recognized the incapacity in only 42% of affected patients. Beyond the bedside, geriatric and memory clinics assess financial and testamentary capacity, psychologists and neuropsychologists write reports for guardianship and conservatorship petitions, and designated assessors complete statutory forms in the provinces that require them. Choose the document by the question: a capacity evaluation when one identified decision is at issue, a neuropsychological report when the referral needs cognitive characterization across domains, and a mental status exam to record the bedside cognitive findings that feed either one. In hospital work the capacity opinion often lands inside a consultation-liaison note rather than a stand-alone report. ## Capacity evaluation structure: what goes in each section Referral question and the decision at issue. Who asked, the specific decision or transaction in question, and the legal standard or court form that governs it: a statutory test, GC-335, Form 4, or the treating-consent standard. One decision per opinion; list multiple decisions separately. Pitfall: accepting a global "is this person competent?" referral; capacity is decision-specific, and a global answer is the first thing a reviewing court discounts. Notification and consent for the assessment. What the person was told about the purpose and the report's recipients, and how they responded, including a refusal where the statute gives them that right (Ontario's guardianship assessments carry rights-information requirements). Pitfall: no record of what was explained; in guardianship work the assessment itself can be challenged before the opinion is ever reached. Background, diagnosis, and functional evidence. The relevant history and diagnosis, plus the real-world events that prompted the referral, dated and sourced: missed medications, unpaid bills, a lease signed twice. Pitfall: letting the diagnosis do the work; dementia, psychosis, and intellectual disability are context, not conclusions, and no diagnosis equals incapacity by itself. Method, sources, and setting. Interview dates, place, and duration; instruments used and why each fits this decision; records reviewed; collateral contacts. No statute in the US, Canada, or Australia requires a specific instrument, so state why the tools you chose serve the question. Pitfall: presenting the MacCAT-T or a cognitive screen as legally required or self-sufficient; instruments aid the judgment, they are not the judgment. The four abilities, with decision-specific evidence. Understanding, appreciation, reasoning, and expressing a choice, each supported by what the person actually said or did with the information relevant to this decision, paraphrases in their own words included. Pitfall: conclusory ability labels ("understanding intact") with no quoted or observed evidence; unstructured clinical judgment agreed only 56% of the time across physicians rating the same mild dementia patients, and evidence is what closes that gap. Supports, undue influence, and less restrictive alternatives. Whether education, aids, a supporter, or better timing would restore capacity for the decision; whether anyone stands to gain from the outcome; and what short of guardianship would meet the need, since modern statutes require the least restrictive option. Pitfall: silence on alternatives; a published audit found 56.4% of guardianship applications failed to link the powers sought to the person's actual vulnerabilities. Opinion, limits, and declaration. The decision-specific opinion, its date, what could change it, a reevaluation trigger, and your qualifications in the form the jurisdiction requires, dated within any statutory window (Alberta's Form 4 must sit within six months of the application). Pitfall: opining that the person "is incompetent"; legal status is the court's call, and a strong report says so on its face. ### Blank template (copy and adapt) CAPACITY EVALUATION REPORT (CONFIDENTIAL) Person evaluated: ____________________ DOB/age: _______________ Referral source(s): __________________ Report date: ___________ Decision(s) at issue (one opinion per decision): _______________ Legal standard or court form, if any: __________________________ NOTIFICATION (purpose explained, report recipients, response, right to refuse where the statute provides one) ________________________________________________________________ BACKGROUND, DIAGNOSIS & FUNCTIONAL EVIDENCE (dated events, sourced; diagnosis is context, not the conclusion) ________________________________________________________________ ________________________________________________________________ METHOD, SOURCES & SETTING (interview dates, place, duration; instruments and why each fits; records; collateral) ________________________________________________________________ DECISION-SPECIFIC ABILITIES (evidence for each, in the person's own words or observed performance) Understanding: _________________________________________________ Appreciation: __________________________________________________ Reasoning: _____________________________________________________ Expressing a choice: ___________________________________________ SUPPORTING DATA (cognitive testing with limits of scores stated) ________________________________________________________________ SUPPORTS, UNDUE INFLUENCE & LESS RESTRICTIVE ALTERNATIVES ________________________________________________________________ OPINION & LIMITS (decision-specific, dated; what could change it; reevaluation trigger; the court decides legal status) ________________________________________________________________ Examiner qualifications & declaration: _________________________ Signature and credentials: ____________________ Date: _________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample capacity evaluation Scenario: an 81-year-old widow with mild vascular neurocognitive disorder is referred for two decisions: consent to a proposed anticoagulant, and ongoing management of her finances. A licensed psychologist evaluates her at home. The sample is condensed but structurally complete; a full report runs 2 to 8 pages. All details are fictional. Capacity Evaluation Report (Confidential). Person evaluated: E.H., 81 · Referral: A. Osei, MD (cardiology) and R. Feld, JD (Ms. H.'s attorney, at her request) · Exam dates and setting: 07/21/2026 and 07/23/2026, at her home · Report date: 07/27/2026 · Examiner: N. Reyes, PsyD, licensed psychologist Referral questions and decisions at issue: (1) Does Ms. H. currently have the capacity to consent to, or refuse, the anticoagulant her cardiologist has proposed for newly diagnosed atrial fibrillation? (2) Does she currently have the capacity to manage her financial affairs, and if not, what is the least restrictive arrangement that would meet the need? No guardianship petition is on file; her attorney raised the second question while reviewing her power of attorney, and Ms. H. agreed to both questions being assessed. Notification: At the start of the first visit I explained the purpose of the evaluation, that a written report would go to Dr. Osei, Mr. Feld, and Ms. H. herself, that the evaluation is not treatment, and that she could pause or decline at any point. She restated the purpose accurately ("you are checking whether I can decide about the blood thinner and the money myself") and chose to proceed. Background, diagnosis, and functional evidence: Mild vascular neurocognitive disorder, diagnosed 03/2026 following a hospitalization for heart failure. Records and collateral (her daughter, interviewed 07/22/2026 with Ms. H.'s written consent) document two duplicate charity payments in May 2026, a property tax notice in June 2026 that went unpaid until her daughter found it, and difficulty using the bank's online system since her branch closed. She fills her own weekly medication organizer without errors, keeps her apartment, and cooks for herself. Method, sources, and setting: Semi-structured four-abilities interviews at both visits (framework chosen as an aid; no instrument is legally required), a financial-management discussion using her own statements with her consent, the MoCA, review of hospital and primary care records, and the collateral interview. Both visits took place at her kitchen table, mid-morning, at her preference. Decision 1, consent to anticoagulation: Understanding: she explained in her own words that the medication "thins the blood so a clot is less likely to cause a stroke" and named bleeding as the main risk. Appreciation: she connected the choice to her own condition ("my heartbeat is the reason") and to her sister's stroke. Reasoning: she weighed the bleeding risk against stroke prevention and asked how monitoring would work. Choice: she wants the medication, stated consistently across both visits. Opinion: Ms. H. currently has the capacity to make this treatment decision. Decision 2, financial management: Understanding of her income, regular bills, and approximate balances was adequate. On functional tasks with her own documents, she could not locate or reconcile the duplicate May payments, attributed the unpaid tax notice to a bank error after reviewing it, and could not complete a two-step allocation task she chose herself. Reasoning about routine spending was intact and consistent with her values. Opinion: Ms. H. currently lacks the capacity to manage complex finances (account tracking, taxes, large or novel transactions) and retains the capacity for routine daily spending. Supporting data: MoCA 21/30, with losses in delayed recall and serial subtraction, consistent with the 03/2026 diagnosis. The score is supporting context only; the opinions above rest on the decision-specific evidence. Supports, undue influence, and less restrictive alternatives: Automatic payment of recurring bills, paper statements, and a standing weekly review with her daughter under the existing financial power of attorney would meet the identified need, and Ms. H. endorsed that arrangement ("the help, not the takeover"). Screening identified no indicators of pressure or influence by any party, and her stated preferences match long-held values per collateral. On these findings, guardianship is not indicated. Opinion, limits, and declaration: The opinions are decision-specific and current as of 07/27/2026. Contributing factors (recent hospitalization, the transition to online banking) may change; I recommend reassessment of financial capacity in six months, or sooner if new errors emerge. Whether any legal arrangement should change is a question for Ms. H., her attorney, and, if ever petitioned, a court; this report informs that judgment and does not make it. I am a psychologist licensed in this state, and I examined Ms. H. in person on the dates and in the setting above. Signed: N. Reyes, PsyD, licensed psychologist, 07/27/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works One decision, one opinion. The two referral questions get two separate ability analyses and reach opposite conclusions, which is exactly what decision-specific means and what reviewing courts look for. Every ability carries evidence. Understanding, appreciation, reasoning, and choice each quote the person's own words or observed performance rather than a conclusory label. The score supports, it does not decide. The MoCA sits in supporting data with its limits stated, and the functional evidence carries the opinion. Less restrictive alternatives are analyzed before guardianship is even mentioned, with an existing power of attorney and named supports doing the work, and the influence screen documented. The opinion states its own limits. Dated, contributing factors named, a reevaluation trigger set, and the court's role expressly left to the court. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A capacity report travels further than most clinical documents: to attorneys, courts, tribunals, and the person evaluated, sometimes years later in cross-examination, so write for that audience and confirm the release of information pathway for anyone the referral does not already cover. Check who may lawfully perform the assessment before accepting it, because authorship is regulated in ways clinicians routinely get wrong. In Ontario, guardianship assessments under the Substitute Decisions Act may only be done by designated capacity assessors, members of listed colleges with required training and at least 1 million dollars in liability insurance, who must follow the province's assessment guidelines ; treatment-consent capacity under the Health Care Consent Act, by contrast, is assessed by the practitioner proposing the treatment, no designation required. California's GC-335 accepts a physician or psychologist and is signed under penalty of perjury; Alberta's Form 4 accepts a physician, a College of Alberta Psychologists member, or a designated assessor, and must be dated within six months of the application. Retention follows your record rules, not the court's: HIPAA's six-year rule at 45 CFR 164.530(j) covers required administrative documentation, while patient-record retention comes from state and provincial law, commonly 10 years in Ontario and seven years, or to age 25, under Victoria's health-records legislation. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). Billing turns on one question: is the evaluation part of diagnosis or treatment, or is it for a legal proceeding? Treatment-linked assessment and any medically necessary testing are billable in the ordinary way, subject to Medicare's testing LCD (L34646). Court-ordered and forensic capacity work generally is not: Medicare excludes services that are not for diagnosis or treatment, and 42 CFR 411.4 adds the exclusion for services a legal process is responsible for arranging. The practical route is the one coding forums describe: code the encounter with Z04.6, the ICD-10 code for an evaluation requested by an authority, obtain an advance notice where Medicare might otherwise be billed, and bill the requesting party directly under a fee agreement. Report preparation for a third party (90889) is not a Medicare benefit, and record review (90885) is not separately payable. Australia mirrors the split: medico-legal reports sit outside MBS treatment items, and the Professional Services Review has ordered repayment of roughly 100,000 dollars where records did not support what was billed. Capacity work is legal-adjacent by nature: when a guardianship petition, subpoena, or contested assessment is in play, involve your attorney or licensing board before acting, because state and provincial rules vary. ↑ Back to the template and downloads ## Common capacity evaluation errors auditors flag The audit trail for capacity work runs through courtrooms as much as payers. When researchers reviewed 183 guardianship applications for a 2016 BJPsych Bulletin study, clinicians had failed to apply the statutory capacity definition explicitly in 47.5% of reports, and 56.4% of applications did not link the powers sought to the person's actual vulnerabilities; Ontario built its assessment guidelines partly because assessors applying the wrong legal test were producing inappropriate findings. On the payer side, a federal OIG review of psychological testing in nursing homes found 39% of the testing medically unnecessary, an audit finding from the 1999 to 2001 reviews that still shapes contractor scrutiny of testing claims. The reliability literature explains why evidence matters: five physicians judging the same mild Alzheimer's patients agreed on capacity only 56% of the time, against 98% for healthy controls. The BastionGPT Clinical Advisory Board sees the same errors most often in capacity evaluation reviews: A global verdict instead of a decision-specific opinion. The report answers "is this person competent?" in one line. Capacity holds for some decisions and not others; tie each opinion to one named decision, or the whole report gets discounted. A diagnosis or a screening score doing the work of a functional analysis. "Dementia, therefore incapable" and "low MoCA, therefore incapable" are the same error: cognitive screens have limited diagnostic value for capacity except at the extremes, and the four abilities must be evidenced, not inferred. The statutory test never applied. Nearly half the audited guardianship reports (47.5%) did not use the legal definition of incapacity explicitly. Name the standard that governs your jurisdiction and organize the findings against it. No analysis of supports or less restrictive alternatives. Modern statutes require the least restrictive option, and 56.4% of audited applications failed to link the powers sought to the person's vulnerabilities. An opinion with no alternatives analysis reads as advocacy for the petition. The wrong signer or a stale assessment. An Ontario guardianship assessment by a non-designated clinician, an Alberta Form 4 older than six months, or a GC-335 completed outside the authorized professions fails on formal grounds before anyone reads the content. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on capacity evaluations. Draft the report from your interview notes and instrument summaries, with the four abilities organized decision by decision and every ability tied to its evidence. Summarize records and collateral into the background section: dated functional events, each with its source, kept separate from conclusions. Check the draft before you sign: statutory standard named, supports and less restrictive alternatives addressed, opinion inside the referral question, dates within any required window. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Who can perform a capacity evaluation, a psychologist or only a physician? It depends on the jurisdiction and the decision. California's GC-335 Capacity Declaration accepts a physician or a licensed psychologist. Ontario splits it: guardianship assessments under the Substitute Decisions Act require a designated capacity assessor drawn from listed professions with mandatory training, while treatment-consent capacity under the Health Care Consent Act is assessed by the practitioner proposing the treatment. Alberta's Form 4 accepts a physician, a College of Alberta Psychologists member, or a designated assessor. Check the governing statute before you accept the referral. What is the difference between capacity and competency? Capacity is a clinical judgment: a clinician's assessment of whether the person can make a specific decision now. Competency, or legal capacity, is a legal status that only a court or tribunal can change. A capacity evaluation informs the court; it never removes anyone's rights by itself, and a well-written report says so. The VA draws the same line explicitly in its informed-consent policy. Do I have to use the MacCAT-T or another instrument? No. No US, Canadian, or Australian statute reviewed for this page mandates the MacCAT-T, the Aid to Capacity Evaluation, or any other instrument, and none requires formal cognitive testing. Structured tools earn their place on reliability: the MacCAT-T interview takes about 30 minutes, the ACE about 10 to 20, and structure sharply narrows examiner disagreement in borderline cases. Choose the aid that fits the decision and record why you chose it. Can I bill insurance for a capacity evaluation? Only when it is part of diagnosis or treatment. A bedside evaluation of consent capacity during care is clinical work, billable in the ordinary way. A court-ordered or attorney-requested evaluation generally is not: Medicare excludes services that are not for diagnosis or treatment and services a legal process is responsible for arranging, so practitioners code the encounter Z04.6, use an advance notice where relevant, and bill the requesting party directly under a fee agreement. Is a low MMSE or MoCA score enough to decide capacity? No. Capacity is decision-specific and functional, and the systematic review behind modern practice found cognitive screens carry limited diagnostic value for it except at the extremes. A score belongs in supporting data, framed by what the person actually did with the information relevant to the decision. Two people with the same score can differ on the same decision, and one person can hold capacity for treatment consent while lacking it for finances. How long is a capacity evaluation report? Most stand-alone reports run 2 to 8 pages (1,000 to 4,000 words) and take 2 to 6 hours of interview, testing, collateral, and writing. Bedside treatment-consent determinations are usually shorter and live inside a consultation-liaison note or progress note, while statutory forms like GC-335 and Alberta's Form 4 set their own structure. How is a capacity evaluation different from a neuropsychological evaluation? A neuropsychological evaluation characterizes cognitive strengths and deficits across domains; it often feeds a capacity opinion but does not answer it, because capacity turns on how the person handles one identified decision, not on scores alone. Order a capacity evaluation when a specific decision is in question; order a neuropsychological evaluation when the referral needs the cognitive map. How recent must a capacity assessment be for court? Where the statute sets a window, follow it: Alberta's Form 4 must be dated within six months of the guardianship or trusteeship application. California sets recency by local court practice rather than a single statewide number, and other jurisdictions vary. Everywhere, the working rule is the same: the assessment must reflect the person's current abilities, so a material change in condition means reassessment regardless of the calendar. Can BastionGPT help write a capacity evaluation report? Yes. Give it the referral question, your interview notes, and any instrument summaries, and it drafts the report with the four abilities organized decision by decision, then checks that every ability carries evidence, the statutory standard is named, and less restrictive alternatives are addressed. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Mental Status Exam (MSE) Note Hospital Consult & Liaison Note Release of Information (ROI) Authorization Fitness-for-Duty Evaluation Forensic Psychological Evaluation Report Disability Evaluation & Medical Source Statement Court Report & Treatment Progress Update MoCA Score Documentation MMSE Score Documentation Mini-Cog Score Documentation SLUMS Score Documentation Trail Making Test Score Documentation Clock Drawing Test Score Documentation PCL-R Report Write-Up Glasgow Coma Scale (GCS) Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/forensic-evaluation-report All note templates # Forensic Psychological Evaluation Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A forensic psychological evaluation report is the written product of an evaluation conducted to answer a legal question, such as competency to stand trial, criminal responsibility, or parenting capacity. A court or attorney is the client, the person examined is an evaluee rather than a patient, and confidentiality is limited. Psychologists and psychiatrists write them; length varies by matter, and no statute sets a maximum. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Forensic psychologists and psychiatrists; competency work requires a licensed, court-designated examiner in most jurisdictions Audience Judges, attorneys for both sides, review boards; copies flow by statute or court order, not by patient request Typical length 2,500 to 10,000 words (roughly 8 to 35 pages) · 10 to 40 hours of interviews, testing, records, collateral, and writing, varying by matter (clinical team estimate); no rule sets a maximum length Format family Forensic mental health assessment (FMHA) report (compare: psychological evaluation report, capacity evaluation, violence risk assessment) When it's used When a court or attorney poses a legal question: adjudicative competency, criminal responsibility, sentencing, civil damages, custody and parenting, guardianship Standards context Structure is convention (Grisso's model, 2013 APA and 2015 AAPL guidelines); a few statutes mandate content, led by 18 U.S.C. 4247(c) On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a forensic psychological evaluation report? A forensic psychological evaluation report is the written product of an assessment conducted to answer a legal question posed by a court or a retaining attorney: competency to stand trial, criminal responsibility, sentencing mitigation, psychological damages in civil litigation, parenting capacity, or guardianship. The person examined is an evaluee, not a patient: there is generally no treatment relationship, confidentiality is reduced and explained up front, and the report is written to be read by judges and lawyers rather than clinicians. The modern structure descends from Thomas Grisso's Evaluating Competencies (1986) and the forensic mental health assessment (FMHA) framework, carried forward by the 2013 Specialty Guidelines for Forensic Psychology and the 2015 AAPL practice guideline . Courts and clinicians also call it a forensic mental health assessment, a competency or fitness-to-stand-trial report, a criminal responsibility or sanity evaluation, a child custody or parenting evaluation, or, in US federal work, a 4247(c) report. Two boundary lines organize the type. First, forensic is not clinical: a psychological evaluation report serves diagnosis and treatment of your own patient and is confidential and usually billable, while a forensic report answers a legal question about an evaluee and is neither. Because a retained, non-treating examiner is generally not acting as the evaluee's health care provider, the report typically sits outside the HIPAA designated record set, and access runs through the court rather than a patient-access request. Second, the format is convention with narrow statutory islands. No US statute dictates the internal structure of forensic reports generally, but where a statute speaks it is specific: 18 U.S.C. 4247(c) requires a federal mental-condition report to state the person's history and present symptoms, the tests employed and their results, the examiner's findings, and opinions on diagnosis, prognosis, and the legal question, with copies to counsel for both sides, and Texas Art. 46B.025 says a competency report "must state an opinion on a defendant's competency or incompetency to stand trial" or explain why the examiner cannot give one. ## Who uses forensic psychological evaluation reports and when Criminal courts generate the highest volume: published national estimates of US competency-to-stand-trial evaluations have ranged from roughly 19,000 to 94,000 per year ( Morris and McNiel, 2021 ), and more recent commentary describes continued growth. Beyond competency work, forensic psychologists and psychiatrists write criminal responsibility and sentencing evaluations, child custody and parenting evaluations for family courts, and civil reports on psychological damages and testamentary issues. Choose the document by the question and the client: when a court or attorney needs a legal question answered, the forensic report is the vehicle; when the goal is diagnosis and treatment planning for your own patient, write a psychological evaluation report or a psychiatric diagnostic evaluation ; when the referral needs cognitive characterization, a neuropsychological report can serve either context depending on who is asking. A violence risk assessment can stand alone in clinical care or feed a forensic opinion, and a capacity evaluation answers the narrower question of whether a person can make one specific decision. ## Forensic evaluation report structure: what goes in each section Identifying data and the referral question. Who ordered or retained the evaluation, under what statute or rule, the case caption, and the specific psycholegal question or questions. One opinion per question; list multiple questions separately. Pitfall: accepting a vague referral ("please evaluate the defendant"); every later section inherits the vagueness, so pin the legal question in writing before the first interview. Notification of purpose and limits of confidentiality. What the evaluee was told: who requested the report, who will receive it, that the usual treatment confidentiality does not apply, and how the evaluee responded. Pitfall: reflexively writing "informed consent obtained"; a court-ordered evaluee often cannot decline, so the accurate record is notification given and understanding demonstrated, which guidelines treat as its own report section. Sources of information. Every interview with date and duration, every record reviewed, every collateral contact, every instrument administered, ideally with the time devoted to each. Pitfall: an undisclosed source surfacing on cross-examination; if it informed the opinion it belongs on the list, and if you chose not to rely on it, be ready to say why. Relevant history. Psychosocial, medical, psychiatric, and legal history at the depth the question requires, with every fact attributed to where it came from. Pitfall: repeating the evaluee's self-report as established fact; write "Mr. R. reported" and corroborate against records, because unattributed history reads as endorsement. Mental status and clinical findings. Observations across contacts and formal mental status exam findings, plus a statement addressing response style and the consistency of the presentation across interviews, records, and collateral. Pitfall: silence on validity; if the report never says how the evaluee approached the evaluation, the first cross-examination question writes itself. Tests and instruments, with results. Name each instrument, why it fits this question, and the results in language a judge can use. Forensic instruments and clinical tests answer different questions, so say which role each one plays. Pitfall: scores reported but never connected to the legal standard, which leaves the court to guess at the link the examiner was retained to supply. Psycholegal analysis and opinions. The reasoning section: link each clinical finding to the specific legal capacity at issue, then state each opinion with its reasoning, its limits, and its currency, observing any ultimate-issue restriction that applies. Pitfall: a conclusion with no stated rationale, the defect audits flag most; in one published Hawaii review, 63% of competency reports failed to state one. ### Blank template (copy and adapt) FORENSIC PSYCHOLOGICAL EVALUATION REPORT (CONFIDENTIAL COURT DOCUMENT) Evaluee: ______________________ DOB/age: ______________________ Court or retaining party & case no.: ___________________________ Referral question(s) & governing statute or rule: ______________ Report date: ___________ Exam date(s) & total time: ___________ NOTIFICATION GIVEN (purpose, who receives the report, limits of confidentiality, evaluee's response) ________________________________________________________________ SOURCES OF INFORMATION (each interview with date and duration; each record reviewed; each collateral contact; each test) ________________________________________________________________ ________________________________________________________________ RELEVANT HISTORY (psychosocial, medical, psychiatric, legal; attribute every fact to its source) ________________________________________________________________ ________________________________________________________________ MENTAL STATUS & CLINICAL FINDINGS (observations across contacts; response style and consistency addressed) ________________________________________________________________ TESTS & INSTRUMENTS (why each fits the question; results in plain language) ________________________________________________________________ PSYCHOLEGAL ANALYSIS (link each finding to the legal standard; state the reasoning, not just the conclusion) ________________________________________________________________ ________________________________________________________________ OPINION(S) (one per referral question; limits and currency stated; the legal determination belongs to the court) ________________________________________________________________ Examiner qualifications & declaration: _________________________ Signature and credentials: ____________________ Date: _________ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample forensic psychological evaluation report Scenario: a court-ordered competency-to-stand-trial evaluation of a 29-year-old defendant with schizophrenia, stabilized on medication in detention and facing a nonviolent felony charge. The sample is condensed but structurally complete; a full report runs longer. All details are fictional. Forensic Psychological Evaluation Report (Confidential Court Document). Evaluee: D.R., 29 · Matter: State v. D.R., Case No. 26-CR-0412, felony property charge · Ordered by: District Court, under the state competency statute · Exams: 07/14/2026 and 07/16/2026 at the county detention center, 2 hours 45 minutes total · Report date: 07/22/2026 · Examiner: L. Marsh, PhD, licensed psychologist, court-designated competency examiner Referral question: Whether D.R. has a factual and rational understanding of the proceedings against him and sufficient present ability to consult with his attorney, under the state competency statute. The court's order poses that single question, and this report answers no other. Notification: At the start of each contact I explained who ordered the evaluation, that a written report goes to the court and to counsel for both sides, that the evaluation is not treatment, and that what he tells me is not confidential in the treatment sense. D.R. restated the purpose accurately ("you check whether I can work with my lawyer, and the judge reads it") and participated in both interviews. Sources of information: Interviews of D.R. on 07/14/2026 (90 minutes) and 07/16/2026 (75 minutes, including a structured competency interview); charging documents; detention medical record, 03/2026 to present; state hospital discharge summary, 11/2023; collateral contact with detention nursing staff regarding medication adherence (07/15/2026, with facility authorization); telephone contact with defense counsel regarding attorney-client interactions (07/15/2026, 15 minutes, with counsel's agreement). Relevant history: Records document schizophrenia first diagnosed at age 23, with hospitalizations in 2021 and 2023; the 11/2023 discharge summary describes stabilization on medication. Detention medical records show a long-acting injectable antipsychotic since 03/2026 with documented adherence, corroborated by nursing staff. D.R. reported completing a GED and three years of warehouse work, consistent with the record. Available records show no prior criminal history. Mental status and clinical findings: Across both contacts he was alert, oriented, and adequately groomed, with organized speech and goal-directed thought. He described residual auditory hallucinations as "quieter and easy to ignore" on current medication; direct inquiry elicited no delusional content bearing on the case, the court, or his attorney. Attention and memory were adequate for sustained interview. His account was consistent within and across interviews and matched the records and collateral; nothing in his presentation suggested exaggeration or minimization. Psycholegal analysis: Factual understanding: he named the charge and its felony status, described the roles of judge, jury, prosecutor, and defense counsel in his own words ("my lawyer speaks for me; the prosecutor tries to prove it"), and explained the plea options and their general consequences. Rational understanding: he weighed the pending plea offer against trial in concrete, self-referenced terms ("taking the offer means less time than losing at trial; I want my lawyer to try for less") and connected his choices to what he understands the evidence to be. Assisting counsel: he described the content and frequency of meetings with his attorney, disagreed with counsel on one strategic point and explained his reasoning calmly, and defense counsel independently described him as engaged and able to track their discussions. His conduct in both interviews, sustained attention, responsive answers, and relevant questions, was consistent with that account. Opinion and reasoning: In my opinion, stated to a reasonable degree of psychological certainty as the court's order directs, D.R. currently has a factual and rational understanding of the proceedings and sufficient present ability to consult with his attorney. The schizophrenia diagnosis is documented and treatment continues; the opinion rests not on the diagnosis but on the demonstrated abilities above, each observed directly and corroborated by records and collateral. The opinion is current and assumes continued treatment; a significant disruption in medication or functioning before trial would warrant reevaluation. Whether D.R. is competent to stand trial is a determination for the court; this report informs that determination. Declaration: I am a psychologist licensed in this state and designated by the court to conduct competency examinations; my qualifications are on file with the court. I examined D.R. in person on the dates above and reviewed the sources listed. Signed: L. Marsh, PhD, 07/22/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The legal question is named up front and answered exactly once. Every section serves the referral question, and the report says on its face that it answers no other. Notification replaces treatment-style consent. The report records what the evaluee was told, his response, and the limits of confidentiality, which is what the guidelines expect for a court-ordered exam the evaluee cannot decline. Every source carries a date and a duration, so cross-examination cannot surface an unlisted interview, record, or phone call. The reasoning from findings to opinion is stated, not implied. Each competency ability is evidenced by what the evaluee said or did, closing the gap audits flag most: in one published review, 63% of competency reports failed to state a rationale for the opinion. The diagnosis is context, not the conclusion. A defendant with schizophrenia is found competent on demonstrated abilities, the opinion states its limits and currency, and the legal determination is expressly left to the court. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A forensic report does not live in the treatment record and does not follow treatment rules. Distribution is set by the statute or order that created it: 18 U.S.C. 4247(c) requires the federal report filed with the court with copies to counsel for both sides, and state statutes and court orders set their own lists. Because a retained, non-treating examiner is generally not acting as the evaluee's health care provider, the report typically sits outside the HIPAA designated record set, and the evaluee's route to a copy runs through the court rather than a patient-access request; collateral contacts outside the order's scope still need a documented release of information or the court's authorization. Raw test data have their own rule: under APA Ethics Code 9.04, absent a release, psychologists provide test data "only as required by law or court order," and courts have ordered production of both data and materials under valid subpoenas, so plan for protective orders rather than refusal. Retain the file under your professional rules, not the court calendar: the APA record-keeping guideline suggests seven years after the last service for adults, several state boards codify comparable periods, and Ontario's college requires ten, so keep the file to the longest applicable rule and at least through the end of appeals. Forensic work also reaches people in crisis. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). Billing is the simplest part once the frame is right: public payers do not cover forensic evaluation, so there is nothing to deny. Medicare excludes services that are not for diagnosis or treatment (42 U.S.C. 1395y(a)(1)(A); 42 CFR 411.15(k)), and Canadian provincial plans and Australian Medicare exclude medico-legal work on the same logic, so the fee is a private matter: set it in a written agreement with the court or retaining party before the first interview, and do not submit it to a health plan. The content requirement, by contrast, comes from the strictest instrument that applies to the matter, so treat that instrument as the checklist: 18 U.S.C. 4247(c) for a federal mental-condition exam; the state competency statute where one enumerates elements (Texas Art. 46B.025 requires the opinion or the explanation, and on a finding of incompetency the nature and severity of the deficits, a restoration estimate, and treatment options, inside the 30-day deadline of Art. 46B.026); and Fed. R. Civ. P. 26(a)(2)(B) for a civil retained expert, which demands all opinions with their basis and reasons, the facts and data considered, qualifications, four years of prior testimony, and compensation. The format is a convention; the content requirement is real and jurisdiction-specific. Forensic work is legal work: when a subpoena, a records demand, or a complaint arrives, involve your attorney or licensing board before acting, because state rules vary. ↑ Back to the template and downloads ## Common forensic evaluation report errors auditors flag No payer audits these reports, because no public payer covers them; the quality trail runs through peer-reviewed studies, courtrooms, and licensing boards, and it is unusually consistent. When researchers scored 150 Hawaii competency reports against a quality model ( Robinson and Acklin, 2010 ), the mean quality score was 68.95, only a quarter of reports reached 80% of the maximum possible score, and the authors described "pervasive mediocrity"; an earlier Hawaii study (Acklin and colleagues, 2005) found 63% of competency reports and 49% of criminal-responsibility reports failed to state a rationale for the psycholegal opinion. A Virginia review of 1,126 court-ordered competency reports (2020) found generally strong statutory compliance but "marked variability in forensic conclusions." Discipline follows the same defects: in a 2019 Western Australian tribunal decision, a Family Court single expert was reprimanded, fined 20,000 AUD, and barred from single-expert work after attributing psychopathy traits on insufficient data, and a 2023 review of two decades of Kentucky board cases found a high share involved children and custody work. The BastionGPT Clinical Advisory Board sees the same errors most often in forensic evaluation report reviews: A conclusion with no stated reasoning. Three decades of audits converge on this defect: the report announces an opinion without showing how the findings compel it. Write the analysis so a reader can trace each opinion to specific data; if the chain cannot be traced, the report is not ready to sign. Opinions outside the referral question. Gratuitous diagnoses, opinions about people never examined (the defect in a British Columbia custody discipline case), or an ultimate-issue statement where FRE 704(b) bars one; in the other direction, failing to state a competency opinion a statute directs the examiner to state or to explain why none can be given. Treatment habits imported into forensic work. Calling the evaluee a patient, promising confidentiality the setting does not allow, skipping the notification section, or writing in an advocate's voice. The role distinction is the defining feature of the document, and blurring it discounts everything that follows. Unattributed facts. The evaluee's self-report repeated as established history, sources listed without dates or durations, collateral described vaguely. Attribution is what lets a court weigh the evidence; its absence reads as endorsement. Instrument misuse. Traits attributed on insufficient data, checklists scored without proper administration (the basis of the 2019 Western Australian discipline decision), or clinical tests presented as if they answered the legal question by themselves. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on forensic psychological evaluation reports. Draft report sections from your interview notes, test summaries, and record chronology, with every fact attributed to its source and self-report kept separate from corroborated history. Organize the psycholegal analysis so each referral question is paired with the findings that bear on it, keeping the reasoning from data to opinion visible on the page. Check the draft before you sign: referral question answered and nothing more, notification documented, sources listed with dates and durations, reasoning stated for every opinion, statutory elements present. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a forensic psychological evaluation report be? No statute or rule in the US, Canada, or Australia sets a maximum length. Length tracks the matter: competency reports are often comparatively short, while criminal responsibility and custody reports run much longer. The only completeness mandate is for civil retained experts, where Fed. R. Civ. P. 26(a)(2)(B) requires all opinions with their basis and reasons, the facts and data considered, qualifications, prior testimony, and compensation. Write until the reasoning is complete, not to a page target. Is there a legally required format for forensic evaluation reports? Not in general: structure is professional convention descended from Grisso's model and carried by the 2013 APA and 2015 AAPL guidelines. Narrow statutory islands exist. 18 U.S.C. 4247(c) enumerates content for US federal mental-condition reports (history and symptoms, tests and results, findings, and opinions on diagnosis, prognosis, and the legal question), and state competency statutes such as Texas Art. 46B.025 mandate specific elements on a 30-day deadline. Build each report to the strictest instrument that applies. Will Medicare or insurance pay for a forensic evaluation? Generally no, and not as a denial: court-ordered and medico-legal evaluation is categorically outside coverage because it is not for diagnosis or treatment of the examinee (42 U.S.C. 1395y(a)(1)(A); 42 CFR 411.15(k)). Canadian provincial plans and Australian Medicare exclude it on the same logic. The fee is billed privately to the court, the retaining party, or the examinee under a written agreement settled before the evaluation begins. Can I give an opinion on the ultimate legal issue? More often than the folklore says. FRE 704(a) permits ultimate-issue opinions; 704(b) bars only an expert opinion on whether a criminal defendant had a mental state that is an element of the charged crime or a defense. Competency, dangerousness, and custody best-interests opinions are not barred, and some statutes require one: Texas directs the examiner to state a competency opinion or explain why none can be given. Guidelines still counsel restraint where no statute requires the opinion. What is the difference between a forensic and a clinical psychological evaluation? The client and the question. A clinical psychological evaluation report serves diagnosis and treatment of your own patient: confidential, patient-accessible, usually billable. A forensic report answers a legal question for a court or attorney about an evaluee: confidentiality is limited and explained up front, distribution follows the court's order, and no health plan pays for it. The same clinician can do both kinds of work, but not both roles in the same case. Is the report confidential, and does the evaluee get a copy? It is not confidential in the treatment sense, and telling the evaluee so is part of the evaluation. Distribution is set by statute or court order: 18 U.S.C. 4247(c) requires filing with the court with copies to counsel for both sides, and state and family courts set their own lists. Because a retained, non-treating examiner is generally not the evaluee's health care provider, HIPAA's patient-access right usually does not attach; the evaluee's copy comes through counsel or the court. Do I have to release raw test data to attorneys? Under APA Ethics Code 9.04, absent the evaluee's release, psychologists provide test data "only as required by law or court order." Test materials carry separate test-security duties under 9.11, but courts have repeatedly ordered production of both under valid subpoenas. The practical rule: negotiate protective orders that route data to a qualified professional, and do not treat the ethics code as a shield after a court has ruled. How long do I need to keep forensic evaluation records? Forensic files follow your professional record rules, not the court's calendar. The APA record-keeping guideline suggests seven years after the last service for adults, longer for minors; several state boards codify comparable periods, and Ontario's college requires ten years. With litigation in play, keep the file at least through the end of any appeal, and when rules conflict, retain to the longest applicable one. Can BastionGPT help write a forensic evaluation report? Yes, as a drafting and checking layer under your judgment: it drafts sections from your notes and summaries with every fact attributed to its source, organizes the psycholegal analysis so each opinion traces to findings, and checks statutory elements before you sign. The opinions remain yours. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychological Evaluation Report Neuropsychological Evaluation Report Psychoeducational Evaluation Report Violence & Homicide Risk Assessment Psychiatric Diagnostic Evaluation Consultation Response & Report Fitness-for-Duty Evaluation Capacity & Competency Evaluation Court Report & Treatment Progress Update SASSI Documentation PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/culturally-safe-assessment All note templates # Aboriginal & Torres Strait Islander Culturally Safe Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated July 2026 A culturally safe assessment is a clinical assessment conducted and documented so that Aboriginal and Torres Strait Islander clients experience care as culturally safe, a judgment that belongs to the client, not the clinician. Australian psychologists, GPs, and multidisciplinary teams apply the standard to intake, mental health, and cognitive assessment. No law prescribes a form; the record evidences a culturally safe process. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, GPs, psychiatrists, nurses, social workers; Aboriginal Health Workers and Health Practitioners often contribute Audience The client and family, treating team, Aboriginal health staff, referrers; accreditation assessors for organisations Typical length 600 to 1,200 words · 60 to 120 minutes including unhurried yarning time (clinical team estimate) Format family Practice standard applied to any assessment format (compare: intake note, biopsychosocial assessment, cultural formulation interview) When it's used Any assessment with an Aboriginal or Torres Strait Islander client: intake, social and emotional wellbeing review, cognitive assessment, the item 715 health check Standards context Cultural safety is embedded in the National Law and the Psychology Board's 2025 Code of Conduct; no rule mandates a form or note fields On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a culturally safe assessment? A culturally safe assessment is any clinical assessment of an Aboriginal or Torres Strait Islander client conducted so the client experiences it as safe, and documented so the record shows how. Cultural safety began with Māori nurse educator Dr Irihapeti Ramsden (Ngāi Tahu, Rangitāne) as Kawa Whakaruruhau in late-1980s Aotearoa New Zealand, and its defining feature crossed the Tasman intact: safety is determined by the person receiving care, never declared by the practitioner. The National Aboriginal and Torres Strait Islander Health Plan 2021-2031 puts it plainly: cultural safety "is about how care is provided, rather than what care is provided." In Australian practice you will also hear culturally informed assessment, social and emotional wellbeing (SEWB) assessment, or yarning-based assessment; each names the same obligation applied to assessment work. Three distinctions prevent most of the confusion. First, cultural safety is a practice standard embedded in law, not a form: 2022 amendments wrote it into the Health Practitioner Regulation National Law state by state, and since 1 December 2025 the Psychology Board's Code of Conduct and Professional competencies impose a positive duty to practise in a culturally safe way as defined by Aboriginal and Torres Strait Islander Peoples. No statute, Board standard, or MBS rule prescribes an assessment form, required note fields, or a mandatory instrument. Second, the validated Indigenous-specific tools (the adapted aPHQ-9, WASC-Y and WASC-A, KMMS, HANAA, Strong Souls) are optional screening instruments used inside an assessment, not the assessment itself, and the SEWB framework of Gee, Dudgeon, Schultz, Hart and Kelly (2014) is a conceptual map, not a questionnaire. Third, because the recipient decides whether care was safe, a clinician cannot self-certify cultural safety in the note: the record can evidence a culturally safe process (consent, participation, consultation, the client's own words) but never the outcome. That paradox shapes every section below, and it is what separates this document from a generic intake note or biopsychosocial assessment . ## Who uses culturally safe assessments and when Every Australian registered practitioner assessing an Aboriginal or Torres Strait Islander client is bound by the standard, and the workforce numbers explain why it cannot be delegated: only 1.3% of Australia's 959,858 registered health practitioners identify as Aboriginal and/or Torres Strait Islander (Ahpra Annual Report 2024/25), so most assessments are conducted by non-Indigenous clinicians. The settings span private psychology and psychiatry, general practice running the annual health check (MBS item 715), community mental health, youth services, aged care, and Aboriginal Community Controlled Health Organisations, where Aboriginal Health Workers and Health Practitioners share the work. The stakes are access itself: in the 2018-19 national survey year, of 243,663 Aboriginal and Torres Strait Islander people who did not access needed health care, 32% cited cultural reasons. Choose the vehicle by the job: a structured DSM-based module is the cultural formulation interview , a general first-contact record is the intake note , and a funding-focused functional report is the NDIS functional capacity assessment . The culturally safe assessment is not an alternative to any of them; it is the standard governing how each is done with Aboriginal and Torres Strait Islander clients. ## Culturally safe assessment structure: what goes in each section Identifying details and Indigenous status. Ask every client the standard question ("Are you of Aboriginal or Torres Strait Islander origin?") and record the self-reported answer: Aboriginal, Torres Strait Islander, both, neither, or not stated. Status is never assumed from appearance, name, or address, and never altered by anyone but the client. Pitfall: not asking. The AIHW national guidelines describe refraining from asking any client as "an act of discrimination"; asking respectfully is the baseline, and declining to answer is itself a valid, recordable answer. Consent and participation. Record who the client wants involved (family and kinship contacts, an Aboriginal Health Worker or liaison officer, an interpreter), consent for each, and what the client agrees may be recorded. Consent here is a negotiated, ongoing conversation, not a signature line. Pitfall: writing down cultural information the client shared but asked you not to record; the client's wishes govern the record, and the defensible move is documenting that a request was made, not the content withheld. Presenting concern in the client's words and priorities. Open with what the client says matters, in their words, before the referral question. Unhurried, yarning-style conversation is the expected pace; the referrer's question gets answered, but it does not get to reorder the client's priorities. Pitfall: importing the referral question as the story; an assessment the client experiences as an interrogation about someone else's agenda has failed before any finding is written. Social and emotional wellbeing domains. Map the picture across the SEWB domains: connection to body; mind and emotions; family and kinship; community; culture; Country; and spirituality and ancestors, set against social, cultural, historical, and political determinants. Record strengths and connections with the same care as problems. Pitfall: running the domains as a checklist. The framework (Gee and colleagues, 2014) is a map for organising this client's account, not boxes to tick or assumptions to import. Screening and testing, with tool choice explained. Name each instrument and why it fits: culturally validated options first where they exist (the adapted aPHQ-9, which at a cut-point of 10 reaches 84% sensitivity and 77% specificity; WASC-Y and WASC-A; the Kimberley Mum's Mood Scale in perinatal work; HANAA; Strong Souls; KICA for older adults), with a written note on the limits of mainstream norms whenever a standard instrument is used. Pitfall: reporting a mainstream score as definitive; no tool is mandatory, but presenting results normed on other populations without stating the limitation misleads every later reader. Cultural consultation and supports. With the client's consent, record that consultation with an Aboriginal Health Worker, liaison officer, Elder, or community member occurred and what it changed clinically. Pitfall: recording third-party identifying detail beyond what the clinical point needs; document the consultation's bearing on your reasoning, not a transcript of it. Culturally informed formulation. Situate the presentation in the client's own account across the SEWB domains and determinants, naming strengths, protective connections, and what the client wants restored. Pitfall: group-level attributions ("common in this community") that stereotype instead of formulate; every cultural statement in the formulation should trace to something this client said or agreed with. Plan agreed with the client. Actions the client and family helped shape, referrals that include Aboriginal Community Controlled options where the client wants them, cultural supports alongside clinical ones, and a review date. Close the loop by asking how the process felt and recording the answer. Pitfall: a plan the client had no hand in; because the client is the judge of safety, an imposed plan undermines the very process the note is meant to evidence. ### Blank template (copy and adapt) CULTURALLY SAFE ASSESSMENT (ABORIGINAL AND TORRES STRAIT ISLANDER CLIENT) Client: __________ Date(s) & setting: __________ Clinician: __________ Indigenous status (self-reported, standard question asked): [ ] Aboriginal [ ] Torres Strait Islander [ ] Both [ ] Neither [ ] Not stated / declined Interpreter (language, used?): _________________________________ CONSENT & PARTICIPATION (who the client wants involved; consent for family or health-worker involvement; what may be recorded) ________________________________________________________________ PRESENTING CONCERN & PRIORITIES (the client's own words first) ________________________________________________________________ ________________________________________________________________ SOCIAL & EMOTIONAL WELLBEING (body; mind & emotions; family & kinship; community; culture; Country; spirituality & ancestors; determinants; strengths recorded with the same care as problems) ________________________________________________________________ ________________________________________________________________ SCREENING & TESTING (tool and why; culturally validated first: aPHQ-9, WASC-Y/A, KMMS, HANAA; limits of mainstream norms noted) ________________________________________________________________ CULTURAL CONSULTATION & SUPPORTS (Aboriginal Health Worker, liaison officer, Elder or community consultation, with consent; clinical bearing on the assessment) ________________________________________________________________ CULTURALLY INFORMED FORMULATION (anchored to this client's account; strengths and connections; no group-level attributions) ________________________________________________________________ ________________________________________________________________ PLAN AGREED WITH THE CLIENT (actions shaped with client and family; referrals incl. ACCHO options; supports; review date) ________________________________________________________________ How the client experienced the process (asked and recorded): ________________________________________________________________ Clinician signature/credentials: ______________ Date signed: ______ Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample culturally safe assessment Scenario: initial assessment of an Aboriginal man referred by his GP after the annual health check, conducted at the local Aboriginal health service with an Aboriginal Health Worker present at his request. All details are fictional. Culturally Safe Assessment. Client: D.W., 42 · Dates: 16/07/2026 and 21/07/2026 · Setting: consulting room at the local Aboriginal health service · Clinician: R. Patel, Clinical Psychologist · Referral: GP mental health treatment plan following the item 715 health check Indigenous status: Standard question asked at the first appointment. D.W. identifies as Aboriginal, a Wiradjuri man living away from Country for work; recorded as self-reported. Consent and participation: D.W. asked that K.T., Aboriginal Health Worker at the service, sit in on both yarns, and agreed I may talk with her afterwards. Interpreter not needed. He wants his partner told the plan in outline only. One family matter he asked to keep out of the record; the request is honoured and noted here without its content. Presenting concern, in his words: "I have not been right since we lost my uncle in May. Flat, snappy, not sleeping. I want to be right for my kids." Sorry business has meant three trips back to community since May; his supervisor is "running out of patience," which worries him more than the mood itself. Social and emotional wellbeing: Body: sleep 4 to 5 hours, appetite down; GP follow-up arranged. Mind and emotions: flat mood most days, irritability, guilt about missing the final day of sorry business. Family and kinship: strong bond with partner and two sons; his uncle was "the one who taught me everything." Community: football club involvement lapsed this season; describes community as supportive but distant lately. Culture and Country: names being away from Country as a heaviness; a September trip back is planned and wanted. Spirituality and ancestors: finishing grief the right way matters to him and feels unfinished. Strengths: steady work history, six years alcohol-free, sons as motivation, clear cultural identity. Screening, with tool choice: aPHQ-9 chosen over the standard PHQ-9 with K.T.'s input, given its validation with Aboriginal and Torres Strait Islander adults; score 13, above the recommended cut-point of 10. Risk items reviewed with him directly: no thoughts of self-harm or suicide, no safety concerns identified; routine monitoring agreed. Cultural consultation: With consent, spoke with K.T. after the first yarn. Clinical bearing: the family's sorry business is recent and significant, and the men's group restarting at the service next month would be an acceptable support. No third-party detail recorded beyond this. Formulation: Grief responses and depressive symptoms in a 42-year-old Wiradjuri man, best understood through disrupted connection: to family (his uncle's death), to Country (away for work, return trip pending), and to community (lapsed involvement), against workplace pressure that penalises cultural obligation. Consistent with the GP's provisional diagnosis of a depressive episode; grief is the organising thread in his telling. His identity, family motivation, and sobriety are the platform the plan builds on. Formulation reflects his account; no group-level assumptions added. Plan agreed with D.W.: Fortnightly sessions timed around the September trip to Country; grief work paced by his sense of the right way, with K.T.'s continuing support; rejoin the men's group when it restarts (his suggestion); GP review of sleep on 24/07/2026; workplace letter offered, declined for now; review together on 18/08/2026. Process check and signature: Asked at the end of the second yarn how the assessment felt: "better than the last one I did; nobody rushed me." Recorded with his OK. R. Patel, MPsych(Clin), Clinical Psychologist. Date signed: 21/07/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Status is asked and recorded as self-report, meeting the AIHW standard and the expectation behind NSQHS Action 5.8, with no assumption from appearance or address. Consent reads as a process, not a signature: who is involved, what may be shared with whom, and a record-exclusion request honoured by noting the request rather than the content. The tool choice is explained: a culturally validated instrument selected with Aboriginal health staff input, scored against its published cut-point, with risk items reviewed directly. Consultation appears with its clinical bearing and nothing more, so the record shows what changed in the reasoning without exposing third-party detail. The formulation and plan trace to the client's own words, strengths included, and the note records how the process felt to him, which is as close as a record can come to evidencing safety the client defines. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Cultural safety sits differently in the Australian record from most documentation duties: it is a registration-level practice standard, not a note-content rule. The 2022 amendments embedded it in the Health Practitioner Regulation National Law state by state (New South Wales through a substituted section 3A; Western Australia legislates separately), and from 1 December 2025 the Psychology Board's Code of Conduct and Competency 8 of the Professional competencies created a positive, enforceable duty for psychologists. Ahpra can act on cultural-safety grounds without any separate clinical error: an October 2023 tribunal decision, the first major application of the racism-elimination amendments, reprimanded a doctor and disqualified him for 12 months over culturally unsafe conduct toward an Aboriginal patient. Because the National Scheme definition makes the client the judge, write the record to evidence process, and write it for the client's eyes: assessments sit in the accessible record, and a note the client could read without feeling diminished is both good practice and good evidence. Retention follows ordinary health-records law: in private practice, 7 years from last contact for adults and until age 25 for clients seen as minors under state Acts, with public services following state retention instruments. Four moving pieces are worth watching over the next year: the successor to Ahpra's 2020-2025 cultural safety strategy, the NSQHS third edition in development, the strengthened Aged Care Quality Standards that commenced 1 November 2025, and the Cultural Respect Framework 2016-2026 reaching the end of its term. The payer layer is blunter: no MBS item conditions payment on cultural-safety documentation. Item 715's explanatory notes require the health-check elements (history, examination, an overall assessment, a documented management plan, consent recorded), and the Better Access items require their own referral, plan , and review chain; neither adds a cultural field. What the Professional Services Review does police is individualisation: in a December 2024 outcome, a GP who billed item 715 at rates exceeding 99% of peers with non-individualised template records was reprimanded, ordered to repay more than $480,000, and disqualified from certain items for 12 months, and a March 2025 outcome disqualified another practitioner from item 715 for 24 months. The compliance lesson runs parallel to the cultural one: a cloned assessment fails Medicare because it shows no individual patient, and it fails the client because it shows no individual person. The practice standard is law; the note format is yours; the content that protects both the claim and the relationship is this client's own story. ↑ Back to the template and downloads ## Common culturally safe assessment errors auditors flag The audit trail here is new and sharpening. Ahpra's review of racism-related notifications, released in July 2026, counted 477 notifications of racism and discrimination between 1 July 2023 and 28 February 2026, about 1.5% of all notifications in the period; 36 concerned racism toward Aboriginal and Torres Strait Islander people, and unlike the largely social-media-driven remainder, those arose mostly in clinical practice settings. Ahpra itself cautions that the figures understate what Aboriginal and Torres Strait Islander people experience. On the Medicare side, Professional Services Review outcomes on the item 715 health check include a repayment above $480,000 with a 12-month disqualification and a separate 24-month disqualification, in both cases for inadequate, non-individualised records. The BastionGPT Clinical Advisory Board sees the same errors most often in culturally safe assessment reviews: Status never asked, or assumed. The AIHW guidelines direct the standard question to every client and treat refraining from asking as discriminatory. Assuming from appearance or postcode, or leaving the field blank, is the most common data failure and the easiest fixed. Template text that fits everyone. The defect PSR sanctions in item 715 records, non-individualised templates, is the same one that empties an assessment of the person; if the wellbeing section would read true of any client, it evidences no one. Self-certification. "Culturally safe care was provided" is a conclusion the definition reserves to the client. Reviewers read it as boilerplate; the process facts (consent, participation, consultation, the client's feedback) are what the record can honestly hold. Mainstream scores reported without their limits. A standard instrument normed on other populations, presented as definitive, with no applicability note and no consideration of a culturally validated alternative. Stereotype in the formulation. Group-level attributions and cultural explanations the client never gave. Every cultural statement should trace to this client's account, and consultation should appear with its clinical bearing rather than third-party detail. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on culturally safe assessments. Draft the assessment from your yarning notes, dictation, or a transcript, keeping the client's own words where you marked them and the social and emotional wellbeing domains organised. Restructure an existing intake or biopsychosocial note into a SEWB-informed format without losing clinical content. Check the draft before you sign: status recorded as self-report, consent and participation documented, tool limits stated, formulation free of group-level attributions, plan agreed and dated. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a required form for a culturally safe assessment in Australia? No. Cultural safety is a practice standard embedded in the Health Practitioner Regulation National Law by 2022 amendments and, for psychologists, in the Board's Code of Conduct and competencies from 1 December 2025. It binds how you practise, but no statute, Board standard, or MBS rule prescribes an assessment form, mandatory note fields, or a required instrument. There is no set length either; most run 600 to 1,200 words with unhurried yarning time. The record's job is to evidence a culturally safe process. How do I record Indigenous status if a client declines to answer? Ask every client the standard question and record the answer they give, including declining: "not stated / inadequately described" is the standard entry. Status is self-reported only; it is never assumed from appearance, name, or community, and never changed by anyone but the client. Not asking at all is the failure the AIHW guidelines single out as an act of discrimination. Can a non-Indigenous clinician conduct a culturally safe assessment? Yes. Cultural safety is a standard for every registered practitioner, and with only 1.3% of Australia's registered health workforce identifying as Aboriginal and/or Torres Strait Islander, most assessments are conducted by non-Indigenous clinicians. The standard asks for reflection on your own biases and power, and use of the supports the client wants: Aboriginal Health Workers, liaison officers, interpreters, family. Whether the assessment was safe remains the client's judgment. Can Ahpra act on a cultural safety complaint if the clinical care was sound? Yes. Since the 2022 National Law amendments, cultural safety and the elimination of racism are guiding principles of the scheme, and an October 2023 tribunal decision reprimanded a doctor and disqualified him for 12 months for culturally unsafe conduct toward an Aboriginal patient. Conduct can be sanctioned on cultural-safety grounds without a separate clinical error. Which screening tools are validated for Aboriginal and Torres Strait Islander clients? The adapted aPHQ-9 (use the cut-point of 10, where sensitivity is 84% and specificity 77%, rather than the diagnostic algorithm), the Westerman Aboriginal Symptom Checklists (WASC-Y for ages 13 to 17, WASC-A for adults), the Kimberley Mum's Mood Scale for perinatal screening, HANAA's yarning-based interview, Strong Souls for youth, and KICA for cognitive assessment in older adults. None is mandatory; when you use a mainstream instrument instead, document the limits of its norms. Does MBS item 715 or Better Access require cultural-safety documentation? No. Item 715's explanatory notes require the health-check elements: history, examination, an overall assessment, a documented management plan, and recorded consent. The Better Access items require their referral, plan , and review chain. Neither conditions payment on cultural content, a cultural section, or any instrument. Where practitioners get in trouble with item 715 is individualisation: PSR outcomes include a repayment above $480,000 and disqualifications of 12 and 24 months for template-driven records. How is this different from the DSM-5 Cultural Formulation Interview? The Cultural Formulation Interview is a scripted 16-question DSM-5 module about how any client understands their problem; it travels across cultures. The SEWB framework is a specifically Aboriginal and Torres Strait Islander map of wellbeing across domains from body to Country to ancestors. The culturally safe assessment is neither: it is the practice standard governing how any assessment is conducted, and in Australia it is the only one of the three embedded in law. How long do I keep culturally safe assessment records? The same as other Australian health records: in private practice, 7 years from the last contact for adults, and for clients seen as minors until age 25, under state health records laws (New South Wales and Victoria legislate this expressly); public services follow state retention instruments. Cultural material the client asked you not to record never enters the record in the first place; a note that a request was made is enough. Can BastionGPT help write a culturally safe assessment? Yes, as a drafting layer under your judgment: give it yarning notes, dictation, or a marked-up transcript and it drafts the assessment with the client's words preserved, the SEWB domains organised, and the pre-sign checks run: status as self-report, consent documented, tool limits stated, no group-level attributions. The clinical and cultural judgment stays with you and the client. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Cultural Formulation Interview (CFI) Intake Intake Note Biopsychosocial Assessment Better Access Mental Health Treatment Plan Referral NDIS Functional Capacity Assessment Report Mental Status Exam (MSE) Note NDIS Plan Reassessment & Progress Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/narrative-progress-note All note templates # Narrative Progress Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A narrative progress note is a free-text progress note that documents a therapy session in flowing prose instead of the labeled sections of SOAP or DAP. Therapists use it when a session's chronology and reasoning matter more than a template, and it is fully compliant as long as the required content is present. Most narrative notes run 150 to 500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, counselors, and psychologists, especially in solo practice and long-term or insight-oriented work Audience Treating therapist, auditors, legal reviewers Typical length 150 to 500 words · 10 to 25 minutes by hand (clinical team estimate) Format family Free-text progress note (compare: SOAP, DAP, BIRP, PIE) When it's used After any therapy session; strongest for complex sessions where chronology and clinical reasoning matter Standards context The oldest progress note convention; no US, Canadian, or Australian regulator or payer mandates a note format On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a narrative progress note? A narrative progress note is a free-text progress note: it documents one therapy session in flowing prose rather than in the labeled sections of a structured format. It is also the oldest form in the family. Narrative charting was the standard method of the source-oriented medical record, and Lawrence Weed introduced the problem-oriented record and its SOAP note (published in 1964, popularized in 1968) specifically to impose structure on narrative notes he found hard to interpret. So narrative is not a loose variant of SOAP; SOAP was invented as the correction to narrative. You will also hear it called a free-text note, an unstructured note, a prose note, or simply a progress note written in paragraphs. Two boundary lines do the real work. First, the format is a convention, not law: no statute, regulation, or payer in the US, Canada, or Australia mandates SOAP over narrative, and Medicare's Program Integrity Manual states that progress notes "may be in any form or format." Every enforceable rule attaches to content, authentication, and timeliness instead. Second, a narrative progress note is not a process note or HIPAA psychotherapy note, however loosely clinicians use those words. The protected psychotherapy-note category is defined by content and separate storage under 45 CFR 164.501, and a note that carries diagnosis, symptoms, progress, or session times is an ordinary progress note , part of the standard record, wherever you file it. ## Who uses narrative progress notes and when Narrative is most at home in solo and small private practices, psychodynamic and insight-oriented work, and long-term therapy: settings where the treating clinician is the primary reader and the arc of the work matters. It beats structured formats when a session will not decompose cleanly into fields: several intertwined threads in one hour, a disclosure that changes the treatment picture, a rupture and repair that only makes sense told in order. Chronological prose is also what legal reviewers and licensing boards read most easily when a record is examined years later, because it preserves reasoning, not just conclusions. The tradeoff is discipline: no field prompts you for time, intervention, or risk, so the writer has to carry the checklist. Clinicians who want light structure without full SOAP often land on DAP , programs that must demonstrate intervention and response each session use BIRP , and problem-indexed team settings use PIE . ## Narrative progress note structure: the five content anchors A narrative note has no required sections; that is the point, and also the risk. Auditors, payers, and boards read free text looking for the same content they would find in any structured note. Write in whatever order the session demands, but before you sign, every narrative should hit five anchors, framed by the header facts: date, service and code, start and stop times for time-based codes, diagnosis, and a signature with credentials. Presentation. What the client brought: current symptoms, functioning, and change since last session, with any scores (PHQ-9, GAD-7) and your observations woven into the prose. Pitfall: opening with logistics ("client arrived on time, homework was reviewed") and never stating symptom status; a reviewer should find how the client is doing in the first third of the note. Intervention. What you did, by name: cognitive restructuring, behavioral activation, exposure planning, grief work. A named technique is what makes the note evidence of a psychotherapy service rather than a conversation. Pitfall: letting "discussed" and "explored" carry the whole note; unnamed interventions are among the most common narrative omissions because no field asks for them. Response. How the client responded to the intervention, and where that leaves progress toward a specific treatment plan goal. This is the golden thread that ties the session to the plan and the plan to medical necessity. Pitfall: a vivid account of what happened with no sentence connecting it to a goal; in prose, the medical-necessity chain breaks silently. Risk. Current risk status, stated either way when clinically indicated, in a sentence a reader can find in seconds. Documented monitoring beats silence, and a short labeled line beginning "Risk:" keeps it findable without breaking the narrative voice. Pitfall: burying a risk denial mid-paragraph or omitting it entirely; a risk statement nobody can locate does the same work as one that was never written. Plan. What happens next: homework, changes in approach, coordination, and the next appointment date. Pitfall: prose that trails off with "will continue to work on goals"; end the story with dated, specific next steps the following note can be checked against. ### Blank template (copy and adapt) NARRATIVE PROGRESS NOTE Client: [initials] Date: Session #: Service: [individual/telehealth] Code: Diagnosis (ICD-10): Start/stop time: Narrative (any order; hit all five anchors before you sign): [Presentation: symptoms, functioning, change since last session, scores, observations] [Intervention: what you did, named] [Response: client response + progress toward goal #___] [Risk: current status, stated either way] [Plan: specific next steps, dated] _______________________________________________________________ _______________________________________________________________ _______________________________________________________________ _______________________________________________________________ _______________________________________________________________ _______________________________________________________________ Pre-sign check: [ ] time [ ] named intervention [ ] response tied to a goal [ ] risk stated [ ] dated plan Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample narrative progress note Scenario: adult client, major depressive disorder, CBT with behavioral activation, session 8 of a planned 16. One flowing narrative instead of labeled sections. All details are fictional. Client: R.L., 41 · Date: 08/03/2026 · Session: 8 · Service: Individual psychotherapy (90834), 47 min, in office · Start/stop: 2:03 pm to 2:50 pm R.L. attended his eighth CBT session for major depressive disorder, recurrent, moderate (F33.1). He described the past week as "flatter than the last two," rating average mood 4 of 10, and reported skipping two of three planned gym sessions after a critical performance review at work on Tuesday. Sleep is holding at 6 to 7 hours with the earlier wind-down routine. Today's PHQ-9 was 11, down from 16 at intake but up 2 points from two weeks ago. He arrived on time and groomed, affect congruent but restricted, and his speech was slowed early in the session before normalizing as he engaged. The session focused on the setback. We reviewed his behavioral activation log, then used cognitive restructuring on the thought "this review proves I am sliding back to useless," which he linked on his own to the all-or-nothing pattern identified in session 5. A responsibility pie exercise redistributed the review feedback across workload, staffing, and his own contribution; he generated the alternative thought "one critical review is data, not a verdict" and re-rated his belief in the original thought from 90 to 55 percent. He agreed to restart the gym plan at a reduced, specific level, two sessions this week, rather than abandoning it. This is direct progress on Goal 2 of his treatment plan, maintaining activity scheduling through mood dips, and the first time he has applied the relapse-prevention frame without prompting. Risk: R.L. denied suicidal ideation, intent, or plan and denied self-harm urges; he named his brother as a support he contacted twice this week; no risk indicators were observed in session; continue routine monitoring. Plan: continue weekly CBT with a behavioral activation focus; homework is the two scheduled gym sessions plus a thought record on work triggers; re-administer the PHQ-9 at session 10; next appointment 08/10/2026. Signed A. Morales, LPC, 08/03/2026, 6:05 pm. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works All five anchors are findable in prose. Symptom status and a PHQ-9 score sit in the opening, the interventions are named, the response ties to a numbered goal, risk gets its own labeled sentence, and the plan is dated and signed with credentials the same day. Time support lives in the header. Start and stop times back the time-based code; undocumented psychotherapy time was a named failure behind the OIG's $580 million improper-payment estimate. The golden thread survives free text. One sentence, "direct progress on Goal 2 of his treatment plan," keeps the narrative connected to medical necessity. Risk is stated, not buried. The denial, the protective contact, and the monitoring decision sit in a single labeled sentence a reviewer or attorney can find in seconds. The detail is unclonable. A specific setback, a specific automatic thought, and belief ratings from 90 to 55 percent individualize the note; none of it could be copy-forwarded to another session without showing. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A narrative progress note is part of the designated record set: clients can request it, payers can audit it, and lawyers can subpoena it. It is not a HIPAA psychotherapy note, and that boundary is law, not convention. 45 CFR 164.501 defines psychotherapy notes by content and separate storage, excluding diagnosis, symptoms, treatment plan, progress, and session times, which is the exact material a progress note exists to carry; filing prose in a private folder does not convert it into a protected note. Timeliness is the other structural expectation. Medicare payer policy expects documentation created at the time services are rendered, with late entries and corrections handled under its amendment rules, and Australia's MBS explanatory notes count only clinical details "recorded at the time of attendance" toward the consultation time. The universal 24 to 48 hour completion deadline repeated on vendor pages exists in no federal rule; real deadlines come from payer contracts, Medicaid agencies, accreditors, and agency policy, so find yours and write it down. For insurance work the burden is content: diagnosis linkage, a named intervention, the client's response and progress against the treatment plan, duration or start and stop times for the time-based codes (90832, 90834, 90837), and a signature with credentials. The format is a convention; the content is the requirement. Narrative's specific weakness is that nothing prompts you: prose makes a missing time statement or an unnamed intervention easy to commit and hard to self-detect, and cloned narrative is exactly what payer text-comparison algorithms catch, because free text is supposed to differ from session to session. A pre-sign checklist of the five anchors closes most of that gap, and the choice stays reversible note by note: many clinicians write narrative for the complex sessions and SOAP for the routine ones, because the record is judged by what it contains, never by its shape. ## Common narrative progress note errors auditors flag No audit has ever faulted a note for being written in prose; every finding is about missing content, and free text simply makes the omissions easier to commit. In the OIG's nationwide audit of Medicare psychotherapy services ( report A-09-21-03021, 2023 ), only 84 of 216 sampled enrollee days fully met Medicare requirements, undocumented psychotherapy time and missing signatures were named failures, and improper payments were estimated at $580 million. In CMS's 2024 reporting-period data for outpatient psychiatric services, the improper payment rate was 16.1%, about twice the overall Medicare rate, with insufficient documentation behind 78.3% of it. The BastionGPT Clinical Advisory Board sees the same errors most often in narrative progress note reviews: No time anywhere in the prose. Duration or start and stop times for 90832, 90834, or 90837 never made it into the note: the single most-cited audit failure and the leading driver of 90837 downcoding and recoupment. Unnamed interventions. "Discussed stressors" and "explored feelings" describe a conversation, not a psychotherapy service; without a named technique the note cannot support the code billed. A broken golden thread. The narrative never states progress toward a treatment plan goal, so the chain from diagnosis to plan to session fails, and prose gives no empty field to warn you. Buried or missing risk language. A risk denial folded into the middle of a paragraph, or omitted entirely; in later legal review, a risk statement nobody can find works like one that was never written. Cloned narrative. Copy-forward prose recycled across sessions is flagged by payer text-comparison algorithms, and identical free text reads worse to a reviewer than an identical checkbox, because narrative is supposed to vary. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on narrative progress notes. Draft a complete narrative note from a few bullets, a dictation, or a pasted session transcript, written as flowing prose that covers all five anchors. Summarize a long, winding session into a focused narrative that keeps the chronology and the clinically load-bearing detail. Check a finished note before you sign: time support, a named intervention, goal linkage, a findable risk statement, and a dated plan. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Can I just write a paragraph instead of a SOAP note? Yes. No statute, regulation, or payer in the US, Canada, or Australia mandates a note format. Medicare's Program Integrity Manual says progress notes "may be in any form or format," and Ontario's CRPO tells registrants they are "free to choose whichever record-keeping system is best suited to their needs." A narrative paragraph is fully compliant when it contains the required content: diagnosis, a named intervention, the client's response, medical-necessity linkage, time for timed codes, and a signature with credentials. How long should a narrative progress note be? No regulation anywhere specifies progress-note length or word count; the 150 to 300 word "standards" you see on vendor pages are conventions, not rules. Most narrative session notes run 150 to 500 words and take 10 to 25 minutes by hand. The right length is functional: long enough to support the billed service and hit all five anchors, short enough that you will actually write it the same day. Will insurance accept narrative notes? Yes. Payers audit content, not format. Reviewers look for medical necessity, a documented intervention, the client's response and progress against the treatment plan, and time support for time-based codes. Structured formats reduce audit risk only because their fields prompt those elements; a narrative note written against a pre-sign checklist gets the same protection. Do I need start and stop times in a narrative note? For the time-based psychotherapy codes (90832, 90834, 90837), documented duration or start and stop times is the most audited element in the note, and its absence is the leading driver of downcoding and recoupment. Undocumented psychotherapy time was a named failure in the OIG's 2023 nationwide psychotherapy audit. Australia applies the same pressure from another angle: MBS explanatory notes count only clinical details recorded at the time of attendance toward the consultation time. Put times in the header and the issue disappears. What is the difference between a narrative note and a DAP or BIRP note? Structure. DAP folds observations into a Data section but still labels Data, Assessment, and Plan. BIRP forces an explicit intervention and response pair, which is why community mental health and substance-use programs favor it. A narrative note drops labels entirely and tells the session in chronological prose. No regulator prefers any of them; choose the format you complete fastest and most completely. Is a narrative progress note the same as a process note or psychotherapy note? No, and the confusion has legal weight. HIPAA defines psychotherapy notes (45 CFR 164.501) by content and separate storage: the category excludes diagnosis, symptoms, treatment plan, progress, and session start and stop times. A narrative note carrying those elements is a progress note, part of the standard record and accessible to clients and payers, no matter where you file it. Keep true process notes physically separate and keep billable content out of them. When is a narrative note better than a structured format? When the session will not decompose into fields: several intertwined threads, a disclosure that reframes the treatment, a rupture and repair that only makes sense told in order. Chronological prose preserves clinical reasoning, which is what legal reviewers and licensing boards read for when a record is examined years later. The cost is that nothing prompts the required elements, so pair narrative writing with a pre-sign checklist of the five anchors. How long do I need to keep narrative progress notes? The same as any clinical record; format changes nothing. In the US, state law controls retention, commonly 6 to 10 years for adults and longer for minors; HIPAA's six-year rule covers compliance documents, not charts, and APA guidelines suggest seven years after the last service for adults. Ontario's CRPO requires at least 10 years from the last interaction, or from the client's 18th birthday, whichever is later. In Australia the common baseline is 7 years after the last service, or until age 25 for records made when the client was a child. Can BastionGPT write a narrative progress note from a session transcript? Yes. Paste a transcript, dictate, or give it bullets, and it drafts a flowing narrative in your voice that covers all five anchors, ready for your review. It can also tighten a long narrative or check a finished one for missing time, intervention, or risk language before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychotherapy Progress Note SOAP Note DAP Note BIRP Note PIE Note Canada Telepsychology & Virtual Care Note Start-Stop Time Attestation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/canada-telepsychology-note All note templates # Canada Telepsychology & Virtual Care Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A Canadian telepsychology and virtual care note is an ordinary progress note with telepresence elements added: session modality, the client's physical location, consent to virtual care, an emergency plan naming local resources, and a technology-failure fallback. Canadian psychologists and psychotherapists write one after every video or telephone session. No college mandates a template; expectations come from provincial standards and the CPA telepsychology guidelines. Most run 200 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, registered psychotherapists, social workers, and counsellors delivering virtual care in Canada Audience Treating clinician, the client on request, supervisors, college record reviewers, extended-health and federal payers Typical length 200 to 400 words · 10 to 20 minutes by hand (clinical team estimate) Format family Progress-note variant with telepresence elements (compare: telehealth therapy note, SOAP, DAP) When it's used After every video or telephone session where the clinician and client are in different locations Standards context Provincial college standards plus CPA telepsychology guidance (2023); no college mandates a template On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a Canadian telepsychology and virtual care note? A Canadian telepsychology and virtual care note is the ordinary clinical progress note for a session delivered by video or telephone, with a telepresence layer added: modality, both parties' locations, a consent reference, an emergency plan tied to the client's location, and a technology fallback. The national guidance behind it runs from the ACPRO Model Standards for Telepsychology (2011) to the current CPA Guidelines on Telepsychology , approved in June 2023, which define telepsychology as "a method of providing psychological services through electronic means" when the psychologist and client are not in the same physical location. Clinicians and payers also call it a virtual care note, teletherapy note, telehealth progress note, or, in Quebec, a télépsychologie or télépratique record. The framing that carries this page: almost none of the note's contents are mandated. No Canadian college prescribes a telepsychology template, a field list, a note-completion deadline, a written consent form, or Canadian data residency; those are conventions. Even the foundational ACPRO Model Standards state that they "have no force outside of official adoption by a member jurisdiction," and Ontario adopted them only as a non-binding advisory. What is law: provincial health-information statutes (Ontario's PHIPA, Alberta's HIA, Quebec's privacy act as amended by Law 25), enforceable college record-keeping standards, and, most consequentially, the client-location rule codified by the ACPRO Memorandum of Understanding on interjurisdictional telepsychology (in effect 1 April 2024): in eight provinces and territories you need registration or a licence where the client is sitting. The note is where you show you checked. Two boundaries matter: the telehealth therapy note covers the US-facing session record with its place-of-service and modifier logic, and the standalone telehealth consent is the one-time agreement this note references session by session. ## Who uses Canadian telepsychology and virtual care notes and when Psychologists write most of them, alongside registered psychotherapists, social workers, and counsellors seeing clients by video or phone; a 2017 survey cited in the CPA guidelines found 47 percent of Canadian psychologists already used technology to deliver care, and the pandemic made the virtual session routine. The setting is usually private practice, because provincial public health insurance rarely covers private-practice psychology; the readers are extended-health carriers, federal programs such as NIHB and Veterans Affairs Canada, workers' compensation boards, supervisors, and college reviewers. Use the virtual-care framing whenever the session is remote: an in-person progress note simply omits the telepresence elements, and a US-billed session adds the Medicare layer covered on the telehealth therapy note page. Telephone-only sessions count as telepsychology; record that the session was audio-only and why. ## Canadian telepsychology and virtual care note structure: what goes in each section Header and modality. Client identifier, date, session number, service type, start and stop times, and the modality: secure video or telephone. If the session was audio-only, say so and note why. No Canadian rule requires naming the platform (convention); recording that a secure platform was used is prudent, and US Medicare billing needs the audio-only reason on the record. Pitfall: logging "telehealth" generically; when a payer or college asks whether video was used, a generic label answers nothing. Locations and jurisdiction check. Where the client physically is for this session, where you are, and the conclusion: you are authorized to practise where the client sits. The field itself is convention (no Canadian regulation codifies it), but it evidences a law-level question: under the ACPRO MOU, eight jurisdictions including Ontario, BC, and Alberta require registration or a licence where the client is located, while Quebec, Nova Scotia, and New Brunswick regulate through your home registration. Pitfall: capturing an address at intake and never asking again; the rule turns on where the client is today, which is exactly what changes with snowbirds and students. Identity, privacy, and setting. Identity verification (first session, new device, or new location), confirmation the client is in a private space, and who else is present with the client's agreement. The CPA guidelines suggest agreeing on a code word the client can use if the space stops being private (convention). Pitfall: assuming privacy; a client joining from a car, a kitchen, or a workplace changes what can safely be discussed, and the note should show you asked. Consent to virtual care. A dated reference to the standing consent conversation (risks, benefits, technology, privacy limits) plus a session-level confirmation and anything that changed. Consent itself is law: Ontario's PHIPA s 18 and its provincial analogues govern it. Written form is not: verbal consent is valid everywhere in Canada, and no regulation sets a re-consent interval (both convention). Pitfall: treating a signed form as the finish line; the form proves one conversation happened once, and the note is what shows consent stayed live. Emergency plan and local resources. A support person near the client, the local crisis line (9-8-8 in Canada), and the emergency department nearest to where the client actually is this session. This element is convention: no Canadian rule codifies it, but the CPA guidelines direct psychologists to discuss emergency procedures, local back-up assistance, and local emergency resources. Pitfall: an emergency plan built around the clinician's city; if the client joins from a cottage two hours away, every named resource is wrong. Technology contingency. The agreed fallback if the connection fails (usually: you call the client's phone), and documentation of any failure and how contact resumed. Convention, from the CPA guidelines' direction to plan an alternate means of contact. Pitfall: no documented fallback; when the video drops mid-disclosure, the record must show what happened next, not leave a gap where risk information should be. Clinical content. The progress note proper: the client's report, your observations including any limits the medium put on them, assessment against treatment goals, a risk statement, and a dated plan. Record-keeping content standards are enforceable: in Ontario the CPBAO Standards are deemed standards of the profession under O Reg 801/93. Pitfall: a complete telepresence checklist wrapped around a thin note; colleges discipline record-keeping and consent deficiencies, and the virtual layer does not substitute for clinical substance. ### Blank template (copy and adapt) CANADIAN TELEPSYCHOLOGY / VIRTUAL CARE NOTE Client: ____________ Date: __________ Session #: ____ Service: [individual/couples/family] Start/stop: ______________ Modality: [ ] secure video [ ] telephone (audio-only; reason: ________) LOCATIONS AND JURISDICTION CHECK Client's physical location this session: ________________________________ Clinician's location: ___________________________________________________ Authorized to practise where the client is located: [ ] yes If client is outside your home jurisdiction, basis: _____________________ IDENTITY, PRIVACY, SETTING Identity confirmed: [ ] yes Private setting confirmed: [ ] yes Others present (with client agreement): _________________________________ CONSENT TO VIRTUAL CARE Standing consent on file: date ________ [ ] verbal [ ] written Reconfirmed this session: [ ] yes Changes discussed: __________________ EMERGENCY PLAN (for the client's location TODAY) Local support person and contact: _______________________________________ Local crisis line: 9-8-8 (Canada) Nearest emergency dept: _____________ Code word for compromised privacy: ______________ TECHNOLOGY CONTINGENCY If the connection fails, clinician will: ________________________________ Failure this session / how contact resumed: _____________________________ CLINICAL CONTENT Client report / session focus: __________________________________________ Observations (note any limits of the medium): ___________________________ Assessment and risk: ____________________________________________________ Plan and next appointment: ______________________________________________ Clinician signature/credentials: ______________ Date signed: __________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample Canadian telepsychology and virtual care note Scenario: an Ontario psychologist in Toronto sees an adult client with generalized anxiety disorder by secure video at the client's home in Kingston, Ontario. CBT, session 5. All details are fictional. Client: D.B., 41 · Date: 08/12/2026 · Session: 5 · Service: Individual psychotherapy, secure video, 50 min · Start/stop: 14:02 to 14:52 Telepresence: Secure video session; connection stable throughout. Client location confirmed at the start: home office, Kingston, Ontario, unchanged from prior sessions. Clinician location: private practice office, Toronto, Ontario. Writer registered in Ontario (CPBAO); client in Ontario, so no additional authorization required. Identity verified at intake; same account and device today. Client alone, door closed, using headphones. Consent: Verbal consent to virtual care documented 05/06/2026, covering risks, benefits, technology, and privacy limits. Reconfirmed today; no changes. Emergency plan: Local support: sister S.B., about 10 minutes away, contact on file. Nearest emergency department confirmed for the Kingston address. 9-8-8 reviewed at intake. Code word in place if privacy is compromised. Technology fallback: If video fails, clinician calls the client's phone on file. Not needed this session. Client report: "Better than the spring, but review season is coming." Average anxiety 5/10, down from 7/10 at session 3. Practised paced breathing on 4 of 7 days; one prolonged worry episode Sunday, about 90 minutes, tied to a quarterly review. Sleep 6 to 7 hours. Denied suicidal ideation, self-harm urges, and substance use. Observations: On camera: groomed, appropriately dressed, home office setting. Speech normal in rate and tone; affect mildly anxious, brightening when describing skills use. Engaged in agenda setting. GAD-7 completed by screen share: 10, down from 14 at intake. Video and audio quality adequate for observation; head and shoulders only in frame, so psychomotor observations were limited to what the frame showed. Assessment and risk: Symptoms consistent with GAD, improving. Progress toward Goal 1 (reduce daily worry duration): client initiated a coping skill without prompting. Risk: denies ideation; no indicators observed on screen; routine monitoring continues. Virtual delivery remains suitable; no barriers noted. Plan: Continue weekly CBT by video. Introduce worry postponement next session. Homework: thought records plus one scheduled 15-minute worry period daily. Re-administer GAD-7 at session 8. Next appointment 08/19/2026; client expects to join from the same Kingston location. Signed: R. Leduc, C.Psych., 08/12/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The jurisdiction question is answered, not implied: the client's location this session, the clinician's registration, and the explicit conclusion that no additional authorization was needed. The emergency plan is local to the client, not the clinician: the support person, crisis line, and emergency department all attach to the Kingston address the client joined from. Consent is dated and kept live: the standing verbal consent (valid across Canada) is referenced by date, and the note records that it was reconfirmed with nothing changed. The clinical note stands on its own: a tracked score, goal linkage, a risk statement, and a dated plan would survive even if the telepresence layer were ignored. The medium's limits are documented: the framing constraint on observation is stated, which protects the mental-status findings if the record is ever reviewed. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The virtual care note is ordinary clinical-record content under provincial health-information law: clients can request it, colleges can review it in a complaint or practice review, and it carries none of the special segregation US psychotherapy notes do. The enforceable layer is your college's record-keeping standard, and the clocks genuinely differ: Ontario's CPBAO Standards expect records kept at least 10 years after the last clinical contact, or 10 years after a minor client turns 18, Alberta works on a 10-year framework, and Quebec's regulation sets a 5-year floor from the last professional service. Supervision is also enforceable: under the CPBAO Standards a supervisor "must co-sign all documents which may be reasonably relied upon" in decisions affecting client care, which reaches a provisional registrant's virtual session notes. For clients with elevated risk, the emergency-plan element of this note is a pointer, not a substitute: build and reference a full safety plan whose resources match wherever the client joins from. The payer layer is policy, not law, and in Canada it is fragmented. Extended-health carriers such as Sun Life, Manulife, and Canada Life reimburse registered providers practising within scope and warn that claims failing documentation standards can be delayed or denied; NIHB covers up to 22 hours of counselling per eligible client per calendar year (more case by case) from providers in good standing with a legislated regulator, with virtual sessions eligible on secure platforms; Veterans Affairs Canada, WSIB, and the workers' compensation boards each set their own record rules. None of them prescribes a note format: the format is a convention, the content is the requirement. If your caseload crosses the border, the US layer is different in kind: place-of-service and modifier rules drive Medicare claims, behavioral health telehealth (including audio-only to the home) is a permanent Medicare benefit, and a Canadian licence does not authorize practice into a US state; PSYPACT requires a US home licence, so it is not a workaround for Canadian registrants. ## Common Canadian telepsychology and virtual care note errors auditors flag Canada's enforcement picture has a distinctive shape: as of mid-2026 there is no widely reported Canadian psychology-college discipline decision turning specifically on telepsychology-note deficiencies as the sole ground. Record-keeping and consent deficiencies are recurrent discipline themes, with telepractice appearing as context rather than the charge, and payer review by carriers, NIHB, and Veterans Affairs Canada focuses on provider eligibility, scope, and record adequacy. The US shows what concentrated scrutiny looks like: the HHS Office of Inspector General's telehealth integrity review (OEI-02-20-00720, September 2022) flagged 1,714 of roughly 742,000 telehealth-billing providers as high risk, with $127.7 million in Medicare fee-for-service payments, and in July 2022 the Department of Justice charged 36 defendants in telemedicine schemes alleging more than $1.2 billion in fraud. The BastionGPT Clinical Advisory Board sees the same errors most often in Canadian telepsychology and virtual care note reviews: Location captured once, at intake. The client-location rule turns on where the client sits during each session; a note that never re-verifies cannot show that a snowbird or student session was authorized. An emergency plan local to the wrong place. Resources named for the clinician's city, or for the client's usual address rather than the cottage, dorm, or hotel they actually joined from. Consent treated as a filing event. A form from two years ago with no session-level confirmation and no record of what was discussed; verbal consent is valid everywhere in Canada, but only the note proves it happened. Audio-only sessions logged as generic telehealth. The modality disappears from the record, and with it the reason video was not used, the detail US Medicare reviewers key on: place-of-service mix-ups, wrong modifiers, and audio-only claims without that reason are the documented US denial drivers. A complete telepresence checklist around a thin clinical note. No risk statement, no goal linkage, no observations; colleges discipline record-keeping substance, and the virtual wrapper does not supply it. ### Telepsychology and virtual care notes in Canada, the US, and Australia Aspect Canada United States Australia Status No mandated template; college record-keeping standards are enforceable (Ontario's CPBAO Standards are deemed standards of the profession under O Reg 801/93), while the CPA telepsychology guidelines (2023) and ACPRO model standards are guidance unless a college adopts them Telehealth billing is governed by regulation and payer policy (42 CFR 410.78); behavioral health telehealth, including audio-only to the home, is a permanent Medicare benefit Ahpra telehealth guidance expects clear, accurate records of each consultation; MBS telehealth determinations govern billing Terminology Telepsychology note, virtual care note; télépratique or télépsychologie in Quebec Telehealth progress note, teletherapy note Telehealth consultation note, case note What changes Registration where the client is located (the ACPRO MOU's three-way split), a per-session location check, and an emergency plan local to the client Place-of-service codes (02 vs 10), modifiers 93, 95, and FQ, and a recorded reason when audio-only is used; patient and clinician generally both in the US Referral chain under Better Access; the 12-month prior-relationship rule binds the referring GP, not the psychologist, and any eligibility exemption used must be recorded in the clinical notes Retention Provincial: Ontario expects 10 years after the last contact, or 10 years after a minor client turns 18; Alberta works on a 10-year framework; Quebec's floor is 5 years from the last professional service No federal patient-chart retention rule; HIPAA's 6-year rule covers required documentation such as policies, and state law governs charts 7 years from last contact and until age 25 for minors, statutory in NSW, Victoria, and the ACT; best practice elsewhere The clinical note travels; the regulatory layer does not. Before a cross-province or cross-border session, the live questions are whether you may practise where the client sits, which college's retention clock applies, and what the payer expects to see. The specifics move (the ACPRO MOU was updated in October 2025, and Yukon regulated psychology after it was drafted), so re-check your college's guidance and the MOU annually. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on Canadian telepsychology and virtual care notes. Draft the complete note, the telepresence layer plus the clinical content, from a few bullets, a dictation, or a pasted session transcript. Check a finished note before you sign for the elements reviewers look for: client location this session, consent reference, local emergency plan, technology fallback, and a risk statement. Adapt the same session record across jurisdictions: flag what changes when the client joins from another province, a US state, or under Australian MBS rules. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to record my client's physical location in every session note? No Canadian regulation makes the client's location a mandatory note field, so the field itself is convention. It earns its place because it evidences a question that is law: whether you are authorized to practise where the client is sitting, which is the rule most provinces apply. Confirm the location at the start of each virtual session and record it in one line; the sessions where it matters are exactly the ones where the answer changed. Do I need to be registered in my client's province to see them by video? It depends on the province. The ACPRO Memorandum of Understanding (in effect 1 April 2024) codifies a three-way split: eight jurisdictions (BC, Alberta, Saskatchewan, Manitoba, Ontario, PEI, Newfoundland and Labrador, and the Northwest Territories) require registration or a licence where the client is located, and most offer a Limited Telepsychology Practice certificate, Ontario's running up to 12 months. Quebec, Nova Scotia, and New Brunswick regulate through your home registration, and Yukon is not a signatory. The split is also profession-specific: CRPO "does not require practitioners who reside in another jurisdiction to become registered" to see an Ontario psychotherapy client, the opposite of the psychology position in the same province. Does consent to virtual care have to be in writing? No, nowhere in Canada. Ontario's PHIPA s 18 and its provincial analogues permit express or implied, verbal or written consent, and no regulation sets a re-consent interval. What matters is the record: the date and mode of the consent conversation, the risks and technology discussed, and a session-level confirmation in the note. The standalone telehealth consent is the document this note references. Can I keep seeing a client who winters in Florida or studies in another province? Treat the two cases differently. Inside Canada, apply the three-way split: a student spending term in an ACPRO-MOU province needs you to hold at least a limited telepsychology authorization there, while Quebec, Nova Scotia, and New Brunswick accept your home registration. Into the US, generally no without authorization from the destination state: US practice is governed by the client's location, a Canadian licence does not carry, and PSYPACT requires a home licence in a US compact state, so it is unavailable to Canadian-licensed psychologists. Some state boards offer narrow temporary or consulting exceptions; check before the client travels. How long should a telepsychology note be? The same as the underlying progress note plus a short telepresence block: most run 200 to 400 words and take 10 to 20 minutes by hand. No Canadian psychology college sets a note-completion deadline or a length rule; the enforceable standard is contemporaneous, accurate, complete records. When the telepresence facts are stable, a few lines confirming location, consent, and the emergency plan carry the layer. How is this different from a telehealth therapy note? The clinical core is identical; the wrapper changes with the regulatory layer. The telehealth therapy note page covers the US-facing record, where place-of-service codes, modifiers, and audio-only justifications drive Medicare claims. The Canadian note leads with the registration question, provincial retention, and a payer set of extended-health carriers and federal programs. Australia's telehealth layer sits under Ahpra's telehealth guidance , and its multidisciplinary coordination record is the separate MBS case conference note . What goes in the emergency plan for a virtual session? Four things, all anchored to where the client is that day: a local support person who can physically reach them, the local crisis line (9-8-8 anywhere in Canada), the nearest emergency department, and an agreed code word the client can use if the space stops being private. No Canadian rule codifies the element; it is convention driven by the CPA guidelines, which direct psychologists to discuss emergency procedures and local back-up before care begins. Rebuild the plan whenever the client's location changes. Do Canadian insurers require a specific note format for virtual sessions? No. Extended-health carriers such as Sun Life, Manulife, and Canada Life reimburse when the provider is registered with a recognized provincial regulator and the service is in scope, and they warn that claims failing documentation standards can be delayed or denied. NIHB covers up to 22 hours of counselling per eligible client per calendar year, with more available case by case, on secure virtual platforms; Veterans Affairs Canada and the workers' compensation boards publish their own rules. All of this is payer policy: the format is a convention, the content is the requirement. How does BastionGPT help with telepsychology documentation? Give it bullets, a dictation, or a session transcript and it drafts the full virtual care note: the telepresence layer and the clinical content underneath it. Before you sign, it checks the elements a college reviewer or payer looks for: the client's location this session, the consent reference, a local emergency plan, the technology fallback, and a risk statement. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Telehealth Psychotherapy Note Australia Telehealth & MBS Case Conference Note Safety Plan Telehealth Consent Narrative Progress Note Indigenous & First Nations Wellness Assessment Addendum Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/risk-screening-note All note templates # General Risk Screening Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A mental health risk screening note documents a brief, standardized screen for suicide, violence, abuse, substance use, and psychosis risk: the instrument used, the result for each domain, a positive or negative determination, and the follow-up action. Therapists and intake staff complete it at intake or when concerns arise, often inside the intake or progress note. Most run 50 to 150 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, intake staff, and trained front-line staff, with licensed clinician review of positive screens Audience Treating clinician, care team, supervisors, accreditation surveyors, auditors Typical length 50 to 150 words · 2 to 5 minutes by hand (clinical team estimate) Format family Short structured screen or checklist (compare: suicide risk assessment, intake note) When it's used At intake, at re-screening intervals the organization sets, and when clinical status changes Standards context Accreditation standard in accredited settings (Joint Commission, NSQHS); convention in private practice, usually embedded in the intake or progress note On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a mental health risk screening note? A mental health risk screening note records a brief, standardized screen for safety concerns: suicide risk, self-harm, violence toward others, abuse or safety at home, substance use, and indicators of psychosis. It captures the instrument used, the result for each domain, an explicit positive or negative determination, and the follow-up action taken. The instruments that standardized this note type are the Columbia Suicide Severity Rating Scale (Posner and colleagues, 2011), whose short Screener version exists for routine triage, item 9 of the PHQ-9 (Kroenke, Spitzer, and Williams, 2001), and the NIMH's four-question ASQ (Horowitz and colleagues, 2012), which takes about 20 seconds to administer. You will also hear the record called a suicide risk screen, an SI/HI screen, a safety screen, or, loosely, a risk assessment. It rarely stands alone: most practices embed it in the intake note or a progress note, or keep it as a one-page screening flowsheet. The distinction that carries this page: a screen is not an assessment. A screen is a triage step that answers one question, whether further evaluation is needed, and its job is routing. A negative screen is documented honestly and the visit moves on; a positive screen triggers a full suicide risk assessment , and the note shows that handoff. No federal statute, and essentially no state statute, requires an outpatient therapist to administer a formal screen at every session. The force behind screening is accreditation: Joint Commission standard NPSG.15.01.01, an accreditation standard rather than a law, has required accredited hospitals and behavioral health organizations since July 2019 to screen individuals aged 12 and older who present with a behavioral health condition as their primary reason for care, using a validated tool. Even there, the Joint Commission's R3 Report states the goal "does not require universal screening for suicidal ideation" of medical patients. In private practice, screening at intake is a professional convention backed by the malpractice standard of care, not a statutory mandate. ## Who uses risk screening notes and when Outpatient therapists and group practices screen at intake and when something shifts: new stressors, a status change, or ideation surfacing mid-treatment. Accredited settings screen on a schedule: hospital intake, emergency departments, community mental health centers, and integrated care clinics run validated screens under Joint Commission, Accreditation Canada, or NSQHS requirements and re-screen at intervals the organization sets. The C-SSRS Screener and the ASQ were designed so intake coordinators, nurses, and other front-line staff can administer them after brief training, with a licensed clinician taking over the moment a result is positive. Reach for this note when you need a fast, documented answer to whether further evaluation is needed. When the concern is specific from the start, skip the general screen and go straight to the deeper document: a full suicide risk assessment for suicide concerns, the violence risk assessment when the risk runs toward others, or the self-harm assessment for non-suicidal self-injury. ## Risk screening note structure: what goes in each section Header and context. Client identifier, date and time, setting (intake, routine visit, or re-screen), and who administered the screen. Time matters more here than in most notes: when a positive screen escalates, the record has to show how quickly the next step happened. Pitfall: no time recorded; a same-morning escalation you cannot place on a clock reads like a delay to a reviewer. Instrument and version. Name the tool exactly: C-SSRS Screener, ASQ, PHQ-9 with item 9 reviewed, or your clinic checklist with its version date, and file or reference the completed instrument. Accreditors expect a validated tool, and a validated tool means a named one. Pitfall: a narrative line like "denies SI/HI" with no instrument behind it; to a surveyor or a plaintiff's expert, that line is not evidence a screen occurred. Domains and results. One line per domain screened: suicide risk, self-harm, violence toward others, abuse or safety at home, substance use, psychosis indicators. Record a result for every domain, including the negatives. Pitfall: documenting only the domain that came back positive; silent negatives leave no proof the other questions were asked. Determination. One explicit line: negative screen, or positive screen with the domain named. This is the triage output everything downstream depends on, and the place to note when clinical judgment overrides a negative instrument result, which NIMH's ASQ guidance expressly allows. Pitfall: raw responses filed with no stated determination, leaving the next reader to score the screen themselves. Follow-up action and routing. What happened because of the determination: no further evaluation indicated, a full suicide risk assessment completed or scheduled with clinician and date, a safety plan, a referral, or an escalation with a name and a time. Every positive screen needs a documented next step. Pitfall: the screen-without-follow-up gap; in published EHR reviews, between one in five and one in three positive screens had no documented provider follow-up. Sign-off. Name, credentials, signature, and date and time signed. If a non-clinician administered the screen, add the licensed clinician who reviewed the result. Pitfall: missing or late authentication; missing provider signatures were among the most common failures in the OIG's national psychotherapy audit. ### Blank template (copy and adapt) GENERAL RISK SCREENING NOTE Client: [initials] Date: __________ Time: __________ Setting: [ ] intake [ ] routine visit [ ] re-screen Administered by (name, role): _________________________________________ INSTRUMENT (name and version; file or reference the completed tool) _______________________________________________________________________ DOMAINS SCREENED (one result per domain) Negative Positive Declined Suicide risk [ ] [ ] [ ] Self-harm [ ] [ ] [ ] Violence toward others [ ] [ ] [ ] Abuse or safety at home [ ] [ ] [ ] Substance use [ ] [ ] [ ] Psychosis indicators [ ] [ ] [ ] DETERMINATION [ ] Negative screen [ ] Positive screen, domain(s): _________________ Clinical judgment note (may override a negative result): ______________ FOLLOW-UP ACTION (required for every positive screen) [ ] No further evaluation indicated [ ] Full assessment (type, clinician, date/time): _____________________ [ ] Safety plan [ ] Referral: _______________________________________ [ ] Escalated to (name, time): ________________________________________ Administered by (signature/credentials/date/time): ____________________ Reviewed by licensed clinician (if applicable): _______________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample risk screening note Scenario: adult intake at an outpatient therapy practice. The screen is negative on five domains and positive on one, and the positive result routes to a full suicide risk assessment the same morning. All details are fictional. Risk Screening Note. Client: R.L., 41 · Date: 08/05/2026 · Time: 10:20 · Setting: Intake, outpatient behavioral health · Administered by: M. Sandoval, Intake Coordinator Instrument: C-SSRS Screener administered verbally per clinic intake protocol; completed screener filed in the chart with this note. Clinic intake safety checklist (v2.3, March 2026) used for the remaining domains. Domains and results: Suicide risk (C-SSRS Screener): POSITIVE Self-harm: negative Violence toward others: negative Abuse or safety at home: negative Substance use: negative Psychosis indicators: negative Determination: Positive screen, suicide risk domain; all other domains negative. Clinical judgment consistent with the instrument result. Follow-up action: Escalated at 10:26 to C. Reyes, PhD, licensed psychologist on site. Client remained with intake staff until handoff at 10:31. Full suicide risk assessment completed by Dr. Reyes beginning 10:40 and documented separately in the assessment note of 08/05/2026. Intake resumed after the assessment per Dr. Reyes' disposition. Administered by: M. Sandoval, Intake Coordinator, signed 08/05/2026, 10:28 Reviewed by: C. Reyes, PhD, 08/05/2026, 11:05 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The instrument is named, the completed tool is filed, and the checklist has a version date, so the record shows a validated screen rather than a summary of one. Every domain carries an explicit result, including the five negatives, so the note proves what was asked, not just what was found. The determination is a single unambiguous line, kept separate from the clinical formulation that belongs to the assessment. The positive screen has a complete routing trail: escalation time, named clinician, the assessment it triggered, and where that assessment lives in the chart. Both the administrator and the reviewing clinician sign with credentials and times, the authentication details national payment audits flag most often. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Locate yourself on the regulatory map first, because the rule that reaches a hospital intake desk does not reach a solo practice. In accredited organizations, screening is enforceable: Joint Commission NPSG.15.01.01 has required validated-tool screening since July 2019, a positive screen must trigger an evidence-based assessment, and the overall risk level and mitigation plan must be documented. That is an accreditation standard; it functions like law through deemed status, but it is not a statute. From January 1, 2026, hospitals and critical access hospitals carry the requirement as National Performance Goal 8 under the Joint Commission's restructured goals, while behavioral health organizations remain on NPSG 15. Canada reaches the same place through an Accreditation Canada Required Organizational Practice dating to 2008, now formalized as CAN/HSO 5064:2023, and Australia through NSQHS Comprehensive Care Standard 5 (Actions 5.31 and 5.32), with the Psychology Board's enforceable Code of Conduct adding prescriptive record-keeping from December 1, 2025. For unaccredited private practice, no statute in any of the three countries requires a per-visit screen. What some US states legislate instead is training: Washington's RCW 43.70.442 requires six hours of suicide-assessment training every six years for listed professions, a competency law, not a documentation law. The outpatient convention that holds up under review: screen at intake and when clinical status changes, and document negative results as carefully as positive ones. The billing layer is narrower than vendor guides suggest, and it is payer policy, not law. Many commercial payers accept 96127 for a brief standardized instrument, with per-day unit caps and bundling edits that usually block separate payment on the same day as psychotherapy or a 90791 evaluation. Medicare's G0444 covers annual depression screening only in primary care settings and cannot be billed by psychologists or clinical social workers, one of the most repeated billing mistakes in behavioral health. In Australia, no separate MBS item attaches to a screen performed inside a Better Access session. None of this changes the documentation duty. The screening note is part of the standard clinical record, it follows the retention rules of whatever record it lives in, and the format is a convention while the content is the requirement: instrument, result per domain, determination, follow-up, signature, and time. Keep the chain navigable: a reviewer should be able to walk from a positive screen to the assessment it triggered and on to any resulting safety plan without guesswork. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ## Common risk screening note errors auditors flag Two bodies of evidence frame this list. On the payment side, the HHS Office of Inspector General's national audit of Medicare psychotherapy services estimated $580 million in improper payments in a single year, with missing provider signatures among the most common failures. On the clinical side, published EHR reviews quantify the screening-specific failure: a 2018 study of 2,134 screened adolescents found documented follow-up for only 83% of positive screens, and a 2024 quality-improvement review found that just 67.7% of positive suicidal-ideation screens had documentation acknowledging the result. The BastionGPT Clinical Advisory Board sees the same errors most often in risk screening note reviews: "Denies SI/HI" as the entire screen. No named instrument, no recorded responses, no determination. In an accredited setting that fails the validated-tool expectation on its face; in private practice it gives a reviewer, or a plaintiff's expert, nothing to stand on. A positive screen with no documented next step. The 17 to 32 percent follow-up gap above is the most measured failure in screening documentation. If the assessment happened but is filed elsewhere, the screening note still needs the pointer to it. "Screen" and "assessment" used interchangeably. Calling a triage screen a risk assessment sets an audit expectation the note cannot meet; calling a genuine assessment a screen understates the work and weakens the defense it would have provided. Match the label to what was done. Silent negative domains. Only the concerning domain appears in the note, so the record cannot show the other domains were screened at all. A result per domain, or the filed instrument, closes the gap. The wrong screening code on the claim. G0444 billed by a psychologist or outside primary care, or 96127 stacked on same-day psychotherapy without checking bundling edits. Both are documented denial patterns, and both are payer policy questions, not documentation law. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on general risk screening notes. Draft the note from a dictated recap or a few bullets: instrument, a result for every domain, the determination, and the follow-up line, structured and ready for your signature. Pull the screening content out of a longer intake conversation or transcript, so the screen stands as its own defensible entry instead of dissolving into narrative. Check a finished note before you sign: named instrument and version, no silent domains, an explicit determination, and a documented next step behind every positive result. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I have to run a formal risk screen at every session? No. No US federal statute, no CMS manual provision, and no accreditation element requires an outpatient private-practice therapist to run a formal screen at every session. Accredited organizations screen patients presenting with behavioral health conditions and set their own re-screening intervals, and the C-SSRS has a shorter Since Last Visit version built for exactly that. The defensible outpatient pattern: screen at intake, then again when status changes, new stressors appear, or ideation surfaces in session. What is the difference between a risk screen and a suicide risk assessment? A screen is a brief triage step with a standardized tool, and it answers one question: is further evaluation needed. An assessment is the comprehensive clinical formulation a positive screen triggers, covering risk and protective factors, history, and a documented overall risk level with a plan. The two carry different audit expectations, so label each note honestly. The suicide risk assessment template covers the deeper document. How long is a risk screening note, and does it need its own page in the chart? Most run 50 to 150 words, because the validated instrument does the heavy lifting: the ASQ takes about 20 seconds to administer, the PHQ-9 one to three minutes, and the C-SSRS Screener a minute or two when symptoms are absent. The note does not need to stand alone. Many practices embed it in the intake note or a progress note, or keep a screening flowsheet; what matters is that the elements are findable: instrument, results, determination, follow-up, signature, and time. How do I document a negative screen defensibly? Name the instrument and version, record the responses or triage result for every domain, state the negative determination in its own line, and sign with credentials, date, and time. NIMH's ASQ guidance is explicit that "clinical judgment can always override a negative screen," so note when your judgment confirmed, or overrode, the instrument result. Does a positive PHQ-9 item 9 mean I have to do a full assessment? Clinically, any endorsement deserves a documented next step. In a study of more than 200,000 questionnaires, patients answering "nearly every day" on item 9 had a 6.37 times higher hazard of a subsequent suicide attempt than those answering "not at all," and their one-year attempt risk was roughly ten times higher in absolute terms. A positive item 9 does not automatically mean hospitalization. It means a documented response: a brief safety assessment, a full assessment, or a clearly reasoned clinical decision, signed and timed. Does the Joint Commission require the C-SSRS specifically? No. The requirement is a validated tool, not a named one. The Joint Commission's own FAQ confirms that no specific instrument is required and that different departments may use different tools; the C-SSRS, the ASQ, and PHQ-9 item 9 are the common choices. Track the 2026 change too: hospitals and critical access hospitals now carry suicide-risk reduction as National Performance Goal 8, part of a restructuring that cut the accreditation manual from 1,551 requirements to 774, while behavioral health organizations remain on NPSG 15. Who can administer the screen, and can it be billed? The C-SSRS and ASQ were designed for administration by non-mental-health staff after brief training, which is why intake coordinators and nurses run them in many settings; a licensed clinician takes over the moment a screen is positive. Billing is a separate, narrower question. Many commercial payers accept 96127 per standardized instrument, with unit caps and edits that usually block separate payment alongside same-day psychotherapy or 90791. Medicare's G0444 covers annual depression screening only in primary care and cannot be billed by psychologists or clinical social workers, and in Australia no separate MBS item attaches to a screen done inside a Better Access session. Verify your own payer's policy before building billing into the workflow. Do Canada and Australia require risk screening notes? The pattern matches the US: accreditation and professional standards, not statutes. Accreditation Canada has treated suicide risk assessment and monitoring as a Required Organizational Practice since 2008, now formalized in CAN/HSO 5064:2023, and provincial colleges govern records, with retention commonly around 10 years. Australia's NSQHS Comprehensive Care Standard 5 (Actions 5.31 and 5.32) requires accredited health services to screen and respond, and the Psychology Board's Code of Conduct, enforceable since December 1, 2025, added prescriptive record-keeping; the retention convention is 7 years from last contact, or until a minor turns 25. Private practitioners in both countries screen by professional convention and standard of care. Can BastionGPT draft a risk screening note? Yes. Dictate the intake recap, paste the relevant part of a transcript, or give it a few bullets, and BastionGPT drafts the structured note: instrument, a result for every domain, the determination, and the follow-up line, ready for your review. It can also check a finished note for silent domains or a missing next step before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Suicide Risk Assessment Violence & Homicide Risk Assessment Self-Harm (NSSI) Assessment Safety Plan Intake Note Eating Disorder Assessment Transfer-of-Care Summary GAD-7 Score Documentation C-SSRS Documentation EPDS Score Documentation CAGE Questionnaire Documentation AUDIT-C Score Documentation DAST-10 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/eating-disorder-assessment All note templates # Eating Disorder Assessment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An eating disorder assessment is a structured intake evaluation that combines a validated screener such as the SCOFF or EDE-Q, DSM-5-TR criterion mapping, eating and weight history, and a medical-risk appraisal with a level-of-care rationale. Therapists, psychologists, and counselors complete it at intake, coordinating with medical and nutrition colleagues. Most run 400 to 1,000 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, and counselors, in coordination with medical and nutrition teams Audience Treating clinician, medical and nutrition team, supervisors, payers, auditors Typical length 400 to 1,000 words · 30 to 60 minutes by hand (clinical team estimate) Format family Structured intake assessment (compare: biopsychosocial assessment, intake note, substance use assessment) When it's used At intake when disordered eating presents or a screener flags it; measures repeated at level-of-care decisions Standards context APA guideline and payer expectations shape content in the US and Canada; Australia's MBS items legislate it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an eating disorder assessment? An eating disorder assessment is a composite clinical document: a diagnostic evaluation focused on eating pathology that combines a brief screener, a dimensional measure, DSM-5-TR criterion mapping, and a medical-risk appraisal with a level-of-care rationale. No single author created the composite, but its components have verifiable lineages: the SCOFF questionnaire (Morgan, Reid, and Lacey, BMJ 1999), the Eating Disorder Examination and its self-report EDE-Q (Cooper and Fairburn in the 1980s; Fairburn and Beglin 1994), and the EAT-26 (Garner and colleagues, 1982), all now read against DSM-5-TR (2022) and the APA practice guideline for eating disorders, 4th edition (February 2023). Clinicians and payers also call it an ED evaluation, an eating disorder intake, or, in US billing terms, a psychiatric diagnostic evaluation (90791) when disordered eating is the presenting problem. In Australia it feeds the Eating Disorder Treatment and Management Plan (EDTMP). The distinction that carries this page: almost nothing about this document is required by law in the United States or Canada. The near-universal beliefs that a clinician must weigh the client, must order labs, and must obtain medical clearance before outpatient therapy are APA guideline recommendations (rated 1C) and payer expectations, not statutes. Australia is the outlier: the MBS eating disorder items, in force since 1 November 2019, write assessment content directly into national payment law, down to a global EDE-Q score of 3 or higher as an eligibility threshold for non-anorexia diagnoses. Everywhere else, the eating disorder assessment is a focused specialization of the biopsychosocial assessment , one level deeper than a general intake note : the same frame, plus eating-specific history, behavior domains, a medical-risk appraisal, and a level-of-care argument. ## Who uses eating disorder assessments and when Psychologists, professional counselors, clinical social workers, and marriage and family therapists complete it at intake in outpatient practices, college counseling centers, and eating disorder specialty programs; psychiatrists and primary care teams run the parallel medical workup, and dietitians write the separate nutrition assessment. Reach for it when disordered eating is the presenting concern, or when a screener flags it partway into treatment. At later decision points, re-administer the measures rather than redoing the whole document (an outcome measure note handles those re-administrations). When substances share the picture, a substance use assessment runs alongside rather than instead: the two documents argue different risks. ## Eating disorder assessment structure: what goes in each section Referral context and presenting concerns. Who sent the client and why, and the concern in the client's own words, quoted briefly. Then onset, course, and prior treatment episodes with response. Pitfall: opening with your formulation instead of the client's words; on a 90791 review, the chief complaint in the client's own language is the first element a payer reviewer looks for. Screening and standardized measures. Name the instrument, the date, who administered it, and where the scored form is filed. A SCOFF at the front desk plus a self-report measure in the waiting room is a common pairing; the interview-based EDE belongs to specialty settings with 45 to 75 minutes to spend on it. Pitfall: a score with no instrument, date, or administrator attached. Australia's Professional Services Review has cited outcome tools that were "incorrectly scored" in mental health services; an unattributed number reads the same way. Eating, weight, and behavior history. Trajectory and pattern at the domain level: restriction, binge episodes, compensatory behaviors, driven exercise, food-repertoire narrowing, and the share of the day spent preoccupied with food, weight, and shape. Record which domains are endorsed and which are denied; the quantified detail lives in the standardized measure and the medical record, referenced by location. Pitfall: reproducing weight numbers and behavior counts in the narrative. The assessment travels further than the measure does (releases, record requests, the client's own portal), so domain-level narrative plus a pointer keeps it clinically complete and safe to share. Medical risk and the handoff. The section that separates a defensible assessment from a liability. Document the medical-risk domains reviewed (vital signs, orthostatic symptoms, labs, ECG when indicated), who owns each one, the referral made, the release signed, and when results are due back. The therapist's duty is the handoff and its follow-through, not the values. Pitfall: transcribing lab values and vitals into the therapy record. "CBC and metabolic panel ordered by Dr. R., results to this office by 08/20" shows the duty discharged; copied numbers add nothing and go stale. Mental status, co-occurring conditions, and suicide risk. A full mental status exam, screening for depression, anxiety, trauma, and substance use, and a suicide-risk assessment with stated reasoning. Eating disorders carry elevated suicide risk, so silence here is the worst omission available. Pitfall: a checked risk box with no reasoning; billing analysts place checkbox risk assessment among the leading documentation-based 90791 denials. DSM-5-TR diagnosis and severity. Map the endorsed findings to criteria, state the diagnosis and specifier, and mark it provisional if medical findings or collateral records are pending. Pitfall: a diagnostic label with no criterion trail. When documented findings cannot be matched to the billed diagnosis, the denial arrives as an information or medical-necessity error (the CARC 16 and 50 family). Level-of-care rationale. The recommended level (outpatient, intensive outpatient, partial hospitalization, residential, inpatient), the criteria considered, and the facts supporting the choice. This is the paragraph payers actually read. Pitfall: arguing level of care from a DSM severity band. Payer criteria weight physiologic instability and rate of change over any static category, so anchor the rationale in current status, trajectory, and the medical team's findings. Collaborative plan and communication. Therapy modality and frequency, the medical monitoring schedule, nutrition referral, releases on file, who communicates with whom, and the review date. Pitfall: "referred to PCP" with no recipient, date, or release. For auditors and licensing boards alike, multidisciplinary communication that is not documented did not happen. ### Blank template (copy and adapt) EATING DISORDER ASSESSMENT Client: [initials] DOB: ______ Date: ______ Clinician: ________ Referral source: ______________ Reason: ___________________________ PRESENTING CONCERNS (client's words): HISTORY (onset, course, prior treatment and response): MEASURES (instrument | date | administered by | filed where) _________________ | ______ | _______________ | _____________________ EATING / WEIGHT / BEHAVIOR HISTORY (domain level; detail in measure) Domains endorsed: ____________________ Denied: _____________________ Weight status and trajectory: assessed by _________ (medical record) MEDICAL RISK AND HANDOFF Domains reviewed (vitals / orthostatics / labs / ECG): ______________ Referral to: ____________ Date: ______ Release signed: [ ] ______ Results due: ______ Priority symptoms flagged: ____________________ MSE AND CO-OCCURRING CONDITIONS: SUICIDE RISK (findings and reasoning): DSM-5-TR DIAGNOSIS (criteria mapped; provisional if pending): LEVEL-OF-CARE RATIONALE (status, trajectory, criteria used): COLLABORATIVE PLAN (therapy / medical / nutrition / review date): Clinician signature/credentials: Date signed: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample eating disorder assessment Scenario: adult outpatient, self-referred after a partner raised concern, evaluated by a psychologist in private practice. Screening instruments are named without scores, and medical values stay with the medical team, deliberately. All details are fictional. Client: R.L., 29 · Date: 08/11/2026 · Clinician: D. Marsh, PsyD · Service: Diagnostic evaluation (90791), 74 min, in office Referral and presenting concerns: Self-referred after her partner raised concern about restrictive eating and withdrawal from shared meals. Client: "I plan my whole day around not eating in front of people." Reports rigid food rules beginning in her mid-20s, a gradually narrowing food repertoire, and growing distress around work lunches. One prior course of general therapy; eating was not addressed. No prior eating disorder treatment. Measures: SCOFF administered verbally at the start of session; screen positive. EDE-QS completed in the waiting room; global score in the clinical range. Scored forms filed in the measures section of the chart and shared with the medical team under the release below; scores are not repeated in this narrative. Eating, weight, and behavior history: Restriction and food-repertoire narrowing endorsed. Episodes of loss-of-control eating endorsed. Compensatory behavior endorsed at intake (a driven exercise pattern); frequencies are documented in the EDE-QS and the medical record, not here. Client describes preoccupation with food and shape occupying most of the day. Weight status and trajectory will be assessed by the client's primary care physician; client declined in-office weighing today, which was accepted and documented, and a records request went out 08/11/2026. Medical risk and handoff: Review of systems notable by client report for lightheadedness on standing and cold intolerance; no fainting. This office does not collect vitals. Same-week appointment confirmed with A. Osei, MD (primary care) for vital signs, orthostatic check, CBC, and comprehensive metabolic panel, with ECG at Dr. Osei's discretion per APA guideline indications. Release of information signed 08/11/2026 for two-way communication. Lightheadedness flagged to Dr. Osei's office as the priority item; results due to this office by 08/20/2026. Mental status, co-occurring conditions, and suicide risk: On time, casually dressed, cooperative, guarded when food is discussed directly. Mood "worn down," affect congruent and constricted. Thought process linear; no psychotic symptoms. Depression and anxiety screens completed and filed with the measures. Alcohol and substance use screened; denies problematic use. Suicide risk: denies current ideation, plan, and intent; no history of attempts; connected to partner and sister and engaged in seeking care. Risk judged low today, with reasoning noted and monitoring planned at each session given the elevated suicide risk that travels with eating disorders. DSM-5-TR diagnosis: Endorsed pattern of restriction, loss-of-control eating, and compensatory behavior mapped against DSM-5-TR feeding and eating disorder criteria in interview. Provisional diagnosis: other specified feeding or eating disorder (OSFED), pending medical findings and prior records; differential includes bulimia nervosa and anorexia nervosa, to be revisited when weight-status data arrives from Dr. Osei. Level-of-care rationale: Outpatient level recommended at this time: stable function at work, no reported fainting or rapid change in weight status, motivation for treatment, and daily support at home. Criteria for a higher level of care (physiologic instability, rapid trajectory change, inability to interrupt behaviors at outpatient intensity) reviewed and not met on today's information. This rationale will be re-argued, not assumed, when the medical results return. Collaborative plan: Weekly individual therapy, CBT-E informed, starting 08/18/2026. Primary care evaluation this week as above; case call with Dr. Osei after results. Referral offered to a registered dietitian experienced in eating disorders; client will decide by next session. Re-administer the EDE-QS at week 8. Partner session offered for meal-support psychoeducation; client considering. Diagnosis and level of care to be reviewed 09/08/2026, or sooner if medical findings require. Signed: D. Marsh, PsyD, 08/11/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The chief complaint is in the client's words, and the history gives onset, course, and prior treatment: the elements Medicare contractor fact sheets list for a 90791. Every measure is named, dated, attributed, and filed, so nothing in the record rides on an orphan score. Medical risk is discharged as a documented handoff: a named physician, the ordered workup, a signed release, the priority symptom flagged, and a date results are due back. The level-of-care paragraph argues from current status and trajectory rather than a static severity category, which is how payer reviewers actually decide. Suicide risk carries reasoning and a monitoring plan, and behavior detail stays at domain level, so the document is safe to release without editing. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations An eating disorder assessment is part of the designated record set: the client can request it, the medical and nutrition team will rely on it, and payers can audit it. That reach should shape what goes in it. Keep behavior detail at domain level with a pointer to the measure, keep weight data with the medical team, and the document stays accurate, useful, and safe to share with a client who may be acutely distressed by numbers. Authorship is a scope question in every jurisdiction: in the US, psychologists, counselors, clinical social workers, and marriage and family therapists complete the evaluation and assign the diagnosis within state scope of practice; in several Canadian provinces, formally communicating a diagnosis is a controlled act under provincial health-professions law; in Australia, only a GP, consultant psychiatrist, or consultant paediatrician may make the EDTMP eligibility decision, whatever assessment data allied health contributes. A dietitian's nutrition assessment is a companion document, not a substitute. Retention is law, and it is local: commonly 6 to 10 years for adults under US state rules (longer for minors), at least 10 years from the last interaction (or 10 years past the client's 18th birthday) in Ontario, and generally 7 years for adults with minors' records kept to age 25 in Australia. For US insurance work, the billing container is the psychiatric diagnostic evaluation (90791), and coverage turns on medical necessity under section 1862(a)(1)(A) of the Social Security Act . Medicare contractor fact sheets expect a complete history, a mental status exam, a diagnosis, and an initial treatment plan; once-per-episode and once-per-year limits are payer rules, not CPT rules; and managed behavioral health organizations make prior authorization the leading initial-evaluation denial (CARC 197, precertification absent). None of that mandates a form: in the US and Canada, the format is a convention and the content is the requirement. Australia inverts this. MBS Explanatory Note AN.36.2 legally enumerates what the plan must document (assessment results, patient needs, goals and actions, referrals, and a review date), eligibility for non-anorexia diagnoses is written into the item rules (a global EDE-Q score of 3 or higher plus behavioral and clinical indicators), and reviews around sessions 10, 20, and 30 gate access to the full 40 psychological and 20 dietetic sessions. Build the assessment so its measures re-administer cleanly (the outcome measure note covers that cadence) and the medical-necessity argument reads in one place. ## Common eating disorder assessment errors auditors flag No OIG or CERT report isolates the diagnostic evaluation codes, and none gives an eating-disorder-specific error rate, so the enforcement record here is analogue and worth reading that way. The nearest US signal is the HHS-OIG psychotherapy audit (A-09-21-03021, 2023), which estimated $580 million in improper Medicare psychotherapy payments in a single year, with missing treatment plans and undocumented time among the causes. The nearest Australian signal is the Professional Services Review's run of mental-health-plan findings: in the April 2023 Director's Update, a practitioner repaid $896,000 where plans "did not include the level of information required" by the item descriptor, and in May 2023 another repaid $330,000 where plans were "template based" with generic goals. The BastionGPT Clinical Advisory Board sees the same errors most often in eating disorder assessment reviews: A diagnosis without a criterion trail. The DSM-5-TR label appears, but the documented findings do not map to it, so a reviewer cannot connect the claim to the record; these return as information and medical-necessity denials (CARC 16 and 50). Checkbox risk assessment. Suicide risk marked denied with no reasoning, in a population where elevated risk is the base rate; billing analysts place this among the leading documentation-based 90791 denials. Orphan measures. A screener mentioned with no instrument name, date, administrator, or filing location, or a score entered wrong; the PSR's March 2023 update records outcome tools "incorrectly scored" in mental health services. Template-shaped narratives. Cloned text and generic goals across clients: the exact finding behind the $330,000 PSR repayment, and the pattern US reviewers read as a service not individually rendered. A missing handoff trail. Medical risk is noted, but no named recipient, referral date, release, or follow-up plan appears. In one Australian tribunal case, a psychologist was disqualified where the tribunal found a failure to "appropriately assess the client and her treatment needs." The duty is not only to spot risk; it is to document who now owns it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on eating disorder assessments. Draft the complete assessment from an intake transcript, a dictation, or your bullets, with the client's words preserved and every domain sectioned. Pull the assessment-relevant history out of referral letters, prior records, and screener results, and merge it with your interview into one coherent document. Check the draft before you sign: criterion trail, risk reasoning, level-of-care rationale, referral and release documentation, and any stray numbers that belong in the measure instead of the narrative. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long does an eating disorder assessment take? The written document usually runs 400 to 1,000 words and takes 30 to 60 minutes to write by hand. Administration time depends on the instruments: the SCOFF takes under a minute, the EDE-Q about 15 to 20 minutes as self-report, and the full EDE interview 45 to 75 minutes, which is why the interview stays in specialty settings. Most outpatient clinicians pair a brief screener with a self-report measure inside a standard intake hour. Do I have to weigh the client or order labs before starting therapy? No law in the United States or Canada requires a therapist to weigh a client, order labs, or obtain medical clearance before outpatient eating disorder treatment. Those expectations come from the APA practice guideline (rated 1C recommendations) and from payer level-of-care criteria: real reasons to arrange them, but a different kind of reason than a statute. If weighing is deferred or declined, document the clinical reasoning and route weight monitoring to the medical team. The Academy for Eating Disorders' medical care standards are the reference most teams hand their medical partners. Can a counselor or social worker diagnose an eating disorder without a physician? In the US, yes, within state scope of practice: psychologists, counselors, clinical social workers, and marriage and family therapists complete diagnostic evaluations and assign DSM-5-TR diagnoses, and the counselor and MFT professions gained permanent Medicare billing in 2024. The medical-risk half of the picture still requires medical involvement, which is why the handoff section matters. In several Canadian provinces, formally communicating a diagnosis is a controlled act. In Australia, allied health can contribute assessment data, but only a GP, consultant psychiatrist, or consultant paediatrician can make the eating disorder plan eligibility decision. What billing code covers an eating disorder assessment in the US? 90791, the psychiatric diagnostic evaluation (90792 when a prescriber adds medical services); the Medicare national non-facility rate for 90791 is $173.35 for 2026. Medical necessity is carried by the content: a chief complaint in the client's words, full history, complete mental status exam, risk assessment with reasoning, a criterion-mapped diagnosis, and an initial plan. Frequency limits such as once per provider per year are payer rules, not CPT rules, and prior authorization is the top managed-care denial for initial evaluations, so check both before the appointment. How is this different from a biopsychosocial or intake assessment? The generic intake note and biopsychosocial assessment survey every domain at even depth. The eating disorder assessment keeps that frame and goes deeper in one place: eating-specific behavioral history, weight-history coordination with the medical team, a medical-risk appraisal, and a level-of-care rationale. When substance use shares the presentation, run a substance use assessment alongside it: withdrawal risk and refeeding risk are different arguments, and payers read them separately. Which screener should I use: SCOFF, EDE-Q, or EAT-26? No US or Canadian regulation names one, so the choice is clinical. The SCOFF is fastest, with pooled sensitivity of 0.86 and specificity of 0.83 across 25 validation studies, and the same review ( Kutz et al. 2020 ) concluded the evidence is insufficient for the full range of DSM-5 eating disorders, so treat a positive as a flag, not a finding. The EDE-Q adds subscale and global scores and is the instrument Australia wrote into its MBS eligibility rules, which makes it the default wherever Australian portability matters. Whichever you pick, document instrument, date, administrator, and filing location. What do Australia's MBS eating disorder items require? The items (90250 to 90257 for GP plans, with psychiatrist, paediatrician, and telehealth equivalents) took effect 1 November 2019 and legislate the content: Explanatory Note AN.36.2 requires the plan to document assessment results, patient needs, goals and actions, referrals, and a review date. Anorexia nervosa qualifies on clinical diagnosis; other eligible diagnoses need a global EDE-Q score of 3 or higher plus behavioral and clinical indicators from the item rules. Only a GP, consultant psychiatrist, or consultant paediatrician can author the plan, reviews fall around sessions 10, 20, and 30, and the scheme caps at 40 psychological plus 20 dietetic sessions in 12 months, one plan per patient per year. How long do I need to keep an eating disorder assessment? Retention is set by state, provincial, and professional rules, not by the document type. US states commonly require 6 to 10 years for adult records and longer for minors. Ontario colleges expect at least 10 years from the last interaction, or 10 years past the client's 18th birthday. Australian boards generally expect 7 years from the last entry for adults, with minors' records kept until age 25. The assessment anchors the diagnosis and the level-of-care decision, so it should survive as long as anything in the chart. Can BastionGPT draft an eating disorder assessment? Yes. Give it an intake transcript, a dictation, or bullets, and it drafts the full assessment: domains sectioned, client quotes preserved, the medical handoff documented, and the level-of-care rationale argued from status and trajectory. It also flags stray numbers that belong in the measure rather than the narrative. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Intake Note Biopsychosocial Assessment Substance Use Assessment Outcome Measure Review Note General Risk Screening Note Case Management Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/collateral-contact-note All note templates # Parent & Caregiver Collateral Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A parent and caregiver collateral note documents information a clinician gathers from someone other than the client, such as a parent, guardian, or caregiver, and what the clinician did with it. Child and adolescent therapists write one after any substantive caregiver contact. The note lives in the client's chart, attributes every statement to its source, and typically runs 100 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Child and adolescent therapists, psychologists, counselors, evaluators, care team members Audience Treating clinician, evaluators, the care team, payers and auditors where billed Typical length 100 to 300 words · 5 to 10 minutes by hand (clinical team estimate) Format family Contact note in the progress note family (compare: school collateral note, care coordination note) When it's used After any substantive contact with a parent, guardian, or caregiver about the client Standards context A documentation convention, common in child and adolescent work; no law prescribes its format On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a parent and caregiver collateral note? A parent and caregiver collateral note (often just called a collateral contact note) documents information a clinician gathers from someone other than the client: a parent, guardian, spouse, or caregiver. It records who the informant was, why the contact happened, what they reported, what the clinician observed and concluded, and what happens next. The format has no single author or founding body. It grew inside the progress note tradition as public behavioral health systems began listing collateral contact as its own note or service type in their documentation manuals, and local labels persist: Massachusetts uses a consultation and collateral contact progress note, some Medicaid systems bill it as a distinct service line, and Australia calls the reimbursable version family and carer participation. The orientation that saves clinicians the most trouble: the collateral is not the client, and the note is not theirs. Everything a caregiver tells you goes into the client's chart, travels with the chart when records are released, and, under the US information-blocking rules in force since April 2021, is generally visible to the patient or their representative by default. A collateral note is also not a family therapy note. Talking with a parent to gather information about the client is documentation; treating the family system in session is a clinical service with its own billing rules, covered on the family and couples therapy note page. Caregiver history taken during a first appointment does not need a separate note at all: it belongs in the intake note . And when the informant is a teacher or school staff member rather than a caregiver, use the school collateral note , which adds the education-records overlay. ## Who uses parent and caregiver collateral notes and when Child and adolescent therapists write these more than anyone: a 9-year-old's chart without caregiver input is missing half the clinical picture. Evaluators use the same structure when caregiver interviews feed a developmental assessment , and clinicians treating adults reach for it when a spouse, adult child, or caregiver calls with observations, which HIPAA permits you to receive. Write one after any substantive contact: a scheduled parent check-in, an unprompted phone call, a caregiver email with clinical content, a conversation in the waiting room that turned clinical. Choose a different document when the contact is really something else: a session that treats the family relationship belongs in a family therapy note , and provider-to-provider logistics belong in a care coordination note, a boundary New York's Medicaid guidance draws explicitly. ## Parent and caregiver collateral note structure: what goes in each section Contact header. Client identifier and date, plus the mode of contact (phone, in person, telehealth, written) and start and stop times or total duration. Time feels bureaucratic for an unbilled call, but it is the first thing a reviewer checks if the contact ever supports a billed service. Pitfall: recording who called but not the mode or time, which leaves the note unable to either support or clearly rule out a billing question later. Informant and relationship. Who provided the information and their relationship to the client: mother, stepfather, grandmother with guardianship, group home staff. Role matters as much as name, because guardianship and custody determine who can consent and who may access the record. Pitfall: assuming the source must always be fully named. A 2024 American Psychiatric Association ethics opinion supports describing a source in more general terms when the source insists on confidentiality and the material is sensitive; decide deliberately, not by habit. Consent and authorization status. One or two lines stating the authority for the contact: for a minor, which parent or guardian consented to treatment and caregiver involvement; for an adult, whether the client agreed to the contact and whether a release covers anything you disclose. Pitfall: conflating listening with disclosing. Receiving information from a family member does not require a release; sharing the client's information with them does. Write down which direction the information flowed. Reason for contact. Why this contact happened and how it ties to the treatment plan: monitoring skill generalization at home, clarifying a reported incident, gathering developmental history. Pitfall: no tie to the plan. New York's CFTSS guidance defines collateral contact as gathering treatment-relevant information and excludes care coordination and logistics; a reason that reads as scheduling belongs in a different note. Information reported. The informant's account, with every statement attributed: mother reports, stepfather states, grandmother describes. One short quote can carry tone. Pitfall: the top failure in collateral documentation is dropped attribution, a caregiver's report written so it reads as the clinician's own finding. Once "struggling at school" appears without a source, the chart asserts it as fact. Clinician observations and assessment. What you directly observed during the contact (the caller's tone, engagement, use of coached strategies) and your appraisal of the information: consistent with in-session presentation or not, clinically significant or not. Pitfall: letting your conclusions drift up into the reported-information section. The report, the observation, and the appraisal must stay visibly separate; that separation is what keeps the note defensible when the client, another provider, or a court reads it. Disclosure, plan, and signature. What, if anything, you shared with the caregiver (kept to the minimum necessary), how the information changes treatment, any follow-up, then signature with credentials and date, with supervisor co-signature where your setting requires it. Pitfall: documenting what you heard but not what you said. The disclosure direction is the regulated one; a note silent on it cannot show the disclosure stayed within consent. ### Blank template (copy and adapt) PARENT / CAREGIVER COLLATERAL CONTACT NOTE Client: [initials] DOB: __________ Date of contact: __________ Mode: [ ] phone [ ] in person [ ] telehealth [ ] written Start/stop or total duration: __________________________________________ Informant name: ______________________ Relationship/role: _____________ Guardianship or custody status (if relevant): __________________________ CONSENT AND AUTHORIZATION STATUS Who consented to treatment / caregiver involvement: ____________________ Release on file for disclosures to this person: [ ] yes [ ] no Scope of permitted disclosure: _________________________________________ REASON FOR CONTACT (tie to treatment plan goal): _______________________ INFORMATION REPORTED (attribute every statement to its source) ________________________________________________________________________ ________________________________________________________________________ CLINICIAN OBSERVATIONS OF THE CONTACT: _________________________________ ASSESSMENT (clinician appraisal, kept separate from the report): _______ INFORMATION DISCLOSED TO INFORMANT (minimum necessary): ________________ PLAN / ACTION TAKEN: ___________________________________________________ Clinician signature/credentials: ______________________ Date: _________ Supervisor co-signature (if required): _________________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample parent and caregiver collateral note Scenario: an outpatient child therapist documents a scheduled phone check-in with the mother of a 9-year-old client in treatment for emotion regulation and disruptive behavior. All details are fictional. Collateral Contact Note. Client: J.R., 9 · Date of contact: 08/04/2026 · Mode: telephone · Start/stop: 3:10 pm to 3:28 pm · Clinician: A. Bhatt, LPC Informant: Client's mother, legal guardian. Consented to treatment 06/10/2026. Consent status: Caregiver involvement agreed at intake and documented in the treatment consent. No third-party release involved in this contact; disclosures limited to home-practice coaching per the plan. Reason for contact: Scheduled biweekly caregiver check-in under treatment plan Goal 2: generalize emotion regulation skills from session to home and school. Information reported: Mother reports J.R. used the stop-and-breathe routine twice this week without prompting, once before homework and once after losing a video game. She reports homework refusal decreased from daily to about twice this week. Mother describes one episode Sunday evening: J.R. slammed his bedroom door when told to end screen time, then calmed on his own within about ten minutes. She states his teacher emailed that he "had a calmer week." Mother reports no changes in sleep or appetite and denies any safety concerns at home. Clinician observation: Mother was engaged and specific, asked how to respond to door slamming, and described using the labeled-praise strategy from the 07/21/2026 caregiver session. Observation of the call only; the client was not present. Assessment: Caregiver report is consistent with in-session progress and suggests skills are beginning to generalize to home. This appraisal rests on reported information, not direct observation of the client. Disclosed to informant: Reviewed the home steps for the stop-and-breathe routine and coached a neutral response to door slamming. No other clinical content disclosed. Plan: Review the Sunday episode with J.R. at the 08/11/2026 session using the repair-and-replay exercise. Continue biweekly caregiver check-ins; next contact the week of 08/17/2026. 18-minute informational contact; not billed as family psychotherapy. Filed in J.R.'s record. Signed A. Bhatt, LPC, 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Every statement has a source. Each item in the reported section starts with "Mother reports" or an equivalent, so nothing a caregiver said can be mistaken for a clinician finding. Consent authority is stated, not assumed. The note names the guardian, the treatment consent date, and the agreed scope of caregiver involvement, so a reviewer can see the basis for the contact in one line. Report, observation, and appraisal stay separate. The clinician's observation is labeled as an observation of the call, and the assessment names its own evidence base as reported information. The disclosure direction is documented. The note shows what flowed back to the caregiver and that it stayed within the minimum necessary for her involvement in care. Time, mode, and billing status are explicit. An 18-minute phone contact marked as not billed heads off the most common collateral billing error before it can happen. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Whose record is it: the client's, always. A collateral note lives in the client's designated record set; the caregiver who provided the information has no chart here and no standing to control the note. That cuts both ways. When records go to another provider, an attorney, or the client, the informant's statements travel with them, and under the Cures Act information-blocking rules most electronic notes are available to the patient or their representative by default. Psychotherapy notes are excluded from that electronic-access default, which makes storage location a genuine decision when a family source shares something volatile. A 2024 American Psychiatric Association ethics opinion warns that a named family source can be discovered by the patient, supports describing a source more generally when confidentiality is requested and the material is sensitive, and cautions against promising a family source confidentiality you cannot guarantee. The receiving side is more permissive than folklore suggests: HHS guidance permits listening to family members, and disclosure to a person involved in the client's care is allowed within minimum-necessary limits; a signed release is what you need to share beyond that, not to receive and document. For minors, the consenting parent or guardian generally holds record access, so write every collateral note knowing a parent may read it, and note that state minor-consent laws vary on who controls the record. Many child clinicians also tell the client, in age-appropriate terms, what they will and will not share with caregivers; that is alliance practice and assent convention, not a legal requirement. The billing layer is payer policy, and it is narrower than most write-ups admit. Medicare pays family psychotherapy codes only when the primary purpose is treatment of the identified patient, and bundling rules already count family members present as informants as part of ordinary psychotherapy, so a brief information-gathering call is generally not separately billable. 90887, the code for explaining findings to the family, exists, but its coverage varies by payer, so verify before relying on it. Some state Medicaid programs pay collateral contacts under their own codes and unit rules (South Dakota uses H0046 in 15-minute units, minimum 15 minutes), and New York's CFTSS guidance draws the clearest boundary: collateral contact means gathering treatment-relevant information, and care coordination does not count. One true legal mandate exists: Louisiana Act 737 of 2024 requires behavioral health providers to accept collateral information offered by a family member, caregiver, or friend and include it in the medical record unless the provider finds it not relevant to care. Nearly everything else, including the 24-or-48-hour signature timeframes many clinics enforce, is organizational or payer convention, not law. The note format is a convention; the accurate, attributed record is the requirement. When a caregiver contact becomes treatment of the family system, document it as a family therapy session instead. ## Common collateral contact note errors auditors flag No auditor publishes an error rate for collateral notes as a distinct document type; the enforcement record sits one level up, in family psychotherapy and psychotherapy documentation generally. A 2023 HHS Office of Inspector General audit estimated $580 million in improper Medicare psychotherapy payments, with 128 of 216 sampled enrollee days failing requirements for reasons like missing time documentation, treatment plans, and signatures, and OIG's work plan keeps family psychotherapy codes 90846 and 90847 on its list of services at heightened documentation risk. A 2025 insurer case study describes a licensed professional counselor disciplined after billing a spouse's participation as a collateral service across 20 sessions with no collateral agreement and no supporting documentation: 12 months of board probation and more than $14,000 in defense costs. The BastionGPT Clinical Advisory Board sees the same errors most often in parent and caregiver collateral note reviews: Unattributed statements. A caregiver's report written as clinical fact. Once "client is aggressive at home" appears without "mother reports," the chart asserts it, the client can read it, and every records release repeats it to downstream readers. Billing an informant call as family therapy. A brief information-gathering contact billed under 90846 or 90847. Bundling rules treat informant participation as part of ordinary psychotherapy, and 90847 billed for a session the client did not attend is a classic reviewer flag. No consent trail. The note never states who authorized caregiver involvement or what any release covers, so neither the listening direction nor the disclosing direction can be defended after the fact. Clinician conclusions inside the report. Appraisal mixed into the reported-information section, so no reader can reconstruct what was heard versus what was concluded, the distinction that matters most if the record reaches a custody dispute. Missing time and mode. No duration, no contact method. Unbilled contacts survive this; billed ones do not, because missing time support is the most commonly documented failure in psychotherapy audits. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on parent and caregiver collateral notes. Draft a complete collateral note from bullets, a dictation, or a quick recap after the call, with every statement attributed to its source. Pull the clinically relevant reports out of a long caregiver email or voicemail transcript and summarize them into an attributed information-reported section. Flag unattributed statements, clinician conclusions that drifted into the report, and source-identifying detail you may want to generalize, before you sign. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a parent or caregiver collateral note be? Most run 100 to 300 words and take 5 to 10 minutes by hand. The signal that matters is completeness of the frame, not word count: informant and relationship, consent status, an attributed report, your observation and appraisal, and the plan. A short note that nails attribution beats a long one that never says who reported what. Do I need a signed release before talking to a client's parent or caregiver? Not to listen. HIPAA permits receiving information from family members, and documenting what they told you is not a disclosure. A release governs the other direction: sharing the client's information with the caregiver beyond what is permitted for involvement in care. For a minor, the consenting parent or guardian generally already holds involvement rights; for an adult client, document their agreement to caregiver contact and the scope of what may be shared. Can I bill for a collateral contact? Usually not separately. Bundling rules count family members present as informants as part of ordinary psychotherapy, and 90846 requires a substantive family psychotherapy session tied to the client's treatment plan, which Medicare covers only when the primary purpose is treating the identified patient. 90887 covers explaining findings to the family, with coverage that varies by payer. Some Medicaid programs pay collateral contacts under state-specific codes such as H0046 in 15-minute units. Document every contact; bill only the ones that meet a code's actual criteria. Can my client see who gave me the information? Assume yes. Under the US information-blocking rules, most electronic progress notes are available to the patient by default, and a released record carries the informant's statements with it. Psychotherapy notes are excluded from that default, so truly sensitive analysis can live there. A 2024 American Psychiatric Association ethics opinion supports describing a source in general terms when the source insists on confidentiality and the material is sensitive, and warns against promising a family source confidentiality you cannot guarantee. Is a collateral contact note required by law? The note type is a convention; no US statute prescribes its format or fields. What law does govern: your general record-keeping duty, and HIPAA's rules on what you may receive and disclose. One statute goes further: Louisiana Act 737 of 2024 requires behavioral health providers to accept collateral information offered by family and caregivers and include it in the record unless it is not relevant to care. Signature timeframes like 24 or 48 hours are organizational or payer policy, not law. What is the difference between a collateral note and a family therapy note? Purpose. A collateral contact gathers information about the client from someone in their life; the informant is not receiving treatment. A family session treats the family system in the room and is documented and billed as psychotherapy; see the family and couples therapy note . When the informant is school staff, the school collateral note handles the education-records overlay, and caregiver interviews inside a structured evaluation belong in the developmental assessment . How long do I keep collateral notes? As long as the chart they live in; there is no collateral-specific retention rule. State law governs, and minors extend the clock: California expects 7 years after therapy ends, or 7 years past age 18 for minors, and Washington runs to age 22 or 8 years, whichever is longer. APA record-keeping guidelines suggest 7 years after the last service or 3 years after a minor reaches majority, whichever is later. Follow the longest period that applies to you. Can BastionGPT write a collateral contact note? Yes. Dictate a recap after the call or paste a caregiver's email, and it drafts the note with every statement attributed to its source, the consent line in place, and your observations kept separate from the report. It can also review a finished draft and flag unattributed statements before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Child & Adolescent Developmental Assessment Intake Note Family & Couples Therapy Note School & Teacher Collateral Note Indigenous & First Nations Wellness Assessment Addendum CANS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/school-collateral-note All note templates # School & Teacher Collateral Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A school and teacher collateral contact note is a therapist's chart record of a conversation with a teacher, school counselor, or other school staff member about a minor client. Child and adolescent clinicians write one after each school contact to capture classroom observations and connect them to the treatment plan. Most run 100 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Child and adolescent therapists, child psychologists, school-linked clinicians Audience Treating clinician, care team, supervisors, parents through record access, auditors where billed Typical length 100 to 300 words · 5 to 10 minutes by hand (clinical team estimate) Format family Narrative collateral note (compare: collateral contact note, service note) When it's used After each phone call, meeting, or email exchange with a teacher or other school staff about a minor client Standards context A convention, not a mandated form; the same information may sit under HIPAA in your chart and FERPA in the school's file On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a school and teacher collateral note? A school and teacher collateral contact note is the chart entry a treating clinician writes after communicating with a school employee (teacher, school counselor, principal, or classroom aide) about a minor client. It belongs to the broader collateral contact note family, and its structure descends from the same problem-oriented record tradition (Lawrence Weed's problem-oriented medical record, 1960s) that produced the SOAP note. No author or standards body defines the school version; it exists operationally, assembled from state Medicaid manuals, agency forms, and EHR conventions. You will also hear it called a collateral contact, contact note, service note, or H0046 note after the Medicaid billing code, and Massachusetts' standardized documentation project publishes a consultation and collateral contact progress note that lists teachers, principals, and guidance counselors among the collaterals. The fact that decides most questions about this note: which privacy law governs the information depends on who holds the record. Under FERPA, records directly related to a student and maintained by a school or a party acting for it are education records (20 U.S.C. 1232g; 34 C.F.R. 99.3), and HHS guidance on student health records (HIPAA FAQ 518) confirms that health information held in a FERPA education record is excluded from HIPAA's definition of protected health information. That is law, not convention: the same observation sits under FERPA in the teacher's file at school and under HIPAA and state law once you write it into your clinical chart. The note itself, by contrast, is a convention. No US statute, Canadian regulator, or Australian instrument mandates a school collateral note as a form, and no regulation requires it to be a separate document rather than a section within a progress note. The law governs the record once it exists and the privacy of what is inside it; the format is yours. ## Who uses school and teacher collateral notes and when Child and adolescent therapists, child psychologists, and clinicians in school-linked or school-based mental health programs write these notes, and they cluster at two moments: intake, when you are assembling a multi-setting picture of a new child client, and active treatment, when you need to know whether gains at home are showing up in the classroom. The volume is no accident. In a 2023 European Psychiatry analysis of England's national child mental health survey, teachers were the most commonly reported professional contact for children with emerging difficulties (39.8%, against 6.5% for specialist child mental health services), so the teacher usually holds the longest observational record of the child outside the home. Write a collateral note when the school contact is an input to your own assessment or treatment. When teacher-reported data is feeding a formal diagnostic product, you are documenting toward a developmental assessment or an ADHD evaluation report instead, and when the document lives on the school side of the line, it is a psychoeducational report , not a chart note. ## School and teacher collateral note structure: what goes in each section Contact header. Client identifiers and age, the date of contact, the method (phone, in person, video, email), and the duration in minutes. Method is not decoration: North Carolina's Medicaid policy, for one, requires each note to record the type of contact, with collateral among the named types. Pitfall: no minutes recorded. If the contact is ever billed under a time-based code such as H0046, South Dakota's 15-minute minimum makes an untimed note unbillable on its face. Who you spoke with. Name, role, and school, plus how you confirmed you were actually talking to that person (a call back through the school's main line, a school email domain). Pitfall: "spoke with the school." Acting on third-party information without verifying the source's identity is a recurring licensing-board theme, and a note that cannot name its source cannot defend the treatment decisions built on it. Authorization on file. The release date, who signed it (the parent or guardian with medical decision-making authority), the named parties, the direction it covers (receive, disclose, or both), and its expiration. Disclosing anything about the client to the school requires authorization; receiving information may not strictly require one, but documenting consent in both directions is the defensible pattern. Pitfall: assuming the intake release covers the school. A release naming the pediatrician does not authorize the teacher call; check the named party before dialing, not after. Purpose tied to a plan goal. One line connecting the contact to a specific treatment-plan goal. New York's CFTSS rules make the stakes concrete for billable collateral: the contact must occur per the treatment plan, and case consultation and brief information requests are excluded. Pitfall: "called to check in." A purpose that names no goal reads as an unplanned courtesy call, which is exactly the category payers refuse. Information gathered, attributed. What the teacher reported, attributed by name and kept observational: on-task minutes, work completion, peer behavior, what helps. Pitfall: absorbing the teacher's report into your own clinical voice. Unattributed third-party material is hard to handle at access time (some laws let you withhold confidential third-party information only if you can find it) and it contaminates your own later findings. Information you shared. An itemized list of what you disclosed, held to the minimum necessary for the purpose. Pitfall: recording only the inbound half of the call. If the release's scope is ever questioned, the disclosure list is your evidence of what actually crossed; a missing list reads as "unknown." Clinical use and next step. What the information changes: a plan update, a rating scale to integrate, a re-contact date. New York's guidance states collateral information "should never serve as a sole resource" for planning, so pair it with your own assessment. Pitfall: collateral that arrives and changes nothing. A chart of gathered-but-unused teacher reports invites the question of why the information was collected at all. Billing status and signature. State plainly whether the contact is billable (usually it is not; some state Medicaid programs pay under H0046) and close with signature, credentials, and date, plus supervisor co-signature for pre-licensed clinicians. Pitfall: routing an information call through 90846. That code is family psychotherapy in service of the identified patient, and repeated 90846 claims without documented clinical justification are a known review trigger. ### Blank template (copy and adapt) SCHOOL / TEACHER COLLATERAL CONTACT NOTE Client: [initials] Age: ______ Date of contact: __________ Method: [ ] phone [ ] in person [ ] video [ ] email Duration: ____ min School contact (name, role, school): ___________________________________ Identity verified by: __________________________________________________ Authorization: release signed ________ by __________________ (guardian) Covers: [ ] receiving info [ ] disclosing info [ ] both Expires: ____ Purpose (treatment plan goal): _________________________________________ INFORMATION GATHERED (attribute to the source): ________________________________________________________________________ ________________________________________________________________________ INFORMATION SHARED (minimum necessary): ________________________________________________________________________ CLINICAL USE / NEXT STEP (plan update, re-contact date): ________________________________________________________________________ Billing: [ ] non-billable collateral [ ] state Medicaid code: ________ Clinician signature/credentials: ____________________ Date signed: ____ Supervisor co-signature (if required): _________________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample school and teacher collateral note Scenario: a licensed professional counselor documents a scheduled phone call with the 5th-grade teacher of a 10-year-old client in treatment for ADHD, combined presentation, three weeks into the school year. All details are fictional. School / Teacher Collateral Contact Note. Client: T.M., 10 · Date of contact: 08/26/2026 · Method: Telephone, 12 minutes · School contact: D. Alvarez, 5th-grade teacher Authorization: Two-way release of information signed 08/14/2026 by T.'s mother (holds medical decision-making authority per intake), naming the school, in effect through 06/30/2027. Call placed through the school's published main number; front office transferred to Ms. Alvarez. Purpose: Baseline classroom data for treatment plan Goal 2 (increase on-task behavior and work completion at school), per the plan's monitoring schedule. Information gathered: Ms. Alvarez reported T. stays on task about 10 minutes before leaving his seat, turned in 3 of 5 assignments last week, and manages mornings better than afternoons. Peer interactions friendly; no aggression observed. A private verbal cue works better than a public redirect. She agreed to complete the teacher rating scale sent 08/20/2026 and return it by 09/04/2026. Information shared: Confirmed T. is in treatment and that his mother requested the contact. Described two strategies in use: a signaled movement break and front-of-room seating. No diagnosis, medication, or session content disclosed. Clinical use: Teacher report converges with parent report of stronger mornings; afternoon decline to be explored at the 08/31/2026 parent session. Rating-scale scores will be integrated at the September plan review. Re-contact planned for the week of 09/21/2026. Billing: Non-billable collateral contact; no claim submitted. Signed: R. Bennett, LPC · 08/26/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The consent trail is complete before anything else happens. Date, signer, authority, named party, direction, and expiration, plus how the teacher's identity was confirmed: the exact chain a board complaint tests. Every classroom observation is attributed. "Ms. Alvarez reported" keeps third-party report separate from clinical finding, which protects both later assessments and access decisions. The outbound half is itemized. What was shared is listed and minimal, so the disclosure can be checked line by line against the release's scope. Purpose and clinical use tie the call to Goal 2. The contact happened per the plan's monitoring schedule and produced a concrete next step, which separates planned collateral from an excluded courtesy check-in. Billing status is stated out loud. "Non-billable, no claim submitted" closes the question an auditor would otherwise have to ask. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Once written, the note is ordinary clinical-record content: part of the designated record set, releasable on request, and generally accessible to the parent or guardian who consented to treatment, subject to state mature-minor rules. Two boundary effects follow from the FERPA and HIPAA split. Anything you receive and rely on clinically becomes part of your record, so a teacher's emailed summary attached to the chart is now chart content, with the access and retention consequences that carries. And the school side is not locked down the way your side is: FERPA's sole-possession exception for a teacher's private memory aid is read narrowly and ends the moment the note is shared, including by emailing it to you, so expect detailed teacher observation notes to be accessible education records at the school and write what you send accordingly. The same two-direction logic, with a parent rather than a school on the other end, is covered on the collateral contact note page. Payment, where it exists, is payer policy rather than law. Medicare sets the tone: a call to gather information from a school is collateral contact, generally not separately billable, and 90846 is family psychotherapy conducted for the identified patient, not an information channel. The exceptions live in state Medicaid manuals. South Dakota's manual pays collateral contact under H0046 with a 15-minute minimum, tied to another covered service, excluded in inpatient settings, with records retained at least 6 years after the last claim; New York's CFTSS rules allow billable collateral only within an existing service definition and per the treatment plan, excluding case consultation and brief information requests; Kentucky covers collateral outpatient therapy for a client under 21 only when the plan of care includes it. The pattern across every program: the format is a convention; the plan linkage, the minutes, and the consent trail are the requirement. For retention, minors' rules run long past the last session (10 years past age 18 in Ontario; to age 25 in Victoria and New South Wales), so when more than one rule could apply, keep the record to the longest. ## Common school and teacher collateral note errors auditors flag The enforcement record around collateral work is mostly about documentation, not clinical judgment. In the FY2021 Medicare supplemental improper-payments data, psychiatry's improper payment rate was 19.4%, with insufficient documentation accounting for 87.5% of those errors, and insufficient documentation and missing signatures remain the top categories across the CERT program year after year (7.66% and $31.70 billion in FY2024). Where children's behavioral health is examined directly, the findings sharpen: a 2026 HHS Office of Inspector General audit of a Pennsylvania behavioral health Medicaid contractor found that not one of a 100-case sample of denied prior-authorization requests met all applicable procedural requirements. The BastionGPT Clinical Advisory Board sees the same errors most often in school and teacher collateral note reviews: An information call billed as 90846. That code covers family psychotherapy in service of the identified patient; a call to collect classroom observations is collateral contact, generally not separately billable, and repeated 90846 without documented justification is a recurring audit finding. Disclosure without a matching release. Anything shared with the school, even confirming treatment, needs guardian authorization naming the school; unauthorized disclosure to school personnel is a recurring licensing-board discipline theme. Third-party report in the clinician's voice. The teacher's observations written as the clinician's own findings, which corrupts later assessments and strips the note of the attribution some access laws depend on. No linkage to the treatment plan. "Called the teacher to check in," with no goal named and no clinical use recorded; rules like New York's CFTSS exclude brief information requests, and collateral information should never be the sole basis for planning. Method and minutes missing. No contact type where state policy requires one, and no duration where a time-based code needs support; a code with a 15-minute minimum cannot be billed from an untimed note. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on school and teacher collateral notes. Summarize the contact however it reached you (a teacher email thread, a dictated call recap, quick bullets) into a dated, attributed collateral note in your format. Rewrite what you plan to send the school in plain, minimum-necessary language: treatment-relevant strategies in wording a teacher can use, with diagnosis, medication, and session content stripped out. Check the note before you sign: authorization documented, teacher report attributed to its source, disclosure list complete, purpose tied to a plan goal, billing status stated. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I need a signed release before I talk to my client's teacher? Split it by direction. To disclose anything about the client to the school, even confirming that the client is in treatment, you need authorization from the parent or guardian who holds medical decision-making authority. To receive information from the school, a release is often not strictly required, but documenting guardian consent for both directions before the call is the defensible pattern, and it is the first thing a reviewer or board investigator asks for. Confirm the release actually names the school; an intake release naming other providers does not cover the teacher. Is my note about a teacher call governed by HIPAA or FERPA? Both laws can touch one conversation, and the split is set by who holds the record, which is law rather than convention. Your note in the clinical chart is governed by HIPAA and state health-privacy law. The teacher's records at school are FERPA education records, and HHS guidance (HIPAA FAQ 518) confirms that health information held in a FERPA education record is excluded from HIPAA's definition of protected health information. FERPA's sole-possession exception is narrower than school staff often assume: a teacher's private memory aid loses its exempt status once it is shared, including by emailing it to you. Can I bill for a school collateral contact? Usually no. Medicare's framing, which commercial payers echo, treats pure information-gathering with a school as collateral contact, not a separately billable service, and 90846 applies only to family psychotherapy conducted for the identified patient's treatment. The exceptions are payer policy at the state Medicaid level: some programs pay collateral contact under H0046 or a state code, with conditions such as South Dakota's 15-minute minimum tied to another covered service, or New York's requirement that the contact occur per the treatment plan. Verify the specific benefit before the call; if the payer excludes collateral, document the contact as non-billable and do not submit a claim. How long should a school collateral note be, and does it need its own form? Most run 100 to 300 words and take 5 to 10 minutes: header, authorization line, attributed teacher observations, what you shared, and the clinical use. No regulation requires the school collateral note to exist as a separate document; it can be a section within a progress note, provided the date, method, participants, consent trail, and duration stay recoverable. Is there a deadline for writing the note after the contact? No law sets one. Agencies commonly require completion within 24 to 72 hours, and that internal rule is worth following, but it is convention: the Joint Commission declines to set a number and leaves the timeframe to each organization (RC.01.01.01). The practical argument for same-day completion is accuracy; attributed detail about who said what degrades faster than session memory. How is this note different from a psychoeducational report or IEP paperwork? Custody and purpose. Your collateral note is a health record in your chart, an input to assessment and treatment. A psychoeducational report , a school psychologist's evaluation, and teacher observations gathered under IDEA are education records on the school side, governed by FERPA and state education rules; New York's special-education regulations , for example, build a classroom observation into the initial evaluation on IDEA's 60-day timeline. Your note crosses that line only if you contribute it to the school team, and the copy in the school file then becomes an education record. Can the client's parents see what the teacher told me? Generally yes. The note is part of the clinical record, and parents or guardians of minor clients typically hold the access right, subject to state mature-minor rules. A few jurisdictions build in a third-party shield: Ontario's PHIPA, for example, permits withholding information that could identify a third party who supplied it in confidence. The practical rule follows: attribute the teacher's report and keep it observational, because the family may well read it, and a note the parent can read without surprise defuses most disputes before they start. Do Canadian or Australian rules change anything for school contacts? The note stays a convention in both countries; retention and billing change. Minors' retention runs long: Ontario requires records kept 10 years past the client's 18th birthday, and Victoria and New South Wales require child health records kept until the client turns 25, not the generic 7 years often quoted. On billing, Australia's Better Access items generally require the patient present, and the family and carer participation items added in March 2023 cover family members and carers, not teachers, capped at two services per year and counted against the patient's own limit. Pure school information-gathering is generally not rebatable. How does BastionGPT help with school and teacher collateral notes? Paste the teacher's email thread or dictate the call, and BastionGPT drafts the attributed note in your format: who reported what, what you shared, and the plan linkage. It can also rewrite what you plan to send the school in plain, minimum-necessary language with clinical detail stripped out, and check the finished note for authorization, attribution, disclosure list, goal linkage, and billing status before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Child & Adolescent Developmental Assessment Psychoeducational Evaluation Report ADHD Evaluation Report Parent & Caregiver Collateral Note Referral Letter Return-to-School Letter Section 504 Clinician Statement IEP Psychoeducational Input Vanderbilt ADHD Scale Documentation PSC-17 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/indigenous-wellness-assessment All note templates # Indigenous & First Nations Wellness Assessment Addendum: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An Indigenous wellness assessment addendum is a culturally grounded section added to a standard mental health intake for First Nations, Inuit, Métis, American Indian/Alaska Native, or Aboriginal and Torres Strait Islander clients. It documents strengths, family and community connections, language and culture, spiritual factors, and community supports. No payer or regulator mandates the format; it reflects wellness frameworks such as the FNMWC. Most run 300 to 600 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Counsellors, psychologists, and social workers in Canadian and US Indigenous-facing services; Indigenous wellness teams Audience Treating clinician, care team, community program partners, payers (NIHB in Canada), auditors Typical length 300 to 600 words · 20 to 40 minutes by hand (clinical team estimate) Format family Culturally grounded assessment addendum (compare: cultural formulation interview, biopsychosocial assessment) When it's used At intake with First Nations, Inuit, Métis, and other Indigenous clients, alongside the standard assessment; revisited as trust builds Standards context A clinical convention aligned with culture-grounded wellness frameworks (FNMWC, SEWB); no payer or regulator mandates the format On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an Indigenous & First Nations wellness assessment addendum? An Indigenous and First Nations wellness assessment addendum is a culturally grounded section that rides alongside a standard mental health assessment when you serve First Nations, Inuit, Métis, American Indian and Alaska Native, or Aboriginal and Torres Strait Islander clients. It documents what standard intakes tend to flatten or miss: strengths, family and community connections, language and culture, spiritual factors at the level the client chooses to share, and community supports. The name is an umbrella, not a credentialed form. Clinicians assemble this documentation from a family of distinct sources: the First Nations Mental Wellness Continuum (FNMWC) Framework, published in 2015 by Health Canada's First Nations and Inuit Health Branch, the Assembly of First Nations, and Indigenous mental health leaders, organized around the wellness outcomes of Hope, Belonging, Meaning, and Purpose; the Native Wellness Assessment, a trademarked strengths-based measure from the Thunderbird Partnership Foundation; medicine-wheel and wholistic assessment formats; two-eyed seeing (Etuaptmumk), brought to the academic community in 2004 by Mi'kmaw Elders Albert and Murdena Marshall with Dr. Cheryl Bartlett of Cape Breton University; the social and emotional wellbeing (SEWB) model in Australia; and, in British Columbia, the First Nations Health Authority's cultural safety and humility standards, which operate at the organizational level. You will also hear "wholistic assessment," "medicine wheel assessment," and "SEWB assessment." The clarification that saves the most confusion: no regulator or payer in the United States, Canada, or Australia defines, certifies, or requires an Indigenous wellness assessment addendum, and the FNMWC Framework specifies no clinical chart fields. Structuring a note around Hope, Belonging, Meaning, and Purpose is CONVENTION: clinical best practice, not a mandate. The hard requirements in this space are generic. PAYER POLICY sets NIHB's assessment, treatment-plan, and session-record rules in Canada, and records LAW (provincial health-information statutes, HIPAA, state retention rules) governs the chart itself. That is why this document is an addendum: it supplements the intake note or biopsychosocial assessment that carries the payer-required elements, and it never replaces them. ## Who uses Indigenous wellness assessment addendums and when Counsellors, psychologists, and social workers serving First Nations and Inuit clients through Canada's NIHB Mental Health Counselling benefit use it most, alongside community-based and urban Indigenous services, IHS and Tribal behavioral health programs in the United States, and Australian clinicians working within social and emotional wellbeing frameworks. Most complete it at or near intake, then revisit it as trust builds, because clients often share more about community, language, and spirituality in session four than in session one. Choosing between neighbors is a matter of purpose: use the Cultural Formulation Interview when the goal is culturally valid diagnosis and the client's explanatory model, use a wellness addendum informed by FNMWC, medicine-wheel, or SEWB thinking when the goal is strengths-based engagement and culturally grounded treatment planning, and keep the biopsychosocial assessment or standard intake as the billable backbone either way, because that is the record payers audit. ## Indigenous & First Nations wellness assessment addendum structure: what goes in each section No regulator mandates a section list, so this structure follows the wellness domains the FNMWC and SEWB frameworks share, ordered so strengths lead. Every section records what the client chooses to share, in their words wherever possible. Header and parent-assessment linkage. Client identifiers, date, clinician, and the standard assessment this addendum accompanies. Pitfall: an addendum that floats free of the chart; NIHB verification and payer audits read the parent assessment and treatment plan first, so the addendum must name the record that carries those elements. Identity and community connection, client-led. How the client describes their identity, Nation or community affiliation, and their relationship to home community and territory, recorded as offered. Pitfall: probing for status-card, membership, or ancestry detail the client did not offer; identity description belongs to the client, and the addendum records what they choose to share, not what a form demands. Strengths and cultural resources. What has kept the client well: cultural practices, skills, roles, relationships, humor, previous help that worked. Pitfall: writing culture into the record only when it appears as a barrier; a deficit-framed addendum inverts the document's whole purpose. Family, kinship, and community context. Kinship and caregiving roles as the client names them, household composition, community involvement and belonging. Pitfall: forcing kinship into a nuclear-family genogram; aunties, grandparents, and chosen kin often carry primary roles, and the record should use the client's terms. Language and culture. Languages spoken, understood, or being reclaimed; preferences for service; cultural practices and learning the client wants connected to care. Pitfall: reducing language to an access checkbox ("English fluent, no interpreter needed"); language connection is itself a wellness domain, and reclamation efforts belong in the strengths picture. Spiritual and land-based factors, at the client's level of sharing. Client-defined spirituality and connection to land, water, and seasons as they relate to wellness. Pitfall: recording ceremony contents or teachings; the chart is releasable, so document that a practice supports the client's wellness, never what happens inside it. Community supports and services. Elders, Knowledge Keepers, cultural programs, and community organizations involved or wanted, and how each connects to care. Pitfall: entering an Elder as a co-author or co-signer; Elders are not NIHB-enrolled providers, so their involvement is documented as collateral or delivered through community-managed programs. Client wellness goals and agreed adaptations. Goals in the client's own terms, mapped to the treatment plan in the parent record, plus the service adaptations you agreed on: pacing, who joins sessions, scheduling around time in community. Pitfall: translating the client's goals into clinician language at the point of capture; keep their words in the addendum and do the clinical mapping in the treatment plan. ### Blank template (copy and adapt) INDIGENOUS / FIRST NATIONS WELLNESS ASSESSMENT ADDENDUM (accompanies the standard intake or biopsychosocial assessment) Client: __________ Date: __________ Clinician: ______________________ Parent assessment (type and date): _____________________________________ IDENTITY AND COMMUNITY CONNECTION (client's own words) Self-description and affiliation (as shared): __________________________ Connection to home community, territory, land: _________________________ STRENGTHS AND CULTURAL RESOURCES What keeps the client well: ____________________________________________ Cultural practices, skills, roles: _____________________________________ FAMILY, KINSHIP, AND COMMUNITY CONTEXT Kinship and caregiving roles (client's terms): _________________________ Community involvement and belonging: ___________________________________ LANGUAGE AND CULTURE Languages spoken / understood / being reclaimed: _______________________ Service preferences (language, family present, other): _________________ SPIRITUAL AND LAND-BASED FACTORS (at the client's level of sharing) Practices and connections supporting wellness: _________________________ COMMUNITY SUPPORTS AND SERVICES Elders, Knowledge Keepers, programs involved (collateral): _____________ Referrals to cultural or land-based supports: __________________________ CLIENT WELLNESS GOALS (client's own terms) ________________________________________________________________________ Agreed service adaptations: ____________________________________________ Mapped to treatment plan goal(s) in parent record: _____________________ Clinician signature / credentials / date: ______________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample Indigenous & First Nations wellness assessment addendum Scenario: adult First Nations client in a regional Ontario city, counselling funded through the NIHB Mental Health Counselling benefit, addendum completed alongside a standard intake assessment. All details are fictional. Indigenous & First Nations Wellness Assessment Addendum. Client: D.K., 41 · Date: 08/11/2026 · Clinician: M. Aubin, RSW · Parent assessment: Intake assessment, 08/11/2026 Identity and community connection (client's words): D.K. describes herself as Anishinaabe and a member of a First Nations community in northwestern Ontario. She has lived in the city since age 19; her mother and younger brother remain in the home community, and she visits each summer, "when I can get there." She prefers that the chart describe her community by region rather than by name; preference honoured. Strengths and cultural resources: Steady employment as a dental assistant for 15 years. Beadwork learned from her grandmother, which she describes as "where my mind gets quiet"; she sells finished pieces at a winter market. Close cousin network in the city. Warm humour; a previous counselling episode in 2021 that she found helpful. Family, kinship, and community context: Raised by her mother and grandmother. Auntie to two nieces in the city; provides after-school care two days a week and names this role as central. Attends a Friendship Centre women's drop-in most months. Identifies late winter, when travel home is hardest, as her recurring low period. Language and culture: Understands conversational Anishinaabemowin and speaks some; joined a weekly online language class in March 2026, "the best hour of my week." Prefers counselling in English. Welcomes cultural connection as an open topic in sessions. Spiritual and land-based factors (as shared): D.K. chooses to share that smudging and time on the water in summer support her wellness. No further detail sought or recorded, at her preference. Community supports and services: Friendship Centre women's drop-in, ongoing. Interested in connecting with an Elder through the centre's cultural support program; contact information provided with her consent. Any Elder involvement will be documented as collateral input; no co-signature. Client wellness goals (her terms): 1. "Feel at home in myself in the city, not just up north." 2. Stay with the language class through the year. 3. Get home twice in the next year, summer and fall. Mapped to treatment plan Goals 1 and 2 in the intake assessment of 08/11/2026. Agreed service adaptations: Biweekly 60-minute sessions with an unhurried opening. Option for her cousin to join one future session as support. Scheduling flexibility around summer travel home. Monthly check-in on cultural supports and the Elder referral. Signed: M. Aubin, RSW, 08/11/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Identity is client-led and minimal. The record holds what D.K. chose to share, including her preference that the chart describe her community by region, which is the core discipline of culturally safe documentation. Strengths lead. Employment, beadwork, kinship roles, and language learning come before any stressor, so the addendum supports wellness planning instead of building a deficit inventory. Spiritual content stays at practice level. The note records that smudging and time on the water support wellness, and explicitly stops there, keeping ceremony detail out of a releasable record. Elder involvement is framed as collateral. The referral is documented and the note states there will be no co-signature, which matches how NIHB actually treats cultural practitioners. The addendum stays tethered to the billable record. It names its parent intake, and the client's goals map to numbered treatment plan goals, so the payer-facing chain stays intact where auditors look. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The addendum is part of the standard clinical record. It sits in the designated record set, the client can request and read it, it travels with releases of the chart, and it is governed by the same privacy law and college standards as any other note. No Indigenous-specific records rule exists in any of the three countries. Retention runs on general LAW: Ontario's CRPO clinical-records standard expects at least 10 years from the last interaction, or 10 years past the client's 18th birthday, whichever is later; British Columbia expects 16 years; US patient charts follow state law plus HIPAA's narrower 6-year documentation rule; Australian records follow state and territory law. Write every line knowing the client will read it: their words, their chosen level of sharing, and no ceremony contents or sacred knowledge, ever. The payer layer is generic, and that is the point. Canada's NIHB Mental Health Counselling Benefits Guide (effective December 4, 2024) is PAYER POLICY: it requires a client assessment, a treatment plan kept on the client's file, and complete session records with date, location, and start and end times, with prior approval needed beyond the first 2 hours, claims within one year, and a standard benefit of up to 22 hours per calendar year. It mandates no cultural documentation format; its only cultural language is that coverage "supports culturally competent mental health counselling." Keep those payer elements complete in the parent assessment and treatment plan, and let the addendum do the clinical work. Traditional healing and Elder services in Canada are funded through separate community agreements, not NIHB fee-for-service billing. In Australia, the MBS item 715 health assessment is a GP item with its own documented management-plan and record requirements, and its descriptor is scheduled to change on 1 March 2026; psychologists and social workers document SEWB content within their own notes instead. Many NIHB clients in remote communities are seen virtually, so pair this addendum with the consent and licensure practices in the Canadian telepsychology note . ## Common Indigenous wellness assessment addendum errors auditors flag The audit record in this space is administrative, not cultural. NIHB's Provider Claim Verification Program recovers payments for ineligible services, wrong-region billing, and services delivered beyond an approval; the Benefits Guide states that where a provider has billed the program inappropriately, "claim payments will be recovered." None of its documented recovery triggers involve cultural content. Australia reads the same way: compliance activity around the item 715 health assessment turns on whether the required elements and personal GP attendance occurred, and published research (Schütze and colleagues, Australian Family Physician, 2016) traces low uptake to under-identification of Indigenous status and billing-knowledge gaps, not to the cultural framing of notes; national uptake rose from about 11% (2010 to 2011) to 29% (2016 to 2017). The BastionGPT Clinical Advisory Board sees the same errors most often in Indigenous wellness assessment addendum reviews: Treating the addendum as a mandated form. No regulator or payer in the US, Canada, or Australia defines or certifies one, and the FNMWC Framework specifies zero chart fields. Presenting the format to an auditor as a legal requirement raises a bar no rule actually sets. Letting the addendum absorb the payer-required record. The assessment, the treatment plan on file, and session records with date, location, and start and end times are what NIHB verification actually reads; a well-written addendum cannot rescue a chart missing them. Recording ceremony contents or sacred knowledge. The chart is releasable and discoverable. Document that a practice supports the client's wellness and stop; the contents of ceremony and teachings do not belong in a clinical record. Entering an Elder or cultural practitioner as author or co-signer. Elders are not NIHB-enrolled provider categories, and under MBS mental health items they are not billable authors; their input is collateral, or it is delivered through community-managed programs. Citing the Native Wellness Assessment as a diagnostic or billing instrument. The NWA is a trademarked, strengths-based outcomes measure, validated on 177 clients across 12 treatment centres in addictions and wellness programs (Fiedeldey-Van Dijk and colleagues, 2017). Administer it at entry, midpoint, and exit with attribution, unaltered, and keep its scores out of diagnostic claims. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on Indigenous wellness assessment addendums. Draft the addendum from a few bullets, a dictation, or a session transcript, holding the client's identity description and goals as direct words rather than paraphrase. Rewrite the completed addendum in plain language for a client copy or a community program referral, free of clinical jargon. Check a draft before you sign: deficit-framed language, clinician wording where the client's terms belong, ceremony detail that should not sit in a releasable record, and a missing link to the parent assessment and treatment plan. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there an official Indigenous wellness assessment form I am required to use? No. No regulator or payer in the United States, Canada, or Australia defines, certifies, or requires an Indigenous wellness assessment addendum. NIHB requires a client assessment, a treatment plan on file, and complete session records, with no culturally specific format attached. The culture-grounded structure is a clinical convention: use your usual clinically sound assessment and document cultural context within or alongside it, at the level the client chooses to share. What is the First Nations Mental Wellness Continuum Framework? The FNMWC Framework was published in 2015 by Health Canada's First Nations and Inuit Health Branch, the Assembly of First Nations, and Indigenous mental health leaders. It organizes mental wellness around four outcomes, Hope, Belonging, Meaning, and Purpose, and the balance of mental, physical, spiritual, and emotional wellness. It is a policy and systems framework that specifies no clinical chart fields, so a note structured around its outcomes reflects convention, not a Framework requirement. What is the Native Wellness Assessment, and can I use it? The Native Wellness Assessment, released in 2015 by the Thunderbird Partnership Foundation , is the only First Nations-specific wellness measure of its type in Canada with published psychometric validation: 177 clients across 12 treatment centres, with quadrant-level internal consistency between 0.91 and 0.97 (Fiedeldey-Van Dijk and colleagues, 2017). It pairs a client Self-Report Form with an Observer Rating Form, covers 66 cultural statements across 4 quadrants and 13 subthemes, takes about 15 minutes, and is administered at entry, midpoint, and exit. It is a strengths-based outcomes measure validated in addictions and wellness programs, not a diagnostic tool or a payer form. Obtain it from Thunderbird with attribution and do not alter it. How is this addendum different from the Cultural Formulation Interview? The Cultural Formulation Interview is a 16-item semi-structured interview from DSM-5 that elicits any client's explanatory model for diagnosis and engagement. It is pan-cultural rather than Indigenous-specific, and it produces a cultural formulation, not a wellness picture. Use the CFI when the goal is culturally valid diagnosis; use a wellness addendum when the goal is strengths-based, culturally grounded treatment planning and outcome tracking. Does NIHB require prior approval for assessment sessions? The first 2 hours of NIHB counselling, typically assessment and treatment planning, generally do not require prior approval; services beyond that do. The standard benefit is up to 22 hours per calendar year, claims must be received within one year of the date of service, and providers must be independently licensed with a provincial or territorial regulatory body. Keep the treatment plan on file and record date, location, and start and end times for every session. Can an Elder or cultural practitioner co-sign or bill the note? Generally no. Elders and cultural practitioners are not NIHB-enrolled provider categories, and under MBS mental health items they are not billable authors either. Document their involvement as collateral input within your own note, or through community-managed programs, which is how traditional healing services are funded in Canada. The clinical record stays under the licensed clinician's signature. In Australia, can a psychologist claim the item 715 health assessment? No. Item 715 is a GP health-assessment item: practice nurses, Aboriginal Health Workers, and Aboriginal and Torres Strait Islander Health Practitioners may assist, but the GP personally delivers the assessment and documents the management plan. Psychologists and social workers document social and emotional wellbeing content within their own notes and meet the culturally safe practice expectations in Ahpra's shared Code of Conduct . How long do I keep an Indigenous wellness assessment addendum? As long as the record it belongs to; there is no Indigenous-specific retention rule. Canadian provinces commonly expect at least 10 years from the last interaction, or 10 years past the client's 18th birthday, whichever is later, and British Columbia expects 16 years. In the United States, state law governs patient charts, with HIPAA adding a narrower 6-year rule for documentation such as policies. Australian retention follows state and territory health-records law. Can BastionGPT draft an Indigenous wellness assessment addendum? Yes. Give it a few bullets, a dictation, or a transcript and it drafts the addendum with strengths leading, the client's own words held as quotes, and spiritual content kept at the level the client shared. It can also rewrite the finished addendum in plain language for a client copy or a community referral. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Cultural Formulation Interview (CFI) Intake Biopsychosocial Assessment Intake Note Parent & Caregiver Collateral Note Canada Telepsychology & Virtual Care Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psychiatric-consultation-note All note templates # Psychiatric Consultation Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A psychiatric consultation note records a psychiatrist's or psychiatric nurse practitioner's opinion on another provider's patient: the consult question, a focused history, mental status findings, an impression that answers the question, and numbered recommendations back to the referrer. Consultants write one after each referred evaluation in outpatient, primary care, and specialty settings. Most run 250 to 800 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Consulting psychiatrists, psychiatric nurse practitioners, and psychologists answering a referral Audience The referring provider first, then the shared care team, payers, and auditors Typical length 250 to 800 words · 20 to 40 minutes by hand (clinical team estimate) Format family Consultation report family (compare: consultation-liaison note, consultation report, referral letter) When it's used When another provider requests a psychiatric opinion on their patient; common in specialty and inpatient settings Standards context Structure is convention; the documented request and the report back to the referrer are payer conditions, and law in Ontario On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a psychiatric consultation note? A psychiatric consultation note is the written record of a psychiatrist's or psychiatric nurse practitioner's opinion on another provider's patient: who asked, what they asked, what you found, and what you recommend. It belongs to the consultation-report family that runs across medicine, with the psychiatric layer (mental status examination, risk assessment, a capacity comment when asked) added by convention. The US billing lineage runs from Medicare Carriers Manual section 15506 into the Medicare Claims Processing Manual, and the familiar three Rs mnemonic (Request, Render, Reply) is a coding-community distillation of that guidance, not verbatim CMS text. The inpatient side of this work is owned by consultation-liaison psychiatry, the subspecialty renamed from Psychosomatic Medicine effective January 1, 2018. You will also hear the document called a psych consult, an initial psychiatric consultation, a psychiatric opinion, or, in Australia, a 291 assessment after the MBS item number. Three documents get conflated here, and the boundary is worth drawing. The psychiatric consultation note is the outpatient or specialty-clinic record that answers a referring provider's question and travels back to them as a report. The consultation-liaison note is the hospital bedside version, written into the shared inpatient chart by a C-L service. The consultation report is the letter a psychologist sends back after a testing referral. What makes the consultation note itself is the consult-question-first structure: the question opens the note and the impression answers it. No US, Canadian, or Australian statute prescribes the headings. What law and payers regulate sits around the note: hospitals must file the results of every consultative evaluation in the medical record, dated and authenticated ( 42 CFR 482.24 ), Ontario law requires the record to carry the referrer's name and address, and Medicare has not paid consultation codes since January 1, 2010, when CMS Transmittal R1875CP made them "no longer recognized for Medicare Part B payment." Most template sites still miss all three points. ## Who uses psychiatric consultation notes and when Outpatient psychiatrists and psychiatric NPs taking referrals from primary care write these constantly, and the same structure serves telepsychiatry consult services, integrated-care practices, geriatric and addiction specialists answering a colleague's question, and psychologists asked for a focused opinion. Write one when another provider asked a question and expects an answer they can act on. If you are starting your own comprehensive workup for a patient entering your care, that is a psychiatric diagnostic evaluation , not a consult. Once you assume ongoing prescribing, the record shifts to a medication management note . If your consulting happens through a registry and caseload review with a care manager, without seeing the patient, that is collaborative care note territory. And the document that opens this exchange from the other side is the referral letter ; a good one hands you the consult question ready-made. ## Psychiatric consultation note structure: what goes in each section Header and referral source. Patient identifiers, consult date, the referring provider with contact details, and how and when the request arrived. In Ontario this is law: the Medicine Act General Regulation requires the consultation record to include the name and address of the referring professional. Pitfall: a consult with no visible request. Reviewers verify the order or referral before they read a word of clinical content, and without it commercial consult codes, OHIP consultations, and MBS 291 do not pay. Consult question. One sentence, in the referrer's terms. If the referral was vague, clarify it with the referrer and record the sharpened version. Pitfall: copying "please evaluate" forward as the question. The impression must answer this exact sentence, so a mushy question guarantees a mushy answer. Focused history. History of the presenting problem plus only the psychiatric, medical, medication-trial, substance, and social history that bears on the question. For medication questions, list each prior trial with agent, dose, duration, response, and adherence evidence. Pitfall: reproducing a full intake. A consult that reads like a comprehensive diagnostic evaluation buries the answer and slows the referrer down. Mental status examination. The findings, with weight on the domains the question implicates: cognition for a capacity question, mood and neurovegetative signs for a depression question. Pitfall: an all-normal templated MSE that contradicts the history. A PHQ-9 of 16 next to "euthymic mood, full affect" reads as cloned text to any reviewer. Risk assessment. Suicide, self-harm, and violence risk, stated either way, dated to the encounter. Pitfall: silence. The referrer treats your note as the specialist's word on risk; leaving it out hands the ambiguity back to a clinician less equipped to resolve it. Impression. Diagnostic impression, with the first sentence answering the consult question directly. Add the capacity comment here when that is what was asked. Pitfall: listing DSM labels without a sentence that speaks to what the referrer asked. That is the single most common structural failure in consult reviews. Recommendations and communication back. Numbered recommendations a non-psychiatrist can execute: agent, dose, titration, monitoring, the re-referral trigger, and who does what. Close with the disposition line: report sent to the referrer (date and route) and whether you assumed care or returned it. Pitfall: no transmission line. In Ontario and Australia the written report is a payment condition; everywhere it is what turns an opinion into a consultation. ### Blank template (copy and adapt) PSYCHIATRIC CONSULTATION NOTE Patient: [initials] DOB: __________ Date of consult: __________ Referring provider (name, contact): _________________________________ Request received (date / route): ____________________________________ CONSULT QUESTION (one sentence, in the referrer's terms) _____________________________________________________________________ FOCUSED HISTORY (only what bears on the question) HPI: ________________________________________________________________ Psychiatric history: ________________________________________________ Medication trials (agent, dose, duration, response, adherence): _____________________________________________________________________ Medical history / labs: _____________________________________________ Substance use: ________________ Social context: ____________________ MENTAL STATUS EXAMINATION Appearance / behavior: ________________ Speech: ____________________ Mood / affect: ________________________ Thought process / content: _ Perception: ___________________________ Cognition: _________________ Insight / judgment: _________________________________________________ RISK ASSESSMENT (suicide, self-harm, violence; state it either way) _____________________________________________________________________ IMPRESSION (first sentence answers the consult question) _____________________________________________________________________ Diagnoses: __________________________________________________________ RECOMMENDATIONS (numbered, executable by the referrer) 1. __________________________________________________________________ 2. __________________________________________________________________ 3. __________________________________________________________________ Re-referral trigger: ________________________________________________ COMMUNICATION AND DISPOSITION Report to referrer (date / route): __________________________________ Role: [ ] one-time opinion [ ] follow-up planned [ ] care assumed Consultant signature / credentials / date: __________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample psychiatric consultation note Scenario: outpatient psychiatric consultation for a 42-year-old man referred by his family physician after two unsuccessful SSRI trials for major depressive disorder. The consultant answers the medication question and returns care with a plan. All details are fictional. Psychiatric Consultation Note. Patient: R.D., 42 · Date of consult: 08/11/2026 · Referring provider: A. Okafor, MD, family medicine · Request received: 08/04/2026, written referral · Consultant: S. Marsh, MD, psychiatry Consult question: "Is a further medication change appropriate for persistent depressive symptoms after two SSRI trials, and can treatment continue in primary care?" Focused history: 42-year-old man with major depressive disorder, recurrent; current episode 10 months. Sertraline titrated to 200 mg daily, 12 weeks at full dose: partial response, then plateau. Escitalopram 20 mg daily, 10 weeks: minimal change. Both trials adequate in dose and duration; adherence supported by refill history. PHQ-9 at referral 16. Reports low energy, anhedonia, early waking; appetite intact. No manic or hypomanic history; MDQ negative today. No psychotic symptoms. Alcohol 1 to 2 drinks weekly; no other substances. Medical: hypertension on lisinopril; TSH and CBC within normal limits 07/2026. Working full time; married; no psychiatric hospitalizations. Mental status examination: Alert, cooperative, casually dressed. Speech normal in rate and tone. Mood "worn down"; affect constricted but reactive. Thought process linear; no delusions; no perceptual disturbance. Cognition grossly intact. Insight and judgment good. Risk: Denies suicidal ideation, intent, or plan; no history of attempts; no concerns identified on direct inquiry. Protective factors include family support and engagement in care. Routine monitoring appropriate. Impression: Answering the consult question: a switch in antidepressant class is appropriate, and care can remain in primary care with the plan below. Major depressive disorder, recurrent, moderate (F33.1), with inadequate response to two adequate SSRI trials. No bipolarity, psychosis, or untreated medical contributor identified. Recommendations: 1. Cross-taper: reduce escitalopram over 1 to 2 weeks while starting bupropion XL 150 mg each morning; increase to 300 mg after 2 weeks if tolerated. No seizure or eating disorder history; monitor blood pressure given hypertension. 2. Repeat the PHQ-9 at 4 and 8 weeks. If the score remains above 10 at 8 weeks on the full dose, re-refer for augmentation options. 3. Refer for structured psychotherapy (CBT or behavioral activation) alongside the medication change. 4. Counsel on regular sleep schedule and alcohol moderation; review at routine hypertension visits. 5. Advise return or contact sooner for mood worsening, emerging suicidal thoughts, agitation, or marked insomnia after the switch. Communication and disposition: Full report faxed and sent by secure message to Dr. Okafor on 08/12/2026; copy filed in this record. One-time consultation; I have not assumed care. Available for re-consultation if the response stalls after an adequate bupropion trial. S. Marsh, MD, psychiatry. Signed 08/11/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The question is quoted at the top and answered in the first sentence of the impression. That is the structural test reviewers, and referrers, apply to a consult. The request is documented. Who referred, when, and how: the element that makes commercial consult codes, OHIP consultations, and MBS 291 payable, and that Ontario law requires the record to carry. The history is proportionate. Two medication trials with agent, dose, duration, response, and adherence evidence, and nothing that does not serve the question. Risk is addressed even though the presentation is routine, so the specialist's word on risk is on the record instead of implied. Numbered recommendations name the agent, dose, titration, monitoring, and re-referral trigger, so the family physician can act without a callback, and the dated transmission line closes the loop. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A consultation note lives in two charts. Your copy sits in your designated record set, and the report you send becomes part of the referrer's record; in Ontario, CPSO expects both the referring and the consulting physician to keep the referral request and the consultation report. Patients can request it, and it travels farther than most psychiatric documentation because it was written to be sent, so write it for the referrer, the patient, and any later reader at once. In US hospitals the filing duty is law: 42 CFR 482.24 requires the results of all consultative evaluations in the medical record, each entry complete, dated, and authenticated by an author identified by name and discipline, with records retained at least 5 years. An inpatient consult also feeds the discharge summary that carries your recommendations out of the hospital, and when staff then help the patient act on those recommendations over the following weeks, that coordination work belongs in a case management note , not in amendments to the consult. The billing layer is where jurisdictions split, and the split is worth labeling. In the US, Medicare payment for consultation codes ended January 1, 2010 (payer policy); you bill the setting-appropriate E/M code, and the consult structure survives as convention because it still answers the clinical need. Some commercial payers still pay 99242 through 99245 and 99252 through 99255 and expect the documented request, the rendered opinion, and the written reply; others, including Anthem since October 1, 2021, have dropped them, so the payer contract decides the code. In Ontario, the referrer's identity in the record is law, and the written report is payer policy: OHIP consultation codes such as A195 and A190 require a written referral and a written report, with a major consultation limited to one per patient per physician in a 12-month period. In Australia, MBS items 291 and 293 tie payment to the referral and "a detailed written report to the referrer," per the MBS explanatory notes. Across all three countries, the format is a convention; the documented request and the reply are the requirements. ## Common psychiatric consultation note errors auditors flag No published audit isolates psychiatric consultation notes; the enforcement record around psychiatric services is specific enough to learn from. HHS-OIG report A-09-21-03021 estimated $580 million in improper Medicare psychotherapy payments in a single pandemic-year audit, and a March 2022 OIG review of one high-volume psychiatrist assessed about $1.1 million in overpayments, citing services by social workers and nurses billed incident-to under the psychiatrist, missing time documentation, and records created months after the encounter. In Australia, the Professional Services Review reported that most inappropriate-practice findings in 2021-22 involved "inadequate clinical notes" and failure to meet MBS item requirements. The BastionGPT Clinical Advisory Board sees the same errors most often in psychiatric consultation note reviews: No documented request. The consult happened, but no order, referral letter, or request date appears in the record. Reviewers verify the request before reading the clinical content, and without it commercial consult codes, OHIP consultations, and MBS 291 do not pay. An impression that never answers the question. Diagnoses are listed, but the sentence the referrer asked for is missing. The reply is what separates a consultation from a duplicate intake. No evidence the report reached the referrer. The note is signed, yet nothing shows a reply was sent, to whom, or when. In Ontario and Australia the written report is a condition of payment; everywhere it is what closes the loop. Consultation codes billed to payers that no longer recognize them. Medicare has paid none since January 1, 2010, and commercial recognition keeps shrinking; the claim needs the setting-appropriate E/M code with documentation to match. Supervision and attestation gaps. A resident's consult without the attending's documented presence and participation, or staff services billed incident-to without meeting the 42 CFR 410.26 conditions; the March 2022 OIG psychiatrist audit was built on exactly these. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric consultation notes. Draft a complete consultation note from the referral letter, chart excerpts, and your dictated findings, with the consult question stated up front and answered in the impression. Summarize a long chart into the focused history a consult needs, then turn the finished note into a plain-language explanation for the patient and a concise reply letter for the referrer. Check the note before you sign: request documented, question answered, risk addressed, recommendations numbered with doses and monitoring, transmission line dated. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a psychiatric consultation note be? Most run 250 to 800 words and take 20 to 40 minutes by hand. The length tracks the question: an inpatient reply in a shared chart sits at the short end, while an outpatient report sent to the referrer with a full medication-trial history sits at the long end. If the note is pushing 1,200 words, it has usually drifted into diagnostic-evaluation territory. Is a specific psychiatric consultation note format required by law? No. No US, Canadian, or Australian statute prescribes a heading set, and there is no Joint Commission deadline specific to consult notes (the 24-hour rule clinicians cite is the history-and-physical rule). What is binding sits around the structure: US hospitals must include the results of all consultative evaluations in the medical record, dated and authenticated ( 42 CFR 482.24 ), and Ontario law requires the record to include the referring professional's name and address. Section order, headings, and style are convention. Can I still bill a consultation code for a psychiatric consult? Not to Medicare. CMS ended recognition of consultation codes on January 1, 2010; bill the setting-appropriate E/M code instead. The AMA deleted 99241 and 99251 in 2023, but 99242 through 99245 and 99252 through 99255 remain active CPT codes that some commercial payers accept when the request, the opinion, and the written reply are all documented (the three Rs). Others have dropped them; Anthem stopped effective October 1, 2021. The payer contract, not the note type, decides the code. What is the difference between a psychiatric consultation note, a consultation-liaison note, and a consultation report? All three answer another professional's request, which is why the names blur. The psychiatric consultation note is the outpatient or specialty-clinic record that answers a referring provider's question and goes back to them as a report. The consultation-liaison note is the hospital bedside version: a C-L service documents in the shared inpatient chart and bills hospital E/M codes. The consultation report is the letter a psychologist writes back after a testing referral. The shared skeleton is question, findings, answer; the setting and the reader change. How is a consultation note different from a psychiatric diagnostic evaluation? The psychiatric diagnostic evaluation (billed 90791, or 90792 with medical services) is your own comprehensive workup for a patient entering, or possibly entering, your care: no outside question, no report-back duty, and typically once per episode of illness. The consultation note is narrower: everything in it is filtered through the referrer's question, and it ends by handing the answer back. The practical test is who asked, and who acts on the answer. If the answer is you, write the evaluation. Do I have to send a written report back to the referring provider? It depends on where you practice, and the rule strength differs. In Ontario the record-keeping side is law and the report side is payer policy: the Medicine Act's record regulation requires the referrer's identity in the record, and OHIP consultation codes require a written referral and a written report. In Australia, MBS items 291 and 293 tie payment to "a detailed written report to the referrer." In the US, no statute requires a separate letter: for inpatient consults the note in the shared record has historically sufficed, and commercial payers that still pay consult codes expect a written reply. Wherever you are, a dated transmission line in the note is cheap insurance. Can a resident or a psychiatric nurse practitioner write the consult note? Yes, with the supervision layer documented. For Medicare payment on a resident's consult, the teaching physician must document presence and participation, and psychiatry has its own exception: 42 CFR 415.184 allows the required presence to be met by observation through a one-way mirror or video. PMHNPs may author and bill consults under their own NPI within state scope of practice, and behavioral health services billed incident-to a physician may now be furnished under general supervision under 42 CFR 410.26. State scope rules vary, so check yours. How long do consultation notes have to be kept? In US hospitals, at least 5 years under 42 CFR 482.24, with many states requiring 7 to 10 years or longer, and longer again for minors. Ontario's CPSO expects at least 10 years from the last entry for adults, and 10 years past the age of majority for minors, with both the referring and the consulting physician keeping the referral and the report. New South Wales, Victoria, and the ACT require 7 years from the last entry for adults and until age 25 for minors. Keep the outbound report with the note; an audit reads them together. Can BastionGPT draft a psychiatric consultation note? Yes. Give it the referral letter, relevant chart excerpts, and your dictated findings, and it drafts the question-first note: focused history, mental status examination, an impression that answers the question, numbered recommendations, and the reply letter to the referrer. It can also condense a long chart into the focused history and produce a plain-language version for the patient. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Hospital Consult & Liaison Note Consultation Response & Report Medication Management Follow-Up Note Collaborative Care (BHI/CoCM) Note Case Management Note Referral Letter Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/case-management-note All note templates # Case Management Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A case management note documents one case management contact: work that helps a client gain access to medical, social, educational, and other services. Case managers, social workers, and community agency staff write one for every contact, including phone calls and collateral contacts, because the note is the evidence behind billed time. Most case management notes run 100 to 300 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Case managers, social workers, therapists, and community agency staff Audience The client's care team, supervisors, the state Medicaid program, auditors Typical length 100 to 300 words · 5 to 15 minutes by hand (clinical team estimate) Format family Per-contact coordination record (compare: care coordination note, CCM note) When it's used After each contact: face-to-face, phone, telehealth, collateral, or attempted; the work is often non-face-to-face and longitudinal Standards context Federal Medicaid law defines the service and four activity categories; note fields come from state Medicaid manuals On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a case management note? A case management note is the contemporaneous record of one case management contact: the work of connecting a client to housing, benefits, medical care, and other needed services, and of following up until the linkage holds. No professional body invented or standardized the note. The service behind it, targeted case management (TCM), was created by Congress in the Consolidated Omnibus Budget Reconciliation Act of 1985 and refined by the Deficit Reduction Act of 2005, and the federal definition at 42 CFR 440.169 covers services that assist individuals in "gaining access to needed medical, social, educational, and other services." You will also hear the note called a TCM note, case note, contact note, service note, coordination note, or, in California Medi-Cal, a case management encounter. Two boundaries do most of the work on this page. First, the payer boundary: T1016 and T1017, the per-15-minute case management codes, are Medicaid codes, and Medicare does not pay them. A Medicare patient's coordination belongs in a different record, usually a chronic care management note or a transitional or collaborative care record. Second, the format boundary: federal regulation names four activity categories (assessment, care-plan development, referral and linkage, monitoring and follow-up) and specifies no note fields, no start and stop time mandate, and no federal signature deadline. The fields your agency treats as universal come from your state Medicaid manual and program contract. A care coordination note records the same work without the billing frame; the case management note is that work made claim-ready. ## Who uses case management notes and when Community mental health centers, county behavioral health programs, ACT and intensive case management teams, child welfare agencies, and aging and disability programs run on case management notes. The writers are case managers and care coordinators, social workers, and the therapists and agency staff who carry linkage duties alongside clinical ones. Write one after every contact that serves a care-plan goal: the housing authority call, the benefits application follow-up, the collateral call to a school counselor, and the attempted contact that reached voicemail. The note also sits in a web of neighbors: the referral it records may produce a psychiatric consultation note , the post-hospital tasks it tracks start from a discharge summary , and when the whole event is a multi-provider meeting, a case conference note or team meeting note fits better. The case management note is for the one-to-one linkage work in between. ## Case management note structure: what goes in each section Header: client, date, program. Client name or initials, Medicaid ID, date of service, the program or benefit the contact belongs to, and the case manager writing the note. The header is what ties the note to a specific claim line, and a valid Medicaid ID is a billing prerequisite. Pitfall: a note that cannot be matched to the claim's client, date, and program; the OIG's Alabama TCM audit found some billed services with no supporting documentation at all, and an unmatchable note fails the same way. Contact type and participants. Face-to-face, phone, telehealth, collateral, or attempted, plus who was present or contacted. For collateral contacts, record the name, title, and agency of the person you spoke with, and keep other clients' identities out of the record. Pitfall: an attempted contact written up like a completed one; states differ on whether attempts are billable at all, so document the attempt honestly and let the state manual decide the claim. Activity category. Federal Medicaid regulation recognizes four case management activities: assessment, care-plan development, referral and linkage, and monitoring and follow-up. Name the one this contact served. Pitfall: describing direct service delivery, counseling in the moment or driving the client somewhere, as case management; federal policy pays for helping the client gain access to services, not for delivering the underlying service itself. Narrative: action, outcome, barrier. What you did, what came back, and the barrier you found, in concrete nouns: the agency, the application, the missing document. Pitfall: "checked in with client, will continue to monitor"; a narrative with no action and no outcome supports no billable activity and gives the next reader nothing to pick up. Care-plan goal linkage. Name the assessed need or numbered care-plan goal this contact served. Case management is longitudinal, and the goal line is what connects one phone call to the plan an auditor reads months later. Pitfall: missing care-plan linkage; it is one of the recurring failure themes across state manuals and OIG findings, because a contact serving no assessed need is unbillable however well described. Time. Total minutes or start and stop times, per your state's rule. T1016 and T1017 are per-15-minute codes, and state manuals commonly require at least 8 minutes before a unit can be claimed. Pitfall: units the narrative cannot plausibly fill; unsupported time is the leading recoupment trigger across the OIG's state TCM audit series, even where the work clearly happened. Non-duplication check. When the client received another billed service the same day or period, state plainly that this activity was separate from it. Medicaid bars payment for case management that is an integral and inseparable component of another covered service. Pitfall: a case management note that mirrors the same-day therapy note; matching narratives are how duplication findings start. Next steps and signature. Each follow-up task with an owner and a due date, then signature or initials with title and credentials, dated. Some states also set review clocks: California Medi-Cal requires periodic review at least every six months and follow-up within 30 days of a referral appointment date. Pitfall: an open-ended "will follow up"; the auditor's next question is whether the linkage was monitored, and only an owner and a date can answer it. ### Blank template (copy and adapt) CASE MANAGEMENT NOTE Client: ____________ Medicaid ID: ____________ Date of service: ________ Program/benefit: ___________________ Case manager: ______________________ CONTACT [ ] face-to-face [ ] phone [ ] telehealth [ ] collateral [ ] attempted (outcome documented below) Present/contacted (name, title, agency for collateral): __________________ ACTIVITY CATEGORY (the four federal categories) [ ] assessment [ ] care-plan development [ ] referral/linkage [ ] monitoring/follow-up NARRATIVE (action taken, outcome, barrier identified) __________________________________________________________________________ __________________________________________________________________________ __________________________________________________________________________ CARE-PLAN GOAL SERVED: ___________________________________________________ TIME Start: ______ Stop: ______ Total minutes: ______ Units: _______ NON-DUPLICATION [ ] Not part of any other billed service this day Same-day services considered: ____________________________________________ NEXT STEPS (task, owner, due date) 1. _______________________________________________________________________ 2. _______________________________________________________________________ Signature/initials, title/credentials: _____________ Date signed: _______ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample case management note Scenario: community mental health case manager in a Medicaid targeted case management program, documenting a phone follow-up on a supportive housing application and a benefits barrier for an adult client with schizoaffective disorder. All details are fictional. Case Management Note. Client: D.R., 42 · Medicaid ID: on file · Date of service: 08/04/2026 · Program: Targeted case management (behavioral health) · Case manager: S. Okafor, BSW Contact: Phone. Collateral call to the county housing authority (M. Reyes, intake coordinator), then call to client; client reached and participated. No other parties involved. Activity category: Monitoring and follow-up on an active referral, with linkage work on the barrier identified. Narrative: Called the housing authority to check the status of the supportive housing application submitted 07/21/2026. M. Reyes reported the application is pending with one item outstanding: current proof of income. Called client. Client stated his SSI award letter was mailed to a prior address and he has no copy, so he cannot document income. Reviewed options; client chose to request a benefit verification letter from the Social Security Administration and agreed to bring it to the 08/11/2026 appointment. Client asked whether the application had been denied; clarified that it remains pending. Client agreed to the plan below. Care-plan goal: Goal 2, obtain and maintain stable housing (care plan dated 06/15/2026, reviewed 07/28/2026). Time: Start 10:12, stop 10:41; 29 minutes total. Supports 2 units of T1017. Non-duplication: No other service was delivered or billed today; this contact was not part of any therapy, medication, or crisis service. Next steps: (1) Client to request the SSA benefit verification letter; reminder call by case manager 08/08/2026. (2) Case manager to send verification to the housing authority within 2 business days of receipt, target 08/13/2026. (3) Case manager to recontact the housing authority if no status change by 08/20/2026. Signed: S. Okafor, BSW, Case Manager, 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The claim is checkable line by line. Client, program, date, contact type, a named collateral contact, and start and stop times totaling 29 minutes that map to 2 units of T1017. The activity stays inside the federal definition. Status check, barrier identification, and linkage to the benefits agency are access work, not direct delivery of a service. The barrier is named and owned. Missing income verification is stated plainly, with the fix the client chose and a date attached, so the next note can show movement. The goal line makes the chain auditable. Tying the call to Goal 2 of a dated care plan shows monitoring of an assessed need rather than a freestanding favor. Non-duplication is explicit and next steps are dated. The note rules out overlap with any same-day billed service and leaves three owned, dated tasks the following note can close. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The case management note is ordinary record content: part of the client's file, releasable on request, read by supervisors and agency quality reviewers, and the primary evidence when the state or its auditors pull a claim. State TCM manuals treat the case record as the support for each billable encounter, so a missing or thin note is functionally a missing service. Because the work is longitudinal and often non-face-to-face, the chain of notes is the product: each entry should let a stranger reconstruct what was tried, what blocked it, and who owns the next step. When the client changes settings, the chain connects to the handoff documents around it, the transfer-of-care summary on the way out and the task list a discharge summary leaves behind. Retention is state business: there is no single federal number for how long TCM records must be kept, 5 to 10 years is the common range under state Medicaid and record-retention rules, and your state plan and payer contract control. Write to the payer, not to a template. The statute auditors quote, Social Security Act section 1902(a)(27), requires records that "fully disclose the extent of the services" and names no fields. The federal layer above it, summarized in CMS's clarifying fact sheet on covered case management, defines the service and its four activity categories; everything operational lives in the state Medicaid manuals, and states genuinely disagree: Arkansas counts time spent recording required documentation as billable, North Dakota allows note writing but not straight data entry of the plan, Nevada excludes travel to and from appointments, and Texas caps service coordination at monthly encounters. The format is a convention; the content is the requirement. Two rulings worth keeping ready: the restrictive 2007 federal case management rule was partially rescinded in 2009 after a congressional moratorium, so guidance citing its single-case-manager and 15-minute restrictions as current federal law is out of date, and T1016 and T1017 are not payable by Medicare, so a Medicare-primary client's coordination belongs in the CCM , transitional care, or collaborative care code families before format even comes up. ↑ Back to the template and downloads ## Common case management note errors auditors flag The audit trail for targeted case management is unusually public. The HHS Office of Inspector General runs a continuing audit series on state Medicaid TCM programs: Alabama's audit (A-07-22-03253, issued April 2024) closed with a refund of $5,039,433 in federal share on $123.4 million in audited TCM claims, Montana's (A-07-21-03246, issued August 2022) recommended a refund of $5,065,966 in federal share from an estimated $7.7 million in unallowable claims, and a Georgia audit projected $4,654,984 in overpayments for clients the benefit excluded. OIG's work plan notes that prior work in one state found 18 percent of TCM claims unallowable and another 20 percent potentially so. The BastionGPT Clinical Advisory Board sees the same errors most often in case management note reviews: Time the record cannot support. Missing or vague duration, units beyond what the documented activity could fill, or no start and stop times where the state expects them. Unsupported services were a named failure category in both the Alabama and Montana findings. Duplication and the integral-service trap. Case management billed for activity that is an integral and inseparable part of another covered service (the prohibition carried in 42 CFR 441.18 and CMS manual guidance), a narrative that mirrors the same-day therapy note, or the same service billed twice. Unqualified authors. Notes signed by case managers whose qualifications were never documented against the state plan's requirements, a category named in both state audits above. Non-billable activity inside the units. Travel time, straight data entry of the plan, receiving faxes and emails, and attempted contacts in states that disallow them; each manual draws these lines differently, and units built on the wrong side come back. Ineligible clients and missing notes. Claims for clients the benefit excludes, such as those covered by the inmate-of-a-public-institution exclusion in the Georgia finding, and billed services with no supporting documentation at all. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on case management notes. Draft a complete case management note from a dictated recap, a few bullets, or a call log, with contact type, activity category, narrative, time, and next steps in place. Pull the open tasks out of a week of notes: every follow-up with its owner, due date, and care-plan goal, so linkage work stops living in memory. Check a finished note before you sign: time support for the units, an activity category, care-plan linkage, and a non-duplication line whenever another same-day service could overlap. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a case management note be? Most case management notes run 100 to 300 words and take 5 to 15 minutes by hand. Length follows the contact: a completed collateral call with a barrier and dated next steps may need 250 words, while a documented attempted contact can be two lines. Auditors do not measure length; they check whether the units billed are supported by what the note describes. Is there a required format for case management notes? No. Federal Medicaid regulation defines the service and four activity categories (assessment, care-plan development, referral and linkage, monitoring and follow-up) and specifies no note fields, no start and stop time mandate, and no federal signature deadline. The fields that feel universal, contact type, units, signature timeframes, come from state Medicaid manuals and agency policy. SOAP or DAP structure is a convention; the elements, not the format, are the requirement. Does Medicare pay for case management billed with T1016 or T1017? No. T1016 and T1017 are Medicaid codes, and Medicare does not pay them; claims sent to Medicare deny as non-covered. Medicare pays for coordination through its own code families: chronic care management, transitional care management, and collaborative care and behavioral health integration. Choose the code family by payer first and the note follows; the CCM note page covers Medicare's monthly model. Does time spent writing the note count as billable case management? It depends on the state, and states disagree. Arkansas's TCM manual counts time spent recording required documentation as billable; North Dakota allows time writing case notes but not straight data entry of the plan. Check your own manual before building documentation time into your units, and do not assume another state's answer travels. Can I bill for travel time or attempted contacts? Travel time: generally no. Federal guidance treats billable units as time delivering a case management service, not time in transit, though a state can build travel costs into its rate, and North Dakota credits the specific minutes when coordination or assessment happens during transport. Attempted contacts: document every attempt in the chart, but several states treat unsuccessful attempts and no-shows as non-billable. What belongs in the record and what belongs on the claim are different questions. What is the difference between a case management note, a care coordination note, and a CCM note? The care coordination note is the generic record: linkage work documented anywhere, tied to no billing code. The case management note is that work inside a Medicaid targeted case management benefit, with activity categories, timed units (typically T1017), and state-manual fields. The CCM note is Medicare's frame: a monthly time-and-care-plan record for patients with chronic conditions. Same underlying work, three payer frames; choose by payer and program before you choose a format. Can two case managers bill for the same client in the same month? Usually not. Many state plans limit targeted case management to one case manager, and some cap billing at one encounter per month; billing more than once for the same service is a recurring OIG recoupment category. The 2007 federal rule that would have made a single-case-manager restriction national was partially rescinded in 2009, so the controlling answer sits in your state plan. When two programs share a client, decide who bills before either does. How long do case management records have to be kept? In the US there is no single federal retention period for targeted case management records; state Medicaid rules and record-retention laws control, and 5 to 10 years is the common range. For cross-border context, Ontario's health-records framework expects at least 10 years (or 10 years past a minor's 18th birthday), and New South Wales and Victoria require 7 years, or until age 25 for minors. Keep care-plan versions with the notes; an audit reads them together. Can BastionGPT draft a case management note from a call recap? Yes. Give it a dictated recap, a call log, or a few bullets, and it drafts the note with contact type, activity category, narrative, time, and owned next steps ready for your review. It can also check a finished note for missing elements, time support, goal linkage, a non-duplication line, before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Care Coordination Note Chronic Care Management (CCM) Note Case Conference Note Multidisciplinary Team Meeting Note Transfer-of-Care Summary Psychiatric Consultation Note Eating Disorder Assessment CANS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/transfer-of-care-summary All note templates # Transfer-of-Care Summary: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A transfer-of-care summary is the document a clinician writes when ongoing care moves to a new provider, giving the receiving clinician the client's current status, active risks, medications and supports, and pending items. Therapists, psychologists, counselors, and care coordinators write one whenever another clinician assumes the client. Most run 150 to 600 words, shorter than a discharge summary. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, and care coordinators (the sending clinician) Audience The receiving clinician assuming care, and the sending clinician's own record Typical length 150 to 600 words · 15 to 30 minutes by hand (clinical team estimate) Format family Care-transition handoff document (compare: discharge summary, referral letter, return-to-primary-care summary) When it's used When ongoing care moves to a new provider who assumes the client: relocations, clinician departures, caseload reassignments, level-of-care changes with a continuing treater Standards context A professional convention, not a legally defined document; can run shorter than a discharge summary On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a transfer-of-care summary? A transfer-of-care summary is the document a clinician writes when ongoing care moves to a new provider who assumes responsibility for the client: a relocation, a clinician leaving the practice, a caseload reassignment, or a change in level of care with a named continuing treater on the other side. No single body created the format. Its lineage runs through SBAR, the handoff structure Michael Leonard and colleagues at Kaiser Permanente of Colorado adapted for healthcare in 2002 from US Navy submarine communication practice, and through the Joint Commission's 2006 handoff patient-safety goal, which became standard PC.02.02.01 in 2010. You will also see it called a transfer summary, care-transition summary, handoff note, transition-of-care (TOC) document, or transfer letter. Two boundaries do the real work. First, the summary is a document, and the warm handoff is not: SAMHSA defines a warm handoff as "a face-to-face introduction of a patient" to the receiving provider, a live event that can accompany the written record but never replaces it. Second, no US, Canadian, or Australian statute defines a therapist's transfer-of-care summary or prescribes its content, format, timing, or signature. The closest legally specified relative is the CMS summary of care record under Promoting Interoperability, which enumerates content (problem list, medications, allergies, care plan with goals, provider contacts) but binds only participants in that program. What makes the document worth writing is the handoff duty underneath it: active risks, current medications, and open items have to reach the person assuming care. That duty separates it from a discharge summary , which closes an episode with no named continuing treater, from a referral letter , which asks a colleague a question while care stays with you, and from a return-to-primary-care summary, the step back down to the GP when specialty work ends. ## Who uses transfer-of-care summaries and when Outpatient therapists and psychologists, counselors in group practices, community mental health teams, and care coordinators write one whenever a named clinician accepts ongoing responsibility: a client relocates, a clinician departs or retires, a caseload is reassigned, or a client moves between levels of care with treatment continuing on the other side. The deciding question is always who holds the client afterward. When specialty work ends and the GP or primary care physician resumes, the return-to-primary-care summary is the better fit. When the move itself takes arranging (calls, waitlists, insurance checks), that activity belongs in a case management note or care coordination note ; the transfer summary is the clinical handoff those notes arrange. It also runs shorter than a discharge summary, because it carries a current picture for a known reader rather than a complete account of an episode. ## Transfer-of-care summary structure: what goes in each section Header: client and both providers. Client identifiers, the sending clinician with credentials and contact details, and the receiving clinician who has accepted the client, with practice, contact information, and the acceptance date. Pitfall: an unnamed receiving provider; "referred to community resources" documents an ending, not a transfer, and continuity complaints start exactly there. Reason for transfer and current status. Why care is moving and the clinical picture as of the summary date: presenting concerns, functioning, most recent measures. Pitfall: re-telling the whole treatment history; the receiving clinician needs today's picture and can request the full record through the release. Diagnoses and treatment response. The current diagnosis or problem list, the approach used, and the part only you can supply: what worked and what did not. Pitfall: a bare code list; without response information the new clinician repeats experiments the client already sat through. Medications, allergies, and supports. The current medication list with the prescriber named and an as-of date, allergies, and other active supports such as psychiatry, groups, or case management. Pitfall: an unattributed medication list; if you do not prescribe, say whose list it is and how current it is, so the summary never reads as your verification. Active risks and safety plan. Risk status stated either way, relevant history, and the safety plan's status and next review date if one exists. This section is the document's reason to exist: risk information must reach the person assuming care. Pitfall: silence when risk is low; "no current ideation reported, assessed as low risk" is a finding, while an empty section is a gap the next clinician cannot interpret. Pending items. Everything half-done at the moment of transfer: reviews coming due, re-assessments, authorizations in flight, coordination waiting on a callback, each with an owner after the transfer and a date. Pitfall: pending items with no owner; an item nobody owns after the handoff is an item nobody does. Recommendations, first appointment, and consent. What you suggest the receiving clinician do first, a recommended window for the first appointment, and the release that authorizes this disclosure with its date and scope, plus a line noting that the original record stays with you. Pitfall: sending the summary before the release covering it is signed; the consent status sits inside the document because the disclosure has to be defensible later. ### Blank template (copy and adapt) TRANSFER-OF-CARE SUMMARY Client: ______________ DOB: __________ Date prepared: ____________ From (sending clinician, credentials, practice, contact): ________________________________________________________________________ To (receiving clinician, practice, contact; acceptance confirmed [ ]): ________________________________________________________________________ Reason for transfer: ___________________________________________________ CURRENT STATUS (picture as of the date above; latest scores) ________________________________________________________________________ ________________________________________________________________________ DIAGNOSES / PROBLEM LIST 1. _____________________________ 2. _____________________________ TREATMENT TO DATE (approach, response, what worked, what did not) ________________________________________________________________________ ________________________________________________________________________ MEDICATIONS & ALLERGIES (prescriber: ______________ list as of: ______) ________________________________________________________________________ ACTIVE RISKS & SAFETY PLAN (state findings either way; next review date) ________________________________________________________________________ ________________________________________________________________________ PENDING ITEMS (item, owner after transfer, due date) 1. _____________________________________________________________________ 2. _____________________________________________________________________ RECOMMENDATIONS & FIRST APPOINTMENT (suggested window) ________________________________________________________________________ CONSENT & RECORDS Release signed: [ ] yes [ ] no Date: __________ Scope: _____________ Original record retained by sender per retention rule: [ ] noted Clinician signature/credentials: ________________ Date signed: ________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample transfer-of-care summary Scenario: outpatient therapist transferring an adult client with generalized anxiety disorder to a new therapist after a relocation. All details are fictional. Transfer-of-Care Summary. Client: A.L., 29 · Prepared: 08/04/2026 · From: D. Okafor, LPC, Ridgeline Counseling · To: S. Whitfield, LMFT, Cedar Counseling Group (acceptance confirmed 07/31/2026) Reason for transfer: Client is relocating for work in mid-August 2026 and will continue weekly therapy with S. Whitfield, LMFT. Final session with this writer was 08/03/2026 (session 19). Current status: GAD-7 on 08/03/2026: 8, down from 14 at intake (01/14/2026). Worry episodes now brief (under 30 minutes most days) and managed with worry postponement and thought records used independently. Sleep 6 to 7 hours nightly. Work functioning stable; client describes the relocation as "wanted but stressful." Diagnoses and treatment to date: Generalized anxiety disorder (F41.1). 19 sessions of weekly CBT since 01/14/2026: cognitive restructuring, worry postponement, graded reduction of reassurance seeking. Response has been strongest with structured between-session practice; a brief stretch of unstructured supportive sessions in March produced less movement, and structure was resumed. Medications and allergies: Sertraline 50 mg daily, prescribed and managed by the client's primary care physician, Dr. J. Marsh (list current as of the 07/28/2026 PCP visit summary). No known drug allergies. This writer does not prescribe. Active risks and safety plan: No current suicidal ideation reported at the 08/03/2026 session; no history of self-harm or attempts. During an acute stress period in March 2026, client reported passive ideation without plan or intent; a written safety plan was completed 03/09/2026 and the ideation resolved by early April. The plan's six-month review (due 09/09/2026) falls after transfer and remains outstanding. Current risk assessed as low; routine monitoring recommended. Pending items: 1. Safety-plan review due 09/09/2026; owner: S. Whitfield, LMFT (copy of the plan included in the records packet). 2. GAD-7 re-administration at the first appointment (last score 8 on 08/03/2026). 3. New prescriber needed after relocation; Dr. Marsh will continue sertraline refills through 10/31/2026. Recommendations and first appointment: Continue weekly CBT with a relapse-prevention focus and maintain worry-postponement practice through the move. First appointment recommended by 08/31/2026 to keep the between-session gap under four weeks. Review and re-date the safety plan at intake. Consent and records: Release of information signed 07/28/2026 authorizing disclosure of this summary and the treatment record to S. Whitfield, LMFT (scope: full clinical record excluding psychotherapy notes). Original record retained by this practice per state retention requirements. D. Okafor, LPC. Signed 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Responsibility visibly passes. Both clinicians are named with practices and an acceptance date, so the record shows a transfer rather than an ending that hopes for one. Risk travels with an owner. Risk is stated even though it is low, the safety plan's outstanding review date is a numbered pending item, and the receiving clinician owns it in writing. The medication list is attributed. The prescriber and the as-of date are named, so a non-prescribing therapist's summary cannot be misread as medication verification. Response, not just history. The note says what worked and what worked less well, which is the one thing the new therapist cannot get from a diagnosis code. The disclosure is defensible. The release's date and scope sit inside the document, and the record notes that the original chart stays with the sender. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The summary is part of the standard clinical record: the client can request it, a board or court can read it, and the receiving clinician will rely on it. Two mechanics matter more than any format choice. First, the record does not travel with the client. The original stays with you under your jurisdiction's retention rule (commonly 6 to 10 years under US state law, with HIPAA's 6-year rule covering required documentation rather than charts; 10 years from the last entry in Ontario under CPSO and CPBAO standards; commonly 7 years for adults and until age 25 for minors in Australia), and you disclose what a signed release of information authorizes, nothing more. Second, substance use disorder content changes the consent math: under the 42 CFR Part 2 rule , with compliance required since February 16, 2026, a single consent can cover future treatment, payment, and operations disclosures, but SUD counseling notes carry separate protection and need their own consent before they ride along in a transfer packet. Label the rules honestly when you build your workflow. LAW governs the privacy of what you send (HIPAA, Part 2, state and provincial privacy acts), not the document itself: no statute in the US, Canada, or Australia specifies a transfer summary's content, format, deadline, or signatory. PAYER POLICY appears only where the handoff underpins a billed service: Medicare's transitional care management codes require interactive contact within 2 business days of discharge and a face-to-face visit within 7 or 14 days, with medication reconciliation documented (a physician and qualified-practitioner service most independent therapists cannot bill; see the TCM note ), and the CMS summary of care record sets a content list for Promoting Interoperability participants. Accreditation sits in between: Joint Commission standard PC.02.02.01 requires accredited organizations to run a handoff process, not to use any particular form. Ontario's CPSO Transitions in Care policy is the most explicit Canadian instrument and binds physicians, not psychologists, and Australia's Ahpra code requires practitioners to "promptly facilitate the transfer or management" of health records ( Ahpra guidance ) without prescribing a format. Everything else, including the summary's structure and its timing, is CONVENTION: document as soon as practicable and build the content from the strongest available list rather than a blog's guess. ↑ Back to the template and downloads ## Common transfer-of-care summary errors auditors flag Where transfer documentation meets billing, the numbers are specific. The HHS Office of Inspector General's audit of transitional care management claims (A-07-17-05100, issued July 2021) found almost $1.7 million in overpayments on 13,577 claims out of $249.5 million paid for calendar years 2015 and 2016, driven by duplicate and overlapping billing, and there is no TCM-specific denial code: documentation-deficiency denials arrive under the generic CARC 16 "lacks information" code with a case-specific remark code. Where it meets licensing, the record is disciplinary: in August 2023 the North Carolina Psychology Board conditioned a license after an abrupt termination in which the documentation "did not include summary content of each therapy session" and notes were "duplicated from session to session." The BastionGPT Clinical Advisory Board sees the same errors most often in transfer-of-care summary reviews: No named receiving clinician. "Referred out" with no accepted, named provider documents an ending, not a transfer; abandonment rules such as New York Education Law 6530(30) treat discontinuing care without arranged continuation as professional misconduct. Risk that does not travel. The safety plan, risk history, or current risk status stays in the old chart; the one piece of content the handoff exists to move is the piece most often left out. The warm handoff treated as the record. A live introduction happened, so no summary was written; the verbal event and the written record are complementary, and only one of them can be produced for a board or an auditor later. Disclosure wider than the release. The full chart, or Part 2-protected substance use content, goes out under a release that authorized a summary; match what you send to the signed scope. History instead of status, with nothing pending. Pages of treatment narrative, no current picture, and no list of open items with owners; the receiving clinician then learns about the overdue safety-plan review from the client, months later. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on transfer-of-care summaries. Summarize a full course of treatment into a transfer-ready current picture: status, treatment response, medications, active risks, and pending items, at 150 to 600 words. Merge the scattered sources, recent progress notes, the treatment plan, the prescriber's medication list, and the safety plan, into one coherent handoff document. Check a draft before it goes out: named receiving clinician, risk stated either way, pending items with owners and dates, and content matched to the release's scope. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is a transfer-of-care summary legally required? For most therapists, no. No US, Canadian, or Australian statute names the document or prescribes its content, format, deadline, or signature. The enforceable duties sit around it: ethics codes and licensing boards require continuity and prohibit abandonment, privacy law governs what you may disclose, and payer policy adds requirements only where the handoff underpins a billed service such as Medicare transitional care management. Treat the summary as the professional convention that satisfies those surrounding duties. How is a transfer-of-care summary different from a discharge summary or a referral letter? Follow the responsibility. A transfer-of-care summary hands ongoing care to a named clinician who assumes the client, and active risks and pending items travel with it. A discharge summary closes an episode when care at your setting ends. A return-to-primary-care summary steps the client back to the GP or primary care physician after specialty work. A referral letter asks another provider a question or service while you keep the client. How long should a transfer-of-care summary be? Most run 150 to 600 words, deliberately shorter than a discharge summary. It is a current picture, not a history: status, diagnoses and treatment response, medications, active risks, pending items, and a first-appointment recommendation. The receiving clinician can request the full record through the release; the summary's job is to be readable before the first session. Does a warm handoff replace the written summary? No. SAMHSA defines the warm handoff as a face-to-face introduction of the client to the receiving provider: a live event, not a record. The Joint Commission's handoff standard requires a process with opportunity for discussion, and the written summary is what remains afterward. The evidence on warm handoffs is also more mixed than most guidance suggests; the best-controlled study (Annals of Family Medicine, 2018) found no attendance benefit. Do the introduction when you can, and write the summary either way. Do my records go with the client when care transfers? No. The original record stays with you for your jurisdiction's full retention period: commonly 6 to 10 years under US state law, 10 years from the last entry in Ontario, and commonly 7 years for adults or until age 25 for minors in Australia. What moves is what the client's signed release authorizes: usually this summary, plus specific records on request. Keeping the original is part of the duty; sending it all away is a records-management failure, not a courtesy. How quickly do I need to complete a transfer-of-care summary? No behavioral-health rule sets a deadline; the convention is as soon as practicable, ideally before the receiving clinician's first appointment. The comparators clinicians borrow are payer and physician rules: Medicare transitional care management requires interactive contact within 2 business days of discharge, and Ontario's CPSO requires physicians to complete discharge summaries within 48 hours to bill a same-day visit. Neither binds a therapist's transfer summary, but both show the direction reviewers think in: fast enough that nothing falls in the gap. What if the summary contains substance use disorder content? Then 42 CFR Part 2 applies on top of HIPAA. Under the 2024 Final Rule, with compliance required since February 16, 2026, a single consent can cover future treatment, payment, and operations disclosures, but SUD counseling notes carry separate protection and need their own consent. Check the release's scope before the packet goes out, and when in doubt send the summary without the Part 2 content and handle that disclosure separately. Can I be disciplined over a transfer-of-care summary? Yes, even though no rule prescribes the format. In 2023 the North Carolina Psychology Board conditioned a license after an abrupt termination in which the transfer documentation lacked session summary content, and a companion order required a departing psychologist to transfer all client records to another licensed psychologist. New York lists abandoning a patient without arranging continuation of care as professional misconduct. Boards enforce the continuity duty; the summary is your evidence that you met it. Can BastionGPT write a transfer-of-care summary from my chart? Yes. Paste or summarize the recent notes, the treatment plan, the medication list, and the safety plan, and BastionGPT drafts a transfer-ready summary: current status, treatment response, risks stated either way, pending items with owners, and a first-appointment recommendation. It can also check a finished draft for a named receiving provider and release-scope match before you send. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Discharge Summary Return-to-Primary-Care Summary Care Coordination Note Transitional Care Management (TCM) Note Case Management Note Referral Letter General Risk Screening Note Glasgow Coma Scale (GCS) Documentation Braden Scale Documentation Morse Fall Scale Documentation MEWS & NEWS2 Early Warning Score Documentation Pressure Injury Staging Documentation OASIS Assessment Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/referral-letter All note templates # Referral Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A mental health referral letter is a signed, dated request from one clinician to another to assess or treat a named client. It states the referral question, relevant history, current medications, symptoms and measures, risk status, and what the referrer keeps doing while care continues. GPs, therapists, psychologists, and psychiatrists exchange them whenever care crosses providers. Most run 150 to 500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any referring clinician: GPs, primary care physicians, therapists, psychologists, psychiatrists, NPs Audience The receiving provider; the client's record on both ends; payers when a rebate or claim depends on it Typical length 150 to 500 words · 10 to 20 minutes by hand (clinical team estimate) Format family Clinical correspondence (compare: consultation report, transfer-of-care summary) When it's used When you ask another provider to assess or treat a client while your own care continues Standards context Naming and service systems vary by country; only Australia's Better Access prescribes the content by law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a mental health referral letter? A mental health referral letter is a written, signed, dated request from one clinician to another to assess or treat a named client: a GP asking a psychologist to provide therapy, a therapist asking a psychiatrist to evaluate for medication, a psychiatrist directing a patient into a specialist program. It carries the clinical information the receiving provider needs in order to say yes, and in payer-linked systems it is the document that makes the resulting care billable. Nobody invented it; it is a professional convention that each country's payer has formalized differently, most fully in Australia, where the Better Access initiative made the referral a condition of Medicare rebates in 2006. You will also hear it called a referral, a GP referral, a letter of referral, a consult request, or, in newer Australian templates, a request for services letter. Get the letter's status right, because it points in opposite directions in different countries. In the United States, no referral is generally required for outpatient therapy: under Original Medicare, clinical psychologists, clinical social workers, and, since January 1, 2024, marriage and family therapists and mental health counselors treat without a physician referral or order ( 42 CFR 410.71 and 410.73); where a referral requirement exists, it is a Medicare Advantage or commercial-plan contract term, not law. In Australia the letter is load-bearing: a valid signed, dated referral is a precondition of every Better Access rebate, and the GP mental health treatment plan is expressly not a referral, so a practice that sends the plan without the letter has referred nobody. Canada sits between: provincial plans pay for psychiatry on a physician or nurse practitioner referral, while private therapy needs none. There is, in short, no single standard format. Only Australia prescribes the content by law; everywhere else the letter is judged by whether the receiving clinician can act on it. ## Who uses referral letters and when Referral letters move in every direction across mental health care: GPs and primary care physicians refer into therapy and psychiatry (in Australia, the only route to a rebated psychology session), therapists and counselors refer clients to psychiatrists for medication evaluation, psychologists refer for testing, psychiatrists and paediatricians refer directly into Better Access, and nurse practitioners refer for consultations in Canada. Reach for a referral letter when you are asking another provider to take on a defined task while your own care continues. When you are ending your role and handing over responsibility, write a transfer-of-care summary instead; when an inpatient or program episode closes, the discharge summary does the handing off; and the specialist's written answer to your question is its own document type, covered on the psychiatric consultation note page. ## Referral letter structure: what goes in each section Header, addressee, and identifiers. Date, your name, credentials, and practice details (in Australia the prescribed particulars are your name, your practice address or provider number, and the date; for US plan-required referrals, the NPI that will go on the claim), then the client's name, date of birth, and contact details, all three required on a Better Access referral. Address it to a named clinician, a practice, or simply "Dear colleague." Pitfall: assuming the letter must name a specific clinician. Australia expressly allows a generic address, and the client may take the referral to any eligible professional of the same discipline; Ontario accepts a clinic name when the consultant is unknown. Referral question. One or two sentences, placed first: what you are asking the receiver to do (assess and advise, evaluate for medication, deliver a course of therapy), for whom, and why now. Everything else in the letter exists to serve this sentence. Pitfall: "please assess" with no answerable question. Incomplete referrals are the main reason letters bounce; one audit cited in the referral-quality literature found more than 40% of referrals to specialist mental health services were not accepted. Relevant history and medications. Working diagnosis or impression, treatment to date and response, prior episodes, relevant medical history, and current medications with doses. Select for relevance: include what would change the receiver's decision, leave out the rest. Pitfall: omitting the medication list. It is one of the content fields a Better Access referral must include, and for a medication-evaluation referral it is the first thing the psychiatrist looks for. Current symptoms and measures. The presentation now, with dated scores where you have them (PHQ-9, GAD-7, K10, DASS-21) and the functional impact on work, study, or relationships. Australia requires the symptoms or diagnosis to be stated. Pitfall: severity carried by adjectives alone. "Significant distress" gives the receiving clinician nothing to triage on or retest against; a dated score does both. Risk status. One explicit line either way, with when you last assessed it. Pitfall: silence on risk. The receiving service sets urgency from this line, and a letter that never mentions risk forces them to assume it was not assessed and to re-triage from nothing. Your ongoing role. State what continues with you: session frequency, what you will keep working on, who the client contacts between appointments, and how you want to coordinate. Pitfall: leaving your role unstated. A referral hands over a defined task while care continues with you; a letter that never says so reads as a transfer of care, and the receiver may assume they now hold the whole case. Urgency, sessions, attachments, and signature. Mark the priority and any timeframe, state the number of sessions requested (in Australia, a referral that does not specify a number is taken as six), whether an MHTP or PAMP has been prepared, and what is attached; then sign and date it, on paper or electronically. Pitfall: an unsigned or undated letter. Signature and date are conditions of a valid Australian referral, and a plan-required US referral missing the referring provider's identifier denies with CARC 183 or RARC N286. ### Blank template (copy and adapt) MENTAL HEALTH REFERRAL LETTER Date: __________ Priority: [ ] routine [ ] urgent, needed by: ________ From: ____________________________ Credentials: ________________________ Practice address / provider no. (AU) / NPI (US plan referrals): _________ To: _____________________________________________________________________ (a named clinician, a practice, or "Dear colleague") Client: ____________________ DOB: __________ Phone: ___________________ Address: ________________________________________________________________ REFERRAL QUESTION (what you are asking the receiver to do, and why now) _________________________________________________________________________ RELEVANT HISTORY (working diagnosis, treatment to date and response, relevant medical history) _________________________________________________________________________ CURRENT MEDICATIONS, WITH DOSES (required on Better Access referrals) _________________________________________________________________________ CURRENT SYMPTOMS AND MEASURES (dated scores: PHQ-9, GAD-7, K10, DASS-21) _________________________________________________________________________ RISK STATUS (state it either way, with the date last assessed) _________________________________________________________________________ MY ONGOING ROLE (what continues with you; contact between appointments) _________________________________________________________________________ Sessions requested (AU: six assumed if unspecified): ____________________ MHTP or PAMP prepared? [ ] yes, date: ________ [ ] no [ ] n/a Attachments: [ ] release/consent [ ] treatment plan [ ] score log [ ] other: _________________________________________________ Report back requested: [ ] after assessment [ ] end of course Signature (electronic accepted in AU and US): ___________________________ Name / credentials: ______________________ Date signed: ________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample mental health referral letter Scenario: a therapist in US private practice refers an adult client with recurrent depression to a psychiatrist for a medication evaluation while weekly therapy continues. All details are fictional. Mental Health Referral Letter. From: S. Okafor, LCSW, Riverbend Counseling Group · To: A. Meloni, MD (psychiatry) · Date: 08/04/2026 · Re: D.R., DOB 05/14/1997 Dear Dr. Meloni, Referral question: I am referring D.R., a 29-year-old client in weekly cognitive behavioral therapy with me since March 2026, for psychiatric evaluation of recurrent major depressive disorder and a recommendation on pharmacotherapy. Please assess whether medication is indicated, and either initiate and manage prescribing or advise me and her primary care physician. Routine priority; she can attend weekday afternoons. History: First depressive episode at age 22, resolved without treatment. The current episode began in January 2026 after a job loss. Fourteen weekly CBT sessions since 03/09/2026 (behavioral activation, cognitive restructuring) produced early gains that have plateaued over the past six weeks. No prior psychotropic trials, no hospitalizations, no history of mania or psychosis. Medical history: hypothyroidism, stable on levothyroxine 50 mcg daily; TSH in range 05/2026 per PCP labs. No other medications, no known drug allergies. AUDIT-C 2 at intake; she reports no current alcohol or drug concerns. Current presentation: PHQ-9 on 08/03/2026: 14 (moderate), down from 19 at intake and stable between 13 and 15 across the last six weekly administrations. Persistent early waking, low energy, and reduced concentration affecting work performance. GAD-7: 6. Risk: D.R. denies suicidal ideation, plan, and intent, screened at every session including 08/03/2026; no history of self-harm. I assess current risk as low and review it weekly. Ongoing care: I will continue weekly CBT and remain her primary mental health contact between appointments. I am asking for evaluation and medication management, not a transfer of care. I would appreciate a copy of your consultation report and am glad to discuss by phone. Attachments: signed release of information, current treatment plan, PHQ-9 and GAD-7 score log. Thank you for seeing her. Sincerely, S. Okafor, LCSW (license 12847) Riverbend Counseling Group · (555) 014-2276 · signed electronically 08/04/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The referral question is answerable in a single appointment and states the decision being requested (evaluate, then manage or advise), so the receiving office can accept and book without a clarifying call. History is filtered for what changes the psychiatrist's decision: episode course, therapy response, the absence of prior medication trials, thyroid status, and substance use screening, rather than a chart export. Measures are dated and trended (PHQ-9 19 to 14, stable for six weeks), giving the receiver a baseline to retest against and evidence that the plateau is real. Risk is stated affirmatively, with method and recency, so triage does not stall on an unknown. The ongoing-care and attachments lines mark this as a referral rather than a transfer, name who holds care between appointments, and show that consent to share is on file. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A referral letter is ordinary clinical correspondence, which means it lives in the record twice: your copy in the sending chart, the receiver's copy in theirs, both part of the record set the client can request and read. Write it accordingly. Because the letter carries clinical content to another practice, settle consent before it travels: the clean practice in any jurisdiction is a documented consent conversation, and a signed release of information when records travel with it. Retention follows the strictest rule that applies to you rather than any referral-specific one. In the US, HIPAA's six-year rule covers required documentation while state law governs charts, and Medicare providers are generally advised to keep records 7 to 10 years. Ontario's regulated professions generally keep records 10 years from last contact, or from the age of majority for minors. NSW, Victoria, and the ACT require 7 years from the last entry for adults and until age 25 for minors, and Australian Medicare separately requires documents an MBS item creates, referrals included, to be kept for 2 years under the Health Insurance Act 1973. On the receiving side, an Australian professional must hold the referral at the first consultation for the rebate to stand. The payer layer is where the three countries split. No US or Canadian regulator publishes an element list for a mental health referral letter; the format is convention, and what is regulated is the billing that hangs on it. Under Original Medicare a referral functions as an exemption rather than a prerequisite: a clinical psychologist or clinical social worker who bills Part B must offer, with the client's consent, to consult the client's physician, unless that physician referred the client, which waives the duty ( 42 CFR 410.73 ; 42 CFR 410.71(e)), and neither party may bill for that consultation. The genuine US requirements are narrower: psychological and neuropsychological testing by an independently practicing psychologist needs a physician order, and Medicare Advantage plans (42 CFR 422.112 permits referral gates) and commercial plans may require referrals by contract, in which case the claim carries the referring provider's NPI or denies with CARC 183 or RARC N286. In Ontario, a consultation is billable only on a written request naming the referring physician or nurse practitioner and carrying their OHIP billing number under section 38.4 of Regulation 552; without it, the service pays as a lesser assessment. In Australia the letter is the rebate instrument itself: the plan is not a referral, the psychologist must hold the letter at the first consult, and the loop closes with a written report to the referrer after each course of treatment, the clinician-to-clinician reply covered on the consultation report page. ↑ Back to the template and downloads ## Common referral letter errors auditors flag The referral letter occupies an odd place in the audit record: in the US it is rarely the document that costs money. The OIG's national psychotherapy audit (A-09-21-03021, 2023) estimated that $580 million of roughly $1 billion in Medicare psychotherapy payments from March 2020 to February 2021 was improper, and the defects sat in the psychotherapy documentation itself: missing start and stop times, treatment plans, and evidence of the service furnished, not referral letters. Where the referral does bite is at the claim edge and abroad. US plan-required referrals deny with CARC 183 (referrer not eligible to refer) or RARC N286 (missing or invalid referring-provider identifier). Ontario rejects claims with codes AC4, ARF, or ERF when the referring OHIP number is missing or the referrer ineligible, and pays the consultation as a lesser assessment without a valid written request. Australia's Professional Services Review keeps finding mental health plans and referrals that cannot be located for the date of service; a case in the PSR Director's July 2025 update ended in repayment and disqualification from a range of MBS items. The referral-quality literature adds the clinical cost: audits of referrals into specialist mental health care repeatedly find them incomplete, one cited audit reporting more than 40% not accepted. The BastionGPT Clinical Advisory Board sees the same errors most often in referral letter reviews: A question the receiver cannot act on. "Please assess and treat as appropriate" gives the receiving service nothing to triage, book, or answer, and incomplete referrals are the leading reason letters are bounced back or rejected. The plan sent instead of the referral. Australian practices mail the GP Mental Health Treatment Plan and consider the client referred; the plan is not a referral, and no Better Access rebate is payable until a separate signed, dated referral exists. Missing referrer identifiers. A plan-required US referral without the referring provider's NPI denies with CARC 183 or N286; an Ontario consultation claim without the referrer's OHIP billing number rejects or downgrades; an Australian referral without the referrer's name, practice address or provider number, date, and signature fails the prescribed particulars. Required content fields skipped. A Better Access referral must state the client's symptoms or diagnosis, current medications, the number of sessions, and whether an MHTP or PAMP has been prepared; the medication list and the plan statement are the two fields most templates omit. Risk, urgency, and role left unstated. A letter silent on risk reads as not assessed, a letter silent on urgency defaults to routine scheduling, and a letter silent on your ongoing role reads as a transfer, leaving no clear holder of care between appointments. ### Referral letters in the US, Canada, and Australia Aspect United States Canada Australia Status Direct access is the default: Original Medicare requires no referral for therapy by psychologists, clinical social workers, MFTs, or MHCs; Medicare Advantage and commercial plans may require one by contract Provincial plans pay for psychiatry on a physician or NP referral; private therapy, psychology, and social work need no referral (self-pay or private insurance) A valid written, signed, dated referral is a precondition of every Better Access rebate, set by the section 3C Determination and MBS notes; e-signatures are acceptable Terminology Referral; referring or ordering provider Referral, consultation request, written request Referral, GP referral, request for services letter What changes Plan type decides everything: when a plan requires a referral, the claim carries the referring provider's NPI or denies (CARC 183, RARC N286); a physician referral also waives the psychologist's offer-to-consult duty In Ontario, a consultation is billable only on a written request carrying the referrer's OHIP billing number; without it the service downgrades to a lesser assessment The referral must state symptoms or diagnosis, current medications, session count, and whether an MHTP or PAMP exists; from 1 November 2025, MHTP referrals sit with the MyMedicare or usual GP Retention No federal chart rule; HIPAA's 6-year rule covers required documentation, state law governs records, and Medicare providers are generally advised to keep 7 to 10 years Provincial; Ontario's regulated professions generally keep 10 years from last contact, or from the age of majority for minors NSW, Victoria, and the ACT require 7 years from last entry (to age 25 for minors); Medicare separately requires referral documents kept 2 years under the Health Insurance Act 1973 The letter itself is portable; what changes across borders is who must write it, what it must contain, and what payment hangs on it. A US or Canadian clinician can adopt the Australian content list wholesale and lose nothing: it is simply a complete referral letter. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on referral letters. Draft a complete referral letter from a few bullets, a dictation, or pasted chart notes, with the referral question up front and the history, medications, measures, risk status, and your ongoing role behind it. Rewrite the letter in plain language for the client so they know what they are being referred for and what happens next, or trim a long draft to the facts that change the receiver's decision. Check a finished letter before you sign: answerable question, medication list, risk line, signature and date, and the jurisdiction fields (Better Access particulars, referring NPI, OHIP billing number). See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a mental health referral letter be? Most run 150 to 500 words, a single page. No US, Canadian, or Australian rule sets a minimum or maximum length, requires letterhead, or mandates a coding system on the letter itself; Australia's Better Access rules are about content (symptoms or diagnosis, medications, session count, an MHTP statement), not length. A letter the receiver can read in two minutes, with an answerable question in the first paragraph, beats a chart export every time. Do clients need a referral to see a therapist in the United States? Under Original Medicare, no. Clinical psychologists, clinical social workers, and, since January 1, 2024, marriage and family therapists and mental health counselors furnish outpatient mental health services without a physician referral or order. A referral changes one thing: it waives the psychologist's or social worker's duty to offer to consult the client's physician. The genuine requirements live elsewhere: physician orders for testing by independently practicing psychologists, and Medicare Advantage or commercial plan contracts, which may require a referral and a referring NPI on the claim. Is the GP Mental Health Treatment Plan the same as a referral? No, and this is the most consequential mix-up in Australian practice. The MHTP is the GP's assessment and planning document; the referral is a separate signed, dated letter, and Medicare pays no Better Access rebate without it. GPMHSC and the Department of Health state plainly that the plan is not a referral, and the psychologist must hold the referral at the first consultation. The referral must come from a GP or prescribed medical practitioner managing the client under the plan, a psychiatrist under an assessment and management plan, or directly from a psychiatrist or paediatrician. The plan side is covered on the Better Access treatment plan page. How many sessions does a Better Access referral cover, and does it expire? If the referral does not specify a number, the psychologist may assume six. An initial course is up to six individual services, and the yearly cap is 10 individual and 10 group. The referral does not expire at calendar-year end: unused sessions carry into the new year and count against that year's cap, despite the annual-expiry claim on many template sites. A new referral is needed for each further course, after the psychologist has reported back and the GP has assessed the need. AN.15.6 even includes a lost-referral provision: the service can proceed on a lost, stolen, or destroyed referral, with "lost referral" recorded in place of the particulars. Does a referral letter have to name a specific clinician? No jurisdiction requires it. An Australian referral may be addressed generically, and the client may take it to any eligible professional of the same discipline; Ontario's Ministry accepts naming a clinic when the individual consultant is unknown. What causes denials is not the addressee but the referrer's own side of the letter: a missing signature, date, provider number, NPI, or OHIP billing number, depending on the country. What is the difference between a referral letter and a transfer-of-care summary? A referral asks a question and hands a defined task to another provider while your care continues: you keep the client, and the receiver reports back to you. A transfer-of-care summary ends your role and passes ongoing responsibility to the next clinician. The distinction has payer consequences: a transfer does not set off Australia's report-back loop or Ontario's consultation payment rules. If your letter never says you are continuing care, it reads as a transfer. Does the receiving clinician have to send anything back? Often, yes. In Ontario, the consultant must send written findings, opinions, and recommendations to the referrer, and keep a copy of the request, to bill a consultation. In Australia, the treating professional must give the referrer a written report after each course of treatment; no approved form exists. In the US nothing attaches to the referral itself, though the 2025 codes G0546 to G0551 mirror CPT 99446 to 99452 and pay for some clinician-to-clinician consults where the consulting practitioner cannot bill an E/M visit. The reply is its own document type: see the consultation report and psychiatric consultation note . Is an electronic signature valid on a referral letter? In Australia, yes: MBS note AN.15.6 cites the Electronic Transactions Act 1999, so an electronically signed Better Access referral is valid. In the US, electronic signatures are generally acceptable under CMS signature guidance. What no jurisdiction excuses is the absence of a signature and date: both are prescribed particulars of an Australian referral, and the treating professional must hold the signed letter by the first consultation. Can BastionGPT draft a referral letter from my notes? Yes. Give it a few bullets, a dictation, or the relevant chart notes and it drafts the letter with the referral question up front, then history, medications, measures, risk status, and your ongoing role. It can also produce a plain-language version for the client and check the jurisdiction fields before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Consultation Response & Report Discharge Summary Better Access Mental Health Treatment Plan Referral Transfer-of-Care Summary Psychiatric Consultation Note School & Teacher Collateral Note STOP-Bang Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/informed-consent All note templates # Informed Consent for Treatment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An informed consent for treatment form is the intake document where a therapist explains the nature of therapy, its risks, benefits, and alternatives, fees, and the limits of confidentiality, and the client agrees to treatment, usually by signature. Therapists, counselors, psychologists, and social workers review it before care begins. Most forms run 800 to 2,000 words in plain language. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, and social workers; the practice drafts it, the client signs it Audience The client or guardian, the treating clinician, licensing boards, auditors Typical length 800 to 2,000 words · 90 to 180 minutes by hand (clinical team estimate) Format family Clause-based intake form (compare: telehealth consent, NPP acknowledgment, financial policy) When it's used At intake, before treatment begins; re-reviewed when modality, fees, or confidentiality terms materially change Standards context Required as a process by every ethics code; a signed form is law only in some states; terminology varies by discipline and jurisdiction On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an informed consent for treatment form? An informed consent for treatment form is the intake document, and the underlying process, through which a therapist discloses the nature and course of therapy, its risks, benefits, and alternatives, fees and third-party involvement, and the limits of confidentiality, and through which the client voluntarily agrees to treatment, usually by signature. The doctrine behind it is older than any therapy modality: it traces to the Nuremberg Code (1947) and to US case law such as Canterbury v. Spence (1972). The psychotherapy-specific written format arrived in 1988, when Handelsman and Galvin published a form built as a "series of questions that clients have a right to ask," updated by Pomerantz and Handelsman in 2004 and adapted for telepsychology in 2016. You will also see it called a consent to psychotherapy, a psychotherapist-patient services agreement, a client services agreement, a professional disclosure statement, or, in Colorado's statute, a mandatory disclosure statement. The clarification that does the most work: this is one of four intake documents that generic templates routinely weld together, and each has a different legal basis. The consent to treatment is the clinical and ethical instrument. The Notice of Privacy Practices acknowledgment is HIPAA's document: 45 CFR 164.520 requires giving the privacy notice and making a good-faith effort to obtain written acknowledgment of receipt, and HIPAA nowhere requires a consent-to-treat form. The release of information is an authorization for records to leave the practice. And federally assisted substance use programs use the separate 42 CFR Part 2 consent for their records. Keep the four apart: they trigger at different moments, and an authorization merged into other paperwork can be invalid. ## Who uses informed consent for treatment forms and when Every behavioral health discipline keeps a version: psychologists, professional counselors, clinical social workers, marriage and family therapists, and psychiatrists, in settings from solo practice to community mental health, with terminology that shifts by discipline (counselors often say professional disclosure statement). Training clinics lean on the form hardest, because the supervision disclosure lives in it. Present it at intake, before treatment starts, and walk through it out loud: counseling's ethics code expects consent reviewed in writing and verbally, and every code treats consent as ongoing rather than a one-time signature. This form covers routine outpatient treatment of an adult who can consent. Switch to a minor consent when a parent or guardian consents for a child, add a telehealth consent when care goes remote, and keep money terms in the financial policy this form points to. ## Informed consent for treatment structure: what goes in each section Practice and clinician identification. Who is providing care: name, license type and number, contact details, and, for pre-licensed clinicians, supervised status and the supervisor's name. APA Standard 10.01(c) makes the trainee disclosure explicit. Pitfall: an associate's form that never names the supervisor; a board reads the omission as concealment of trainee status, and the supervisor typically holds legal responsibility for the care anyway. Nature and course of services. What therapy is, your approach, session length and frequency, what a typical course looks like, and the honest line that outcomes are not guaranteed. Pitfall: one boilerplate modality description reused by every clinician in a group practice; the clause has to describe the treatment this client will actually receive. Risks, benefits, and alternatives. Both sides of the ledger: therapy can reduce symptoms and improve functioning, and it can also bring temporary increases in distress, painful memories, and strain on relationships. Name reasonable alternatives: a different approach, a medication evaluation, a support group, or no treatment now. Pitfall: benefits-only drafting; every major ethics code expects risks and alternatives in the disclosure, so a form without them documents a partial disclosure. Limits of confidentiality. The clause the whole form exists for. Name the exceptions: suspected abuse or neglect of a child or vulnerable adult, a serious threat of harm to self or others, court orders, professional consultation, and insurer access to diagnosis and dates when insurance is billed. Pitfall: the vague version ("confidentiality has some limits") is the version boards discipline; reporting thresholds are state law, so the wording must track your state's statutes, not a national template. Communication, emergencies, and after-hours contact. How clients reach you between sessions, expected response times, your email and text policy and its privacy limits, and exactly what to do in a crisis. Pitfall: a phone number with no after-hours plan; the form that never says what a client should do when you are unreachable is the gap an availability complaint points at first. Fees and billing pointer. State that fees, missed-session charges, and insurance terms live in the financial policy provided alongside this form, and note what billing insurance means for privacy. Pitfall: pasting the full fee schedule into the consent form, so every fee change strands a stale signed version in the chart; keep one source of truth and point to it. Client rights, refusal, and withdrawal. The client may ask questions at any time, decline any technique, seek a second opinion, and end treatment at any time, with a line on how endings are handled and referrals offered. Pitfall: consent language that reads like a commitment to finish a course of treatment; the freedom to withdraw is what makes the consent voluntary. Consent statement and signatures. A short statement that the client has read the form, had questions answered, and agrees to treatment, with signature and date lines for the client or guardian and the clinician, plus space to record later re-reviews. Pitfall: an unsigned, undated form in the chart; in most states that is a documentation gap, and in Colorado it becomes a statutory violation after the second visit. ### Blank template (copy and adapt) INFORMED CONSENT FOR TREATMENT Practice: _____________________________ Phone: ______________________ Clinician: ____________________________ License type/#: _____________ If pre-licensed: supervised by ________________________ (name, license) 1. SERVICES. Approach, session length and frequency, expected course: ___________________________________________________________________ Outcomes are not guaranteed; progress is reviewed with the client. 2. RISKS, BENEFITS, AND ALTERNATIVES. Possible benefits: _________________________________________________ Possible risks (e.g., temporary increase in distress): _____________ Alternatives discussed (other approaches, referral, no treatment): ___________________________________________________________________ 3. LIMITS OF CONFIDENTIALITY. Reviewed with the client, including: [ ] suspected abuse or neglect of a child or vulnerable adult [ ] serious threat of harm to self or others [ ] court order or other legal requirement [ ] professional consultation and supervision [ ] information shared with an insurer when insurance is billed State-specific duties: _____________________________________________ 4. COMMUNICATION AND EMERGENCIES. Between-session contact, response time, email/text policy, and the emergency plan: ___________________________________________________________________ 5. FEES AND BILLING. Fees, cancellation terms, and insurance are in the financial policy, provided separately. Client received it: [ ] 6. CLIENT RIGHTS. The client may ask questions at any time, decline any technique, and end treatment at any time. 7. CONSENT. I have read this form, had my questions answered, and consent to treatment as described. Client (or guardian) signature: ______________________ Date: ________ Clinician signature/credentials: _____________________ Date: ________ Consent re-reviewed (date/initials): __________________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample informed consent for treatment form Scenario: the intake consent form of a fictional solo private practice, reviewed and signed at a first appointment for adult outpatient psychotherapy. All details are fictional. Lantern Hill Counseling, PLLC · Clinician: M. Okafor, LPC (License #C-4471) · Client: J.C. · Reviewed together and signed: 08/04/2026 1. What counseling here is like. Therapy at Lantern Hill is talk-based counseling, usually 50 minutes once a week. I draw mainly on cognitive behavioral and acceptance-based approaches. We will set goals together in the first few sessions and review them about every three months. Therapy works best when it is honest and collaborative, and I cannot guarantee a particular outcome. 2. Benefits, risks, and other options. Many people find therapy reduces symptoms and improves relationships and coping. It also has risks: sessions can bring up painful memories, you may feel worse for a stretch before you feel better, and change can strain existing relationships. Other options include a different type of therapy or therapist, a medication evaluation with a prescriber, a support group, or deciding not to start right now. I am glad to talk through any of these with you. 3. Confidentiality and its limits. What you share here stays between us, with exceptions the law requires me to act on. I must break confidentiality if I suspect abuse or neglect of a child or a vulnerable adult, if I believe there is a serious threat of harm to you or to someone else, or if a court orders me to release information. I may consult other licensed professionals about my work; when I do, I leave out details that identify you. If you use insurance, your insurer receives your diagnosis and dates of service. 4. Reaching me between sessions. You can leave a voicemail any time; I return calls within one business day, Monday through Friday. Email is for scheduling only, because it is not a secure channel for clinical matters. I do not provide 24-hour crisis coverage. If you are in crisis, call or text 988, call 911, or go to the nearest emergency room. 5. Fees and billing. My fees, late-cancellation policy, and insurance details are in the Lantern Hill financial policy, which you received with this form and sign separately. Please ask me about anything in it. 6. Your rights, and ending therapy. You may ask about my training, my methods, or your progress at any time. You may decline anything I suggest, seek a second opinion, and end therapy whenever you choose. If I believe you need care I cannot provide, I will tell you and offer referrals. If the terms in this form change in a meaningful way, we will review it together again. 7. Consent. I have read this form, had the chance to ask questions, and agree to begin treatment. Signed: J.C. (client), 08/04/2026 · M. Okafor, LPC (clinician), 08/04/2026 This sample is fictional and for educational purposes. It does not describe a real practice or client. Check your own state's requirements before adapting it. ↑ Back to the template and downloads ### Why this sample works The exceptions are named. Reporting duties, threat of harm, court orders, consultation, and insurer access appear in plain words; the vague version of this clause is the consent failure boards actually discipline. Consent stays voluntary on the page. The right to decline any technique, seek a second opinion, and stop at any time sits in its own clause, which is what makes the signature meaningful. It promises re-review. The form commits to revisiting its terms when they materially change, matching the ongoing-consent standard in every major ethics code. Money lives elsewhere. The fees clause points to the standalone financial policy, so a fee change never strands a stale signed consent in the chart. The register is plain and the dates are fixed. Short sentences in second person, no statute citations aimed at the client, and both signature lines carry dates, so the record shows when disclosure happened. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The signed form is ordinary clinical-record content: part of the designated record set, releasable to the client on request, and discoverable, with none of the special protections that segregated psychotherapy notes carry under their own authorization rules . Keep it as long as the chart it belongs to: state law, commonly 7 to 10 years for adults and longer for minors in the US, with Canadian provinces ranging from 5 years (Quebec) to 10 (Ontario, Alberta). Label the legal layers correctly when you build yours. The disclosure process is an ethics requirement in every discipline: the APA Ethics Code (Standard 10.01) requires informing clients "as early as is feasible" about the nature of therapy, fees, third parties, and confidentiality limits, with the process documented. A signed form on a deadline is law in a minority of states, Colorado the clearest: a written mandatory disclosure statement signed "not later than the second visit with the psychotherapist" (Colo. Rev. Stat. 12-245-216, content updated effective August 7, 2024). Everywhere else the signature is convention, worth keeping because it dates the disclosure, which is the first thing a board panel looks for. No payer conditions payment on this form. Medicare's psychotherapy documentation requirements reach treatment plans, time, diagnosis, and medical necessity, and payer audits cite those items, not consent paperwork, so write the form for your client and your board rather than for a claims reviewer. What sits beside the consent form still needs its own instrument: a separate telehealth consent where state rules or Medicare telehealth policy expect one; the written 42 CFR Part 2 consent in federally assisted substance use programs, including the single-consent option with its February 16, 2026 compliance date; the Notice of Privacy Practices acknowledgment , which HIPAA asks you to attempt in good faith and retain for 6 years; and the money documents, the financial policy and, for self-pay clients, the good faith estimate . One combination is prohibited outright: a HIPAA authorization must stand alone and cannot be combined with the privacy notice. Separate documents are what let a client understand what each signature means, and they keep one document's update cycle from invalidating another's signature. ## Common informed consent for treatment errors auditors flag By payer-audit standards this form is nearly invisible. In the two largest federal psychotherapy audits, OIG report A-09-21-03021 (an estimated $580 million in improper Medicare payments) and report A-02-21-01006 (a single New York City provider, $1,118,789 in estimated overpayments, 100 of 100 sampled treatment plans out of compliance), every cited deficiency concerns treatment plans, time documentation, or signatures on treatment plans, never the consent form. Consent failures surface somewhere else: licensing-board discipline. The BastionGPT Clinical Advisory Board sees the same errors most often in informed consent form reviews: One signature treated as permanent consent. Every major ethics code treats consent as ongoing (APA 10.01, ACA A.2.a, CRPO 3.2, AHPRA's 2025 code). A form signed years ago, before your current fees and your telehealth practice, documents consent to a treatment that no longer exists; re-document at each material change. Confidentiality limits alluded to, not named. "There are some exceptions" discloses nothing. Failure to disclose the limits of confidentiality is the consent failure boards discipline, and the clause has to name reporting duties, threat-of-harm exceptions, court orders, and insurer access in your state's terms. Four documents welded into one. Consent to treat, NPP acknowledgment, release of information, and Part 2 consent merged into a single signature packet. Each has a different legal basis, HIPAA requires the authorization to stand alone, and a client signing a stack cannot tell what they agreed to. No supervision disclosure. Pre-licensed clinicians whose forms never say so. APA 10.01(c) requires telling the client the therapist is in training and naming the supervisor; the omission reads as concealment in a board file. A reading level that defeats the consent. Dense legal boilerplate the average client cannot parse. Readability affects what clients recall (Handelsman and Martin, 1992), and a disclosure the client could not understand fails at its one job: informed agreement. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on informed consent for treatment forms. Draft a complete, plain-language consent form from bullets about your practice: state, discipline, modalities, communication policy, and how fees are handled. Rewrite the form you use now at an easier reading level, clause by clause, without dropping a required disclosure. Check a draft against the elements ethics codes expect (risks and alternatives, named confidentiality limits, withdrawal rights, the supervision line) before a client sees it. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do clients legally have to sign an informed consent form? It depends on where you practice. Every major ethics code (APA Standard 10.01, NASW 1.03, ACA A.2.a) requires the informed consent process and documentation of it, but the codes themselves do not mandate a signature on a specific form. Some states convert that into law: Colorado requires a written mandatory disclosure statement signed "not later than the second visit with the psychotherapist" (Colo. Rev. Stat. 12-245-216). Everywhere else, a written, signed form is the risk-management standard rather than a statute, because a dated signature is the cleanest evidence the disclosure happened. Does Medicare or insurance require an informed consent form? No. No payer, Medicare included, conditions payment for psychotherapy on a signed consent-to-treatment form. Medicare's documentation requirements concern treatment plans, time, diagnosis, and medical necessity, and the major federal psychotherapy audits cite exactly those items, never the consent form. The practical read: this form protects you with your licensing board and in malpractice disputes, not in a payer audit, so give the treatment plan and time documentation the audit-defense effort. How often does informed consent need to be re-signed? Consent is ongoing, not annual. The ACA code puts it directly: "informed consent is an ongoing part of the counseling process." Ontario's CRPO consent standard treats continued attendance as implied ongoing consent, with express consent for specific situations. Re-document when something material changes: the treatment approach, fees, confidentiality terms, or the modality, such as moving to video sessions, which calls for a separate telehealth consent . A calendar-driven re-signing every January adds paper, not protection. Is the informed consent form the same as the HIPAA privacy form? No. The HIPAA document is the Notice of Privacy Practices acknowledgment : you provide the privacy notice at first service and make a good-faith effort to obtain written acknowledgment that the client received it. It is not consent to treatment, and neither document authorizes records to leave the practice; that takes a release of information , which HIPAA requires to stand alone. Substance use programs add the separate 42 CFR Part 2 consent . Four documents, four legal bases. What goes in the limits-of-confidentiality section? Name the exceptions instead of alluding to them: suspected abuse or neglect of a child or vulnerable adult, a serious threat of harm to self or others, court orders, professional consultation, and what an insurer receives when you bill (diagnosis and dates of service). Reporting thresholds are state law and vary, so match the wording to your state's statutes rather than a national template. A vague limits clause is the consent failure that most often surfaces in board discipline. Psychotherapy notes sit outside this discussion entirely; they carry their own authorization rules . How long should a consent form be, and at what reading level? No regulation sets either one. Most forms run 800 to 2,000 words; past that, comprehension is the casualty. The readability research points one direction: Handelsman and Martin (1992) found recall improved when consent forms were more readable, and a 1993 analog study found clients rated therapists who used informed consent as more expert and trustworthy. Plain language is not a style preference here: a form the client could not understand documents a disclosure that did not really happen. How long do I need to keep signed consent forms? As long as the clinical record they belong to. In the US that is state law, commonly 7 to 10 years for adults and longer for minors; HIPAA's 6-year retention rule covers privacy documentation such as the NPP acknowledgment effort, not the chart. Canadian periods vary more than two to one by province: 10 years in Ontario and Alberta, 7 in British Columbia, 5 in Quebec. Australian state rules commonly run 7 years for adults and to age 25 for records of minors. Check your board's current standard before setting a destruction schedule. I am pre-licensed. Whose name goes on the consent form? Both names, and the client must know which is which. APA Standard 10.01(c) requires that a client of a trainee be told the therapist is supervised and be given the supervisor's name, and counseling and Canadian standards expect the same disclosure. The supervisor typically carries legal responsibility for the care. A form that hides associate status reads to a board like concealment, and practicing without disclosing trainee status is one of the consent failures that shows up in discipline cases. Can BastionGPT draft an informed consent form for my practice? Yes. Give it bullets about your practice (state, discipline, modality, communication policies, how fees are handled) and it drafts a complete plain-language form. It can also rewrite your current form at an easier reading level without dropping a disclosure, and check a draft for missing elements such as alternatives, named confidentiality limits, or the supervision line. Confirm state-specific wording against your board's rules before use. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Psychotherapy Notes Authorization Consent for Treatment of a Minor Financial Policy & Fee Agreement Telehealth Consent HIPAA Notice of Privacy Practices Acknowledgment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/telehealth-consent All note templates # Telehealth Consent: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A telehealth consent form documents that a client agreed to receive care by video or phone after learning the modality's risks, privacy limits, emergency procedures, and technology-failure backup plan. Clinicians obtain and document it before or at the first virtual session. Most US states require some form of telehealth consent by statute, regulation, or Medicaid policy, and documented verbal consent usually satisfies them. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, counselors, and any clinician delivering care by video or phone; trainees under supervision Audience The client or guardian and the virtual-care provider; later, payers, auditors, and licensing boards Typical length 400 to 800 words · 5 to 10 minutes to review with the client (clinical team estimate) Format family Short consent form (compare: informed consent for treatment, NPP acknowledgment, release of information) When it's used Before or at the first telehealth session; revisited when the platform, modality, or client location materially changes Standards context Required by law or Medicaid policy in most US states; board standards govern in Canada and Australia; a standalone signed form is a convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a telehealth consent form? A telehealth consent form documents that, before care began by video or phone, the clinician explained the modality, its risks and limits, the privacy arrangements, and the plan for emergencies and technology failure, and the client agreed to proceed. No single body authored it. The closest thing to a shared lineage in mental health is Guideline 4 of the Guidelines for the Practice of Telepsychology, developed by the joint task force of the American Psychological Association, the Association of State and Provincial Psychology Boards, and the APA Insurance Trust, adopted July 31, 2013 and revised in 2024, alongside state statutes led by California Business and Professions Code section 2290.5, which began as the Telemedicine Development Act of 1996, a written-consent mandate, and reached its current verbal-or-written, documented standard with AB 809 in 2014. You will also see the document called a telemedicine consent, telepsychology consent, teletherapy consent, telemental health consent, or virtual care consent. The disambiguation that matters: in nearly every jurisdiction the legal event is obtaining and documenting consent, not collecting a signature, and no jurisdiction requires a standalone telehealth form as a blanket rule. A general informed consent for treatment that specifically names telehealth is expressly sufficient in California Medi-Cal, New York, and Indiana. The telehealth consent is also not a privacy document: the Notice of Privacy Practices acknowledgment records receipt of privacy practices, while this form records agreement to a modality. And the most repeated myth runs the other way: traditional Medicare requires no consent at all for standard telehealth video visits; its annual verbal-consent rule applies only to communication technology-based services such as virtual check-ins and e-visits. ## Who uses telehealth consent forms and when Everyone delivering virtual behavioral health keeps one: solo therapists and psychologists, group practices, telepsychiatry services, hospital and health-system virtual care programs, and platform-based practices. The share of psychologists' clinical work done by telepsychology rose from about 7 percent before the pandemic to more than 85 percent during it (Pierce and colleagues, American Psychologist, 2021), and the consent form followed the migration. Obtain and document consent before or at the first telehealth session, then revisit it when something material changes: a new platform with different privacy characteristics, a shift from video to audio-only, or a client in a new state. For minors, the guardian consents and the minor assents, with the same telehealth disclosures layered onto the minor consent . This form records the agreement to the modality; the per-session clinical record lives in the telehealth therapy note , which is where each session's modality, platform, and client location belong. ## Telehealth consent form structure: what goes in each section Modality and platform. Name the service types offered (video, phone when clinically appropriate) and the specific platform, with a line on when in-person care is still expected. Pitfall: a form that says "video or phone" but never names the platform; Ontario's CPSO expects fresh consent when the technology materially changes, and an unnamed platform leaves nothing to compare a change against. Risks and limits of telehealth. State the tradeoffs in behavioral health terms: possible technology failure mid-session, reduced nonverbal cues, and situations (crisis, some assessments) that call for in-person care. Pitfall: pasting general telemedicine risk language about the absence of a physical exam while skipping the risks that matter in psychotherapy, such as an interrupted session during emotionally intense work. Privacy and the home setting. Cover both ends of the call: what the practice does (encrypted platform, business associate agreement in place, no session recording) and what the client controls (a private room, headphones, who else is home). Standard confidentiality limits still apply and belong here. Pitfall: a paragraph on encryption with silence about the client's own room; household members overhearing a session is a more common privacy failure than an intercepted video stream. Identity and location verification at each session. The form commits both parties to a start-of-session routine: confirm identity, confirm the client's physical address today, and confirm the client can speak privately. Location drives licensure and emergency response. Pitfall: treating location as intake data; a client who quietly takes sessions from another state puts the clinician outside licensure, a problem no consent form cures after the fact. Emergency plan and local resources. Record an emergency contact who can physically reach the client, the client's local emergency number, and the nearest emergency department, and state when the clinician will use them. Pitfall: a plan that reads "call 911"; a clinician dialing 911 reaches dispatch near the clinician, not the client, so the plan has to run through the client's local resources and named contact. Technology-failure backup plan. Script the failure: who calls whom, on which number, after how many minutes, and what happens if reconnection fails (reschedule, switch to phone, move to the emergency plan). Pitfall: no ordering of steps, which turns a dropped connection with a distressed client into an improvised judgment call and, afterward, an undocumentable one. Recording and AI tools. State the default (no recording by either party) and that any recording, including AI scribe or transcription processing, needs its own separate consent. Pitfall: stretching the telehealth consent to cover recording; recording consent is a distinct object under the APA Ethics Code (Standard 4.03) and many state Medicaid rules, and some states add all-party consent under recording law. Right to refuse or withdraw. The client can decline telehealth, withdraw consent at any time, and ask about in-person alternatives without penalty. California ties this right directly to its consent statute, and Kentucky's telepsychology rule requires the right to refuse plus a description of alternatives. Pitfall: framing telehealth as the only way to access the practice; the withdrawal right is one of the disclosures state rules name explicitly. Licensure line, consent statement, and documentation. Close with the clinician's license number and state(s), a line that services are available only while the client is physically in a jurisdiction where the clinician holds a license, and the consent statement itself: a written signature or a note that verbal consent was given, by whom, and on what date. Pitfall: a filed signature with no documented consent event; California's standard is that consent, verbal or written, "shall be documented", and a bare signature page shows agreement without the required conversation. ### Blank template (copy and adapt) TELEHEALTH CONSENT FORM Client: ____________________________ DOB: ____________ Clinician: _________________________ License # and state(s): ____________ Practice phone: ____________________ Portal/email: ______________________ MODALITY AND PLATFORM Services offered by: [ ] secure video [ ] phone (when appropriate) Platform: __________________________ RISKS AND LIMITS I understand telehealth carries risks, including technology failure, interruption, and reduced nonverbal cues, and that some situations may require in-person or local care instead. PRIVACY The practice uses an encrypted platform and does not record sessions. Privacy also depends on my own space and device. The standard limits of confidentiality still apply. EACH SESSION My clinician will confirm my identity, my physical location (address), and that I can speak privately before we begin. EMERGENCY PLAN Emergency contact (name and phone): _____________________________________ Local emergency number: ___________ Nearest ER: ________________________ Local crisis line: ______________________________________________________ IF TECHNOLOGY FAILS If the connection drops, my clinician will call me at: __________________ If we cannot reconnect within ____ minutes, the plan is: ________________ RECORDING Neither party records sessions. Any recording or AI processing would require my separate written consent. LOCATION AND LICENSURE Services are available while I am physically located in: ________________ I will tell my clinician before joining from a different state, province, or country. AGREEMENT I have had the chance to ask questions. I may decline or withdraw from telehealth at any time and ask about in-person options. Consent: [ ] written (signed below) [ ] verbal, documented in the record Client/guardian signature: ___________________ Date: ___________ Clinician signature: _________________________ Date: ___________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample telehealth consent form Scenario: a licensed marriage and family therapist in private practice completes telehealth onboarding with a new adult client who will attend weekly video sessions from home. All details are fictional. Telehealth Consent. Client: R.L., 29 · Clinician: D. Okafor, LMFT (Colorado license 1847) · Completed: 08/04/2026, reviewed together by video Modality and platform: Weekly individual psychotherapy by secure video on ClearPath Telehealth, with phone backup at the cell number on file. In-person sessions remain available at the Lakewood office by arrangement. Risks and limits: Reviewed possible connection failure, interruption, and reduced nonverbal cues. R.L. understands that a crisis may require local, in-person, or emergency care rather than video. Privacy: Platform is encrypted with a business associate agreement in place; neither party records sessions. R.L. will join from a home office with a closed door and headphones, and will reschedule rather than hold a session with others in the room. Standard confidentiality limits reviewed and unchanged. Each session: Clinician confirms identity, street address, and ability to speak privately at the start of every session. Emergency plan: Emergency contact: T.L. (sister), (555) 014-2276, aware she is listed. Local emergency number: 911. Nearest emergency department: Riverview Medical Center, about 4 miles from home. Crisis line 988 saved in R.L.'s phone. If safety requires, clinician will contact emergency services near the address confirmed at the start of session. If technology fails: Clinician calls R.L.'s cell within 5 minutes of a dropped connection. If not reconnected by 15 minutes, the session is rescheduled. Any safety concern moves straight to the emergency plan. Recording: No recording by either party. Any future recording or AI processing requires separate written consent. Location and licensure: Services are provided while R.L. is physically in Colorado, where D. Okafor is licensed. R.L. will give notice before joining from any other state; those sessions may need to be rescheduled. Agreement: Questions invited and answered (R.L. asked about work travel; in-state trips are fine, out-of-state trips need advance notice). R.L. may decline or withdraw from telehealth at any time and ask about in-person options. Signed electronically by R.L. 08/04/2026; countersigned D. Okafor, LMFT, 08/04/2026; review documented in the chart. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The consent event is documented, not just signed. The record shows who reviewed the form, how (together by video), the client's question, and the date, which is what verbal-or-written consent statutes actually test. The emergency plan names a person, a facility, and a location. A named contact who knows she is listed, the nearest emergency department, and a per-session address check give the clinician something to act on from another city. The technology-failure plan has an order of operations. Who calls whom, on which number, after how long, and when the emergency plan takes over; a dropped call follows a script instead of a guess. The licensure line anticipates travel. Consent is tied to the client's physical presence in a licensed state, with advance notice for exceptions, the constraint that binds hardest when clients move around. Recording and AI tools are closed off explicitly. The default is no recording, and any future recording or scribe processing is routed to its own separate consent instead of riding on this one. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Sort the rules by strength before you build the form. In the United States, the consent event itself is LAW in most states: the Center for Connected Health Policy reports that the majority of states put "some sort of telehealth specific informed consent requirement" in statute, administrative code, or Medicaid policy (CCHP publishes no numeric tally; commonly cited secondary counts land at 45 states plus the District of Columbia). California's Business and Professions Code section 2290.5 is the model: verbal or written consent, documented, and noncompliance "shall constitute unprofessional conduct". PAYER POLICY is a separate layer: some state Medicaid programs require written consent (Alabama and Georgia among them), and traditional Medicare requires consent only for communication technology-based services, once a year and verbal, with the cost-sharing discussion documented. Everything else is CONVENTION: the standalone form, the annual re-consent ritual, and the signature itself are practice choices rather than legal requirements, and a few states (Florida, Hawaii, and Illinois among them) have no telehealth-specific consent rule at all, leaving general consent law in charge. Outside the US, board standards carry the weight. Ontario's CRPO electronic practice standard (updated effective January 1, 2024) requires consent specific to electronic services, and Ontario's CPSO adds the rare explicit answer on platform switches: re-consent when the modality or technology materially changes (more in the Canada telepsychology note ). Australia's AHPRA telehealth guidance holds virtual care to the in-person standard, and the Psychology Board of Australia Code of Conduct, effective December 1, 2025, treats consent as an ongoing process to revisit after a reasonable period, widely read as about 12 months: the closest thing to an annual rule in any of the three countries. Wherever you practice, the consent record is ordinary clinical-record material: clients can request it, disclosure to third parties runs through a release of information , and retention follows general record law rather than any telehealth-specific clock. If a client carries elevated risk, align the consent's emergency section with the safety plan so the address check, contacts, and local resources match instead of drifting apart. ↑ Back to the template and downloads ## Common telehealth consent form errors auditors flag Read the audit record carefully, because the famous telehealth numbers are not consent findings. The HHS Office of Inspector General's South Carolina Medicaid telemedicine audit (A-04-18-00122, April 2020) found 97 of 100 sampled payments unallowable and recommended a $1,524,536 federal refund, but for 95 of those claims the failure was missing start and stop times or missing consulting-site location, not consent; vendor pages recycle it as a consent statistic anyway. The parallel 2018 Medicare telehealth audit failed 31 of 100 sampled claims on originating-site and eligibility grounds. There is also no denial code specific to telehealth consent: a consent-shaped denial routes through generic codes such as CARC 16 with an N-series remark. Where consent failures actually surface is licensing exposure, since statutes like California's make undocumented consent unprofessional conduct. The BastionGPT Clinical Advisory Board sees the same errors most often in telehealth consent form reviews: A signature on file, no consent event in the record. The chart holds a signed PDF but never says who obtained consent, when, how (verbal or written), and what was discussed; documented-consent statutes test the event, not the artifact. Location verified once, at intake. The client travels, the record cannot show which state any session came from, and the clinician cannot show sessions stayed inside licensed jurisdictions; site and location documentation is exactly what the OIG telemedicine audits flagged. An emergency plan built for an office. "Call 911" with no per-session address confirmation, no local emergency department, and no named contact who can physically reach the client. Medicare rules imported where they do not apply. Annual re-consent rituals copied from the virtual check-in rules onto standard video visits, while the coverage condition that does apply since October 1, 2025 (an in-person visit within the 6 months before the first telemental health service, then at least annually, with exceptions) goes untracked. The consent stretched over recording and AI tools. Session recording or scribe processing treated as covered by the telehealth consent, when it needs its own separate consent and, in some states, all-party agreement under recording law. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on telehealth consent forms. Draft a complete telehealth consent from your practice details: platform, modalities, backup steps, emergency workflow, and the states or provinces you serve, ready for attorney review. Rewrite dense consent language in plain English; analyses of US consent forms find most are written above recommended reading levels, and BastionGPT brings the level down while keeping every disclosure intact. Check a finished form before you file it: licensure and location line, emergency plan completeness, technology-failure steps, recording clause, and the documented-consent line. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does a telehealth consent form need a signature, or is verbal consent enough? In most US states, documented verbal consent is legally sufficient: California, Arizona, Texas, and New York all accept it, and Canadian college standards and Australian board guidance do the same. Write down who consented, when, and how. Written consent is the exception, required by a minority of state Medicaid programs (Alabama, Georgia, and Idaho among them) and in some recording situations. A signature is belt and suspenders, not a universal legal requirement. Does Medicare require telehealth consent? Not for standard telehealth video visits. The annual verbal-consent requirement applies only to communication technology-based services such as virtual check-ins and e-visits, where cost sharing must be disclosed; that consent can be collected by auxiliary staff under general supervision. See the CMS Telehealth Services fact sheet . Separately, since October 1, 2025 Medicare telemental health carries an in-person visit requirement (within the 6 months before the first telehealth service, then at least annually, with exceptions); that is a coverage condition, not a consent rule, but it belongs on the same compliance checklist. Do I need a separate telehealth consent form, or can my intake consent cover it? No jurisdiction requires a separate document as a blanket rule. A general informed consent for treatment that specifically names telehealth and includes the required disclosures is expressly sufficient in California Medi-Cal, New York, and Indiana. Use a standalone form when a payer requires a discrete telehealth acknowledgment or when you want a single auditable record of the telehealth disclosures. Do I need to renew telehealth consent every year? No US telehealth statute requires annual re-consent. The annual idea is imported from two narrower rules: Medicare's once-a-year verbal consent for virtual check-ins and e-visits, and Australia's Psychology Board Code of Conduct (effective December 1, 2025), which treats consent as ongoing and worth revisiting after about 12 months. The trigger that matters everywhere is material change: a new platform with different privacy characteristics, a switch from video to audio-only, or a client in a new jurisdiction. Ontario's CPSO makes re-consent on material change an explicit requirement. What if my client travels or moves to another state? Licensure, not consent, is the binding constraint: you generally must be licensed where the client is physically located at the time of the session. Some states offer short windows (Arizona allows up to 90 days per year of practice by out-of-state providers; Utah has a 45-day transitional allowance), and a new state can also mean new consent reflecting that state's disclosures. Confirm and document the client's location at each session, and have clients agree in the consent to give notice before joining from somewhere new. Do audio-only phone sessions need their own consent? Address audio-only explicitly rather than assuming video language covers it. Texas Medicaid requires documented consent (verbal is fine) before audio-only behavioral health services, Medicare expects the practice to have video capability and a documented reason audio-only was used, and Kentucky's telepsychology rule excludes routine audio-only telephone care except in limited-connectivity situations. A line in the consent naming phone sessions, when they are used, and their added limits covers most of this. Does HIPAA require a telehealth consent form? No. Under 45 CFR 164.506 , consent to use health information for treatment, payment, and operations is optional, and HIPAA imposes no telehealth-consent duty. HIPAA's role is different: it governs how the platform and the records are secured, which is why the Notice of Privacy Practices acknowledgment and the telehealth consent are separate documents. Telehealth consent requirements live in state law, Medicaid policy, and licensing board standards. How long do I keep telehealth consent forms? As long as the clinical record they belong to. Retention comes from general record law, not telehealth rules: Texas, for example, requires telehealth records including consent be kept at least 5 years , or the program's required period if longer, and state and provincial minimums elsewhere often run longer. HIPAA's six-year clock covers compliance documentation such as policies, not the chart, so do not treat it as your retention rule. Can BastionGPT write my telehealth consent form? Yes. Give it your platform, modalities, emergency workflow, and the states or provinces you serve, and it drafts the form, rewrites it in plain language, and checks a finished version for the disclosures reviewers look for: location and licensure line, emergency plan, backup steps, recording clause, and the documented-consent line. Have your attorney or board resources confirm state-specific wording. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Telehealth Psychotherapy Note Release of Information (ROI) Authorization Consent for Treatment of a Minor Informed Consent for Treatment HIPAA Notice of Privacy Practices Acknowledgment Canada Telepsychology & Virtual Care Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/npp-acknowledgment All note templates # HIPAA Notice of Privacy Practices Acknowledgment: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A Notice of Privacy Practices acknowledgment is the intake form that documents a patient received a provider's HIPAA privacy notice. Front desk staff collect it at or before first service delivery. Federal law requires a good-faith effort to obtain it, not a patient signature, and refusal to sign never blocks treatment. Most forms run 100 to 250 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Front desk and intake staff; the patient or personal representative provides the acknowledgment Audience Practice compliance file, privacy officer, OCR reviewers, patient or personal representative Typical length 100 to 250 words · 1 to 3 minutes by hand (clinical team estimate) Format family Administrative intake form (compare: informed consent, release of information, Part 2 consent) When it's used At or before first service delivery, once per patient; deferred in emergencies Standards context US-specific: the good-faith effort is law (45 CFR 164.520), the signature is convention; no signed-receipt equivalent abroad On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a Notice of Privacy Practices acknowledgment? A Notice of Privacy Practices acknowledgment (also called an NPP acknowledgment, acknowledgment of receipt, HIPAA acknowledgment form, or informally the HIPAA sign-off) is the intake record showing that a patient received a covered provider's HIPAA privacy notice. It came from HHS rulemaking, not from a professional association: the original December 2000 Privacy Rule required providers to obtain prior written consent for routine uses of health information, and the August 2002 modification replaced that consent requirement with today's lighter standard. Under 45 CFR 164.520(c)(2)(ii) , a provider with a direct treatment relationship must make a good-faith effort to obtain a written acknowledgment of receipt no later than first service delivery, and when it cannot, document the effort and the reason. The disambiguation that matters: this form is not the notice, and it is not consent. HHS wrote in the August 2002 final rule that "the final Rule does not require an individual's signature" on the notice. The patient has no legal duty to sign, a refusal never blocks treatment, and the completed form authorizes nothing: a disclosure outside treatment, payment, and operations still needs a release of information , and agreement to the treatment itself lives in the separate informed consent . What the law demands from the practice is the effort and, when the signature does not happen, the documentation. ## Who uses NPP acknowledgments and when Every US practice that qualifies as a HIPAA covered entity collects this form: group practices, community mental health centers, hospital outpatient clinics, and any solo therapist who bills insurance electronically. The person handling it is usually front desk or intake staff, not the clinician; it is an administrative record, and supervision rules for pre-licensed clinicians do not attach to it. It appears once per patient, at or before the first visit, inside the intake packet next to the informed consent and, where care is virtual, the telehealth consent . Cash-pay practices that never send an electronic claim or eligibility check are generally not covered entities and do not need an NPP or an acknowledgment at all, though state privacy law still applies to them. ## NPP acknowledgment structure: what goes in each section Practice header and notice version. The practice or provider name and the effective date of the NPP the patient received. No regulation requires naming the version, but the form exists to prove timing, and a receipt that cannot say which notice was received proves very little. Pitfall: handing out a form that points at an outdated notice; after the February 16, 2026 Part 2 content update, an acknowledgment referencing a 2019 notice advertises that the notice itself is stale. Acknowledgment statement. One or two plain sentences: the patient received, or was offered, a copy of the Notice of Privacy Practices. Keep it readable; the only published study of HIPAA privacy notices (Hochhauser, 2003) measured them at a grade 14.5 reading level, and this statement is the one piece of the paperwork you fully control. Pitfall: consent-flavored wording such as "I agree to the uses described in the notice"; the 2002 rule deliberately removed consent from this transaction, and agreement language turns a receipt record into a document it is not. Patient identification and signature block. Patient name, date of birth, signature, and date signed. The signature is a convention sitting on top of a law that never asks for one. Pitfall: building the workflow so the form is only complete when signed; the legally required artifact is the good-faith effort, and a signature-or-nothing form fails on exactly the patients who decline. Personal representative section. Name, signature, and relationship or authority when someone other than the patient acknowledges: a parent for most minors, a guardian, a health care agent. HHS guidance treats providing the notice to the personal representative as satisfying the distribution duty. Pitfall: a parent signature with no relationship recorded; state minor-consent laws can shift who the personal representative is for specific services, and the form should show whose authority was relied on. Staff good-faith-effort block. The section that does the legal work when there is no signature: when and how the notice was provided or offered, by whom, and the reason no acknowledgment was obtained, with checkboxes for the common reasons (declined to sign, emergency at first service, no response to a mailed or portal notice, other). Pitfall: leaving this block blank behind an unsigned form; an unsigned acknowledgment with no documented effort is the only compliance failure this form can produce. Filing and retention line. Where the form lives (the administrative record, not the clinical note) and for how long: six years from creation or the date last in effect under 45 CFR 164.530(j)(2). Pitfall: routing the form to a supervisor for co-signature; it is not a clinical document, and no supervision rule asks for one. ### Blank template (copy and adapt) NOTICE OF PRIVACY PRACTICES: ACKNOWLEDGMENT OF RECEIPT Practice/provider: ______________________ NPP effective date: ________ I acknowledge that I received, or was offered, a copy of this practice's Notice of Privacy Practices, which describes how my health information may be used and disclosed and how I can access it. Patient name: ___________________________ DOB: __________ Patient signature: ______________________ Date: __________ Personal representative (if applicable): _______________________ Relationship / authority: ______________________________________ Signature: ______________________________ Date: __________ STAFF USE ONLY: GOOD-FAITH EFFORT (complete if no signature above) Notice provided/offered on: __________ by: ____________________ Method: [ ] in person [ ] mail [ ] portal/email [ ] other Acknowledgment not obtained because: [ ] Declined to sign [ ] Emergency at first service [ ] No response after attempt(s) [ ] Other: ___________________ Notes: _________________________________________________________ Staff signature/credentials/date: ______________________________ File to the administrative record. Retain 6 years from creation (45 CFR 164.530(j)(2)). Treatment is never conditioned on signing. Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample NPP acknowledgment Scenario: first intake visit at a small group practice. The client takes a copy of the notice but prefers not to sign, and the front desk documents the good-faith effort. All details are fictional. Notice of Privacy Practices: Acknowledgment of Receipt. Practice: Riverbend Counseling Associates · NPP effective date: 02/16/2026 · Form date: 08/04/2026 Acknowledgment: I acknowledge that I received, or was offered, a copy of the Riverbend Counseling Associates Notice of Privacy Practices (effective 02/16/2026), which describes how my health information may be used and disclosed and how I can get access to it. Patient: D.K., DOB 03/14/1989. Patient signature: not signed. Date: blank. Personal representative: not applicable. Staff use only, good-faith effort: Notice provided in person at check-in on 08/04/2026 by A. Romero, patient services coordinator. Method: in person, printed copy. Acknowledgment not obtained: declined to sign. Client stated he prefers not to sign forms he has not read at home and took the printed copy with him. Not an emergency visit. Intake session proceeded as scheduled with L. Tran, LCSW. Filing: Scanned to the administrative record 08/04/2026. Retain through 08/2032 (six years from creation, 45 CFR 164.530(j)(2)). Staff signature: A. Romero, patient services coordinator, 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The good-faith effort is the record. Date, method, named staff member, and reason: with no patient signature anywhere, this form still fully satisfies 45 CFR 164.520(c)(2)(ii). The refusal reason is specific. "Declined to sign" plus what the client actually said and did reads as a real event, not a checkbox filled in after the fact. Care visibly went ahead. The form records that the intake session proceeded on schedule, which is documented proof that treatment was never conditioned on signing. The notice version is pinned. Tying receipt to the 02/16/2026 notice shows the practice distributed its Part 2 updated NPP and can prove first-service timing years later. Retention is computed, not implied. A keep-until date six years out matches the 164.530(j)(2) floor without anyone doing math during a review. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations File the acknowledgment as administrative intake documentation, not clinical content. It supports no billing code, appears on no claim, and no denial or remark code exists for it, because it is not a billing document. It needs no clinician author and no supervisory co-signature. The federal retention floor is six years from creation or the date last in effect (45 CFR 164.530(j)(2)); your state's clinical-record rules govern the chart, but this document answers to HIPAA's documentation rule. Electronic workflows count: HHS treats an electronic return transmission, such as a portal checkbox, as a valid written acknowledgment, and for a first contact that happens by phone or video, HHS FAQ 336 says mailing the notice with a return form the same day is enough: keeping "a file copy of the form sent to the patient" documents the good-faith effort even if nothing comes back. Fold the acknowledgment into the same packet as your telehealth consent for virtual-first intakes, and remember that state privacy law can add intake-documentation expectations on top of HIPAA. Keep the legal layers straight, because they run in an unusual direction here. LAW: provide the notice by first service delivery, make a good-faith effort to obtain written acknowledgment, document the effort and reason when you cannot, and retain for six years; in an emergency, the effort is excused until reasonably practicable. CONVENTION: the signature itself, the patient-name and date lines, and the notice-version reference; no regulation prescribes the form's format or wording at all. There is no payer-policy layer, which is rare in this library. Enforcement follows the same shape: OCR's published penalties around the NPP concern notice content and distribution, never the acknowledgment provision, and OCR even proposed eliminating the written acknowledgment in a January 2021 proposed rule that remains unfinalized as of August 2026, so the requirement still stands. The live obligation is different: NPPs needed 42 CFR Part 2 content by February 16, 2026, a deadline that survived Purl v. HHS, so pair a current notice with this acknowledgment and with your Part 2 consent where substance use disorder records are in play. The notice also describes rights patients may exercise later through a records request or a record amendment . ## Common NPP acknowledgment errors auditors flag The enforcement record here is a finding in itself. In OCR's Phase 2 desk audits (2016 to 2017, reported December 2020), only 2% of the 166 audited covered entities fully met the NPP content requirements and 98% were missing required content, while the acknowledgment provision was not audited at all. Across all published OCR resolution agreements and civil monetary penalties, 45 CFR 164.520(c)(2)(ii), the acknowledgment rule, has never been the basis for an enforcement action. The risk sits in the notice, not the receipt. The BastionGPT Clinical Advisory Board sees the same errors most often in NPP acknowledgment reviews: Treating the signature as mandatory. Front desk scripts that will not release the clipboard until the patient signs. The law asks for a good-faith effort, and the American Dental Association's HIPAA guidance puts it plainly: "You cannot refuse treatment solely because a patient refuses to sign." An unsigned form with a blank staff block. No signature plus no documented effort and reason is the one failure this form can produce. The refusal costs nothing; the missing documentation is the violation. Re-collecting every year. The acknowledgment is a one-time, first-service obligation. When the notice changes, redistribute and repost the revised notice; nothing requires new signatures from existing patients. Consent creep. Acknowledgment wording that says "I agree," or a form pulled out later to justify a disclosure. It authorizes nothing: a disclosure outside treatment, payment, and operations needs a release of information , and HIPAA bars combining the notice with an authorization in one document. A perfect acknowledgment stapled to a failing notice. Practices polish the receipt form while the NPP itself is missing required content, the element OCR audits and the one that carried a live February 16, 2026 Part 2 deadline. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on notice of privacy practices acknowledgment forms. Turn a one-line front-desk recap into a complete good-faith-effort entry: date, delivery method, staff name, and a specific reason, ready to file behind an unsigned form. Check your acknowledgment against what the rule actually asks: receipt-only wording with no consent language, a documented-effort path for refusals, a personal-representative line, and the six-year retention note. Rewrite your acknowledgment statement and notice summaries at a grade 6 to 8 reading level; the only published study of HIPAA privacy notices measured them at grade 14.5. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does the patient have to sign the NPP acknowledgment form? No. Federal law (45 CFR 164.520(c)(2)(ii)) requires the practice to make a good-faith effort to obtain a written acknowledgment, and to document the effort and the reason when it is not obtained. HHS said it directly in the August 2002 final rule: "the final Rule does not require an individual's signature." The signature line is convention; the documented effort is the requirement. What if a client refuses to sign? Treat them exactly as planned. Record the good-faith effort: when and how the notice was provided, by whom, and the reason, such as "declined to sign." That documentation fully satisfies the rule. Refusal has no consequence for the client, and conditioning treatment on a signature is the one way a practice turns this form into a compliance problem. Do I need a new acknowledgment every year, or when I update my notice? No. It is a one-time obligation tied to first service delivery. When the NPP changes, as it did for the February 16, 2026 Part 2 content update, you redistribute and repost the revised notice; HHS guidance does not require re-collecting acknowledgments from existing patients on revision or on any schedule. Does a patient-portal checkbox count as a written acknowledgment? Yes. HHS treats an electronic return transmission from the individual as a valid written acknowledgment, so a portal checkbox or e-signature qualifies. For phone or telehealth first contacts, sending the notice the same day and keeping a file copy of the acknowledgment form documents the good-faith effort even if the patient never returns it. Is the acknowledgment the same as consent or an authorization? No, and the difference has teeth. The acknowledgment records receipt of a notice and authorizes nothing. Agreement to treatment lives in the informed consent . Permission to disclose records outside treatment, payment, and operations requires an authorization, the release of information , with its own required elements. The original 2000 Privacy Rule did require consent for routine uses; the 2002 amendment replaced it with this acknowledgment. Do cash-pay therapists need an NPP acknowledgment? Often not. HIPAA covers providers who transmit standard transactions electronically, such as e-claims or electronic eligibility checks. A cash-pay practice that never does is not a covered entity and is not required to maintain an NPP or collect acknowledgments. One electronic standard transaction changes that for the whole practice, and state privacy law applies regardless. Who signs for a minor client? The personal representative, usually a parent or guardian, receives the notice and provides the acknowledgment. State minor-consent laws complicate this: where a minor can lawfully consent to a service themselves, the minor may be the one who acknowledges for that care. Record the representative's relationship and authority on the form. How long do we keep acknowledgment forms? Six years from creation or the date last in effect, under HIPAA's documentation rule (45 CFR 164.530(j)(2)). That federal floor applies to this administrative paperwork, including good-faith-effort documentation for patients who never signed. Your clinical records follow state retention rules, which often run longer. How does BastionGPT help with NPP acknowledgments? Give it a one-line front-desk recap and it drafts the good-faith-effort entry with the date, method, staff, and reason. It can also check your intake packet for consent language hiding in the acknowledgment, a missing personal-representative line, or a stale notice date. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Release of Information (ROI) Authorization Records Request & Chart Copy Packet Record Amendment & Correction Response 42 CFR Part 2 Consent Informed Consent for Treatment Telehealth Consent Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/diagnosis-list-update All note templates # Diagnosis & Problem List Update (ICD): Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A diagnosis and problem list update is the dated, authored entry a clinician makes when adding, refining, correcting, or resolving a diagnosis in the medical record, including the ICD-10-CM code that flows onto claims. Clinicians, coders, and CDI teams use it whenever the active diagnosis list changes. Most entries run 50 to 200 words: a diagnostic statement plus the clinical support for it. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The treating clinician who establishes or changes the diagnosis; coders and CDI staff query rather than author Audience Treating clinician, care team, coders and CDI staff, payers, auditors, and downstream EHR systems Typical length 50 to 200 words · 3 to 8 minutes by hand (clinical team estimate) Format family Structured list plus rationale (compare: amendment entry, progress note assessment) When it's used Whenever a diagnosis is added, refined, corrected, or resolved on the problem list or a claim Standards context ICD-10-CM (US, adopted under HIPAA), ICD-10-CA (Canada), ICD-10-AM (Australia); the amendment format is payer policy, not statute On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a diagnosis and problem list update? A diagnosis and problem list update is the entry a clinician makes when a diagnosis is added, made more specific, corrected, or resolved in the patient record, along with the code change that follows it onto claims. It is not a separately standardized legal document in any country. It is a functional entry that sits on two older document families: the problem list, introduced by Lawrence Weed's problem-oriented medical record (first published in 1964, brought to prominence in 1968, the same framework that produced the SOAP note), and the amendment, addendum, and correction entry. In the US, the problem list entered federal policy through the HITECH-era meaningful use program, where CMS defined it as "a list of current and active diagnoses" and paid clinicians to keep it current. You will also hear the entry called a problem list reconciliation, diagnosis addendum, dx change, diagnosis clarification, provider query response, or, in risk-adjustment circles, HCC recapture documentation; Ontario's Cumulative Patient Profile is the provincial cousin. The distinction that carries this page: the amendment rules everyone treats as law are not law. In the US, the controlling instrument for changing a diagnosis entry is the Medicare Program Integrity Manual, and the CMS transmittal behind it states plainly: "There are no regulatory, legislative, or statutory requirements related to this CR." What it requires, as payer policy enforced through audit, are three properties for any amendment, correction, or delayed entry: identify it as such, date and author it, and preserve the original content without deletion. Two things are stronger than policy. The ICD-10-CM code set itself is adopted under HIPAA, which is law, and coding convention (ICD-10-CM Guideline I.A.19) bases code assignment on "the provider's diagnostic statement that the condition exists," which is why the written statement, not the EHR picklist, is the load-bearing element. The update entry is also distinct from the psychiatric diagnostic evaluation (90791, 90792), the comprehensive service that first establishes a diagnosis: choose that when a new episode begins, and this entry when an existing diagnosis changes. ## Who uses diagnosis and problem list updates and when Anyone who maintains a diagnosis list writes them: physicians and psychiatric nurse practitioners in prescribing practices, psychologists and therapists in insurance-based care, primary care teams managing behavioral health conditions, and the coding, CDI, and HIM staff who query providers in hospitals. In behavioral health the trigger is usually one of three events: an unspecified intake diagnosis reaches full criteria and gains specificity, a scheduled review changes status (when scores drive the change, pair the entry with an outcome measure note ), or a payer document needs the current diagnosis restated, from the billing note that ties a service to it to the treatment plan built on it. The update is a record of fact about the diagnosis itself; forward-looking changes to goals and services belong in the plan, and the initial diagnostic case belongs in the evaluation. ## Diagnosis and problem list update structure: what goes in each section Entry header and type. Date of entry, author with credentials, and what kind of entry this is: new diagnosis, refinement, correction or amendment, delayed entry, status change, or resolution. Under the CMS amendment principles the label is the first requirement; a change that is not identified as a change fails before its content is read. Pitfall: editing an earlier note in place. The EHR audit trail will show it, and an unlabeled alteration reads as concealment rather than correction. Diagnostic statement. The diagnosis in the provider's own words, with severity, episode, and course specifiers. This sentence, not the picklist, is what codes are assigned from: ICD-10-CM Guideline I.A.19 bases assignment on the provider's statement that the condition exists. Pitfall: a code with no statement behind it. A bare F-code on the problem list gives a coder nothing to assign from and an auditor nothing to verify. Onset, status, and rule-outs. When the condition began or was established, whether it is active, improving, in remission, or resolved, and any rule-outs still pending. Pitfall: rule-out language left on the list so long it hardens into the diagnosis. A rule-out is a question, and the record should show when it was answered. Clinical support at this encounter. For any diagnosis carried on the day's claim, evidence it was monitored, evaluated, assessed, or treated that day: a score, an examination finding, a medication decision. Pitfall: the copied-forward chronic condition. The recurring root cause OIG cites in risk-adjustment audits is past conditions reported as current, and a diagnosis with no current support is that exact pattern. Coding mapping. The ICD-10-CM code (or the code your coder will assign), the code it replaces, and the effective dates of each. In behavioral health, DSM-5-TR labels map to ICD-10-CM codes, and the claim needs the ICD-10-CM code. Pitfall: leaving an unspecified code such as F41.9 after the documentation supports specificity; specificity is where medical necessity and risk adjustment are decided. Problem list action. What happened to the list itself: added, status changed, resolved with a date, or moved to history with the applicable Z-code. Keep the list a list of current and active diagnoses, which is the definition federal policy gave it. Pitfall: history-of used as a parking space. ICD-10-CM directs you not to code conditions previously treated that no longer exist, so a resolved condition either leaves the list or becomes a history code; it does not stay current. Attestation. Signature, credentials, and date, at or near the time of the change, with the original entry left readable whenever this entry corrects one. Pitfall: signing or amending at audit time. CMS RADV reviewer guidance treats a record amended when the auditor calls as failing the timeliness requirement outright. ### Blank template (copy and adapt) DIAGNOSIS AND PROBLEM LIST UPDATE Patient: ____________ DOB: __________ Date of entry: __________ Author (name, credentials): _____________________________________ Entry type: [ ] new diagnosis [ ] refinement (specificity) [ ] correction/amendment [ ] delayed entry [ ] status change [ ] resolution Linked encounter date: __________ DIAGNOSTIC STATEMENT (your words; the code is assigned from this) _________________________________________________________________ _________________________________________________________________ Onset/established: __________ Status: [ ] active [ ] improving [ ] in remission [ ] resolved Rule-outs pending: __________ CLINICAL SUPPORT AT THIS ENCOUNTER (for any diagnosis on a claim) Monitored / evaluated / assessed / treated today, shown by: _________________________________________________________________ CODING MAPPING New or confirmed code: __________ Replaces: __________ Effective date: __________ DSM-5-TR label (if used): __________ Claim affected: [ ] yes [ ] no PROBLEM LIST ACTION [ ] added [ ] status updated [ ] resolved (date: __________) [ ] moved to history (Z-code: __________) If correcting a prior entry: original preserved and identified [ ] Signature/credentials: ________________ Date/time: ____________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample diagnosis and problem list update Scenario: outpatient behavioral health practice; a psychiatric nurse practitioner refines an intake anxiety diagnosis and resolves an adjustment disorder at a follow-up visit. All details are fictional. Diagnosis and Problem List Update. Client: R.L., 41 · Date of entry: 08/12/2026 · Author: A. Novak, PMHNP-BC · Linked encounter: office visit, 08/12/2026 Entry type: Refinement of an active diagnosis and resolution of a second diagnosis. New dated entry; no prior note altered. Diagnostic statement (refinement): Generalized anxiety disorder, moderate. Excessive and difficult-to-control worry across work, health, and family domains, present more days than not since January 2026, with restlessness, sleep-onset difficulty, and irritability. Duration and symptom criteria are now met. Replaces other specified anxiety disorder, documented at intake on 03/04/2026 when the duration criterion was not yet met. Diagnostic statement (resolution): Adjustment disorder with anxiety, resolved. The precipitating stressor (interstate relocation) ended in February 2026, and no stressor-linked symptoms have been reported or observed since the 05/2026 review. Clinical support at this encounter: GAD-7 administered today: 14, from 17 at intake. Worry content, duration, and controllability assessed in session against DSM-5-TR criteria. Treatment addressed today: sertraline continued at 100 mg daily; worry-postponement practice reviewed and assigned. Coding mapping: F41.1 active for today's encounter and forward claims, replacing F41.8, which is end-dated 08/12/2026. F43.22 end-dated 08/12/2026; no further claims will carry it. Original intake entries remain unchanged and readable. Problem list action: Generalized anxiety disorder added as active with January 2026 onset; other specified anxiety disorder closed as refined; adjustment disorder marked resolved with today's date and retained in past psychiatric history. Attestation: Entered and signed at the time of the change. A. Novak, PMHNP-BC. 08/12/2026, 4:40 pm. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The statement carries the code, not the reverse. Each change begins with a diagnostic statement in the author's words; under ICD-10-CM Guideline I.A.19 the code is assigned from that statement, so a coder can work from this entry without a query. Change arrives as a new entry and the originals survive. The intake diagnosis is end-dated, not overwritten, which satisfies all three CMS amendment principles: identified, dated and authored, original preserved. Every carried diagnosis has same-day support. The GAD-7 score, the criteria review, and the medication decision are dated evidence the condition was evaluated and treated at this encounter, the element risk-adjustment audits test first. Resolution actually resolves. The adjustment disorder leaves the active list with a date and a reason instead of riding forward as a current condition on future claims. Same-day attestation. Signed with credentials at the time of the change, meeting the at-or-near-the-time-of-service expectation and foreclosing any appearance of after-the-fact addition. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The update entry is ordinary medical-record content: part of the designated record set, releasable to the patient on request, and read by every downstream system that trusts the problem list, from clinical decision support to registries to risk-adjusted payment. Retention is where clinicians most often assume a rule that does not exist. HIPAA sets no retention period for clinical records; its 6-year rule ( 45 CFR 164.316 ) covers compliance documentation such as policies. What governs is state and program law, and the spread is wide: Medicare Advantage and Part D records 10 years, Massachusetts hospital records 20 years, Washington hospital records 26 years from creation under a 2025 amendment. Ontario requires by regulation 10 years from the last entry, or 10 years after a minor turns 18. Victoria and New South Wales require 7 years from the last service, or until age 25 for records made while the patient was a child. Build the practice retention policy to the strictest rule that touches you. On the payer side, the entry anchors the medical-necessity chain: the diagnosis you record here is the one payers read on a prior authorization request before approving care, and the one that decides code-level denials (CARC 11, in X12's words "The diagnosis is inconsistent with the procedure"). The format is a convention; the content, a dated, authored, supported diagnostic statement, is the requirement. Follow the CMS amendment discipline for every change even though it is payer policy rather than statute, and never amend a record after an audit request arrives: RADV reviewer guidance treats audit-time amendments as untimely on their face. Timing otherwise has no federal hour count; the expectation is entries completed and signed at or near the time of service, the same contemporaneity standard behind a start and stop time attestation , with late entries labeled as delayed. A diagnosis change should also ripple forward into the next treatment plan review so the plan and the list tell one story. In Canada and Australia the coding mechanics invert: ICD-10-CA (maintained by CIHI) and ICD-10-AM (published by IHACPA) are assigned by hospital coders from your documentation, not entered by clinicians on claims, so an outpatient diagnosis change reaches the coded dataset chiefly when an admission ends and a discharge summary is abstracted. In Australian behavioral health the diagnosis lives inside the GP Mental Health Treatment Plan, and MBS item 2715 guidance says the plan can be updated at any time while a new plan should generally not be written within 12 months: record the diagnosis change as an update, not a new plan. ## Common diagnosis and problem list update errors auditors flag The audit record around diagnosis documentation is deep because risk-adjusted payment rides on it. CMS's FY2025 improper-payments data put the Medicare Part C improper-payment rate at 6.09%, or $23.67 billion, most of it attributed to supporting documentation that failed to substantiate the diagnosis data submitted for payment. OIG risk-adjustment audits repeat one finding: in a May 2026 audit (report A-02-23-01020), none of the 97 sampled high-risk acute stroke diagnosis codes were supported by the medical records, with potential net overpayments of $462 million for payment year 2021, and the recurring root cause OIG cites is past conditions reported as current. Behavioral health sits on the same tape: a national OIG audit of psychotherapy claims found 128 of 216 sampled enrollee days failed documentation requirements, an estimated $580 million improper. The BastionGPT Clinical Advisory Board sees the same errors most often in diagnosis and problem list update reviews: Carried-forward diagnoses with no current support. Chronic conditions copied from visit to visit with no evidence they were monitored, evaluated, assessed, or treated at the billed encounter; the root cause OIG names most often. A code with no statement behind it. A picklist ICD-10-CM code appears on the list or the claim, but no dated, authored diagnostic statement supports it. The code follows the statement; a reviewer who cannot find the statement treats the code as unsupported. Corrections that erase. The original entry is overwritten or deleted instead of preserved with the change identified, dated, and authored. On paper that means a single-line strike-through leaving the original readable; in an EHR, a distinct entry the audit trail can separate from the original. Back-adding after the encounter closed. A diagnosis confirmed by later results is inserted into the earlier visit, or the record is amended when the audit letter arrives. Capture the confirmed diagnosis at the next encounter instead; RADV reviewer guidance treats audit-time amendments as failing the timeliness requirement. Resolved conditions still billed as current. Conditions that no longer exist stay active on the list and ride onto claims, when ICD-10-CM directs they be dropped or moved to a history Z-code. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on diagnosis and problem list updates. Extract every diagnosis mentioned in a session transcript or narrative note, with onset, status, and the supporting language, ready to reconcile against the current problem list. Map a diagnostic statement to candidate ICD-10-CM codes, or a DSM-5-TR label to its ICD-10-CM equivalent, for your verification, and flag unspecified codes the documentation already supports refining. Check an update entry before you sign: date and author present, original content preserved, current clinical support for each carried diagnosis, resolved conditions actually removed. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a required format for a diagnosis or problem list update? No statute prescribes one. CMS's own transmittal for the amendment and correction rules states: "There are no regulatory, legislative, or statutory requirements related to this CR." What Medicare requires as payer policy is that any amendment, correction, or delayed entry be identified as such, dated and authored, and preserve the original content. Canadian colleges and the Australian boards require accurate, dated records but no particular format either. The structured entry on this page is a convention that makes those properties easy to prove. Can a coder bill a different ICD-10-CM code than the one the provider clicked in the EHR? Yes, when the documentation supports it. Under ICD-10-CM Guideline I.A.19, code assignment is based on the provider's diagnostic statement, and auditors compare the documentation to the claim, not to the provider's picklist selection. A coder may assign the supported code on the claim without the provider editing the note. An addendum is needed only when the documentation itself does not support the intended code. Can I add a diagnosis to a closed encounter after test results come back? Generally no. If the condition was not established at that encounter, code the signs and symptoms that were present that day and capture the confirmed diagnosis at the next encounter. The exception is a true correction: the diagnosis was documented at the encounter and the code was simply mis-entered. Back-adding diagnoses to closed encounters is the pattern risk-adjustment audits target, and CMS RADV guidance treats records amended at audit time as failing the timeliness requirement. How is a diagnosis update different from a psychiatric diagnostic evaluation? The psychiatric diagnostic evaluation (90791, 90792) is the comprehensive service that first establishes a diagnosis. The update entry modifies one that already exists: refining specificity, changing status, correcting an error, or resolving it. Billing a new full evaluation to re-establish an existing diagnosis invites denial; a dated update entry costs nothing and keeps the record current. Is a DSM-5 diagnosis enough, or do I need the ICD-10-CM code? Claims require the ICD-10-CM code, the code set adopted under HIPAA. DSM-5-TR prints the corresponding ICD-10-CM code beside each diagnosis, so the crosswalk is built in, but the billable code is the ICD-10-CM one. Write the diagnostic statement with its specifiers; the code follows the statement. What does a denial saying the diagnosis is inconsistent with the procedure mean? That is CARC 11, a code-level mismatch: the diagnosis on the claim does not support the billed procedure under the payer's coverage rules. It often travels with remark codes such as M76 (missing or invalid diagnosis), and it is distinct from CO-50, a medical-necessity denial after clinical review. The durable fix is documentation: a more specific code the record already supports, or an updated diagnostic statement at the next encounter, never a code swapped in without support. Should resolved conditions stay on the problem list as history? Not as current diagnoses. ICD-10-CM directs coders not to code conditions previously treated that no longer exist; where the history matters to care, the personal-history Z-codes carry it. On the list itself, a resolved condition gets a resolution date and leaves the active list. Keeping everything active for context feels safe, and it is exactly what produces carried-forward unsupported diagnoses, the most common audit finding. How long do diagnosis records have to be kept? HIPAA sets no retention period for clinical records; its 6-year rule covers compliance documentation such as policies. US retention comes from state and program law: Medicare Advantage and Part D records 10 years, Massachusetts hospital records 20 years, and Washington hospital records 26 years from creation under a 2025 amendment. Ontario requires 10 years from the last entry, or 10 years after a minor turns 18. Victoria and New South Wales require 7 years from the last service, or until age 25 for records made during childhood. Can BastionGPT keep diagnosis and problem list updates audit-ready? Yes. Paste a transcript or narrative note and it extracts each diagnosis with onset, status, and the supporting language; it maps diagnostic statements to candidate ICD-10-CM codes for your verification; and it checks the finished entry for what auditors read first: date, author, preserved original content, current clinical support, and resolved conditions actually removed. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Treatment Plan Psychiatric Diagnostic Evaluation Outcome Measure Review Note Start-Stop Time Attestation Prior Authorization Request Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/start-stop-attestation All note templates # Start-Stop Time Attestation: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A start-stop time attestation is the notation in a psychotherapy progress note that records when the session began and ended, or the total face-to-face minutes, to support time-based codes such as 90832, 90834, and 90837. Clinicians add it to every timed session; payers and auditors use it to verify the billed code. Most versions run one to two lines inside the note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Therapists, psychologists, psychiatrists, and prescribers billing time-based psychotherapy or testing codes Audience Treating clinician, billing staff, payers, auditors Typical length 10 to 40 words · 1 to 2 minutes by hand (clinical team estimate) Format family Timed-service data element inside the progress note (compare: billing note, progress note) When it's used Every session billed under a time-based CPT or MBS code, including same-day E/M plus psychotherapy visits Standards context US Medicare policy accepts start and stop times or total time; in Australia meeting the stated duration is a condition of the MBS benefit On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a start-stop time attestation? A start-stop time attestation is the line in a psychotherapy progress note that records the elapsed face-to-face time of the session, either as clock times (3:02 pm to 3:55 pm) or as total minutes (53 minutes), to substantiate the time-based code billed. It is a data element, not a standalone document: no US statute, Canadian health-records act, or Australian MBS item defines a separate time-attestation form. The timed-code framework dates to January 1, 2013, when the AMA's CPT overhaul replaced the old range-based psychotherapy codes (90806 covered a 45 to 50 minute range) with the specific-time codes 90832, 90834, and 90837 governed by the CPT time rule. Clinicians and billers also call it time in/time out, session start and stop, total time, length of session, or simply the time statement. The clarification that matters most: the claim that both a start and a stop time are mandatory is a myth for US Medicare. The controlling guidance, CMS Billing and Coding Article A57520 , states that time "may be documented with start and stop times or with total time." Start and stop clock times are the more audit-resistant convention, not a federal requirement. The attestation lives inside the psychotherapy progress note , which also carries the diagnosis, interventions, and plan; the claim itself is assembled in the billing note , and reviewers verify time against the clinical note, not the claim form. ## Who uses start-stop time attestations and when Any clinician who bills a time-based code needs one in every session note: therapists and counselors billing 90832, 90834, or 90837, prescribers adding psychotherapy to an E/M visit, and psychologists delivering time-based testing services. The stakes rise in three situations: 90837, because the 53-minute floor makes the minutes decisive; same-day E/M plus psychotherapy, because the two clocks must be documented separately; and any provider whose billing pattern draws outlier analytics. Group and individual work follow the same rule, so a group therapy note and an individual therapy session note each carry their own time support. In Australia the duty is stronger: a Better Access item is payable only if the consultation ran the stated minimum, so the recorded duration is the eligibility test itself. ## Start-stop time attestation structure: what goes in each section Date of service and service type. Anchor the time statement to the date and the service furnished (individual, group, telehealth, E/M plus add-on) so the minutes attach to one identifiable encounter. Pitfall: a time entry that floats free of the service description; a reviewer matching the claim line to the note needs both in the same place. Start and stop clock times. The moment face-to-face psychotherapy began and ended, such as 3:02 pm to 3:55 pm. Under Medicare policy this format is optional, but it is the strongest evidence when minutes are questioned. Pitfall: copying the appointment slot from the schedule; the booked 50-minute block is not the clinical time, and a late start makes the two diverge. Total face-to-face minutes. The computed duration, written out (= 53 minutes). Medicare accepts total time alone; record it even when you use clock times so the arithmetic is visible. Pitfall: a stated total that does not match the clock times; that internal contradiction is the first inconsistency an auditor checks. The code the time supports. Under the CPT time rule, 16 to 37 documented minutes support 90832, 38 to 52 support 90834, and 53 or more support 90837; under 16 minutes, psychotherapy is not separately billable. Pitfall: rounding up through a threshold; a 52-minute session billed as 90837 is upcoding, however close the miss. A separate statement for combined E/M visits. When a psychotherapy add-on (90833, 90836, 90838) accompanies an E/M service, document the psychotherapy time on its own, excluding every E/M minute, and select the E/M level by medical decision-making rather than time. Pitfall: one blended total for the whole visit; in the national OIG audit, 6 of the 60 time-documentation failures were combined visits where psychotherapy minutes were never isolated. Signature, credentials, and date signed. Medicare's Program Integrity Manual expects records that are "legible, complete, dated, timed, and authenticated," and contractors expect notes signed when services are rendered. Pitfall: a signature added weeks later with no explanation; if a reviewer requests a signature attestation, it is due within 20 calendar days. ### Blank template (copy and adapt) PSYCHOTHERAPY TIME ATTESTATION (lives inside the progress note) Date of service: __________ Service: [individual / group / telehealth] Psychotherapy face-to-face: start _______ stop _______ = _____ minutes (face-to-face therapy time only; excludes intake paperwork, scheduling, documentation time, and any E/M time) Code supported: [ ] 90832 (16 to 37 min) [ ] 90834 (38 to 52 min) [ ] 90837 (53+ min) [ ] add-on with E/M: _______ Same-day E/M? [ ] No [ ] Yes: E/M documented separately; no minute counted twice; E/M level selected by medical decision-making Clinician signature/credentials: _______________ Date signed: __________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample start-stop time attestation Scenario: two time attestations as they appear inside progress notes, one standalone psychotherapy session and one same-day E/M visit with a psychotherapy add-on. All details are fictional. Example 1: standalone psychotherapy. Client: R.L., 41 · Date of service: 08/11/2026 · Service: Individual psychotherapy, in office Time attestation: Psychotherapy face-to-face: 3:02 pm to 3:55 pm = 53 minutes. Time reflects psychotherapy only; check-in, scheduling, and post-session documentation excluded. Code supported: 90837. Signed: J. Alvarez, LPC, 08/11/2026. Example 2: E/M visit with psychotherapy add-on. Client: D.K., 58 · Date of service: 08/13/2026 · Service: Medication management with psychotherapy E/M portion: Documented in its own section; level selected by medical decision-making, not time. Time attestation: Psychotherapy face-to-face: 10:12 am to 10:50 am = 38 minutes, exclusive of all E/M time. No minute counted toward both services. Codes supported: E/M service plus psychotherapy add-on 90836. Signed: M. Osei, PMHNP, 08/13/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works Clock times and the computed total both appear, so either accepted format stands on its own if the other is questioned. The arithmetic is internally consistent: 3:02 to 3:55 is 53 minutes and 10:12 to 10:50 is 38, so the first check a reviewer runs is already passed. The code follows the documented minutes, and the 90837 entry clears its 53-minute floor exactly where auditors look for rounding. The combined visit isolates psychotherapy minutes from E/M time, the specific failure mode the OIG flagged in same-day visits. Non-therapy time is excluded by name, and each entry is signed with credentials on the day of service, closing the authentication loop. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Session times are part of the disclosable record, and federal privacy law says so directly: 45 CFR 164.501 lists "counseling session start and stop times" among the items excluded from HIPAA's protected psychotherapy notes. That cuts one way: you cannot withhold session times from a payer as process-note material, so keep them in the progress note where they belong. The attestation only earns its keep when the rest of the medical-necessity chain surrounds it: a current diagnosis (kept accurate through a diagnosis list update ), interventions tied to goals, and a plan revisited on schedule in a treatment plan review . Retention follows the note that contains it: commonly 6 to 10 years in the US, with Medicare reviews reaching back 6, at least 10 years in Ontario under CRPO rules, and 7 years or until age 25 for minors in Australia. Label the obligations by strength, because they differ sharply by jurisdiction. In the US the time requirement is payer policy, not statute: Medicare contractor guidance sets it, requirements vary by contractor (during the national OIG audit two contractors had no psychotherapy coverage policy at all), and CMS told the OIG "there was no requirement for uniformity across all jurisdictions for LCDs." In Canada the duty is professional convention: Ontario college record-keeping standards ask for the "duration or timing of services," not clock times. In Australia duration is law: a Better Access item such as 80110 is payable only if the session ran at least 50 minutes, and the MBS explanatory notes add that only clinical details recorded at the time of attendance count toward the consultation time, so contemporaneous entry is part of the requirement. Telehealth changes none of the arithmetic; a telehealth therapy note adds modality and location fields around the same time statement. And when a payer questions necessity across an episode rather than one session's minutes, that is prior authorization territory, a different document with a different clock. ## Common start-stop time attestation errors auditors flag Time documentation is the most measured failure in psychotherapy auditing. In the HHS Office of Inspector General's national audit of Medicare psychotherapy claims (May 2023), 128 of 216 sampled enrollee days failed Medicare requirements, and time was the largest single gap: 60 of the 216 days lacked time documentation, 54 with no time in either format and 6 where a combined E/M visit showed total encounter time but never isolated the psychotherapy minutes. The OIG extrapolated roughly $580 million in improper payments from those patterns, most of it a documentation-sufficiency finding rather than proof that services were not rendered. The BastionGPT Clinical Advisory Board sees the same errors most often in start-stop time attestation reviews: No time in either format. The note describes the session but never states minutes or clock times; 54 of the OIG's 60 time failures were this omission alone. A blended clock in combined visits. One total covers E/M plus psychotherapy. Psychotherapy minutes must stand alone, and no minute may count toward both services. The appointment block passed off as clinical time. EHRs auto-fill the scheduled slot; late starts, check-in tasks, and early ends make the booked time and the face-to-face time different numbers. Minutes and code that disagree. A 52-minute session billed as 90837, or clock times whose arithmetic does not produce the stated total; both read as upcoding rather than as a slip. Identical times across the caseload. Every note reading 10:00 to 10:53 looks templated, and payer analytics flag outlier patterns, including providers whose billed minutes exceed a working day. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on start-stop time attestations. Pull the time statement out of a dictation or session transcript: start, stop, and computed minutes, formatted for the note. Check a finished note before you sign: time present in an accepted format, arithmetic consistent, and psychotherapy minutes separated from any E/M service. Confirm the documented minutes support the code on the claim, and flag a threshold miss such as a 52-minute 90837 before a reviewer does. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do I need both a start and a stop time, or is total minutes enough? For US Medicare, total time is enough. CMS Billing and Coding Article A57520 states that time may be documented with start and stop times or with total time, and that wording was added as a deliberate 2021 clarification. Start and stop clock times remain the most audit-resistant format, so many practices standardize on writing both: start, stop, and the computed total on one line. How do I document time when I bill an E/M plus a psychotherapy add-on? Keep two independent statements. Document the psychotherapy time separately, as start and stop times or as total minutes, excluding every E/M minute, and select the E/M level by medical decision-making rather than time. In the OIG's national audit, combined visits that showed only whole-encounter time were counted as failures because the psychotherapy portion was never isolated. Does time spent on notes, scheduling, or phone calls count toward the session? No. Only face-to-face psychotherapy time counts toward the timed code; pre-service and post-service work is built into the code's value. That is also why the appointment block in your calendar is not the billable time: the attestation should record when therapy actually began and ended, not when the room was booked. Can I bill 90837 for a 52-minute session? No. Under the CPT time rule, 53 minutes is the floor for 90837; 38 to 52 minutes supports 90834, 16 to 37 supports 90832, and under 16 minutes psychotherapy is not separately billable. Rounding a near miss upward is upcoding, and threshold sessions are exactly where reviewers compare the clock arithmetic against the code. Are start and stop times protected psychotherapy notes I can withhold from an insurer? No. HIPAA's definition at 45 CFR 164.501 expressly excludes counseling session start and stop times from protected psychotherapy notes, along with modality, frequency, diagnosis, and progress. Session times belong in the regular progress note , which payers may review. Is the start-stop time attestation a separate document? No. It is a data element inside the progress note, usually one line in the header or footer. No US statute, Canadian records act, or Australian MBS item defines a standalone time-attestation form. The clinical note carries the time; the billing note translates it into codes and units on the claim. If the time line is all you have, you do not yet have a note. What do Canada and Australia require? In Canada, no statute names session times; Ontario college record-keeping standards ask for the duration or timing of services, a professional convention enforced through the colleges. In Australia the duty is statutory: a timed MBS item is payable only if the session met its minimum duration (item 80110 requires at least 50 minutes), and only clinical details recorded at the time of attendance count toward the consultation time. How long must I keep notes containing time attestations? Retention follows the progress note. US practices commonly keep records 6 to 10 years, and Medicare reviews reach back 6. Ontario's CRPO expects at least 10 years from the last interaction, or 10 years from the client's 18th birthday. Australian guidance is 7 years for adults, or until age 25 for minors, and Medicare there expects referrals kept for 24 months. Can BastionGPT handle time documentation in my notes? Yes. Give it a dictation, a transcript, or bullets and it drafts the note with the time statement in place: start, stop, computed minutes, and the code the time supports. It can also check finished notes for missing time, blended E/M minutes, or a threshold miss before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Claim-Support Billing Note Psychotherapy Progress Note Individual Therapy Session Note Treatment Plan Review & Update Diagnosis & Problem List Update (ICD) Narrative Progress Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/prior-authorization-request All note templates # Prior Authorization Request: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A prior authorization request is the clinical justification packet a provider submits to a health plan to get a mental health service approved for coverage before it starts or continues. Practice staff or the clinician assemble it from the record: diagnosis, symptom severity, functional impairment, treatment history, risk, and a plan with measurable goals. No universal form exists; each payer or carve-out vendor sets its own. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Practice staff (billers, care coordinators) or the treating clinician Audience The payer or its behavioral health carve-out vendor: utilization review staff and peer reviewers Typical length 300 to 600 words plus attachments · 20 to 45 minutes by hand (clinical team estimate) Format family Payer utilization management form plus attachments (compare: prior authorization letter, insurance appeal letter) When it's used Before a plan-gated service starts, and at concurrent review before approved sessions run out Standards context Payer-specific by contract, no universal form; law sets the decision clocks; denial risk is high On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a prior authorization request? A prior authorization request is the clinical justification packet a provider submits to a payer to get coverage approved before delivering or continuing a mental health service. It is a payer-facing utilization management document assembled from the clinical record, and it is not the claim: the authorization establishes that a service will be covered, while the claim asks for payment after the service is delivered. Nobody designed it. It grew out of managed care utilization review in the 1980s and 1990s and remains a payer-industry artifact, which is why it answers to so many names: prior authorization or PA, pre-authorization, precertification, treatment authorization request (TAR, the Medicaid term), service authorization request, outpatient treatment report (OTR, the older behavioral health carve-out term), and, for continuing care, concurrent review or continued-stay review. Two boundaries carry this page. First, the request is not the prior authorization letter : the request is the payer's structured form or portal submission, while the letter is the clinician's narrative support that travels with it when the form cannot hold the complexity. Second, the obligation to obtain authorization is almost never law. It is payer contract and policy, plan by plan, and traditional Medicare imposes none of it on outpatient behavioral health: GAO reported in 2025 that CMS "does not currently require prior authorization for any behavioral health services" in traditional Medicare (GAO-25-107342). What the law does regulate is how payers run the process: decision clocks, licensed review of denials, and mental health parity. And for ongoing outpatient therapy, the operative version is usually a concurrent review triggered after a session threshold rather than an upfront gate, because outpatient care is already the least restrictive level. ## Who uses prior authorization requests and when Solo and group outpatient practices, intensive outpatient and partial hospitalization programs, psychological testing practices, and psychiatry clinics requesting services such as transcranial magnetic stimulation all meet the request, and the person completing it is as often a biller or care coordinator as the treating clinician. The gate sits with Medicare Advantage plans, Medicaid managed care organizations, and commercial plans, usually operated through a behavioral health carve-out vendor (Optum, Carelon, Magellan, Evernorth) whose portal and criteria control the process regardless of the logo on the member's card. The gate is also narrower than most clinicians assume: in GAO's 2025 review, 8 of 9 selected Medicare Advantage organizations required prior authorization for behavioral health, almost always for higher levels of care, and none required it for in-network outpatient counseling and psychotherapy. What outpatient clinicians actually meet is concurrent review once length or frequency passes the plan's threshold, and it uses the same packet. Use this template for the form-driven submission itself. When the form cannot carry the clinical argument, attach a medical necessity letter ; when the payer's decision comes back, it arrives as a utilization review summary , which tells you what any appeal will have to rebut. ## Prior authorization request structure: what goes in each section No universal form exists: no federal rule prescribes one for outpatient psychotherapy, so every payer and carve-out vendor builds its own. The sections below are the superset those forms draw from. Treat them as the checklist and map them onto whatever labels your payer's portal uses. Member, provider, and service header. Member name, ID, and plan; the behavioral health carve-out vendor if one manages the benefit; provider name, credentials, license, and NPI; and the request stated in billing terms: CPT codes, units, frequency, and the date range. Pitfall: authorization obtained from the medical plan when a carve-out vendor actually manages behavioral health; a number from the wrong entity does not bind the one paying the claim. Diagnosis. The specific ICD-10-CM code the record supports, written as code plus name. Pitfall: a Z-code or rule-out diagnosis standing alone; reviewers treat both as failing to anchor medical necessity, and it is one of the most common behavioral health documentation failures. Symptom severity and functional impairment. Current symptoms with severity, current measure scores (PHQ-9, GAD-7), and what the client cannot do this week: work, school, relationships, self-care. Pitfall: severity adjectives with no function. "Severe anxiety" approves nothing; "missed three of five workdays" is something a reviewer can defend approving. Level-of-care justification. Why this setting and intensity, argued against the criteria set the reviewer will score: LOCUS for adults, CALOCUS-CASII for children and adolescents, ASAM for substance use, or the plan's InterQual or MCG criteria. Pitfall: asserting the level ("weekly outpatient is appropriate") instead of scoring against the named criteria; the reviewer is completing a criteria worksheet, so write to it. Treatment history and prior response. What has been tried and what happened: prior episodes, medication trials, earlier courses of therapy, partial responses, and the failed or insufficient alternatives that make the requested service the right next step. Pitfall: omitting response to treatment so far; in a continued-care request, measured progress plus remaining impairment is the whole argument, and its absence reads as care without direction. Risk factors. Current risk status stated plainly, either way, with protective factors and the monitoring plan. Pitfall: silence on risk; a packet that never mentions it invites the reviewer to assume it was never assessed. Treatment plan, goals, and requested course. Measurable goals tied to the impairments above, the interventions that address them, and the requested frequency and duration restated as the plan's arithmetic. Pitfall: vague, untracked goals ("improve coping skills") with no baseline or target; reviewers list this among the classic behavioral health documentation failures. Urgency. Standard or expedited, chosen deliberately. Expedited review exists in law: 72 hours under Medicare Advantage rules and CMS-0057-F, and 72 hours for urgent care claims under ERISA plans. Pitfall: filing routine when a delay could seriously harm the client; the standard clock can run 7 to 15 days depending on plan type, and the expedited flag is the only thing that shortens it. ### Blank template (copy and adapt) PRIOR AUTHORIZATION REQUEST (payer-neutral worksheet) Member: ____________________ ID: ____________ DOB: __________ Plan: ______________________ BH carve-out vendor: _____________ Provider: __________________ Credentials/License: _____________ NPI: ____________ Phone/fax/portal: ___________________________ REQUESTED SERVICE CPT code(s): ____________ Units: ______ Frequency: ___________ Start date: __________ End date: __________ [ ] Initial request [ ] Concurrent / continued-stay review Urgency: [ ] Standard [ ] Expedited (delay risks serious harm) DIAGNOSIS (specific ICD-10-CM code plus name; no rule-out alone) ________________________________________________________________ SEVERITY AND FUNCTIONAL IMPAIRMENT (scores plus this week's function) Measures: PHQ-9 ____ GAD-7 ____ Other: ____________ Impairment: ____________________________________________________ LEVEL-OF-CARE JUSTIFICATION (LOCUS / CALOCUS-CASII / ASAM / plan criteria) ________________________________________________________________ TREATMENT HISTORY AND PRIOR RESPONSE (what was tried, what happened) ________________________________________________________________ RISK FACTORS (current status, protective factors, monitoring) ________________________________________________________________ TREATMENT PLAN AND GOALS (measurable, with baseline and target) ________________________________________________________________ ATTACHMENTS: [ ] treatment plan [ ] recent progress note [ ] score sheet [ ] medical necessity letter Clinician signature/credentials: ______________ Date: _________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample prior authorization request Scenario: concurrent (continued-stay) review request submitted to a commercial plan's behavioral health carve-out vendor, seeking 12 more weekly psychotherapy sessions for an adult with recurrent major depressive disorder. All details are fictional. Prior Authorization Request (concurrent review). Member: R.D., 41 · Plan: [commercial HMO], behavioral health managed by [carve-out vendor] · Submitted: 08/04/2026 · Provider: T. Alvarez, LCSW, NPI [on file] Requested service: Individual psychotherapy, 90834, 12 sessions at 1 per week, 08/11/2026 to 11/03/2026. Concurrent review of authorization A26-01187 (16 sessions approved 04/2026; 14 used as of this request). Urgency: standard. Diagnosis: F33.1, major depressive disorder, recurrent episode, moderate. Confirmed at intake 04/07/2026; unchanged at treatment plan review 07/29/2026. Severity and functional impairment: PHQ-9 on 07/29/2026: 12, down from 19 at intake. GAD-7: 9. Client has reduced missed workdays from about 2 per week in April to 1 in the past month, but continues to avoid team meetings, has resumed only one of three previously dropped activities, and reports early waking twice weekly with 5 to 6 hours of sleep. Level of care: Outpatient individual therapy remains the indicated level under LOCUS dimensions: moderate symptom severity, no acute risk indicators, engaged and adherent, recovery environment supportive, functional recovery incomplete. No indicators for a higher level of care. Step-down to biweekly is planned once PHQ-9 remains below 10 for four consecutive weeks. Treatment history and prior response: Second lifetime episode; 2019 episode remitted after approximately 6 months of CBT. Current episode began 01/2026. Sertraline 100 mg daily prescribed and monitored by PCP since 03/2026, adherent. Fourteen CBT sessions to date with partial response: PHQ-9 19 to 12, behavioral activation targets met, cognitive restructuring of self-critical beliefs in progress. Risk factors: Denies current suicidal ideation; passive ideation reported at intake resolved by session 4 and has not recurred. No history of attempts. Protective factors: engaged in treatment, stable housing, supportive partner. Screened each session; monitoring continues. Treatment plan and goals: Goal 1: PHQ-9 below 10 by 11/2026 (baseline 19, current 12). Goal 2: full workweek attendance for 4 consecutive weeks. Goal 3: resume two of three dropped activities. Interventions: weekly CBT with behavioral activation and cognitive restructuring; medication coordination with PCP; re-administer PHQ-9 and GAD-7 every 4 weeks. Attachments: Updated treatment plan (07/29/2026), PHQ-9/GAD-7 score sheet, most recent progress note. Signed: T. Alvarez, LCSW, 08/04/2026. This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The diagnosis is a specified F-code with severity. No Z-code, no rule-out, so medical necessity has an anchor the reviewer can match to criteria. Impairment is measured twice. Dated scores with a trend, and concrete function (workdays, meetings, activities): the pairing reviewers look for and the one most packets miss. The level of care is argued, not asserted. The packet scores against LOCUS, the criteria set the carve-out's reviewer completes, and names a step-down plan, which shows the course has an end. Prior response is quantified. PHQ-9 19 to 12 across 14 sessions makes the continued-care argument in one line: treatment is working and is not finished. The request matches claim arithmetic. Code, units, frequency, and dates that the eventual claims must mirror, filed before the existing authorization runs out rather than after. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The packet is assembled from the designated record set and then leaves your hands, so know what a payer can and cannot demand. Payers can require the medical-necessity elements of the treatment record: diagnosis, severity, functional impairment, the treatment plan , and progress. Psychotherapy process notes kept separate from the record carry heightened HIPAA protection and generally cannot be required as a condition of payment, so build the packet from progress notes, the plan, and outcome measures, never from process notes. File the request, the authorization number, and the payer's response alongside your billing note trail, because an approval is not payment: it confirms medical necessity was accepted, and the claim can still fail on coding, eligibility, timely filing, or place of service. Retention of the packet has no uniform national rule in any of the three countries this library covers; state or provincial licensing, privacy law, and the payer contract set it, and the longest applicable period wins. The clocks around the request are law even though the request itself is not. Medicare Advantage plans must decide within 14 days standard and 72 hours expedited under 42 CFR 422.568 and 422.572, and an approved authorization is valid for the enrollee's full course of treatment. Employer group plans under ERISA owe answers in 15 days pre-service and 72 hours urgent under 29 CFR 2560.503-1 . From 2026, the CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F) holds Medicare Advantage, Medicaid, CHIP, and federal exchange plans to 7 calendar days standard and 72 hours expedited, with a specific denial reason and public reporting of authorization metrics. Mental health parity adds one more layer: a plan that imposes prior authorization on mental health benefits must be able to show the limit is no more stringent than on comparable medical benefits, though the 2024 tightening of those rules sits under a non-enforcement posture while litigation runs. So work the process, not just the form: identify the carve-out vendor before the first session, run a session counter that flags at 75 percent of approved visits, file the concurrent request 3 to 5 business days before the authorization expires with a current treatment plan review and re-administered measures, and if the answer is no, the next document is an insurance appeal letter , not a resubmission. ↑ Back to the template and downloads ## Common prior authorization request errors auditors flag The denial data here is unusually good, and it points at process failures more than clinical ones. The HHS Office of Inspector General found that 13 percent of sampled Medicare Advantage prior authorization denials were for care that met traditional Medicare coverage rules (OEI-09-18-00260, 2022), and that Medicaid managed care organizations denied 1 in 8 requests in 2019, more than double the Medicare Advantage rate. Per KFF's analysis of CMS data , nearly 53 million prior authorization requests went to Medicare Advantage insurers in 2024, only 11.7 percent of 2023 denials were appealed, and yet more than eight in ten appeals overturn the denial. The lesson sits in that gap: most denials are never challenged, and most challenges win. The BastionGPT Clinical Advisory Board sees the same errors most often in prior authorization request reviews: Wrong entity, wrong number. The request goes to the medical plan while a carve-out vendor manages the behavioral benefit, or the claim cites an authorization number the paying entity never issued. Either way the claim returns as CO-197, and the CO group code bars billing the patient for it. A diagnosis that cannot carry weight. A Z-code or rule-out standing alone where a specified ICD-10-CM code belongs; reviewers read it as necessity never established. Severity without function, and no numbers. Adjectives in place of impairment, no PHQ-9 or GAD-7 trend, and interventions described only as "supportive": three of the most commonly cited behavioral health documentation failures. Authorization arithmetic ignored. Sessions delivered past the approved count or date range (denied as CO-198), claim codes or units that do not match the authorization, or 90837 billed without the documented 53-minute minimum and start and stop times that drive its elevated audit attention. The clock unmanaged. A routine filing where the clinical picture supported expedited review, a concurrent request submitted after the authorization already lapsed, or an appeal window (commonly 30 to 180 days) allowed to close unchallenged. ### Prior authorization requests in the US, Canada, and Australia Aspect United States Canada Australia Status Payer policy, not statute; traditional Medicare requires none for outpatient behavioral health, while Medicare Advantage, Medicaid managed care, and commercial plans gate services by contract inside legally set decision clocks Structurally absent from public plans: psychotherapy outside hospitals and physician practices is mostly privately insured, and private plans reimburse claims rather than pre-approve care No clinical prior authorization; Better Access is a rebate-eligibility gate (diagnosis, GP Mental Health Treatment Plan, referral), and private hospital cover runs an administrative eligibility check, not clinical review Terminology Prior authorization, precertification, TAR (Medicaid), OTR, concurrent or continued-stay review Preapproval or predetermination (uncommon for therapy), extended health care claim, treatment plan request Mental Health Treatment Plan, referral, eligibility check (ECLIPSE), Informed Financial Consent What changes Whether a gate exists at all varies by plan and carve-out vendor; higher levels of care are gated most, and outpatient therapy usually meets concurrent review instead Annual dollar maximums and eligible regulated-provider rules replace clinical review; some insurers request a treatment plan or physician referral for intensive or higher-cost claims Rebate caps (10 individual and 10 group services per calendar year) and referral validity replace payer review; clinical necessity stays with the treating clinician Retention No uniform national rule for the packet; state licensing, privacy law, and the payer contract govern Provincial privacy law and college standards govern the underlying record (for example Ontario's PHIPA) State rules and the Psychology Board's Code of Conduct (effective December 1, 2025) govern the record; no authorization-specific rule Only the US column describes true clinical prior authorization, and even there the gate is contractual and concentrated in managed care. Read any claim about "prior authorization" against the country and plan type it names. In Canada and Australia this template still organizes a clean medical-necessity summary, but no payer requires it in this form. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on prior authorization requests. Draft the packet from the chart: paste the treatment plan, recent notes, and scores, and it assembles diagnosis, severity, impairment, history, risk, and goals into the payer's fields. Pull the necessity elements out of a long record: prior episodes, medication trials, and measured response, extracted and dated instead of retyped. Check the request before it goes out: specified diagnosis code, function paired with scores, level-of-care criteria named, CPT codes and units matching the treatment plan, and an expedited flag where the clinical picture supports one. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Do therapists need prior authorization for outpatient psychotherapy? Usually not upfront. In GAO's 2025 review, none of the nine selected Medicare Advantage organizations required prior authorization for in-network outpatient counseling and psychotherapy, and traditional Medicare requires none for any outpatient behavioral health service. The gates concentrate at higher levels of care: inpatient, partial hospitalization, transcranial magnetic stimulation. What outpatient clinicians actually meet is concurrent review once sessions pass the plan's threshold, which uses the same packet. Verify plan by plan; the answer lives in the contract, not in a general rule. Is a prior authorization request required by law? No. The obligation to obtain authorization is payer contract and policy; no statute prescribes the packet, and no universal form exists. What the law regulates is the payer's side of the process: decision timeframes, licensed-clinician review of denials, and mental health parity, which requires plans to show that authorization rules on mental health benefits are no more stringent than on comparable medical benefits. The concept itself is US managed care: Canada's public plans have no equivalent gate for psychotherapy, and Australia's Better Access is a rebate-eligibility gate rather than payer review. How long does the payer have to decide? The clocks are law and depend on plan type. Medicare Advantage: 14 days standard, 72 hours expedited. Employer plans under ERISA: 15 days pre-service, 72 hours urgent, and 24 hours for a concurrent care extension requested at least 24 hours before approved care ends. From 2026, CMS-0057-F holds Medicare Advantage, Medicaid, CHIP, and federal exchange plans to 7 calendar days standard and 72 hours expedited, with a specific reason for any denial. A June 2025 industry pledge also commits most large insurers to real-time electronic approvals for most requests by 2027. If a delay could seriously harm your client, request expedited review and say why. What is the difference between a prior authorization request and a prior authorization letter? The request is the payer's structured form or portal submission: fields, codes, criteria. The prior authorization letter is the clinician's narrative support letter that travels with it when the form cannot capture the situation: an atypical presentation, extended sessions, an out-of-network exception. If coverage has already been denied, neither is the right tool; that calls for an insurance appeal letter , which runs on post-denial deadlines and preserves external review rights. Does 90837 require prior authorization? On most current plans, no upfront authorization, but it draws elevated scrutiny. UnitedHealthcare and Optum required authorization for 90837 until 2018 to 2019 and then dropped it on many plans; some plans still limit extended-session frequency. Because 90837 pays more than 90834 and requires a documented minimum of 53 minutes, missing start and stop times is a frequent clawback trigger in post-payment audits. Document the time and the clinical reason the longer session was necessary. What happens if sessions occur before the authorization is approved? They risk denial under CO-197, the authorization-absent code, and because of the CO group code the balance cannot be billed to the patient. Some plans allow retroactive authorization within a short window or for urgent situations; many do not, and it is never guaranteed. Verify benefits and the authorization requirement before the first gated session. Even with an approval in hand, payment still depends on coding, eligibility, timely filing, and the claim matching the authorized codes, units, and dates. Can the payer demand my psychotherapy notes? Payers can require enough of the treatment record to establish medical necessity: diagnosis, severity, functional impairment, treatment plan, and progress. Psychotherapy process notes kept separate from the record have heightened protection under HIPAA and generally cannot be required as a condition of payment. Build the packet from progress notes, the treatment plan, and outcome measures, and keep process notes out of it. How do I handle concurrent review for continued sessions? Track the approved session count and flag at 75 percent. Submit the continued-stay packet 3 to 5 business days before the authorization expires, with an updated treatment plan review , re-administered outcome measures, and a current diagnosis list . If it is denied, request a peer-to-peer review with a same-specialty reviewer and appeal inside the plan's window, commonly 30 to 180 days. The appeal math favors you: only about one denial in nine is appealed, and more than eight in ten appeals overturn the denial. Can BastionGPT draft a prior authorization request? Yes. Give it the treatment plan, recent notes, and scores, and it drafts the packet: specified diagnosis, severity paired with function, level-of-care justification against the criteria you name, prior response, risk, and measurable goals, ready to map onto the payer's form. It can also turn a denial into the outline of an appeal. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Prior Authorization Support Letter Utilization Review Summary Treatment Plan Medical Necessity Letter Insurance Appeal Letter Diagnosis & Problem List Update (ICD) Peer-to-Peer Review Note Y-BOCS Score Documentation MADRS Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/medical-necessity-letter All note templates # Medical Necessity Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A medical necessity letter (also called a letter of medical necessity or LMN) is a clinician-authored narrative explaining why a specific service, level of care, or item is medically necessary for a particular patient. Therapists, psychologists, and psychiatrists send one to payers, FSA and HSA administrators, schools, and benefit programs to support coverage, reimbursement, or accommodations. Most run 1 to 3 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating clinicians: therapists, psychologists, psychiatrists, NPs, physicians Audience Payers and utilization reviewers, FSA and HSA administrators, schools, employers, benefit programs Typical length 300 to 900 words (1 to 3 pages) · 20 to 45 minutes by hand (clinical team estimate) Format family Clinician-authored justification letter (compare: prior authorization request, insurance appeal letter) When it's used Pre-service authorization, continued-care review, appeals, FSA/HSA claims, and school, employer, or program requests Standards context A professional convention, not a mandated form; broadly useful across reimbursement domains On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a medical necessity letter? A medical necessity letter is a clinician-authored narrative that explains why a specific service, level of care, item, or continued course of treatment meets the applicable definition of medically necessary for one particular patient. You will also see letter of medical necessity, LMN, LOMN, and, in gender-affirming care, letter of support. It is almost always an attachment rather than a freestanding document: it rides with a prior authorization request, a continued-stay review, an appeal, an FSA or HSA reimbursement claim, or a school or program file. The letter has two separate lineages that vendor templates routinely blur. The tax lineage runs through IRC section 213(d) and IRS Publication 502, which decide what counts as a medical expense for FSA and HSA purposes. The coverage lineage produced the standardized mental health format: Feldman, DeBofsky, Plakun, and Potts published it in the Journal of Psychiatric Practice in July 2021, building the structure on the Mental Health Parity and Addiction Equity Act (MHPAEA) and the Wit v. United Behavioral Health litigation; the American Psychiatric Association hosts a derivative template. Three boundaries do the most work. First, no law requires the letter itself: no federal regulation prescribes a format, length, letterhead, or signature style. What the law governs is the decision the letter supports (parity in how plans set medical necessity criteria, appeal rights under ERISA and the ACA) and the clinical record behind it; the letter is a convention. Second, the medical necessity letter is the broad instrument, addressed to payers, schools, employers, and benefit programs alike. When it supports a payer's pre-service review it travels with a prior authorization request ; after a denial, the argument moves into an insurance appeal letter , which adds the procedural case (the plan's own criteria, the denial reason, appeal rights) a necessity letter does not carry. Third, it is not a certificate of medical necessity (CMN), a structured Medicare form historically tied to durable medical equipment, not a behavioral health narrative. ## Who uses medical necessity letters and when Every behavioral health discipline gets asked for one: therapists and counselors justifying session frequency or continued psychotherapy, psychologists supporting testing or a step up in level of care, psychiatrists and nurse practitioners backing a medication, device, or program placement, and any of them supporting an FSA or HSA claim for a dual-purpose item. The recipient sets the flavor: utilization reviewers want the standards-based clinical argument, FSA and HSA administrators want the diagnosis-to-expense tie with a duration, and schools, employers, and disability programs read the letter as evidence for accommodations or eligibility. Reach for a different document when the job is different: justifying continued care inside a treatment episode is the work of the treatment plan review , and a change in level of care is documented in the discharge summary that closes the old level. ## Medical necessity letter structure: what goes in each section Recipient and reference block. Date, addressee (the plan's utilization review department, the FSA administrator, the school office), patient name and date of birth, member ID, and the claim or authorization reference the letter supports. Pitfall: no authorization or claim number; the letter cannot be matched to the pending request, so the decision clock runs out while your strongest document sits unattached in a scanning queue. Credentials and treating relationship. Your license and specialty, when treatment began, session count and current frequency, and the date of the most recent visit. This paragraph shows that an adequate evaluation sits behind the opinion. Pitfall: writing on the strength of a single visit; boards and reviewers test whether the relationship supports the opinion, and one-session letters are a recurring disciplinary trigger. Diagnosis and functional impairment. The ICD-10-CM code and name, current symptoms with any measured scores, and specific impairment at work, school, home, or in relationships. Pitfall: the code alone; reviewers authorize services against documented impairment, not diagnostic labels, and a bare F33.2 gives them nothing to grant. History and prior treatment. What has been tried, this episode and before: levels of care, therapy frequency, medication trials and response, and what happened at lower intensity. Pitfall: omitting the less intensive alternatives that were tried and fell short; that is the first question in any level-of-care review, and silence reads as untried. The specific request. The service or item, the code if known, frequency, duration, and a re-evaluation point. For FSA and HSA letters, add the condition the item treats, a statement that the expense is not for general health, and the duration of need. Pitfall: requesting continued treatment with no units; a reviewer cannot authorize a sentence, and FSA administrators commonly cap a letter's validity at 12 months, so an undated, open-ended request expires. Clinical rationale. Why this service, at this intensity, for this duration, is consistent with generally accepted standards of care, and what deterioration is expected without it. This is the paragraph the standardized format was built for. Pitfall: arguing from credentials instead of standards; coverage disputes turn on whether criteria match generally accepted standards of care, so a rationale anchored there is the one an internal reviewer, and any later external reviewer, can act on. Signature block. Name, credentials, license number, NPI, direct contact, and date, with a supervisor co-signature where the payer or state requires one. Pitfall: chasing the wrong formalities; letterhead and ink signatures are conventions (Medicare accepts electronic signatures with protections against modification), while a missing license number or NPI stalls verification in a way stationery never fixes. ### Blank template (copy and adapt) MEDICAL NECESSITY LETTER [Letterhead: clinician or practice name, credentials, license #, NPI, address, phone, secure fax or portal] Date: __________ To: [Plan / administrator / program] Attn: [Utilization review / claims] Re: [Patient name] DOB: __________ Member ID: __________ Claim / authorization reference: ______________________________________ Requested service or item: [service, code if known, frequency, duration] To the review team: CREDENTIALS AND RELATIONSHIP I am a [license, specialty]. I have treated [patient] since [date]: [number] sessions, currently [frequency]; most recent visit [date]. DIAGNOSIS AND FUNCTIONAL IMPAIRMENT [ICD-10-CM code and name], with [symptoms, measured scores] and [impairment at work / school / home / relationships]. HISTORY AND PRIOR TREATMENT [Levels of care and treatments tried, with dates and response; what was tried at lower intensity and how it fell short.] REQUEST [Service or item, code, frequency, duration, re-evaluation date. FSA/HSA: condition treated, statement that the expense is not for general health, duration of need.] CLINICAL RATIONALE [Why this service at this intensity fits generally accepted standards of care; expected risk of deterioration or relapse without it.] I am available at [contact] for questions or peer review. Sincerely, _____________________________ Date: __________ [Name, credentials, license #, NPI] [Supervisor co-signature if required: name, credentials, license #] Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample medical necessity letter Scenario: a psychologist in outpatient private practice writes to a commercial plan's utilization review department to support continued twice-weekly psychotherapy for an adult with recurrent major depressive disorder. All details are fictional. Riverline Psychology Group · M. Okafor, PhD, Licensed Psychologist · NY license 000000 · NPI 0000000000 · (000) 000-0000 Date: 08/12/2026 To: Stonebrook Health Plan, Utilization Review Department Re: R.S. (DOB 03/14/1994) · Member ID SB0044178 · Authorization reference UR-26-04471 Requested service: Individual psychotherapy (90837), twice weekly for 12 weeks To the utilization review team: I am a licensed psychologist and R.S.'s treating clinician. I have provided individual psychotherapy since 01/13/2026: 39 sessions to date, weekly through early June and twice weekly since 06/09/2026, most recently on 08/10/2026. This letter supports continued authorization of twice-weekly individual psychotherapy (90837) through 11/06/2026. Diagnosis and current impairment. R.S. meets criteria for major depressive disorder, recurrent, severe (F33.2), with generalized anxiety disorder (F41.1). PHQ-9 on 08/10/2026 was 18, down from 23 on 06/09/2026; GAD-7 was 14. R.S. remains on a reduced schedule at work after missing 11 full days in May, avoids driving beyond the immediate neighborhood, and reports concentration lapses that led to reassignment of two projects. Sleep has improved from 3 to 4 hours nightly in May to 5 to 6 hours now. History and prior treatment. This is the third major depressive episode since age 19. Weekly psychotherapy from January through May produced partial response, then regression in late May with passive suicidal ideation, resolved by mid-June and monitored at every session since. Sertraline at a therapeutic dose, prescribed by the primary care physician since March, continues unchanged. The step-up to twice-weekly sessions on 06/09/2026 was followed by a steady PHQ-9 decline from 23 to 18 and a return to part-time work. Less intensive treatment was tried this episode and did not hold. Clinical rationale. Continued twice-weekly psychotherapy is consistent with generally accepted standards of care for severe recurrent depression with partial response and documented regression at lower intensity. The treatment plan targets a PHQ-9 below 10, full return to work, and independent driving, reviewed against these measures every four weeks. Twice weekly is the least intensive level of care likely to hold the current trajectory; a premature step-down already reproduced regression once this episode. Request. Authorize individual psychotherapy, 90837, two sessions per week for 12 weeks (24 sessions), 08/17/2026 through 11/06/2026, with re-evaluation before any extension. I am available at (000) 000-0000 or through the plan's provider portal for peer review. Sincerely, M. Okafor, PhD, Licensed Psychologist NY license 000000 · NPI 0000000000 Signed electronically 08/12/2026 This sample is fictional and for educational purposes. It does not describe a real patient, clinician, practice, or health plan. ↑ Back to the template and downloads ### Why this sample works It files itself. Member ID, authorization reference, and the requested service sit at the top, so the letter attaches to the pending request instead of drifting in a scanning queue. The diagnosis carries function. F33.2 arrives with PHQ-9 numbers, missed workdays, and concrete daily limits, the material a reviewer can authorize against. Stepped care is documented, not asserted. Weekly frequency was tried, regression at lower intensity is dated, and the response to the step-up is measured, which answers the least-intensive-alternative question in advance. The request has units and an end. Code, frequency, session count, date range, and a re-evaluation commitment give the reviewer something specific to grant. The rationale argues standards, and the signer is verifiable. Generally accepted standards of care frame the ask, and the license number, NPI, and direct contact let verification finish without a call-back loop. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The letter is chart-derived and chart-verified. It joins the record like any other clinical document: the client can request it, the payer can audit the file behind it, and a reviewer will read it against the session notes and the treatment plan it summarizes, because the medical necessity chain starts in the plan, not in the letter. Authorship deserves the same care. Medicare does not allow incident-to billing for mental health services delivered by unlicensed clinicians, and whether an associate may sign a letter is payer- and state-specific, so the safe default is that the rendering associate signs, the supervisor co-signs, and the supervision itself is documented. Charging for letter writing is ethically permitted when your fee agreement disclosed it. Treat animal and accommodation letters as clinical work with board exposure: the California Board of Behavioral Sciences disciplined a marriage and family therapist in 2019 for issuing an emotional support animal letter without a proper assessment, California has required a client relationship of at least 30 days before an ESA letter since 2022, and professional associations including the APA and the American Counseling Association discourage writing them routinely. The payer layer rewards routing before rhetoric. The legal hooks around the letter are indirect: parity law (MHPAEA) governs how plans set and apply the medical necessity criteria the letter argues against, ERISA's claims procedure at 29 CFR 2560.503-1 entitles you to the reasoning and criteria behind an adverse determination and lets you act as the client's authorized representative, and ACA rules at 45 CFR 147.136 add internal and external review. None of them requires a letter; the letter is the convention through which those rights get exercised, and requirements like the FSA 12-month validity cap are payer policy an administrator defines. Traditional Medicare generally does not require prior authorization for outpatient psychotherapy, so a standalone letter is rarely a Medicare condition of payment. Route denials by code: CO-50, not medically necessary, gets a letter and an appeal ; CO-197, authorization absent, gets the authorization corrected, not a clinical argument. The workload case for tight templates is documented: the AMA's 2024 survey found physicians complete an average of 43 prior authorizations a week, consuming about 12 hours of physician and staff time. Two federal changes are moving the ground: CMS-0057-F requires impacted payers, for 2026, to return specific denial reasons and decide expedited prior authorizations within 72 hours and standard ones within 7 days, with electronic prior authorization APIs due by January 2027, and the MHPAEA 2024 Final Rule's newest provisions sit under a stated federal non-enforcement posture while litigation runs, so treat 2026 parity specifics as unsettled. ## Common medical necessity letter errors auditors flag The audit record is blunt about where behavioral health documentation fails, and it is usually not the letter. In the HHS Office of Inspector General's May 2023 audit of Medicare psychotherapy , an estimated $580 million of $1 billion paid was improper; among 216 sampled enrollee-days, 128 had documentation deficiencies such as undocumented psychotherapy time and noncompliant treatment plans, and 54 lacked provider signatures. Denial data points the same way: in KFF's analysis of 2024 HealthCare.gov claims, about 5% of denials were for lack of medical necessity while 9% were for missing prior authorization or referral, so the letter is the fix for a minority of denials and the record is the fix for the rest. The BastionGPT Clinical Advisory Board sees the same errors most often in medical necessity letter reviews: A letter the chart cannot back. Reviewers verify the letter against session notes, the treatment plan, and signatures. A letter describing twice-weekly severity while the chart shows sparse visits, missing time, or an expired treatment plan ends the argument on the payer's terms. A diagnosis code with no functional story. A code approves nothing by itself. The letter needs symptoms, measured scores, and named impairment at work, school, or home tied to that code. A request without units. Continued treatment is not an authorizable request. Name the service, the code, the frequency, the duration, and the re-evaluation point. Authority instead of standards. A rationale resting on credentials alone gives the reviewer nothing to weigh. The standardized mental health format was built on MHPAEA and the Wit litigation precisely so the argument runs through generally accepted standards of care. The wrong tool for the denial code. CO-50, the code for services not deemed medically necessary, is the denial a letter answers, and it is frequently overturned on external review. CO-197, authorization absent, is an administrative failure: correct and resubmit the authorization; a clinical letter does not cure it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on medical necessity letters. Draft a complete letter from the treatment plan, recent session notes, or a dictated summary: diagnosis, impairment, prior treatment, the specific request, and a standards-based rationale in payer-ready order. Rewrite the clinical case in plain language a benefits administrator, school, or employer can act on, without weakening the diagnosis or the request. Check a finished letter before you sign: service with frequency and duration, functional impairment language, reference numbers, and a complete signature block. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a medical necessity letter be? Most run 1 to 3 pages, roughly 300 to 900 words, and expert guidance for outpatient behavioral health suggests 1 to 2 pages. No regulation prescribes a length. Reviewers work queues, so put the diagnosis, the request with units, and the standards rationale where a skimmer finds them, and let supporting history follow. There is no published research timing mental health letter writing, so treat any minutes-per-letter figure you see online as a vendor estimate. Is a letter of medical necessity required by law? No. No federal statute or regulation prescribes a letter of medical necessity format, length, letterhead, or signature style for behavioral health. The law governs the decision the letter supports and the record behind it: MHPAEA regulates how plans set and apply medical necessity criteria, ERISA guarantees claims and appeal procedures, and the ACA adds internal and external review. Plans and FSA administrators can require a letter as payer policy, but its form is theirs to define, not law's. Some states go further on the criteria themselves: California's SB 855 requires commercial plans to cover medically necessary mental health and substance use treatment using generally accepted standards of care. The letter is a professional convention for exercising those rights. What is the difference between a medical necessity letter, a prior authorization request, and an insurance appeal letter? A medical necessity letter is the broad justification document: it can go to a payer, school, employer, FSA or HSA administrator, or benefit program. A prior authorization request is the payer's pre-service transaction, usually a form or portal submission, with the letter riding along as the clinical attachment. An insurance appeal letter comes after a denial and adds the procedural case: the plan's own criteria, the denial reason, and appeal rights. Pre-service, send the request plus the letter; post-denial, send the appeal with the letter attached. Does the letter need letterhead and a handwritten signature? Both are conventions, not legal requirements. Medicare's Program Integrity Manual accepts electronic signatures with safeguards against modification, reserves pen-and-ink for controlled-substance orders, and allows rubber-stamp signatures only as a disability accommodation under the Rehabilitation Act of 1973. What actually stalls a letter is a missing license number, NPI, or direct contact, so spend the effort on the signature block, not the stationery. What does an FSA or HSA letter of medical necessity need to include? Four things: the diagnosed condition, how the specific item or service treats it, a statement that the expense is not for general health or cosmetic purposes, and the duration of need. The phrase letter of medical necessity does not appear in IRS publications; administrators built the practice on IRC section 213(d) and Publication 502, which allow dual-purpose items only when recommended by a practitioner for a specific condition. Most administrators cap a letter's validity at 12 months, so date it and calendar the renewal. Keep this template separate from your insurance-coverage letter; the legal bases differ. Can a pre-licensed or associate clinician sign a medical necessity letter? It is payer- and state-specific. Medicare does not allow incident-to billing for mental health services delivered by unlicensed clinicians, and auditors have penalized practices where supervisee documentation lacked co-signatures and supervision was undocumented. The safe default: the rendering associate signs, the supervisor co-signs with license number, and the supervision itself is documented in the record. Is appealing a denial worth it? The numbers say yes, and almost nobody tries. In KFF's Marketplace data, fewer than 1% of denied claims were appealed, and insurers upheld 66% of the internal appeals that were filed. External review is where results flip: 46% of externally reviewed Medicaid managed care denials were fully or partially overturned, and in Medicare Advantage 81.7% of appealed prior authorization denials were partially or fully overturned. Match the tool to the code first: CO-50, not medically necessary, is the denial a letter answers; CO-197, missing authorization, needs an administrative fix, not a clinical argument. Do Canadian or Australian clinicians write medical necessity letters? Yes, but the regulated documents differ. In Canada, provincial plans rarely cover non-physician psychotherapy, so letters mostly serve private extended-health insurers as payer policy; tax claims run through CRA receipt rules, and the Disability Tax Credit uses Form T2201, a statutory certification a medical practitioner completes, not a letter. In Australia, Medicare rebates hang on the Mental Health Treatment Plan and a valid referral rather than a letter, with Better Access redesigned on November 1, 2025, and NDIS supports letters are a separate family under the NDIS Act. Retention follows local rules either way: Canadian colleges commonly expect about 10 years, and Australian psychologists keep records about 7 years for adults or until a former minor turns 25. In all three countries the letter itself stays a convention. Can BastionGPT write a medical necessity letter? Yes. Give it the treatment plan, recent notes, or a dictated summary, and it drafts the letter in payer-ready order: diagnosis with functional impairment, prior treatment, the specific request, and a standards-based rationale. It can also rewrite the case in plain language for a school or benefits administrator and check the letter for missing units, identifiers, and dates before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Prior Authorization Support Letter Treatment Plan Claim-Support Billing Note Prior Authorization Request Insurance Appeal Letter Peer-to-Peer Review Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/insurance-appeal-letter All note templates # Insurance Appeal Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An insurance appeal letter is a written request asking a health plan to reverse an adverse benefit determination: a denied, reduced, or terminated mental health claim. Clinicians, billing staff, patients, or authorized representatives file it within the plan's appeal deadline, pairing the denial code with a clinical rebuttal. No law prescribes its format; most run 1 to 3 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating clinicians, billing and practice staff, patients, or authorized representatives Audience Payer appeals reviewers; independent review organizations and ombudsman bodies on escalation Typical length 400 to 900 words (1 to 3 pages) · 45 to 90 minutes by hand (clinical team estimate) Format family Payer-facing appeal correspondence (compare: medical-necessity letter, prior-authorization request, utilization-review summary) When it's used After a denial, reduction, or termination of payment or coverage, within the payer's filing deadline Standards context Format is convention; filing deadlines are law, and plan-specific language is often required On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an insurance appeal letter? An insurance appeal letter asks a health plan to reverse an adverse benefit determination: a denial, reduction, or termination of payment or coverage for mental health or substance use care. The letter has no single author or standards body; its skeleton comes from the process the law does prescribe, the ERISA claims-procedure regulation (29 CFR 2560.503-1), the ACA's internal-appeal and external-review overlay (45 CFR 147.136), and Medicare's five-level appeal ladder (42 CFR Part 405). Depending on the payer you will hear it called a first-level or internal appeal, a redetermination (Medicare), a reconsideration (Medicare Advantage), or an external review request. A "parity appeal" is the same letter arguing that the plan judged mental health care by a stricter standard than comparable medical care. One correction does most of the work on this page: no statute, regulation, or payer manual in the US, Canada, or Australia prescribes what the letter must contain. The law fixes deadlines, who may file, reviewer independence, and what the payer must disclose; the letter's structure is professional convention. Medicare accepts a plain written request or optional Form CMS-20027, and courts have treated even a handwritten note saying "I appeal" as enough to preserve ERISA appeal rights. The convention persists because it works: a letter organized around the denial code and the plan's own criteria is easier for a reviewer to grant. Keep the boundary with its closest neighbor straight, too: a medical-necessity letter argues that care is clinically needed and usually rides along as an enclosure; the appeal letter argues that a specific denial was wrong. ## Who uses insurance appeal letters and when Solo therapists and psychologists, group-practice billing managers, hospital utilization and revenue-cycle teams, and patients themselves all file appeals. No US rule requires the treating clinician to author the letter: billing staff or an authorized representative may draft and file it, with the clinician's signature reserved for the medical-necessity statement. Timing picks the tool. Before care, approval runs through a prior-authorization request ; during care, a utilization-review summary answers concurrent review; a peer-to-peer call is a conversation, not an appeal, and should never be allowed to run out the appeal clock. The appeal letter is the post-denial instrument: it responds to a determination already made, packages the clinical evidence, usually the treatment plan and dated measure scores, and asks for a specific reversal inside the filing window. ## Insurance appeal letter structure: what goes in each section Routing and header block. Identify the track before you draft: ERISA employer plan, ACA individual plan, Medicare, Medicare Advantage, or Medicaid managed care, each with its own appeals address, forms, and filing window. The header carries patient and member identifiers, claim numbers, dates of service, the date on the denial notice, and the word appeal. Pitfall: a letter that never names the adverse benefit determination it is appealing can be logged as a grievance, the complaint category 42 CFR 438.400 separates from appeals, and grievances do not overturn denials. The denial, quoted. State what was denied and quote the claim adjustment reason code verbatim from the remittance advice or EOB. CO-50 (not medically necessary) and CO-197 (no prior authorization) are the two codes behavioral health appeals rebut most often, and plans subject to the ACA must print the denial code and its meaning on the notice itself. Pitfall: appealing a CO-16 denial; that code means information was missing, arrives paired with a remark code naming the missing element, and is usually resolved with a corrected claim, not an appeal. Point-by-point rebuttal. Name the plan's medical-necessity policy and answer each criterion it applied, in order, using the plan's own level-of-care guidelines (InterQual, MCG, LOCUS, or ASAM criteria where the plan uses them). Pitfall: a general essay on why therapy helps. Failing to address the stated denial rationale is a standard procedural loss, and a Medicare CO-50 driven by a local coverage determination (remark N115) must be argued against that LCD, not against a commercial medical policy. Clinical evidence summary. The diagnosis with its most specific ICD-10 code, standardized measure scores with dates, functional impairment in concrete terms, treatment history, risk factors, and the link to treatment plan goals. Pitfall: undated scores. A single PHQ-9 with no baseline gives the reviewer nothing to weigh; a dated trend does the arguing for you. Disclosure and parity requests (US plans). Ask in writing for the criteria used and for every document relevant to the claim, which an ERISA plan must provide free of charge on appeal (29 CFR 2560.503-1(m)(8)), and for a mental health denial, the plan's NQTL comparative analysis under section 203 of the Consolidated Appropriations Act, 2021. Pitfall: making parity the whole argument. A parity appeal is a framing layer on the ordinary appeal, not a separate track, and the 2024 parity rule sits under a non-enforcement policy, so the clinical rebuttal leads and the parity demand rides along. Requested action and deadline language. Say exactly what you want: reverse the determination and pay the listed claims, or authorize the requested services. State that the appeal is timely, ask for a written decision, and reserve external review. Pitfall: a letter that documents frustration but never makes a grantable request; reviewers reverse specific determinations, they do not interpret complaints. Signature, authorization, enclosures. The filer signs the letter; the treating clinician signs the medical-necessity statement that rides as an enclosure. Anyone filing for the patient attaches the authorization: Form CMS-1696 for Medicare (a provider representative may not charge a fee for this), the enrollee's written consent for Medicaid managed care, or an authorized-representative designation for ERISA and ACA plans. End with a numbered enclosure list. Pitfall: the missing representative form is among the most common procedural dismissals; obtain it when the appeal is drafted, not after the payer bounces the filing. ### Blank template (copy and adapt) INSURANCE APPEAL LETTER (MENTAL HEALTH CLAIM) [Date] Sent via: [portal / fax / certified mail] To: [Plan name] Appeals Department [Address exactly as shown on the denial notice] RE: Appeal of adverse benefit determination Patient: [name] Member ID: _________ Group #: _________ Claim #(s): _________ Date(s) of service: _________ Denial notice date: _________ Denial code(s): [e.g. CO-50] Filing deadline per notice: _________ [ ] Expedited review requested 1) DETERMINATION APPEALED [Service, units, dates; quote the denial code and reason verbatim] 2) WHY THE DENIAL IS WRONG [Name the plan's medical-necessity policy; rebut each criterion it applied, point by point] 3) CLINICAL SUPPORT [Diagnosis with ICD-10 code; measure scores with dates; functional impairment; treatment history; risk factors; treatment plan goals] 4) RECORDS REQUESTED [Criteria applied; all documents relevant to the claim; for US mental health denials, the NQTL comparative analysis under CAA-2021 section 203] 5) REQUESTED ACTION Reverse the determination and [pay claim(s) / authorize services]. This appeal is timely filed. Please issue a written decision with external review instructions. Signature: ___________ Name/credentials: ___________ Date: ______ Filed by: [ ] patient [ ] provider [ ] authorized representative [ ] Authorization enclosed (CMS-1696 / written consent / designation) Enclosures: [ ] denial notice [ ] medical-necessity statement [ ] treatment plan + latest review [ ] measure log [ ] records Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample insurance appeal letter Scenario: a commercial ERISA group plan denies continued weekly psychotherapy as not medically necessary (CARC CO-50) after a session-count utilization review. The practice's billing manager files a timely first-level internal appeal as the patient's authorized representative, with the treating psychologist's medical-necessity statement enclosed. All details are fictional. First-level internal appeal · Plan type: ERISA group plan · Filed: 08/12/2026, via plan portal and certified mail August 12, 2026 Appeals Department, Northlake Health Plan P.O. Box 9410, Columbus, OH 43218 RE: Appeal of adverse benefit determination. Patient: R.D. · Member ID: NHP 402881 · Group: 7714 · Claims: 2026-0633317 and 2026-0648812 (eight psychotherapy sessions, 90834, dates of service 06/03/2026 to 07/29/2026) · Denial notice dated 08/01/2026, code CO-50, "not medically necessary." This appeal is filed 11 days after the notice, within the plan's 180-day appeal window. To the Appeals Reviewer: I write as the authorized representative of R.D. (signed designation enclosed) to appeal the denial of the claims listed above. The denial issued after a utilization review triggered at session 20 and states that continued weekly psychotherapy is "not medically necessary." The determination fails the plan's own criteria. Northlake Medical Policy BH-112 (Outpatient Psychotherapy) covers continued weekly treatment where a covered diagnosis persists, symptoms or functional impairment remain measurable, and the treatment plan shows expectable improvement toward defined goals. All three criteria are met. R.D. carries a diagnosis of major depressive disorder, recurrent episode, moderate (F33.1), confirmed at intake on 02/25/2026 and at the July treatment plan review. PHQ-9 scores fell from 18 on 03/04/2026 to 12 on 07/22/2026: real improvement that still sits in the moderate range. Missed work days, the impairment that brought R.D. to treatment, fell from four in March to one in July. Dr. S. Whitfield's enclosed medical-necessity statement maps each element of the record to BH-112, section by section. The review letter reasons that improvement shows the sessions are no longer needed. BH-112 says the opposite: expectable improvement is a continuation criterion, not a discharge trigger. The enclosed treatment plan already schedules a taper from weekly to biweekly sessions with a defined endpoint at the 10/2026 review; ending coverage mid-taper invites the relapse the policy's continued-care criteria exist to prevent. Under 29 CFR 2560.503-1(m)(8), please provide, free of charge, all documents relevant to this claim, including the clinical criteria applied and any internal or external review notes. Because this is a mental health claim, we also request the plan's comparative analysis of the nonquantitative treatment limitation applied here, required by section 203 of the Consolidated Appropriations Act, 2021, together with the utilization review criteria the plan applies to comparable outpatient medical care. Requested action: reverse the determination, pay claims 2026-0633317 and 2026-0648812 in full, and confirm coverage of the sessions scheduled under the current treatment plan through the 10/2026 review. If any part of the denial is upheld, please state the specific basis, the criteria applied, and the instructions for external review in your written decision. Sincerely, T. Alvarez, Billing Manager, Lakeshore Psychology Group Authorized representative for R.D. (designation dated 08/06/2026, enclosed) Treating clinician: S. Whitfield, PhD, Licensed Psychologist (NPI on file) Enclosures: (1) denial notice of 08/01/2026; (2) authorized-representative designation; (3) medical-necessity statement, S. Whitfield, PhD; (4) treatment plan and 07/2026 treatment plan review; (5) PHQ-9 log, 03/2026 to 07/2026; (6) session attendance record. This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or health plan. ↑ Back to the template and downloads ### Why this sample works It appeals the code, not the mood. CO-50 is quoted from the remittance advice, and the rebuttal walks the plan's own policy criterion by criterion, the one shape a reviewer can grant. Timeliness is arithmetic on the page. Notice date, filing date, and the 180-day window appear together, so a procedural dismissal has nothing to grab. The representative authorization is enclosed, not promised. The most common procedural own-goal in appeals is closed out in the signature block and the enclosure list. Improvement is framed as evidence for coverage, not against it. The dated PHQ-9 trend and the taper plan with a defined endpoint turn progress into the medical-necessity argument instead of a discharge rationale. The disclosure demands set up the next round. The relevant-documents request and the CAA-2021 comparative-analysis demand mean an upheld denial produces the record an external reviewer will read. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The appeal letter and everything enclosed with it join the claim file and the patient's designated record set: releasable on request, discoverable, and read mostly by non-clinicians, so write for a lay reviewer and enclose the specific record pages that do the arguing rather than the whole chart. Two disclosure rights carry real weight. On an ERISA appeal, the plan must provide, free of charge, all documents relevant to the claim, including the medical-necessity criteria it applied ( 29 CFR 2560.503-1 ); and plans subject to the ACA must state the denial code and its meaning on the notice itself ( 45 CFR 147.136 ), which is why the letter can quote the code back with confidence. Retention follows the chart: HIPAA sets no retention period for medical records (its six-year rule covers compliance documentation), so state law governs, commonly 5 to 11 years and longer for minors; keep the denial notice, the appeal, and every decision letter together with the claim they concern. The format is a convention; the deadline and the track are law. Name the track before drafting: ERISA internal appeal (at least 180 days to file), ACA external review (4 months, only after the internal appeal is exhausted), Medicare redetermination (120 days, the first of five levels), Medicare Advantage reconsideration (60 days, and a plan that upholds itself must forward the case to the independent review entity automatically), or Medicaid managed care (60 days to the plan, then 120 days for a state fair hearing). File the written appeal at the same time as any peer-to-peer request so the clock cannot expire while a call is being scheduled. Then let documents that already exist carry the clinical weight: the latest treatment plan review shows goals moving, and the medical-necessity letter turns the record into the clinical argument, leaving the appeal letter short, procedural, and specific about what it wants. The enforcement record says these fights are winnable: in a 2021 action announced jointly by the New York Attorney General and the US Department of Labor, United Behavioral Health agreed to pay approximately $14.3 million in restitution over practices that included cutting allowed amounts for psychologist and masters-level services and a review program triggered once a member passed 20 psychotherapy visits in six months. ## Common insurance appeal letter errors auditors flag The appeal system is defined by how rarely it is used and how often it works. In KFF's analysis of 2023 HealthCare.gov claims data , consumers appealed less than 1% of denied in-network claims, and insurers reversed about 44% of the appeals actually filed. In Medicare Advantage, 81.7% of appealed prior-authorization denials in 2023 were partially or fully overturned, and Medicaid external medical reviewers found fully or partially for the beneficiary 46% of the time. The appeals that lose tend to lose on procedure. The BastionGPT Clinical Advisory Board sees the same errors most often in insurance appeal letter reviews: Wrong track, dead appeal. The letter goes to the claims address instead of appeals, argues a Medicare denial like a commercial one, or misses the window: at least 180 days for an ERISA internal appeal, 120 for a Medicare redetermination, 60 for Medicare Advantage and Medicaid managed care. Routing and deadline errors, not weak clinical arguments, kill most appeals. No representative authorization. The practice files on the patient's behalf with nothing attached. Medicare needs Form CMS-1696 or an equivalent written appointment, Medicaid managed care requires the enrollee's written consent, and ERISA and ACA plans require an authorized-representative designation; without one, the appeal dies on procedure however strong the clinical case. The code never appears. The letter rebuts a paraphrase of the denial instead of quoting CO-50 or CO-197 from the remittance advice, or appeals a CO-16 missing-information denial that needed a corrected claim. Generic necessity instead of the plan's criteria. No named medical policy, no point-by-point rebuttal, and no written request for the criteria the reviewer used, which an ERISA plan must hand over free of charge on appeal. Parity treated as its own appeal type. The letter leans entirely on the 2024 parity rule, which has been under a federal non-enforcement policy since May 2025, instead of leading with the clinical rebuttal and adding the CAA-2021 comparative-analysis demand as the pressure layer. ### Insurance appeal letters in the US, Canada, and Australia Aspect United States Canada Australia Status Letter content is convention; law fixes deadlines, reviewer independence, and disclosure (ERISA 29 CFR 2560.503-1, ACA 45 CFR 147.136, Medicare 42 CFR Part 405) Convention; the insurer's internal complaint process runs first, then OLHI escalation once a final position letter issues; public-plan appeals are provincial (e.g. Ontario's HSARB) Convention; insurer internal dispute resolution first, then the Private Health Insurance Ombudsman; PHI complaints may even be made by phone Terminology Internal appeal, redetermination (Medicare), reconsideration (Medicare Advantage), external review Complaint, final position letter, OLHI complaint; HSARB appeal (Ontario public plan) Internal dispute resolution (IDR), PHIO complaint, HW051 review of a Medicare decision What changes Track and filing window by plan type: 180 days (ERISA internal), 4 months (ACA external review), 120 days (Medicare redetermination), 60 days (Medicare Advantage and Medicaid); parity framing available No parity statute; OLHI recommendations are non-binding; each province routes public-plan appeals differently No parity statute; PHI complaints go to the PHIO, not AFCA; the ART does not hear a consumer's rejected Better Access claim Retention No federal medical-record retention rule; state law governs (commonly 5 to 11 years, longer for minors); keep the appeal file with the claim records Provincial college and privacy standards govern (commonly 10 years for adults) State and territory health-records law governs The letter travels; the machinery around it does not. Confirm the escalation body and the filing window before adapting a US letter for another country, and leave the parity argument at the border: it has no statutory footing outside the US. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on insurance appeal letters. Draft a complete appeal from the denial notice details, a few bullets, or a dictated summary, built around the specific denial code being rebutted and the plan type's deadline language. Merge the sources into one argument: paste the treatment plan, measure scores, and the medical-necessity statement, and get a point-by-point rebuttal of the plan's criteria with a matching enclosure list. Check a finished draft before it goes out: timeliness stated, code quoted verbatim, requested action explicit, disclosure requests included, representative authorization referenced. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should an insurance appeal letter be? Most run 1 to 3 pages, about 400 to 900 words, and take 45 to 90 minutes to assemble by hand with enclosures. Length does not decide appeals: Medicare accepts a bare written request, and no payer awards points for volume. What the letter needs is the claim identifiers, the denial code quoted verbatim, a rebuttal of the plan's own criteria, a short clinical evidence summary, and a specific requested action. Put the depth in the enclosures and keep the letter procedural. Is there a required format for an insurance appeal letter? No. No statute, regulation, or payer manual in the US, Canada, or Australia prescribes the letter's format or content. The law fixes deadlines, who may file, reviewer independence, and what the payer must disclose. Medicare makes the point explicitly: a redetermination needs only a written request containing identifying details, or optional Form CMS-20027, and courts have treated even a handwritten note saying "I appeal" as preserving ERISA rights. Structure the letter anyway: one organized around the denial code and the plan's criteria is easier to grant. What deadlines apply to a mental health insurance appeal? The filing windows are law and vary by track. ERISA employer plans must give you at least 180 days for the internal appeal. ACA external review must be requested within 4 months of the final internal denial. Medicare redeterminations are due within 120 days, Medicare Advantage reconsiderations within 60 days (receipt of the notice is presumed 5 days after its date), and Medicaid managed care allows 60 days for the plan appeal, then 120 days after its resolution for a state fair hearing. Expedited paths run as fast as 72 hours. The deadline printed on your denial notice controls; calendar it the day the notice arrives. Can I file an appeal on my client's behalf? Only with written authorization. Medicaid managed care requires the enrollee's written consent (42 CFR 438.402(c)). Medicare requires Form CMS-1696 or an equivalent written appointment, valid for one year; a provider who represents a beneficiary may not charge a fee for the representation. ERISA and ACA plans require an authorized-representative designation under the plan's procedures. Attach the authorization to the appeal itself; filing without it is one of the most common procedural dismissals. What is a parity appeal, and should I file one? A parity appeal is an ordinary appeal that adds a comparison argument: the plan applied a stricter standard to mental health or substance use care than to comparable medical or surgical care. It is not a separate legal track, and it has no statutory footing outside the US. The strongest parity move is a records demand: section 203 of the Consolidated Appropriations Act, 2021 requires plans to document a comparative analysis of every nonquantitative treatment limitation and produce it on request, and failure to produce one is itself evidence of a parity problem. Cite the 2013 parity rule and the CAA-2021 duty; the 2024 final rule has been under a federal non-enforcement policy since May 2025, so lead with the clinical rebuttal. What is the difference between an appeal letter and a medical-necessity letter? A medical-necessity letter argues that care is clinically needed; it can be written before care to support a prior-authorization request or after a denial, where it becomes the appeal's central enclosure. The appeal letter argues that a specific adverse determination was wrong, on the plan's own criteria and the record, and asks for a defined reversal. A utilization-review summary defends care mid-course during concurrent review, and a peer-to-peer call is a conversation that is not an appeal at all. When a denial arrives, file the written appeal even if a peer-to-peer is scheduled: the call can fix the problem, but only the filing protects the deadline. Do appeals of mental health denials actually work? More often than the filing rates suggest. In KFF's analysis of 2023 HealthCare.gov claims, consumers appealed less than 1% of denied in-network claims, and insurers reversed about 44% of the appeals that were filed. In Medicare Advantage, 81.7% of appealed prior-authorization denials in 2023 were partially or fully overturned, and Medicaid external medical reviewers found fully or partially for the beneficiary 46% of the time. Filing also does not invite scrutiny of your practice: no primary source links a patient appeal to provider audit selection. Who handles insurance appeals in Canada and Australia? In Canada, exhaust the insurer's internal complaint process and obtain its final position letter, then escalate to the OmbudService for Life and Health Insurance (OLHI), a voluntary industry body whose recommendations are non-binding; public-plan denials follow provincial routes, and Ontario's appeal board upholds OHIP in nearly all decided cases. In Australia, complain to the insurer first, then to the Private Health Insurance Ombudsman , part of the Commonwealth Ombudsman; AFCA handles life and general insurance, not private health. A rejected Medicare (MBS) claim goes to Services Australia on form HW051 within 28 days. Neither country has a mental health parity statute, so leave that argument out. Can BastionGPT write an insurance appeal letter? Yes. Give it the denial notice details and the supporting records, or dictate what happened, and it drafts the appeal around the code being rebutted, with timeliness language, disclosure requests, and the enclosure list in place for your review. It can also check an existing draft against the plan type before it goes out. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Prior Authorization Support Letter Utilization Review Summary Claim-Support Billing Note Prior Authorization Request Medical Necessity Letter Peer-to-Peer Review Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/peer-to-peer-note All note templates # Peer-to-Peer Review Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A peer-to-peer review note is the treating clinician's record of a call with a payer's medical or behavioral health reviewer about the medical necessity of requested services. Clinicians write it right after a payer P2P call, most often around a prior authorization or concurrent review denial. Most run 150 to 300 words and are filed in the general chart, not psychotherapy notes. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating therapists, psychologists, and psychiatrists; staff may log scheduling and attempts Audience The clinical chart, billing and utilization staff, and payer or appeal reviewers later Typical length 150 to 300 words · 10 to 15 minutes by hand (clinical team estimate) Format family Payer communication note (compare: utilization-review summary, prior-authorization request, insurance appeal letter) When it's used Right after a payer peer-to-peer call: before a determination, during concurrent review, or post-denial where allowed Standards context No law prescribes the note; payer policy governs the call, and documenting it is a defensible convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a peer-to-peer review note? A peer-to-peer review note is the treating clinician's written record of an oral discussion with a health plan's medical or behavioral health reviewer about the medical necessity, level, duration, or coverage of requested services. The document has no inventor and no standards body: it grew up inside insurer utilization management workflows, and URAC's health utilization management standards recognize the peer-to-peer conversation as a process the reviewing organization must run, without prescribing any note for the treating clinician's chart. Depending on the payer, the call itself is named a P2P review, peer discussion, physician-to-physician review, medical-director review, or reconsideration discussion. Say "payer peer-to-peer" when you file the note: unqualified "peer review" also names hospital quality review and professional discipline processes, which are legally distinct. Two boundaries do most of the work here. First, the call is not an appeal. No national regulation in the US, Canada, or Australia says a peer-to-peer call constitutes, initiates, extends, or tolls a formal appeal, and payer policy varies: one plan folds the call into its appeal process, another offers it only before the determination, and some Medicare Advantage workflows run post-decision calls that are informational and cannot reverse the denial. When a denial has issued, file the insurance appeal letter on its own clock rather than letting a scheduling queue eat the deadline. Second, your note is not the payer's record. The plan keeps its own utilization management case file with worksheets, criteria, and sometimes recordings; no regulation requires the two records to match, which is exactly why your version of the conversation needs to exist. ## Who uses peer-to-peer review notes and when Any clinician who takes payer calls ends up needing peer-to-peer review documentation: outpatient therapists and psychologists defending a time-based psychotherapy code under concurrent review, prescribers responding to an intent-to-deny, and facility utilization teams working residential and IOP continued-stay requests. Who may conduct the call is payer- and state-specific: Texas requires the utilization review agent to offer a physician of the same or similar specialty, while URAC's clinical peer definition reaches other licensed professionals, so confirm eligibility before the call is booked. The note earns its place among its neighbors by what it preserves. A prior-authorization request opens the administrative record before service, and a utilization-review summary feeds the reviewer current clinical status; the peer-to-peer note is the only document that records what the payer's reviewer actually said, cited, and agreed to. ## Peer-to-peer review note structure: what goes in each section Case and call identification. Member name and ID, plan and product, the payer's case or reference number, any authorization number, the service at issue, and the call logistics: date, start and stop times, platform, whether the call was scheduled, and any missed attempts. Pitfall: batching several patients into one entry or upload. Blue Cross Blue Shield of Michigan warns that combining members in one case submission violates privacy requirements, and a call with no reference number is hard to prove ever happened. Procedural posture. The most consequential field in the note: is the request pending, under intent-to-deny, in concurrent review, already denied, or inside a formal appeal? Name the line of business (commercial, ERISA plan, Medicare Advantage, Medicaid managed care) and record the denial date and the appeal deadline as two separate facts. Pitfall: assuming the call protects the appeal. It does not; some payers run post-denial calls that cannot change the decision, and the deadline runs while the call is being scheduled. Participants. Your name, credential, and license; the reviewer's name, degree, specialty, title, and organization as far as disclosed; anyone else on the line and their role. Pitfall: expecting the reviewer's identity to arrive in writing later. NCQA's 2025 utilization management standards do not require reviewer names on denial or appeal notices, so the call may be your only chance to capture who decided, or to record that they declined to say. Clinical points discussed. What you told the reviewer that bears on the criterion in dispute: current symptoms with acuity, measurable function, risk and protective factors, response to treatment with dated scores, alternatives tried or contraindicated, and why a lower intensity fails today. Pitfall: a diagnosis-and-symptoms summary. Necessity criteria turn on function, risk, and treatment response; a note that omits them records a call that never engaged the actual question. Criteria cited. The guideline or policy the reviewer applied, with edition or version, the specific criterion said to be unmet, and the facts the reviewer relied on; note whether the reviewer separated coverage criteria from clinical advice. Pitfall: leaving the criterion generic. An ERISA plan must disclose the rules it relied on when asked, and an appeal cannot rebut a criterion the note never captured. Outcome and authorization details. The disposition in the payer's own terms: approved, partially approved, modified, upheld, pending more information, or informational only; then the transaction: authorization number, approved dates and units, conditions, and whether written confirmation will follow. Pitfall: scheduling care on an assumed approval. Until the payer's system shows the authorization, an oral yes is a claim risk, so record the confirmation you were promised and chase it. Next steps, owners, deadlines. What happens now: appeal or expedited appeal, records to submit, treatment plan changes, patient notification, the responsible person, and every live deadline. Pitfall: assuming a missed call gets rebooked. At least one major plan states a missed or refused peer-to-peer will not be rescheduled and the provider must appeal; record the attempt and go straight to the appeal calendar. ### Blank template (copy and adapt) PEER-TO-PEER REVIEW NOTE (PAYER CALL) Client: [initials] Member ID: DOB: Payer / plan: Line of business: [commercial/ERISA/MA/Medicaid] Case / ref #: Auth #: Service at issue: [e.g. 90837] Call date: Start / stop: Scheduled: [Y/N] Attempts: Posture: [pending / intent to deny / concurrent review / post-denial] Denial date (if any): Appeal deadline: Participants (name, credential, specialty, organization): Treating clinician: Payer reviewer: Clinical points discussed (function, risk, response, alternatives): Criteria cited (set / policy, edition, criterion at issue): Outcome: [approved / modified / upheld / pending / informational only] Auth # / dates / units: Written confirmation expected: Next steps (action, owner, deadline): Signature / credentials: Date and time of entry: Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample peer-to-peer review note Scenario: an outpatient psychologist documents a scheduled peer-to-peer call with a payer medical director after a concurrent-review denial of continued weekly 90837 sessions. All details are fictional. Client: J.R., 41 · Call date: 08/11/2026 · Entry: 08/11/2026, 3:40 PM · Author: R. Okafor, PsyD Case and call: Caldera Health Plan (commercial ERISA group plan). Member ID CHP 552873, case #CHP-2026-118240, auth #A66412. Service at issue: continued weekly 90837. Scheduled peer-to-peer, 2:00 to 2:22 PM ET, payer teleconference line, first attempt, reference #P2P-40917. Posture: Concurrent-review denial dated 08/06/2026 for sessions after 08/15/2026. Denial notice states a 180-day internal appeal window; appeal deadline calendared for 02/02/2027. Plan offered this call as an optional pre-appeal peer discussion; reviewer confirmed appeal rights are unaffected by the call. Participants: R. Okafor, PsyD, treating psychologist (license and NPI on file). For the plan: M. Sandoval, MD, psychiatrist, associate medical director, behavioral health. Clinical points discussed: Course since 03/2026 intake for major depressive disorder, recurrent, moderate (F33.1). PHQ-9: 21 on 03/10/2026, 17 on 06/02/2026, 14 on 08/04/2026. Passive ideation reported at intake, denied at the last two administrations; risk reviewed each session. Trial of biweekly frequency in 05/2026 was followed by a symptom rebound (PHQ-9 back to 19) and weekly cadence was restored. Discussed why the longer session supports the exposure and behavioral activation work that shorter sessions truncated, and the taper plan tied to the October treatment plan review. Criteria cited: Reviewer applied Caldera medical policy BH-OP-07, Outpatient Psychotherapy, rev. 01/2026, criterion 3: continued weekly frequency requires measurable symptom or functional change within the prior 90 days. Reviewer stated the dated PHQ-9 trend plus the documented relapse on spacing satisfied criterion 3. Outcome: Denial reversed on review. Approved: 12 additional weekly 90837 sessions, 08/16/2026 through 11/07/2026, revised auth #A66412-R1. Written confirmation promised to the provider portal within 3 business days. Next steps: (1) Verify revised authorization in portal by 08/14/2026 (owner: front office). (2) Appeal deadline of 02/02/2027 stays on the calendar until the written authorization posts. (3) Next concurrent review expected near session 10; utilization-review summary due then (owner: R.O.). Signed: R. Okafor, PsyD, 08/11/2026, 3:40 PM. This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or health plan. ↑ Back to the template and downloads ### Why this sample works Posture and line of business are on the page. ERISA plan, concurrent-review denial, optional pre-appeal call: the three facts that decide what this call could legally change and which deadline governs. The appeal deadline survives the good news. The 02/02/2027 date stays calendared until the written authorization posts, so an oral approval that never lands in the payer's system cannot strand the claim. The reviewer is identified at call time. Name, credential, specialty, and title are captured in the one place they are reliably available; no accreditation rule puts them on the denial notice. The criterion is recorded with its version. Policy BH-OP-07, criterion 3, and the facts that satisfied it give any later appeal or audit a specific standard to point at, not a vague recollection. The outcome is a transaction, not a mood. Auth number, 12 sessions, dates, a confirmation window, and named owners for each next step mean billing and scheduling run on the record, not on optimism. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations File the note in the general clinical or administrative record, not in a segregated psychotherapy notes repository: scheduling, billing, and utilization staff need the outcome, and authorization status affects access to care. HIPAA treats the call itself as a payment and operations disclosure that needs no separate authorization, but the minimum-necessary standard still applies, so the note should carry what bears on the coverage criterion and leave intimate session content out. Remember that the plan documents the same call in its own utilization management file; no regulation requires the two records to match, and yours is the version that protects the patient. Retention follows the chart the note lives in: HIPAA's six-year rule covers compliance documentation such as policies, not medical records, so state licensing and record law, payer contracts, and frameworks like Ontario's ten-year client-record standards set the clock. Sort the rules around the call by their real strength. LAW fixes the machinery, not the note: ERISA's claims-procedure regulation ( 29 CFR 2560.503-1 ) guarantees an independent, qualified reviewer on appeal and access to the criteria relied on; Medicare Advantage rules ( 42 CFR 422.584 ) make the payer, not you, reduce an oral expedited reconsideration request to writing in its case file; Texas law requires utilization review agents to offer a discussion with a same-or-similar-specialty physician. PAYER POLICY sets the operational rules that bite: request windows as short as seven business days, calls restricted to medical-necessity denials (a medical director cannot cure an eligibility or benefit-exclusion problem), and post-denial calls that are informational only. Everything else, same-day charting, the field list, signature timing, is CONVENTION: defensible practice, not statute. The practical consequence: when a denial has issued, draft the insurance appeal letter in parallel and let the utilization-review summary carry the clinical detail, so the call can help but can never become the reason the deadline passed. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common peer-to-peer review note errors auditors flag No regulator has audited peer-to-peer notes as a category, and this page will not invent a statistic where none exists; the audit record around the calls is loud enough. A 2022 HHS OIG review of Medicare Advantage prior-authorization denials found 13% of denied requests actually met Medicare coverage rules, and OIG's first behavioral health denial audit , published July 2026, found none of the 100 sampled denials at one Pennsylvania Medicaid plan met every applicable requirement. The AMA's 2025 prior-authorization survey of 1,000 physicians reports 40 prior authorizations per physician per week, 13 combined physician and staff hours spent on them, and only about one third of physicians confident that a licensed, qualified clinician reviewed their medical-necessity denials. That is the environment this note documents. The BastionGPT Clinical Advisory Board sees the same errors most often in peer-to-peer review note reviews: The call substitutes for the appeal. Appeal windows run while staff wait for a discussion. No national regulation makes a P2P call start, extend, or toll an appeal, and some payer policies make a post-denial call informational only, so the written appeal is filed late or never. Wrong denial type on the call. A medical director cannot cure eligibility, benefit-exclusion, referral, or other administrative defects, and payer manuals restrict peer-to-peer to medical-necessity denials; the note should show the denial type was checked before the call was requested. No criterion, no version. The note says the reviewer disagreed but never names the guideline, edition, or the specific criterion at issue, leaving a later appeal nothing concrete to rebut even though an ERISA plan must disclose the rules it relied on. An outcome without a transaction. No authorization number, units, dates, or written-confirmation expectation, so services get scheduled against an approval the payer's system never recorded and the claim denies anyway. A clinical summary that never meets the criterion. Diagnosis and symptoms are recorded, but the function, risk, treatment response, and failed alternatives that necessity criteria actually test are missing, so the note cannot support the next round of review. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on peer-to-peer review notes. Summarize the call the moment it ends: dictate what was said and get a structured note with participants, posture, criteria, outcome, and next steps in their places. Prepare you before the call: condense the chart into the points reviewers test, with dated measure trends, risk factors, failed alternatives, and the criterion language at issue. Check the note before you sign: line of business named, reviewer identified or the refusal recorded, criterion and version captured, authorization details and appeal deadline present. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a peer-to-peer review note be? Most run 150 to 300 words and take 10 to 15 minutes right after the call. Length is beside the point; the note is transactional. Ten short fields, who, when, posture, criteria, outcome, authorization details, next steps, protect more value than a page of narrative. Write it the same day: names, reference numbers, and deadlines are the fastest-decaying facts in the chart. Is there a required format for documenting a peer-to-peer call? No. No statute, regulation, CMS rule, or accreditor in the US, Canada, or Australia prescribes a note format, field list, title, or signature deadline for the treating clinician's record of a payer call. URAC and NCQA standards govern the payer's review process, not your chart. The structure on this page is convention built from payer workflows: defensible because it captures what later disputes turn on, not because any law requires it. Is a peer-to-peer call the same as an appeal? No, and the difference has teeth. No national regulation makes a P2P call start, extend, or preserve an appeal, and payer policy varies: some plans fold the call into an appeal, others offer it only before the determination, and some Medicare Advantage workflows run post-denial calls that are informational and cannot reverse the decision. The relief in the audit record comes from filed appeals: in OIG's 2026 review of Medicare Advantage skilled nursing denials, a different care setting, 95% of appealed denials were overturned. When a denial issues, start the insurance appeal letter immediately, with the medical-necessity letter as its central enclosure, and treat the call as a bonus path. Who can conduct the peer-to-peer call for my practice? Payer, product, and state law decide. Texas law requires the utilization review agent to offer a discussion with a Texas-licensed physician of the same or similar specialty, while URAC's clinical peer definition reaches other licensed clinicians with relevant expertise, and behavioral health plans variously use psychiatrists, doctoral psychologists, or other licensed reviewers. Nothing guarantees a therapist, nurse practitioner, or supervisee will be accepted, so confirm eligibility when the call is scheduled. Supervision adds a layer: no national rule makes a supervisor co-sign a call log, but Ontario, for example, requires co-signature of formal reports and correspondence prepared under supervised practice. What if the reviewer will not give a name, specialty, or the criteria? Document the request and the refusal; that line is evidence. On an ERISA plan you can then demand, in writing and free of charge, the criteria relied on and the identity of medical experts whose advice the plan obtained (29 CFR 2560.503-1). Do not expect the notice to fill the gap: NCQA's 2025 utilization management standards do not require reviewer names or signatures on denial or appeal notices. In the AMA's 2025 survey, only about one third of physicians were confident that a licensed, qualified clinician reviewed their medical-necessity denials, which is exactly why your note captures whatever the reviewer discloses. Can I bill for the time spent on a peer-to-peer call? There is no stand-alone billable service for a payer peer-to-peer call; practice management forums have asked since at least 2008, and the consistent answer is that authorization work is administrative overhead. Whether any of the time can count inside another billed service depends on the specific coding rules, payer contract, and date of service, and documentation alone does not make it billable. Charging the patient instead is not a safe default either: network contracts and consumer protection rules may prohibit it, so get contract-specific advice before creating a fee. How long do I need to keep a peer-to-peer review note? As long as the clinical record it belongs to. HIPAA sets no retention period for medical records (its six-year rule covers compliance documentation such as policies), so state licensing and record law, payer contracts, and malpractice exposure set the clock. In Ontario, the ten-year client-record standards of the psychotherapy and psychology colleges reach any note filed in the client record. In Australia, do not mistake the Medicare two-year referral-retention rule for a record period; state, territory, and Board expectations govern the file. Where should the note live in the chart? In the general clinical or administrative record, alongside documents like the billing note , not in a segregated psychotherapy notes repository. Staff who schedule and bill need the authorization outcome, and HIPAA's special psychotherapy-notes protections would only wall it off from them. The reverse rule matters too: keep intimate session content out of the call note. The payer needs what bears on the coverage criterion, and the minimum-necessary standard applies to what you disclose. Can BastionGPT write a peer-to-peer review note? Yes. Dictate what happened on the call, or paste your scratch notes, and it returns a structured note with posture, participants, criteria, outcome, and next steps in their fields, ready for your review. Before the call, it can condense the chart into the talking points reviewers test. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Prior Authorization Request Insurance Appeal Letter Medical Necessity Letter Utilization Review Summary Claim-Support Billing Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/abn-notice All note templates # Advance Beneficiary Notice (ABN): Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An Advance Beneficiary Notice (ABN), Form CMS-R-131, is the standardized written notice that tells an Original Medicare patient, before a specific service, that Medicare is expected to deny payment, so the patient can choose whether to proceed and accept the cost. Practice and billing staff issue it when a case-specific denial is expected, such as psychotherapy beyond covered frequency. The completed notice is one page. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Front-desk, billing, and practice administration staff; the billing clinician remains responsible for a valid notice Audience The Original Medicare patient or their representative; the billing file; Medicare contractors on request Typical length One page, about 100 to 200 words of practice-entered content · 5 to 10 minutes by hand (clinical team estimate) Format family Standardized CMS payer notice (compare: good faith estimate, financial policy) When it's used Before an Original Medicare service the practice expects Medicare to deny in that specific case, such as psychotherapy beyond a covered frequency Standards context The liability framework is law (section 1879, Social Security Act; 42 CFR 411.404); the form and its completion rules are Medicare payer policy On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an Advance Beneficiary Notice? An Advance Beneficiary Notice of Noncoverage, Form CMS-R-131, is the standardized notice a practice gives an Original Medicare patient before furnishing a service Medicare is expected to deny, so the patient can make an informed choice about receiving it and accepting the potential cost. The mechanism is older than the form: Congress created the limitation-on-liability framework in section 1879 of the Social Security Act in 1972, and the implementing regulation at 42 CFR 411.404 lets appropriate written notice establish that a patient knew Medicare would not pay. CMS unified today's single form in 2008, replacing the earlier general-use and laboratory versions. You will also hear it called an ABN, a Medicare waiver, or a waiver of liability, and the common expansion "Advanced Beneficiary Notice" is technically incorrect: the official adjective is "advance." Two boundaries do the heavy lifting. First, the ABN belongs to Original Medicare fee-for-service only: it has no standing for Medicare Advantage enrollees, who sit under their plans' own determination and notice systems, and it is not a good faith estimate , which serves uninsured and self-pay clients under different law. Second, the official form is the requirement. CMS-R-131 is an approved standard notice (OMB control number 0938-0566, currently approved through March 2029), and Medicare's Claims Processing Manual warns that substituting another notice or materially altering the required language or formatting can invalidate the notice and leave the practice liable. Download the current form and its instructions from CMS; the template on this page is a preparation worksheet that mirrors the required elements so your entries are complete before anyone signs. ## Who uses ABNs and when Any practice that treats Original Medicare patients and bills Medicare: solo therapists and psychologists, group practices, and clinic billing teams. Delivery is usually administrative, since CMS lets an employee or subcontractor present and explain the notice while the billing clinician stays responsible for its validity. The trigger is narrow: issue a mandatory ABN when you expect Medicare to deny an identified service in this particular case for a limitation-on-liability reason, most often frequency beyond coverage limits or an anticipated medical-necessity denial, not as a routine intake form for every Medicare patient. For uninsured and self-pay clients the parallel document is the good faith estimate ; ordinary rates, cancellation charges, and payment terms for everyone belong in the practice financial policy . ## ABN structure: what goes in each section The official CMS-R-131 is a one-page form with a fixed layout, and every element below maps to a field on it. Complete all of them before the patient signs: an incomplete or altered notice may be disregarded, which returns the cost to the practice. Notifier and patient identification. The practice's name, address, and telephone number, and the patient's name. When the notifying and billing entities differ, both can be identified, with billing contact details made clear. Pitfall: the identification field is optional, and the Medicare Beneficiary Identifier and Social Security number should not appear on the ABN. Obsolete copied templates still circulate with a "Medicare #" field; use an internal chart number or nothing. The specific service expected to be denied. Name the particular item or care, not a category. For a continuing course of psychotherapy, state the frequency and expected duration; for a partial denial, identify the excess component, such as the second session each week. Pitfall: broad entries like "all therapy sessions" identify no specific expected denial, and a blanket form signed at intake fails for the same reason. The reason Medicare may not pay. A plain-language, case-specific reason for each listed service, for example that continued sessions at this frequency are not expected to meet Medicare's medical-necessity coverage rules. Pitfall: generic wording such as "Medicare may not pay" is insufficient because it states no basis; CMS's completion instructions require an applicable reason for every listed service. The estimated cost. A good-faith estimate of what the patient would owe. It does not need to be exact, and CMS accepts an estimate that runs substantially higher than the final charge because it does not surprise the patient in the harmful direction. Pitfall: leaving the estimate blank violates the completion instructions; write a number, per session and for the course where relevant. The option boxes. The patient or their representative picks one of three choices: receive the service and have Medicare billed, which preserves Medicare appeal rights; receive it with no Medicare claim, which gives up those rights for the transaction; or decline the service. Pitfall: staff or software preselecting an option invalidates the notice. The choice is the patient's, every time. Signature, date, and delivery. The patient or representative signs and dates after reviewing the completed notice and before the service. Explain the form, answer questions, and give a copy; current CMS sources differ on whether a copy is owed routinely or on request, so handing one over every time and documenting it is the safe practice. Pitfall: no federal ABN rule requires the clinician or a supervisor to sign, and a signature obtained after the session cannot retroactively shift liability. ### Blank template (copy and adapt) ADVANCE BENEFICIARY NOTICE (ABN) PREPARATION WORKSHEET Original Medicare FFS only. The official Form CMS-R-131 from CMS is the notice you must issue; this worksheet readies your entries so that form is valid before anyone signs. NOTIFIER AND PATIENT Practice name / address / phone: _______________________________ Patient name: ____________________ Internal ID (optional): ____ Never enter the Medicare number (MBI) or SSN on the notice. SERVICE EXPECTED TO BE DENIED (specific, never a category) Service or item: _______________________________________________ Course, if repetitive (frequency and duration): ________________ Excess component, for a partial denial: ________________________ REASON MEDICARE MAY NOT PAY (plain language, case-specific) ________________________________________________________________ ________________________________________________________________ ESTIMATED COST (good-faith figure; never leave blank) Per session: $__________ Course total, if stated: $___________ OPTIONS (the patient's choice; never preselect) [ ] Option 1: bill Medicare (Medicare appeal rights preserved) [ ] Option 2: no Medicare claim (no Medicare appeal) [ ] Option 3: patient declines the service Refusal to sign or choose: annotate on the notice and date it. SIGNATURE AND DELIVERY (before the service) Patient or representative signs and dates the official form. Explained by: ______________ Date: ________ Copy given: [ ] CLAIM AND FILE Modifier: [ ] GA (valid ABN on file) [ ] GZ (none obtained) Retain the signed notice 5 years (longer if state law requires) and produce it if the Medicare contractor requests it. Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample ABN Scenario: an Original Medicare psychotherapy patient continues twice-weekly sessions beyond the frequency the practice expects Medicare's coverage rules to support; the practice issues the official form before the next session. All details are fictional. Advance Beneficiary Notice, prepared entries and file note. Cedar Grove Psychological Services · Patient: L.H. · Notice issued: 08/04/2026 · Instrument: official Form CMS-R-131 Service expected to be denied: Individual psychotherapy (90834) at twice-weekly frequency, 08/10/2026 through 10/30/2026. Medicare continues to be billed for one covered session each week; this notice covers the second weekly session only, the component the practice expects Medicare to deny. Reason Medicare may not pay, as entered: Medicare pays for psychotherapy that meets its medical-necessity rules. For your condition, we expect Medicare to decide that a second session each week is more frequent than its coverage rules allow, so we expect it to deny the second weekly session. Estimated cost, as entered: $145 per session, about $1,740 across the 12-week course if every second weekly session is denied. Entered as a good-faith figure before signature; the practice noted the final charge may come in lower. Option and signature: L.H. read the completed notice, chose Option 1 (receive the sessions and have Medicare billed, keeping her Medicare appeal rights), marked the box herself, and signed and dated on 08/04/2026, before the 08/10/2026 session. No option was preselected. Delivery and file note (R. Ortiz, billing coordinator): Notice reviewed with L.H. by phone on 08/03/2026 and in person on 08/04/2026; her questions about the appeal path were answered, and she took a completed copy home. Claims for the second weekly session will carry modifier GA. Signed original scanned to the billing file, retained five years, produced on contractor request. If Medicare pays a comparable claim during the course, the practice will issue a fresh notice rather than rely on this one. This sample is fictional and for educational purposes. It does not describe a real patient or practice. ↑ Back to the template and downloads ### Why this sample works The service entry is specific and scoped. It names the service, the frequency, the duration, and the excess component (the second weekly session), so one notice covers a defined course instead of working as a blanket form. The reason is case-specific and in plain language. It states the expected coverage problem for this patient rather than the generic wording CMS's completion instructions treat as insufficient. The estimate is a written good-faith figure. Per session and for the course, entered before signature, meeting the rule that the cost field never stays blank. The option was the patient's own, chosen before the service. No preselection, signature and date ahead of the session, explanation and questions documented, copy given: the delivery facts contractors test. The claim plan matches the file. Modifier GA is reported only because a valid signed notice exists, the form is retained five years, and the practice knows a later favorable payment weakens the notice for future sessions. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Sort every ABN habit into law, payer policy, or convention. The liability shift itself is law: section 1879 of the Social Security Act and 42 CFR 411.404 decide when a patient can be held to have known Medicare would not pay. How you get there is Medicare payer policy in the Claims Processing Manual, chapter 30: the approved form, the completion rules, the delivery and comprehension expectations, and retention of the signed notice for five years (longer where state law requires). Where the sources go quiet, convention fills the gap: no rule sets a minimum number of hours before the session, no rule prescribes mental-health wording for the reason field, and current CMS sources even differ on whether the patient's copy is owed routinely or on request, so the defensible practice is to deliver early, write plainly, and hand over a copy every time. A signed page is not the finish line either: reporting modifier GA only certifies that a valid notice exists, and a contractor can inspect the form, find it defective, and move the liability back to the practice. The notice also proves nothing about medical necessity; the claim still stands or falls on the clinical and billing record (see the billing note ), and the signed ABN sits in the billing record a patient can obtain through a records request . The form's scope is narrower than most practices assume. It is an Original Medicare fee-for-service instrument: Medicare Advantage plans run their own determination and notice processes, clinicians who have opted out of Medicare set financial terms through private contracts instead, and services Medicare never covers call for a voluntary notice the patient does not need to sign. Special protections override it too: Qualified Medicare Beneficiaries generally cannot be billed for Medicare-covered services regardless of a signed ABN, and CMS's instructions carry specific QMB directions. Keep the neighbors straight as well: the good faith estimate serves uninsured and self-pay clients (a Medicare enrollee generally does not qualify as uninsured for it), while routine charges such as missed-session fees are contract terms for your financial policy and are documented through no-show notes , not ABNs. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common ABN errors auditors flag No CMS, OIG, or CERT publication reports an ABN-specific error rate for outpatient mental health, so treat any percentage you see quoted as borrowed from somewhere else. The scale and the consequences are documented, though. The Office of Management and Budget's 2026 approval package for Form CMS-R-131 models roughly 331.7 million notice responses a year across about 1.7 million respondents, burden estimates that work out to about seven minutes per notice, and Medicare's Claims Processing Manual spells out the audit mechanics: when a claim reports modifier GA, the contractor can request the form, judge its validity, and reassign liability to the practice if the notice fails. The BastionGPT Clinical Advisory Board sees the same errors most often in Advance Beneficiary Notice reviews: Blanket forms at intake. Having every Medicare patient sign one just in case identifies no case-specific expected denial, so the notice establishes no informed knowledge and shifts nothing. A generic reason. Wording like "Medicare may not pay" states no basis. The completion instructions require a plain-language, case-specific reason for each listed service. Preselected options and empty estimates. Staff or software marking an option for the patient invalidates the notice outright, and a blank cost field violates the completion instructions; both are visible on the face of the form. Wrong population or wrong form. Issuing CMS-R-131 to Medicare Advantage enrollees, using a retired edition or a materially reformatted homemade version, or entering the Medicare number because an obsolete template still shows a "Medicare #" field. Timing and delivery failures. A signature captured after the session, no explanation or chance to ask questions, or a recipient who lacked capacity with no representative involved. Each can sink the notice even though the page is signed, and modifier GA cannot cure it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on Advance Beneficiary Notices. Turn chart and schedule details into notice-ready entries: the specific service with frequency and duration, a plain-language case-specific reason, and the good-faith cost arithmetic. Check a prepared notice before signature: specific service, stated reason, estimate present, no preselected option, no Medicare number or SSN, and timing that precedes the service. Map the outcome to claim handling: whether GA, GZ, GX, or GY fits, and what each choice commits you to if the contractor asks for the form. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions When is an ABN required for a Medicare psychotherapy patient? A mandatory ABN is tied to a specific expectation: Original Medicare normally covers the category of service, but you expect denial in this particular case for a limitation-on-liability reason, most often frequency beyond coverage limits or an anticipated medical-necessity denial. It is not required just because payment feels uncertain, and it is not an intake form: CMS requires a case-specific expected denial, and blanket forms signed by every Medicare patient establish no informed knowledge. For services Medicare never covers, a voluntary notice may be offered instead, and the patient does not need to pick an option or sign it. Do I need a new ABN for every session, or a renewal every year? Not necessarily. One notice can cover a defined repetitive course of psychotherapy when it states the service, the frequency, and the expected duration, and it stays effective while the course, the patient's condition, and the applicable coverage policy remain materially unchanged. No Medicare regulation requires renewal solely because twelve months have passed, although CMS permits renewal and a new notice is needed whenever the material facts change. One caution from the manual: if Medicare later pays for a comparable service, the earlier notice may no longer establish that the patient expected denial, so issue a fresh one before relying on it again. What should I write as the reason Medicare may not pay? The actual expected basis, in plain language, for each listed service: for example, that continued sessions at this frequency are not expected to meet Medicare's medical-necessity coverage rules. No regulation prescribes a mental-health-specific phrase, so do not hunt for magic wording, and do not promise that denial is certain. What fails is the generic version: "Medicare may not pay" states no basis at all, and CMS's completion instructions treat it as insufficient. Can we deliver and sign the ABN electronically? Yes. CMS permits electronic issuance and signature when the patient can review the complete form, ask questions, make an uncoerced choice, and sign before treatment, and the patient keeps the right to a paper copy. When in-person delivery is not possible, CMS recognizes alternatives such as a phone discussion combined with mail, fax, or secure electronic delivery. What does not work is a bare portal checkbox: dropping an unsigned form into a portal with no explanation fails the delivery standards even if a signature comes back. Can I use an ABN for a Medicare Advantage patient? No. Form CMS-R-131 is an Original Medicare fee-for-service instrument, and CMS limits it to that program. Medicare Advantage plans run their own organization-determination and notice processes, so contact the plan and use its coverage-determination or liability paperwork. A plan following some Medicare coverage rules does not make the fee-for-service form apply, and a signed CMS-R-131 gives no protection on a Medicare Advantage claim. What if the patient refuses to sign the ABN? Annotate the refusal on the notice itself: that the options were explained, that the patient declined to choose or sign, and the date; a witness signature can be added but is not universally required. Then decide whether to furnish the service, weighing patient safety, potential harm, and liability. Furnishing it anyway means the practice may be unable to establish patient liability if Medicare denies, and the annotation does not guarantee the liability shifted; contractors still assess whether effective notice occurred. What is the difference between an ABN and a good faith estimate? Different patients, different law. The ABN is Medicare payer policy built on the Social Security Act's liability rules: it addresses an expected Medicare denial for an Original Medicare patient and preserves defined appeal choices. The good faith estimate comes from the No Surprises Act and serves uninsured and self-pay clients, with its own timing rules and dispute process; a Medicare enrollee generally does not count as uninsured for it. Neither replaces the other, and your financial policy handles ordinary fees for everyone. How long do we keep signed ABNs? Medicare's Claims Processing Manual says ABNs generally should be retained for five years from discharge or completion of delivery, with electronic retention acceptable, and a longer state records requirement governs when one applies. Keep the form retrievable: the practical trigger is a contractor request after a claim reports modifier GA, and when the notifying and billing entities differ, the billing entity should hold a copy. The signed notice is also part of the billing record a patient can ask for through a records request . Can BastionGPT prepare an ABN for me? Yes, as preparation for the official form. Give it the service, frequency, duration, fee, and the coverage concern, and it drafts the case-specific reason wording, the estimate arithmetic, and a delivery note for your file, ready to transfer onto the current CMS-R-131 from CMS. It can also check a completed draft for the common invalidators and tell you whether GA, GZ, GX, or GY fits the claim. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Good Faith Estimate (No Surprises Act) Financial Policy & Fee Agreement Claim-Support Billing Note Appointment No-Show & Outreach Note Records Request & Chart Copy Packet Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/workers-comp-initial-report All note templates # Workers' Compensation Initial Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A workers' compensation initial report is the first claims-facing report a treating clinician sends after evaluating a claimed work-related mental health condition. It identifies the condition, records the work event, addresses work-relatedness, states functional impact and work status, and proposes treatment for the claims administrator deciding the claim. Every state and scheme names its own form. Most run 600 to 1,500 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, psychiatrists, and other treating clinicians the state scheme authorizes to report Audience Claims administrator or state board, the insurer; work-status sections reach the employer Typical length 600 to 1,500 words · 45 to 90 minutes by hand (clinical team estimate) Format family Claims-facing report, often a mandated scheme form (compare: progress report, IME report) When it's used After the first evaluation of a claimed work-related mental health condition; deadlines can be days Standards context The duty to report is often law; the specific form is payer policy; no national format exists On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a workers' compensation initial report? A workers' compensation initial report is the first claims-facing report in which a treating mental health clinician gives the compensation decision-maker enough information to identify the claimed condition, evaluate its relationship to work, describe current functional and safety consequences, and propose treatment and return-to-work measures. Nobody invented it and no national version exists: workers' compensation is state-administered in the United States, provincial in Canada, and scheme-based in Australia, so each system names its own instrument. California's Doctor's First Report of Occupational Injury or Illness (Form 5021) descends from a regulation adopted in 1980; British Columbia uses the psychology-specific Form 8P; New Brunswick's form is literally titled the Initial Psychology Assessment Report (IPAR) ; New South Wales uses an Allied Health Treatment Request. You will also hear first medical report, doctor's first report, psychologist's report, initial psychological report, initial narrative, or treatment request. Two boundaries keep clinicians out of trouble. First, this is a treating report, not a forensic one: an independent medical examination, QME, or AME is a separate dispute-resolution role with its own rules, and a treating clinician who adopts that voice on the worker's account alone is overreaching. Second, the legal layer and the paper layer are different things. The duty to report is often law: under 8 CCR §9785(e)(1) , the primary treating physician must send Form 5021 within five working days of the initial examination, and British Columbia's equivalent duty runs three days from awareness. The detailed psychology template is usually payer policy, and the polished narrative structure everyone recognizes is convention. What the decision-maker needs is the claims-relevant subset: event, diagnosis, function, risk, plan, and work status, not the whole chart. ## Who uses workers' compensation initial reports and when Any clinician treating an injured worker whose claim includes a mental health condition may owe one: psychologists and psychiatrists diagnosing a stand-alone psychological injury, and in some schemes the clinician treating a psychological condition that follows a physical injury. Washington went further on July 1, 2025, letting psychologists file the initial accident report and serve as attending providers on mental-health-only claims. Write this report at claim start; after that, interval reporting continues on the workers' comp progress report . When an employer wants a safety opinion rather than benefits adjudication, that is a fitness-for-duty evaluation , with the employer as client and a different consent frame. When a disability insurer or Social Security wants functional evidence, that is a medical source statement . And the employer's own record of the event is the workplace incident report , which the claims administrator may read beside yours. ## Workers' compensation initial report structure: what goes in each section Identifiers and claim linkage. Worker, employer, occupation, claim number where one exists, and your credentials with the scheme registration or provider number, plus the dates schemes adjudicate from: date of injury or exposure period, symptom onset, and first treatment. Pitfall: a report the payer cannot match to a claim. California's Form 5021 warns that an incomplete or unreadable report may be returned, late filing can bring a civil penalty, and the certification is signed under penalty of perjury. Work event or exposure history. What happened at work and when: single incident or cumulative exposure, the sequence from event to symptoms, and the first treatment sought. Identify the account as the worker's report. Pitfall: writing the worker's account as verified fact. Official forms treat it as one evidence stream among several; WorkSafeNB's IPAR places the employer's description of the event directly beside the worker's, and the adjudicator reads both. Current symptoms and findings. The presentation now: symptoms, behavioral observations, and a dated standardized measure where your scheme expects one. New South Wales requires at least one outcome measure on the Allied Health Treatment Request ; California's Form 5021 asks for subjective complaints and objective findings but mandates no test battery. Pitfall: severity carried by adjectives. "Severe anxiety" gives the decision-maker nothing to retest at the next report; a dated score becomes the baseline the whole claim is measured against. Diagnosis and its basis. The diagnosis in the framework your scheme expects, usually DSM with the code, and how you reached it: clinical interview, structured interview, testing, records. British Columbia cannot allow a stand-alone mental-disorder claim without a DSM diagnosis from a psychologist or psychiatrist, while WorkSafeNB's form accepts "insufficient information to form diagnosis" when that is the honest answer. Pitfall: overstating certainty to force acceptance. A provisional impression honestly labeled, with the information still needed, is defensible; a definitive diagnosis the record cannot support follows the claim into every later dispute. Work-relatedness. The relationship between your findings and the reported work event, within treating scope: whether the presentation is consistent with the account, the sources you relied on, and what you have not reviewed. Pitfall: forensic voice. Definitive causation percentages or apportionment on the worker's account alone belong to independent evaluations; legal thresholds, such as California's predominant-cause test for psychiatric injury, are the adjudicator's question, not a box your report must resolve. Pre-existing and co-existing conditions. Prior diagnosis and treatment, comorbid conditions, current medications, and non-work contributors, selected for claims relevance. Pitfall: omission reads as advocacy. The claim file usually reaches an independent reviewer eventually, and a first report silent on documented history damages credibility; the fix is relevance, not a full life history the claim does not need. Functional impact and work status. What the worker can and cannot do now, the status call (full duty, modified duty, off work), functional restrictions, and an expected duration or review date. This section often becomes the basis of the return-to-work letter . Pitfall: restrictions phrased as diagnosis. "No lone opening or closing shifts for four weeks, reassessed at review" travels safely to a supervisor; a diagnostic label discloses more than the employer needs. Risk and safety. Risk status either way, with any action taken. WorkSafeNB asks for a suicide-risk rating and a safety-plan description where relevant; Ontario's WSIB assessment asks whether there is imminent danger and what was done. Pitfall: gratuitous detail. Report the risk level, your response, and the implications for work and treatment; the claim file does not need every private disclosure to do its job. Treatment plan, authorization, and signature. Interventions, frequency, the number of sessions requested, measurable goals, barriers, and coordination, then the signature your scheme recognizes, with credentials, provider number, and date. Some schemes couple treatment to this report: WorkSafeNB says not to begin treatment before approval, and Victoria expects its psychology treatment planning form before the sixth session. Pitfall: an unauthorized signer. A supervisee may draft, but a scheme that recognizes only a psychologist, psychiatrist, or statutory physician is not satisfied by a supervisor's cosign. ### Blank template (copy and adapt) WORKERS' COMPENSATION INITIAL REPORT (adapt to your scheme's form; many schemes mandate their own) Report date: __________ Exam date(s): __________ First treatment: ________ Clinician: ____________________ Credentials / license no.: ________________ Scheme registration / provider no.: ________________________________________ Worker: ____________________ DOB: __________ Claim no.: __________________ Employer: ____________________ Occupation: ________________________________ Date of injury / exposure period: ______________ Symptom onset: ___________ WORK EVENT OR EXPOSURE (the worker's account, identified as such; note the employer's account or first report where available) ____________________________________________________________________________ CURRENT SYMPTOMS AND FINDINGS (observations; at least one dated standardized measure where your scheme expects one) ____________________________________________________________________________ DIAGNOSIS AND BASIS (DSM diagnosis with code where required; interview / testing / records; provisional status stated honestly) ____________________________________________________________________________ PRE-EXISTING AND CO-EXISTING CONDITIONS (prior diagnosis and treatment, comorbidities, medications, non-work contributors) ____________________________________________________________________________ WORK-RELATEDNESS (consistent or not consistent with the reported event; sources relied on; what you have not reviewed) ____________________________________________________________________________ FUNCTIONAL IMPACT AND WORK STATUS Current status: [ ] full duty [ ] modified duty [ ] off work Functional restrictions: ___________________________________________________ Expected duration / review date: ___________________________________________ RISK AND SAFETY (risk status either way; action taken if indicated) ____________________________________________________________________________ TREATMENT PLAN (interventions, frequency, sessions requested, measurable goals, barriers, coordination) ____________________________________________________________________________ Signature (scheme-recognized provider): ____________________________________ Name / credentials / provider no.: __________________ Date signed: ________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample workers' compensation initial report Scenario: a licensed psychologist evaluates a bank branch employee three weeks after an armed robbery at her workplace and files the initial report with the claims administrator. All details are fictional. Workers' Compensation Initial Report. Worker: R.V., 41 · Claim no.: WC-2026-48317 · Employer: Meridian Community Bank (teller supervisor) · Date of injury: 07/14/2026 · Exam date: 08/03/2026 · Report date: 08/05/2026 · Clinician: L. Whitfield, PhD, licensed psychologist Work event (worker's account): Ms. V. reports that on 07/14/2026, during her shift as teller supervisor, two individuals robbed the branch; one stood at her window, demanded cash, and indicated a weapon in his waistband. The branch closed for the day after police responded, and she completed the employer's incident documentation that afternoon. She worked one further shift on 07/16/2026, left early after a panic episode at her window, and has not returned to work since. The employer's first report of injury, provided with the referral, describes the same event. This history is Ms. V.'s account except where noted. Current symptoms and findings: Since the event she reports intrusive recollections and nightmares three to four nights weekly, sleep of four to five hours, hypervigilance around entrances and counters, avoidance of the branch and of banks generally, and irritability she describes as unlike her. On examination she was on time, cooperative, and fully oriented, with constricted affect and a visible startle response when the office door opened; speech and thought form were unremarkable. PCL-5 on 08/03/2026: 49. PHQ-9: 11, with no endorsement of item 9. Diagnosis and basis: Posttraumatic stress disorder, acute (F43.10), DSM-5-TR criteria met, based on clinical interview and the PCL-5, with the employer's first report and the referral note from her primary care physician reviewed. No further testing was administered at this stage. Work-relatedness: Ms. V.'s presentation is consistent with the workplace event she describes and the employer's report corroborates. I identified no alternative precipitating event in the history available to me. This opinion rests on my examination and the documents listed; I have not reviewed personnel records or collateral interviews. Pre-existing and co-existing conditions: Brief counseling for adjustment concerns in 2019, resolved without medication. No prior psychiatric treatment or hospitalization. No current medications. AUDIT-C: 1. No non-work contributors identified. Functional impact and work status: Ms. V. manages home responsibilities and drives, but avoids the branch and reports concentration lapses under reminder conditions. Recommended status: off work through 08/28/2026, then graded return on modified duty for four weeks: no lone opening or closing shifts, no single-coverage counter assignment, a second staff member present, hours building from four to eight. Review at the four-week point with an updated report. Risk and safety: She denies suicidal ideation, plan, and intent, denies any self-harm history, and identified her sister as an active support. Risk is low at this time and will be reviewed each session. Treatment plan: Trauma-focused cognitive behavioral therapy, weekly; eight sessions requested for this authorization period. Goals: PCL-5 below 33, return to modified duty as scheduled, independent use of grounding skills at reminder cues. PCL-5 re-administered every four weeks; coordination with her primary care physician by report. Barriers: none identified. A progress report will follow on the scheme's schedule. Signature: L. Whitfield, PhD, licensed psychologist · License PSY 23841 · Scheme provider no. 118276 · Signed electronically 08/05/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The claim identifiers and the dates the scheme adjudicates from (injury, exam, report) sit in the header, so the claims administrator can match and process the report without a call back. The event narrative is labeled as the worker's account and cross-referenced to the employer's first report, which keeps the clinician a witness to findings rather than a fact-finder about the event. Findings carry dated scores (PCL-5 49, PHQ-9 11) that give the claim a baseline every later progress report can be measured against. The work-relatedness opinion says consistent with, names its sources, and states what was not reviewed, staying inside treating scope instead of borrowing forensic authority. Work status is functional and dated, risk is addressed either way, and the requested sessions are tied to measurable goals, which is exactly the subset authorization reviewers check. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The initial report lives a double life: it is part of your clinical record and part of the payer's claim file, and the worker can obtain it through either door. Write it for that audience from the first line. Disclosure is permitted but not unlimited: in the US, HIPAA permits disclosures to workers' compensation insurers and administrators to the extent authorized by law, and the practical discipline is to send the claims-relevant subset the form requests rather than the chart, with psychotherapy notes kept segregated. Retention follows the longest rule that applies to you, because no scheme reviewed publishes a shorter report-specific destruction date: California psychologists keep records at least seven years after discharge under Business and Professions Code §2919, and seven years past age 18 for minors; Ontario's psychology standard runs at least 10 years from the later of last contact or the client turning 18; New South Wales private providers keep adult records seven years from last service and minors' records until age 25. Sort every requirement into three bins before you write: law, payer policy, and convention. The duty to report is often law: California's 8 CCR §9785(e)(1) gives the primary treating physician five working days from the initial examination, British Columbia's reporting duty runs three days from awareness under the Workers Compensation Act, and Ontario's WSIA §37(1) obliges practitioners to provide what the WSIB requires, with the 48-hour initial expectation set by policy. The detailed psychology template is usually payer policy: Ontario's Community Mental Health Program assessment, WorkSafeNB's IPAR (10 business days from completing the assessment, and no treatment before approval), New South Wales' AHTR (a general 21-day insurer response, with a five-working-day automatic-approval pathway for defined approved-provider requests), and Victoria's treatment planning form before the sixth session. The polished narrative everyone recognizes is convention: no reviewed statute or official form mandates SOAP, a fixed mental-status heading, a universal test battery, or an impairment rating in a first treating report. The format is a convention; the content is the requirement. Once treatment is under way, interval reporting continues on the workers' comp progress report , and the work-status section often feeds the return-to-work letter . This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common workers' compensation initial report errors auditors flag The audit record here measures claims handling as much as clinician prose, and both layers decide what happens to your report. SIRA's July 2023 audit of 50 open psychological-injury claims in New South Wales found that only 74% of files showed all available information was considered at the initial liability decision, only 56% of written liability notices contained all relevant details, and follow-up to obtain missing information conformed in 48.83% of applicable files; 25 of the 50 claims were referred for legal advice. On the treatment side, California's independent medical review report analyzing 2024 decisions shows 20.1% of behavioral and mental health treatment denials were overturned, up from 16.8% the year before, against a 12.7% overall rate. And the document-level consequences sit on the form itself: California's Form 5021 may be returned if incomplete or unreadable, late filing can bring a civil penalty, and the certification is signed under penalty of perjury. The BastionGPT Clinical Advisory Board sees the same errors most often in workers' compensation initial report reviews: A report the payer cannot match or process. Missing claim number, date of injury or exposure period, employer, or provider registration leaves the report unmatchable to the claim, and an incomplete first report can be sent back while the benefit clock runs. The worker's account written as established fact. The event narrative is evidence the adjudicator weighs against the employer's account and the file; a report that asserts the event as found fact reads as advocacy and is the first thing a defense review attacks. Forensic overreach on causation. Definitive causation or apportionment opinions on the worker's account alone exceed the treating role; state the consistency of findings with the reported event, your sources, and what you have not reviewed. No measure, vague goals, unjustified frequency. Treatment-request pathways let the insurer approve, decline, partially approve, or demand more information, and requests missing a dated outcome measure, measurable goals, or a rationale linking treatment to recovery and work participation are the ones that bounce. An unauthorized or borrowed signature. Schemes recognize defined providers: the statutory physician in California, a psychologist or psychiatrist for British Columbia's qualifying diagnosis, registered or enrolled providers elsewhere; an associate's report with a supervisor cosign is not a scheme-recognized report. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on workers' compensation initial reports. Draft the complete report from your interview notes, a dictation, or the scheme form's headings, with the event labeled as the worker's account and causation kept in treating scope. Rewrite the work-status section in plain, functional language an adjuster and employer can act on, without disclosing diagnosis where it is not needed. Check the report before you sign: claim identifiers, dated measure, diagnosis basis, risk statement, restrictions with a review date, and the signature block your scheme recognizes. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there a standard workers' comp initial psychological report form? No. Workers' compensation is state-administered in the US and provincial or scheme-based in Canada and Australia, and each system names its own instrument: California's Form 5021 first report, British Columbia's Form 8P, Ontario's WSIB Community Mental Health Program assessment, New Brunswick's Initial Psychology Assessment Report, New South Wales' Allied Health Treatment Request, and Victoria's psychology treatment planning form. New York retired its psychology-specific PS-4 process in July 2022; the report now travels as a required narrative with CMS-1500 billing. The duty to report is often law; the detailed template is usually payer policy. Check your scheme's provider pages before adapting any generic template, including this one. How quickly is the initial report due? Sooner than most clinicians expect, and the clock starts from different events. California requires Form 5021 within five working days of the initial examination under 8 CCR §9785. WorkSafeBC expects Form 8P within three days of becoming aware of a suspected work-related mental condition when the worker wishes to claim. Ontario's WSIB expects initial health-care reporting within 48 hours. WorkSafeNB allows 10 business days from completing the assessment, and Victoria wants its treatment planning form before the sixth session. None of these deadlines runs from the date of injury itself. Who is allowed to sign the initial report? Whoever the scheme's provider definition recognizes, which is narrower than who can provide therapy. In California, a licensed psychologist acting within scope meets the workers' compensation definition of physician, but associate therapists, counselors, and trainees do not become the statutory signer because a supervisor reviews the case. British Columbia requires the qualifying diagnosis to come from a psychologist or psychiatrist. Washington began letting psychologists file the initial report and serve as attending providers on mental-health-only claims in July 2025. When in doubt, the credentialed provider the scheme recognizes should authenticate the report. Do I have to prove causation like an independent examiner? No. The treating report addresses the relationship the form asks about, in treating scope: whether findings are consistent with the reported event, what sources you relied on, and what you have not reviewed. Legal causation tests, such as California's predominant-cause threshold for psychiatric injury, are applied by the adjudicator, and disputed causation goes to an independent medical or qualified medical evaluation, a separate role with its own rules. A treating clinician who issues definitive causation or apportionment opinions on the worker's account alone invites impeachment without helping the claim. Is psychological testing required in the initial report? It depends on the form, not on a universal rule. New South Wales requires at least one standardized outcome measure on the Allied Health Treatment Request. California's Form 5021 asks for subjective complaints and objective findings but mandates no test battery. WorkSafeNB asks what the diagnosis rests on: interview, structured interview, testing, or records, and expressly accepts "insufficient information to form diagnosis." A dated score strengthens any scheme's report because the progress report is measured against it, but "testing is always mandatory" and "testing is never required" are both wrong. Should I send the whole chart with the report? No. In the US, HIPAA permits disclosures to workers' compensation insurers and administrators to the extent authorized by law, but that permission is not an instruction to send everything: official forms request a claims-relevant subset, and minimum-necessary principles apply in relevant circumstances. Send the form and what is lawfully requested, keep psychotherapy notes segregated, and report risk at the level of risk status, response, and work implications rather than reproducing private disclosures. How is this different from a fitness-for-duty evaluation or a medical source statement? The decision being made is different. The initial report serves benefits adjudication and treatment authorization inside a workers' compensation claim. A fitness-for-duty evaluation answers an employer's safety question about a specific job, with the employer as client and a different consent frame. A medical source statement supplies functional evidence for a disability-income determination, where a compensable work event may be irrelevant. Combining the roles in one document creates confidentiality and scope problems; interval reporting inside the claim continues on the workers' comp progress report . How long do I need to keep the initial report? Apply the longest rule that touches you; no scheme reviewed publishes a shorter workers' compensation-specific destruction date, and a form's silence does not override record law. California psychologists keep records at least seven years after discharge, and seven years after a minor turns 18, under Business and Professions Code §2919. Ontario's psychology standard runs at least 10 years from the later of last contact or the client turning 18. New South Wales private providers keep adult records seven years from last service and minors' records until age 25. The payer keeps its own copy in the claim file regardless. Can BastionGPT draft a workers' compensation initial report? Yes. Give it your interview notes, a dictation, or your scheme form's headings and it drafts the report with the event labeled as the worker's account, dated measures in the findings, causation in treating scope, and functional work-status language. It can also produce a plain-language summary for the worker and check claim identifiers before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Workers' Compensation Progress Report Fitness-for-Duty Evaluation Return-to-Work & Accommodation Letter Incident & Adverse Event Report Disability Evaluation & Medical Source Statement Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/workers-comp-progress-report All note templates # Workers' Compensation Progress Report: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A workers' compensation progress report is a periodic report from the treating clinician to the claims administrator, insurer, or state board during an active work-injury claim. It documents the accepted condition, interval symptoms and objective findings, response to treatment, current work restrictions, and return-to-work outlook. California standardizes it as Form PR-2; most run 300 to 700 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The treating clinician; in California, the primary treating physician, a role open to qualified licensed psychologists Audience Claims administrator or insurer, case managers, utilization review, the state board, the employer for work status Typical length 300 to 700 words · 20 to 30 minutes by hand (clinical team estimate) Format family Periodic claim report (compare: initial report, return-to-work letter) When it's used On the scheme's cadence (California: at least every 45 days) and within 20 days of listed changes Standards context No national form; each scheme sets its own: California's PR-2 is regulation, New York and Washington differ On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a workers' compensation progress report? A workers' compensation progress report is the periodic report a treating clinician sends to the claims administrator, insurer, or state board while a work-injury claim is open. It is a claim-management document, not a session note: the reader uses it to decide medical necessity, authorization, payment, wage-loss entitlement, and return to work, so it reports the accepted condition, interval symptoms and objective findings, response to treatment, barriers to recovery, current work restrictions, and the forward plan. There is no national form. The best-known standardized version is California's Primary Treating Physician's Progress Report, Form PR-2, created by the Division of Workers' Compensation through the Administrative Director's regulations: section 9785's history reaches back to 1977, the 45-day periodic-report provision became operative in 2001, and the current two-page form has been in use since October 2015. Other schemes use their own instruments and names: New York pairs a psychology narrative with every CMS-1500 bill, Washington folds the required content into ordinary chart notes, and you will also hear PR-2, PTP report, psychology progress report, or treatment progress report used for the same document family. Two boundaries matter before you write one. First, your ordinary therapy note is not automatically a progress report. In California, a narrative can replace Form PR-2 only when it carries the same report title, states the reason for the report, follows the PR-2 headings in the same order, contains equivalent information, and includes the required declaration under penalty of perjury ( 8 CCR section 9785 ). Second, the progress report sits in the middle of a document sequence: the initial report opens medical reporting on the claim, the progress report tracks the active treatment course, and work-capacity communication to the employer is its own document, covered on the return-to-work letter page. California even forbids using the PR-2 at permanent-and-stationary status, where the PR-3 or PR-4 takes over. ## Who uses workers' compensation progress reports and when Any clinician treating a psychological injury inside a workers' compensation claim writes them: psychologists treating accepted stress, trauma, or depression claims, and therapists delivering authorized care as secondary providers. Who may sign is scheme business. In California, only the primary treating physician issues the report, and Labor Code section 3209.3 admits a licensed psychologist to that role only with two years of clinical experience in a recognized health setting or National Register credentials; associates and master's-level therapists contribute findings to the PTP instead. Washington has allowed psychologists to serve as attending providers on mental-health-only claims since July 1, 2025. Reach for this report on the scheme's cadence and whenever condition, treatment, or work status changes. It is not the note you keep for yourself after each session, and it is not the treatment plan review you run against your own clinical goals; it answers the claim's questions. When the question is a safety-sensitive employment decision outside a claim, that is a fitness-for-duty evaluation . ## Workers' compensation progress report structure: what goes in each section Claim identifiers, report reason, and period. Start with what routes the document: worker name and date of birth, claim number, date of injury, employer, insurer or claims administrator, your name, credentials, and role on the claim, and the exact period the report covers. Then say why the report exists: periodic cadence, a change in condition or work status, a referral, or an insurer request. The official PR-2 makes the reason a checkbox, and a California narrative substitute must identify it too. Pitfall: treating the 45-day cycle as the whole rule. Section 9785 also requires a report within 20 days of listed events, including changes in condition, treatment plan, or work status. Subjective report and interval change. The worker's account of symptoms now, compared with the last report: sleep, mood, intrusions, avoidance, concentration, and what has shifted at home and at work. Anchor every statement to the reporting period. Pitfall: retelling the injury story each cycle. The adjuster already holds the initial report; the file needs what changed in the last six weeks. Objective findings and measures. Psychological reports have objective content: observed presentation and mental status, attendance and engagement, standardized measures with dates and trends, and functional tolerances such as driving, shift length, or task persistence. No California regulation names a mandatory instrument for the PR-2, so pick measures that fit the condition and repeat them; the outcome measure note covers score documentation. Pitfall: severity carried by adjectives. "Still anxious" gives a reviewer nothing to compare; a dated score trend and one concrete functional gain do. Diagnosis and the accepted condition. State the diagnosis with current coding where the scheme asks for it, and tie treatment to the condition the claim accepted. New York's Board warns that a narrative missing causal-relationship or work-status content can be "legally defective." Pitfall: quietly treating conditions the claim has not accepted. The mismatch surfaces in utilization review, where it stalls authorization instead of prompting a conversation with the adjuster. Treatment delivered and response. Sessions held during the period, methods, adherence, and measurable response: score change, functional gains, participation in work. When progress stalls, say so and change something; New York's psychology requirements direct the clinician to reassess diagnosis and plan when the worker is not progressing. Pitfall: the plan copied forward. Across two or three cycles, identical text reads as care on autopilot, the exact profile that draws records requests. Work status, restrictions, and return-to-work outlook. The section most readers act on: current status (off work, modified duty, full duty), restrictions specific enough to schedule against (hours, tasks, environments, supervision), each with a duration or review date, and a projected return-to-work date when one can be estimated. Pitfall: "remains off work" with nothing an employer could offer modified duty against. Wage-loss and accommodation decisions hang on this section, and New York names missing work status as a narrative defect. Plan, prognosis, requests, and signature. Next interventions with frequency and expected duration, barriers to recovery, prognosis, and what you are asking for. Route requests correctly: in California, treatment authorization runs on a Request for Authorization, DWC Form RFA, and the progress report is supporting evidence rather than the request itself. Sign with credentials and date; California reports carry a declaration under penalty of perjury. Pitfall: burying "please approve eight more sessions" in the narrative. A request the scheme's intake process cannot see was never made. ### Blank template (copy and adapt) WORKERS' COMPENSATION PROGRESS REPORT Report date: __________ Period covered: __________ to __________ Reason: [ ] periodic [ ] change in condition [ ] change in work status [ ] referral [ ] insurer request [ ] other: __________________ Worker: ______________________ DOB: __________ Claim #: ______________ Date of injury: __________ Employer: __________________________________ Insurer / claims administrator: _________________________________________ Provider / credentials: ____________________________ Role: [ ] PTP (CA) [ ] attending provider [ ] secondary provider (reports to the PTP) ACCEPTED CONDITION / DIAGNOSIS (current code where required) _________________________________________________________________________ SUBJECTIVE: WORKER'S REPORT AND CHANGE THIS PERIOD _________________________________________________________________________ OBJECTIVE FINDINGS (observed status, dated measures with trend, attendance, functional tolerances) _________________________________________________________________________ TREATMENT DELIVERED AND RESPONSE (sessions, methods, measurable response; if progress has stalled, what changes) _________________________________________________________________________ WORK STATUS AND RESTRICTIONS (status; specific restrictions with duration; projected return-to-work date) _________________________________________________________________________ BARRIERS TO RECOVERY AND NEW INFORMATION _________________________________________________________________________ PLAN AND PROGNOSIS (next interventions, frequency, expected duration; authorization requested via the scheme's channel, e.g. DWC Form RFA) _________________________________________________________________________ Declaration where required (California: under penalty of perjury): _________________________________________________________________________ Signature / credentials: ____________________ Date signed: _____________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample workers' compensation progress report Scenario: a California licensed psychologist, serving as primary treating physician on an accepted psychological-injury claim, files a 45-day periodic report as a PR-2 narrative. All details are fictional. Primary Treating Physician's Progress Report (PR-2 narrative). Worker: R.V., 41 · Claim: WC-26-40917 · Date of injury: 04/22/2026 · Report date: 08/05/2026 · Period: 06/20/2026 to 08/05/2026 · Reason: periodic Accepted condition: Posttraumatic stress disorder (F43.10), accepted 05/28/2026, arising from a workplace robbery in April 2026 at the worker's retail employer. Treating provider: K. Anand, PhD, licensed psychologist, primary treating physician. Subjective complaints: R.V. reports intrusive recollections down from daily to two or three per week, sleep improved from about 5 to 6.5 hours, and less startle at door chimes. She continues to avoid the store's back office and describes anticipatory anxiety before opening shifts. No panic episodes this period. Objective findings: Attended 6 of 6 scheduled sessions. PCL-5 31 on 08/04/2026, down from 44 on 06/19/2026; PHQ-9 9, down from 14. Affect broader in range; speech normal in rate and tone. Completed graded in-vivo tasks: entered the stockroom accompanied (07/10) and alone (07/24). Tolerating 4-hour shifts without leaving the floor. Denies suicidal ideation, screened each session. Treatment and response: Six weekly 60-minute trauma-focused CBT sessions with graded exposure and sleep components. Response is measurable and consistent with treatment-plan milestones. No psychotropic medication on this claim; care coordinated with the primary care physician. Work status and restrictions: Modified duty since 07/06/2026: shifts up to 4 hours, daytime only; no lone opening or closing duties; no cash-office assignment. The employer has accommodated. Recommend extending modified duty 6 weeks, increasing to 6-hour shifts from 08/24/2026 if gains hold. Projected full-duty return: 10/05/2026. Plan and prognosis: Continue weekly sessions for 6 weeks, then re-evaluate for a biweekly taper. Re-administer PCL-5 and PHQ-9 before the next report. Prognosis good. A Request for Authorization for 8 further sessions is submitted separately, with this report as support. Next periodic report due by 09/19/2026. Signed under penalty of perjury as required for California treating-physician reports. K. Anand, PhD (PSY 28417) · 08/05/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The reason and period are stated up front, so the reader knows the 45-day periodic duty is met and which weeks the findings cover. Progress is measured, not asserted: dated PCL-5 and PHQ-9 trends, attendance, and two functional gains give review something to approve against. Work status is operational: specific restrictions with dates, a stepped increase, and a projected full-duty date the employer and adjuster can plan around. The treatment request travels by the right vehicle: the RFA is named as submitted separately, with the report as supporting evidence. The report stays inside the claim: it addresses the accepted condition, keeps psychotherapy process detail out, and closes with credentials, date, and the required declaration. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations This report is written for readers who never sit in the therapy room: the adjuster, a nurse case manager, utilization-review physicians, sometimes the employer's return-to-work coordinator, and, if the claim is disputed, a judge. Write it as a claim document: decision-focused, specific about function and work capacity, and free of psychotherapy process material the claim does not need. Claim reporting does not put your whole chart in the mail. California requires the defined treating reports plus documentation supporting each treatment request, with further records produced on request; it does not attach every session note to every PR-2. Authorship is regulated too: in California only the primary treating physician issues the report, secondary providers send findings to the PTP for incorporation or comment within 20 days, and no rule turns an associate's draft with a supervisor co-signature into the PTP's report. Retention then follows your profession rather than the claim: California's Business and Professions Code section 2919 requires psychologists to keep records at least seven years from discharge (seven years past age 18 for minors), and Washington's Department of Labor and Industries expects claim-related records kept at least five years. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. The payer layer turns on one distinction: the progress report describes care, it does not authorize care. In California, utilization review starts with a complete Request for Authorization; a thorough PR-2 cannot substitute for it, and a request whose supporting information never arrives can end as a conditionally non-certified decision, the category the Division of Workers' Compensation applies when the physician has not supplied information needed for the medical-necessity determination. Note how the usual rule reverses here: for most therapy documentation the format is a convention and the content is the requirement, but workers' compensation often prescribes the format itself. California fixes the narrative's title, reason, heading order, and declaration; New York requires a narrative with work status, causal relationship, and temporary impairment alongside every CMS-1500 bill, mandatory since July 2022; Washington accepts complete chart notes in place of a separate report. Fit the instrument to the scheme, keep measured function at the center, and keep neighboring documents in their lanes: progress against your own clinical goals belongs in the treatment plan review , and safety-based employment questions outside the claim belong in a fitness-for-duty evaluation . ↑ Back to the template and downloads ## Common workers' compensation progress report errors auditors flag Workers' compensation produces unusually concrete denial data because treatment disputes are formally adjudicated. California's Division of Workers' Compensation reports in its 2025 Independent Medical Review annual report that the program received 201,037 applications and issued 152,351 final determinations, with 10.2% of reviewed utilization-review denials overturned overall; behavioral and mental health services were overturned at 18.3%, among the highest of any treatment category, down from 20.1% in 2024. The program's updates also count 6,181 ineligible applications tied to conditionally non-certified decisions, the outcome assigned when the treating provider did not supply requested information. New York's Board warns that a psychology narrative missing required elements may be "legally defective." The BastionGPT Clinical Advisory Board sees the same errors most often in workers' compensation progress report reviews: The report sent where a request was needed. In California, authorization runs on the RFA. A detailed PR-2 with the ask buried in its plan section starts nothing, and a request whose supporting information never arrives ends in conditional non-certification rather than a decision on the merits. Work status the employer cannot act on. "Remains off work, continue therapy" carries no restrictions, durations, or projected return date. Wage-loss and modified-duty decisions stall, and New York treats missing work-status content as a defect in the narrative itself. Progress carried by adjectives. "Improving" without a dated score trend or a functional comparison gives review nothing to approve against; and when the trend is flat, the report should show reassessment and a changed plan, not the previous plan re-signed. An author the scheme does not recognize. An associate's report with a supervisor co-signature is not a PR-2; the primary treating physician must issue it, and under Labor Code section 3209.3 not every licensed psychologist qualifies for the role. Confirm author rules before the report bounces. The periodic cadence treated as the whole rule. California's 45 days is an outer limit; listed events, including work-status changes, require a report within 20 days, and reports that lag authorization boundaries create payment gaps that clinical quality cannot cure. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on workers' compensation progress reports. Summarize a treatment period into a progress report draft: interval change, dated score trends, attendance, response to treatment, and work status pulled from your session notes. Translate clinical content into the plain, decision-focused language adjusters and employers act on, while keeping psychotherapy process detail out of the claim file. Check a finished report before you sign: reason and period, accepted condition, measurable response, restrictions with durations, projected return-to-work date, and the fields your state's instrument requires. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How often do I have to send a workers' compensation progress report? There is no national cadence. In California, section 9785 requires a progress report no later than 45 days after the last one, and within 20 days when a listed event occurs, such as a change in condition, treatment plan, or work status. New York expects a narrative with each CMS-1500 bill. Washington needs no separate report when every chart note carries the required information. The claim's scheme, not a generic rule, sets your clock. Can I send my normal therapy note instead of a separate report? Sometimes, under specific conditions. California accepts a narrative in place of Form PR-2 only if it uses the same report title, states the reason for the report, follows the PR-2 headings in the same order, contains equivalent information, and includes the declaration under penalty of perjury. Washington accepts complete chart notes in place of a separate report. Some payer programs run the other way and refuse narratives, requiring their own form. Confirm the instrument before the first report is due. Does the insurer automatically get my complete psychotherapy notes? No. The defined report is the routine vehicle, and it should carry claim-relevant content: condition, response, function, work status. California requires treating reports and documentation supporting each treatment request, with further records available on request during review; that is not the same as every session note traveling with every report. Keep process-level material in the clinical record and write the report for its claim audience. What counts as objective findings when the condition is psychological? Observed presentation and mental status, behavior in session, attendance and engagement, standardized measures with dates and trends, and functional tolerances: sleep, driving, shift length, task persistence, contact with the workplace. No California PR-2 rule mandates a specific instrument, so choose measures that fit the condition and repeat them; the outcome measure note page covers score documentation that survives review. Can an associate therapist write the report if a supervisor co-signs? Not in California. The primary treating physician must issue the progress report, and Labor Code section 3209.3 admits licensed psychologists to that role only with two years of clinical experience in a recognized health setting or National Register credentials. No regulation converts an associate's draft plus a co-signature into the PTP's report; associates and secondary providers contribute by sending findings to the PTP, who incorporates or comments on them within 20 days. Other states draw their own lines: Washington lets psychologists act as attending providers on mental-health-only claims but does not authorize master's-level therapists to diagnose the compensable condition. Do I have to put CPT codes on the California PR-2? No. The official form describes coding the treatment section as encouraged, and neither the form nor section 9785 makes it mandatory. Codes help the claims administrator match treatment to billing, so include them when you have them, but a report is not defective without them. What is the difference between a progress report and a return-to-work letter? The progress report is the periodic claim document: accepted condition, treatment response, work capacity, and plan, sent on the scheme's cadence. A return-to-work letter answers one question for the employer: what the worker can safely do now, with restrictions and dates and minimal clinical detail. The initial report opens the claim's medical reporting, and a fitness-for-duty evaluation answers a safety question outside the claim. Most open claims eventually need more than one of these. How long do I need to keep workers' compensation progress reports? The claim's filing deadline is not your retention period. The report is part of the clinical record, so professional and state rules govern: California psychologists must retain records at least seven years from discharge under Business and Professions Code section 2919, and seven years past age 18 for minors, while Washington's Department of Labor and Industries expects claim-related records kept at least five years. Apply the strictest rule that reaches you and treat claim documents as part of the chart. Can BastionGPT draft a workers' compensation progress report from my session notes? Yes. Give it the session notes for the period, scores, and work-status details, and it drafts the report with interval change, measured response, restrictions with durations, and return-to-work outlook in claim-facing language. It can also check a draft against your state's field list before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Workers' Compensation Initial Report Return-to-Work & Accommodation Letter Treatment Plan Review & Update Outcome Measure Review Note Fitness-for-Duty Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/return-to-work-letter All note templates # Return-to-Work & Accommodation Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A return-to-work letter is a short, signed statement from a treating clinician telling an employer when an employee can resume work and under what functional restrictions or accommodations. Therapists, psychologists, and social workers write them after a mental health leave or to support a workplace accommodation request. Most run under one page, and the strongest ones describe function, not diagnosis. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating therapists, psychologists, clinical social workers, counselors, psychiatrists, NPs Audience Employer or HR, usually via the client; leave administrators; a copy stays in the chart Typical length 150 to 400 words · 15 to 30 minutes by hand (clinical team estimate) Format family Clinical correspondence (compare: FMLA certification, fitness-for-duty evaluation) When it's used When a client returns after a mental health absence or requests workplace accommodations Standards context No universal form: ADA documentation content is convention; FMLA and workers' comp impose real rules On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a return-to-work and accommodation letter? A return-to-work letter from a mental health clinician is a short, signed statement telling an employer whether, when, and under what functional conditions a client can resume work after an absence. An accommodation letter is its close relative: it explains that a health condition limits specific work functions and connects those limits to workplace changes that would help. One letter often does both jobs (may return September 1 on half days, with two adjustments), and nobody invented the format. The phrase is a vernacular umbrella over four older document families: the medical certificate or sick note, the FMLA fitness-for-duty certification, reasonable-accommodation documentation under the ADA, and prescribed workers' compensation capacity forms. You will also hear work release, release with restrictions, or RTW letter, and the same phrase circulates online for an employee's own notice and for an employer's recall letter after a layoff, which is why this page defines the document by function rather than title. The load-bearing question is which of those families is actually being requested, because the correct author, fields, and legal effect all follow from it. No law prescribes a universal template: the ADA creates a duty to accommodate, not a letter format, and the EEOC's guidance for mental health providers on what helpful documentation contains is enforcement interpretation, expressly not binding law. The genuinely prescribed versions live elsewhere: the FMLA fitness-for-duty certification is governed by 29 CFR 825.312 (see the FMLA certification support note ), and workers' compensation schemes require their own capacity forms with their own eligible signers, so a generic therapist letter cannot substitute for either (start at the workers' comp initial report ). One more reason to write carefully: under the EEOC's reading, a release stating that an employee may return with restrictions can itself count as notice that accommodation is needed. The letter you dash off on a Friday can open a formal interactive process on Monday. ## Who uses return-to-work and accommodation letters and when Any treating mental health clinician gets the request: therapists and counselors when a client returns from a leave for depression or anxiety, psychologists supporting an accommodation request, social workers coordinating a phased return with an employee assistance program. Reach for this letter when the question is ordinary and non-scheme: the client is coming back from a health-related absence, or could work now if something changed, and the employer wants limited capacity information from the clinician who knows them. Its neighbors answer different questions. The FMLA certification support note covers the paperwork that starts protected leave; the fitness-for-duty evaluation is the employer-ordered examination when there is a genuine safety or essential-function question; the workers' comp initial report sits inside a compensation claim with prescribed forms; the medical source statement speaks to disability-benefit eligibility over a sustained period rather than a return date; and for students the parallel document is the return-to-school letter . ## Return-to-work and accommodation letter structure: what goes in each section Date, recipient, and your credentials. Date the letter, identify the client, and state who you are: license, credentials, and the nature and length of the treatment relationship, the two facts the EEOC lists as establishing that you are in a position to know. Decide the route before you write: handed to the client to submit, the letter needs no privacy paperwork; sent by you directly to the employer, it generally needs the client's signed HIPAA authorization first. Pitfall: mailing it to HR yourself as a courtesy. Without a valid authorization, the courtesy is an unauthorized disclosure; give the letter to the client or get the release signed first. Present capacity and return date. The sentence the employer is waiting for: whether the client can return, on what date, and to full or modified duties, anchored to the date of your most recent assessment. Pitfall: writing medically cleared. No regulation requires the phrase, and it can overstate what a therapist assessed, particularly without the job's essential functions in hand; state capacity in functional terms instead. Functional limitations. What the condition currently limits at work, stated as function rather than symptoms: sustained concentration, tolerance for schedule changes, interaction demands, response to high-stimulation environments. The EEOC's mental health provider guidance and the Ontario Human Rights Commission both point documentation here rather than at diagnosis. Pitfall: naming the diagnosis by default. It is usually not required for an accommodation request; a general description of the condition plus its work-relevant limits is ordinarily enough, and routine diagnosis disclosure is the most common over-share in these letters. Restrictions and the graded-return schedule. Make restrictions implementable: dated weeks, hours per day, tasks included and excluded. A graded return is easiest to honor when each step has a start date and an end date. Pitfall: restriction language HR cannot schedule. Light duties and reduced stress hand the employer an interpretation problem; four-hour days through August 28 does not. Accommodations, each tied to a limitation. Recommend the workplace change and say which limitation it addresses; the EEOC treats that connection, not the diagnosis, as what makes documentation sufficient. Leave room for alternatives. Pitfall: presenting one accommodation as the only option. The employer may lawfully choose any effective accommodation, and a letter that admits no alternative stalls the interactive process it was supposed to start. Duration, review date, basis, and signature. Say how long the restrictions run and when you will review them: no rule imposes a 30-, 60-, or 90-day expiry, so the review date is what keeps a temporary restriction from quietly becoming permanent. Close with what the opinion rests on: sessions to date, measures used, whether you reviewed a job description, and which statements are the client's report rather than your observation. Sign and date. Pitfall: client self-report presented as verified fact. Ontario's psychotherapy standard expressly requires labeling objective fact, client-supplied information, and professional opinion separately, and the discipline is worth borrowing wherever you practice. ### Blank template (copy and adapt) RETURN-TO-WORK & ACCOMMODATION LETTER Date: __________ Re (client / employee): ______________________________ From: ____________________________ License / credentials: ______________ Practice contact: _______________________________________________________ Route: [ ] given to client to submit [ ] sent directly; authorization signed and dated: _______________ TREATMENT RELATIONSHIP (nature and length; diagnosis usually not required for an accommodation request) _________________________________________________________________________ PRESENT CAPACITY AND RETURN DATE (full or modified duties; date of your most recent assessment) _________________________________________________________________________ FUNCTIONAL LIMITATIONS (work-relevant function, not symptoms: concentration, schedule tolerance, interaction demands) _________________________________________________________________________ RESTRICTIONS / GRADED-RETURN SCHEDULE (dated steps HR can implement) Weeks 1-2: ______________________________________________________________ Weeks 3-4: ______________________________________________________________ Full duties from: __________ RECOMMENDED ACCOMMODATIONS (tie each to the limitation it addresses; other effective options remain open to the employer) 1. ______________________________________________________________________ 2. ______________________________________________________________________ DURATION AND REVIEW DATE: _______________________________________________ BASIS AND SCOPE (sessions and measures; job description reviewed? client-reported information labeled as such) _________________________________________________________________________ Signature: ____________________ Name / credentials: ____________________ Date signed: __________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample return-to-work and accommodation letter Scenario: a psychologist documents a graded four-week return with two accommodations for an adult client after a mental health leave. The letter names no diagnosis: functional limitations carry the request. All details are fictional. Return-to-Work & Accommodation Letter. From: K. Whitfield, PhD, Cedar Park Psychology · Given to the client for submission to HR · Date: 08/05/2026 · Re: J.M., DOB 03/22/1991 To the Human Resources Department: I am a licensed psychologist and have treated J.M., a project coordinator, in weekly sessions since 04/14/2026 for a health condition that required a leave of absence beginning 06/22/2026. J.M. asked me to provide documentation supporting her return to work and two accommodations. She has reviewed this letter, and I release it with her written authorization. Present capacity and return date: Based on my assessment through 08/03/2026, J.M. is able to return to work on Monday 08/17/2026 on a graded schedule: half days (four hours) in weeks one and two (08/17 to 08/28), six-hour days in weeks three and four (08/31 to 09/11), and her full schedule and duties from 09/14/2026. Functional limitations: Her condition currently limits sustained concentration over long unbroken work periods and her tolerance for short-notice schedule changes. Both are improving with treatment, and I expect these limitations to be temporary. Recommended accommodations: (1) scheduling her concentration-heavy work, such as report preparation, in morning blocks where feasible, which addresses the concentration limitation; and (2) 48 hours' notice of schedule changes where operations allow, which addresses the schedule-tolerance limitation. Other adjustments that address the same limitations may be equally effective, and I am glad to discuss alternatives. Duration and review: I will review these restrictions with J.M. by 10/02/2026 and will provide an updated letter if anything changes before then. Basis and scope: This opinion rests on sixteen weekly sessions, standardized symptom measures administered across treatment (most recently 08/03/2026), and J.M.'s description of her job duties, which is her report; I have not reviewed a written job description. This letter addresses work capacity and accommodations only. Sincerely, K. Whitfield, PhD, Licensed Psychologist (license PSY 23145) Cedar Park Psychology · (555) 013-4471 · signed 08/05/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works No diagnosis appears anywhere in the letter: capacity, functional limitations, and accommodations carry the request, which is the minimum EEOC and Ontario Human Rights Commission guidance support. Every step of the graded return is dated with hours per day, so HR can put it on a roster without a clarifying call and the review letter has something to measure against. Each accommodation is tied to the limitation it addresses, and the letter leaves alternatives open, which keeps the interactive process moving instead of boxing the employer in. The basis-and-scope paragraph separates what the clinician assessed from what the client reported, so nothing reads as verified fact that is not. Authorization, a review date, and a dated signature make the disclosure authorized, the restrictions temporary, and the opinion attributable. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations This letter lives in two records under two rule sets. Your copy is part of the clinical record, subject to client access and your state's retention rules: no federal law sets a universal chart-retention period for it (HIPAA's six-year rule covers required documentation such as signed authorizations, not the chart itself), so state law and licensing-board standards govern, and Ontario's CRPO expects registrants to keep records at least ten years from the last interaction. The employer's copy becomes an employment record with its own law: under 29 CFR 825.500 an employer must keep FMLA medical documentation in separate, confidential files for at least three years. Get the disclosure path right at the start: a HIPAA-covered clinician generally needs the client's valid authorization before sending health information directly to an employer, which is why the cleanest route is handing the letter to the client to submit. Verification has rules too: under 29 CFR 825.307 the client's direct supervisor may never contact you about an FMLA certification, authorized HR or leave-administration staff may only authenticate or clarify, and outside FMLA you still need authorization before discussing the client at all. Sort every request into its regime before you write, because the rules change under your feet. The content of an ordinary accommodation letter is convention: the EEOC's documentation guidance is influential but expressly non-binding, and no statute lists mandatory fields. The FMLA return stage is law, and narrower than HR folklore: an employer may require a fitness-for-duty certification only under a uniformly applied policy, limited to the condition that caused the leave; it may demand an essential-function analysis only if it supplied the function list with the designation notice; it may not seek second or third opinions on the return certification and may not delay reinstatement while authenticating; and for intermittent leave it may ask at most once every 30 days and only with reasonable safety concerns. Who signs is credential-specific: clinical psychologists and clinical social workers are expressly recognized FMLA providers under 29 CFR 825.125, while counselors may qualify through state scope of practice or employer and plan acceptance (details on the FMLA certification page). Workers' compensation is payer policy with teeth: schemes use prescribed capacity forms with defined eligible signers, and a generic therapist letter is not a substitute (start at the workers' comp initial report ). This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common return-to-work letter errors auditors flag This letter occupies an unusual audit position: no Medicare denial code, CERT category, or OIG audit finding specific to therapist-authored return-to-work or accommodation letters was located in current sources, because the letter is an employment communication rather than a billed clinical service. The documented failure modes sit in leave law, scheme rules, and professional discipline instead. Under 29 CFR 825.305 , a certification with blank fields or vague, ambiguous, or nonresponsive answers must be identified in writing, the employee ordinarily gets seven days to cure it, and an uncured deficiency can cost FMLA protection. Ontario's CRPO reports report-writing among its substantial sources of practice inquiries and complaints: delay without reasonable cause, opinions outside competence, missing disclosure authorization, and client report merged into professional opinion. And when an employment opinion leans on a weak chart, the exposure is real: a 2020 HHS OIG audit of one psychotherapy provider found 111 of 120 sampled claims had treatment plans that did not document expected improvement, with at least $3.3 million in estimated overpayments. That was not an audit of these letters, but records that cannot support their own treatment plans will not support a capacity opinion either. The BastionGPT Clinical Advisory Board sees the same errors most often in return-to-work letter reviews: Diagnosis disclosed by default. The employer asked when the client can return; the letter answers with a diagnosis and treatment narrative. EEOC and Ontario guidance direct documentation at functional limitations, and routine diagnosis disclosure exceeds what most requests need. Medically cleared with nothing behind it. The phrase appears in no regulation, and without the job's essential functions in hand it overstates what a therapist assessed. Describe capacity and restrictions; skip the borrowed language. Restrictions HR cannot implement. Light duties and reduced stress give a scheduler nothing to act on. Dated weeks, hours per day, and notice periods turn the same clinical judgment into something an employer can honor and an auditor can verify. One accommodation presented as the only option. The employer may choose any effective accommodation; a letter that admits no alternative invites a standoff instead of the interactive process, and the EEOC expressly warns providers against it. Self-report converted into clinical fact. The letter asserts job conditions and performance the clinician never observed. Ontario's standard requires labeling objective fact, client-supplied information, and professional opinion separately; a letter that blurs them is the first thing opposing counsel reads aloud. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on return-to-work and accommodation letters. Draft a complete return-to-work or accommodation letter from a few bullets or a dictation: present capacity, a dated graded-return schedule, functional limitations, and accommodations tied to each one, in language HR can act on. Flag what should not travel before it does: diagnosis, treatment detail, and history the request does not need, so the letter discloses function and nothing else. Turn clinical findings into plain workplace language, or convert an employer's form questions into functional terms you can answer within scope. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a return-to-work letter from a therapist be? Most run 150 to 400 words and fit on one page. No US regulation prescribes a length, requires letterhead, or demands notarization for an ordinary return-to-work or accommodation letter; those are authenticity conventions. What does the work is specificity: a dated return, restrictions HR can schedule, and accommodations tied to functional limitations. A one-page letter that names dates and hours beats three pages of clinical narrative. Does a return-to-work letter need to include the diagnosis? Usually not. For an accommodation request the EEOC says documentation may describe the general nature of the condition and its functional limitations, and that a specific diagnosis is not always necessary; Ontario's Human Rights Commission likewise directs documentation at limitations rather than diagnosis. An employer may request enough information to establish a disability when it is not obvious, and some insurer or scheme forms legitimately require a diagnosis from a professional authorized to give one. The working rule: disclose the minimum the specific request needs, with the client's authorization. Can a therapist write a return-to-work letter, or does it have to be a doctor? For ordinary accommodation documentation, yes: the EEOC names licensed mental health professionals among those who can provide it. FMLA paperwork is credential-specific: clinical psychologists and clinical social workers are expressly recognized health care providers under 29 CFR 825.125, while licensed counselors are not a named freestanding category and may qualify through state scope of practice or the employer's or plan's acceptance. Workers' compensation schemes publish their own eligible-signer lists and prescribed forms, covered on the workers' comp initial report page. Can an employer require a fitness-for-duty certification before taking my client back? Only within limits. Under 29 CFR 825.312, after FMLA leave for the employee's own condition an employer may require a fitness-for-duty certification only under a uniformly applied policy, limited to the condition that caused the leave. It must have supplied the essential-function list with the designation notice to demand a function-by-function certification, it cannot delay reinstatement while authenticating, it may not seek second or third opinions on the return certification, and the employee bears the cost. For intermittent leave it may ask at most once every 30 days, and only with reasonable safety concerns. Outside FMLA, employer policy and state law govern; no general federal rule requires a note before every return. The leave-side paperwork lives on the FMLA certification page. Can my client's manager or HR call me to verify the letter? Under the FMLA, the client's direct supervisor may never contact you to authenticate or clarify a certification; that contact must come from an HR professional, a leave administrator, a management official, or a health care provider, within the limits of 29 CFR 825.307. Outside FMLA, you still need the client's authorization before discussing their care with anyone at the employer, and authentication requests should stay limited to confirming that you wrote the letter. How is this letter different from a fitness-for-duty evaluation? This letter is a treating clinician's limited statement about capacity, restrictions, and accommodations, resting on the treatment relationship. A fitness-for-duty evaluation is typically an employer-ordered, often independent, examination answering a safety or essential-function question, which the ADA permits only when job-related and consistent with business necessity. And when the question is benefits eligibility over a sustained period rather than a return date, the document is a medical source statement . Keeping the roles separate protects the treatment relationship and the opinion. Does a return-to-work letter expire after 30, 60, or 90 days? No general regulation gives these letters an expiration date; the 30-, 60-, and 90-day limits repeated on template sites are conventions, not law. The exceptions are scheme documents: workers' compensation certificates of capacity carry defined certification periods set by each scheme. For an ordinary letter, put a review date on temporary restrictions. It is the practical substitute for an expiry and keeps a short-term restriction from hardening into a permanent one. Can I bill insurance or charge the client for writing the letter? No nationwide payer policy makes a generic return-to-work letter a separately covered service, and no Medicare denial code exists specifically for one; coding-forum debate on the question reflects practitioner uncertainty, not payer authority. The defensible line is between a medically necessary assessment you actually performed, billed as such, and administrative form completion, charged or not charged to the client under your fee policy, contracts, and state law. Ontario offers a model: CRPO permits a reasonable report fee after an estimate, but a clinician may not refuse a treatment-related report solely because the client cannot pay. Can BastionGPT draft a return-to-work letter from my notes? Yes. Give it a few bullets, a dictation, or the relevant notes and it drafts the letter with capacity and the return date up front, restrictions HR can schedule, and each accommodation tied to a limitation. It can also flag diagnosis and treatment detail the request does not need, and rewrite clinical language in plain workplace terms. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates FMLA & Leave Certification Support Note Fitness-for-Duty Evaluation Workers' Compensation Initial Report Workers' Compensation Progress Report Return-to-School Letter Disability Evaluation & Medical Source Statement Epworth Sleepiness Scale Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/return-to-school-letter All note templates # Return-to-School Letter: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A return-to-school letter is a short note from a treating mental health clinician telling a school when and how a student can resume attendance after a mental health absence or hospitalization. It states the return date, attendance plan, functional limitations, recommended supports, and follow-up, with the minimum detail needed. Families, schools, and clinicians use it during re-entry. Most run under one page. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Child and adolescent therapists, psychologists, psychiatrists, and program or hospital clinicians Audience The family and school staff (administrators, counselors, nurses); a copy stays in the clinical record Typical length 100 to 300 words · 10 to 20 minutes by hand (clinical team estimate) Format family Clinical correspondence (compare: school collateral note, Section 504 statement) When it's used When a student resumes attendance after a mental health absence, assessment, or hospitalization Standards context No US, Canadian, or Australian law prescribes the letter; privacy law limits what it may disclose On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a return-to-school letter? A return-to-school letter is a short piece of clinical correspondence in which a treating mental health clinician tells a school when and under what conditions a student can resume attendance after a mental health absence, emergency assessment, or hospitalization. It typically states a proposed return date, an attendance plan, current functional limitations, a small set of recommended supports, and the follow-up in place. Nobody invented it and no body has ever standardized it: it grew out of hospital-to-school reintegration practice, and the research that touches it studies processes rather than a form. A 2019 national survey of 133 school psychologists examined how varied school re-entry protocols are, and a 2023 research synthesis built a nine-step transition framework from 53 documents; neither produced a standard clinician letter. You will also hear it called a school re-entry letter, a school reintegration letter, a fitness-to-attend letter, or a mental health clearance letter. That last name is the one to resist, because "clearance" promises something no clinician can deliver and no law defines. No US, Canadian, or Australian regulation reviewed for this page defines a state of being "cleared," sets a zero-risk threshold for attendance, or requires a clinician to guarantee safety; a time-limited assessment supports a bounded opinion about current functioning under stated conditions, not a promise about the future. The letter is also not a disability determination. Under 34 CFR 104.35 , a US public school must run its own Section 504 evaluation, drawing on multiple sources through a knowledgeable group, and IDEA assigns eligibility to a team that may not rely on any single measure. Your letter is evidence in those processes, never the decision; the Section 504 clinician statement and IEP psychoeducational input pages cover the documents written for them. ## Who uses return-to-school letters and when The writers are child and adolescent therapists, psychologists, psychiatrists, and clinicians in hospital, residential, partial hospitalization, and intensive outpatient programs whose young clients are heading back to class; the readers are families, school administrators, counselors, and school nurses. Reach for one when the immediate question is attendance: when and how the student returns, on what schedule, with what temporary supports. When the question is ongoing coordination rather than re-entry, the school and teacher collateral note documents the contact; when the student needs formal accommodations that outlast the transition, the Section 504 statement or IEP input serves the school's own evaluation process; and for an adult employee the equivalent document is the return-to-work letter , which follows the same functional logic with an employer audience. ## Return-to-school letter structure: what goes in each section Header, recipient, and student identifiers. Date, the named school contact and their role, the school or program, and the student's name and grade. Use the minimum identifiers the school needs to match its records; a name and grade usually do it. Pitfall: over-identifying. A full date of birth, address, and health numbers add nothing for the school and travel with every copy; privacy law in every reviewed jurisdiction points the same way, and Ontario's PHIPA expressly bars disclosing more than is reasonably necessary. Clinician identity and relationship. Your name, discipline, credentials, license or registration jurisdiction, practice contact details, and your relationship to the student: treating clinician, program clinician, or one-time evaluator. Pitfall: leaving the relationship unstated. A treating opinion and an independent evaluation carry different weight, and the school cannot tell which it is holding unless you say so. Basis of the opinion. The date and type of your most recent contact, and what the opinion rests on: treatment contacts, a discharge assessment, records, collateral information. Pitfall: an undated opinion. A letter that never says when you last saw the student hides how current it is, and professional assessment standards require conclusions to rest on a stated, adequate basis. Return date and attendance plan. The proposed date and the shape of the return: full days, a gradual build, or a reduced schedule with step-up dates. Pitfall: "cleared to return" with no plan. A bare clearance line gives attendance staff nothing to enter and pushes the school to improvise the schedule you should have proposed. Functional limitations relevant to school. What the student can and cannot yet do, in school terms: concentration span, stamina across a full day, testing tolerance, transitions between classes, need for breaks. Pitfall: leading with the diagnosis. Function is what the school can act on; Ontario's human rights guidance says a diagnosis is ordinarily unnecessary where needs can be described without it, and no US federal rule requires one in this letter. Recommended supports. Two to five specific, actionable supports: a pre-agreed break location, staggered make-up work, flexible deadlines, a designated adult, modified testing, brief check-ins, and an agreed response if distress escalates. Pitfall: reproducing the clinical safety plan. A letter circulates further than a restricted crisis plan; give the school observable warning signs and an agreed response, not session content. Duration, review, and signature. How long the plan applies, a review date, the conditions that should trigger reassessment, confirmation that follow-up care is in place, and your signature and date, with the consent that authorized the letter documented in your record. Pitfall: an open-ended plan. A letter with no review date turns a two-week restriction into a permanent feature of the student's file. ### Blank template (copy and adapt) RETURN-TO-SCHOOL LETTER Date: __________ To (name and role at the school): ____________________ School / program: _______________________________________________________ Student: _____________________________ Grade / year: ___________________ (use the minimum identifiers the school needs to match its records) From (treating clinician): __________________ Credentials: _____________ Practice contact: _______________________________________________________ Relationship to student: [ ] treating clinician [ ] evaluator [ ] other BASIS OF OPINION (date and type of most recent contact; what it rests on) _________________________________________________________________________ RETURN DATE AND ATTENDANCE PLAN Return date: __________ [ ] full days [ ] gradual return [ ] reduced Plan and step-up dates: _________________________________________________ _________________________________________________________________________ FUNCTIONAL LIMITATIONS RELEVANT TO SCHOOL (function, not diagnosis) _________________________________________________________________________ RECOMMENDED SUPPORTS (specific; who does what, where, and when) 1. ______________________________________________________________________ 2. ______________________________________________________________________ 3. ______________________________________________________________________ IF DISTRESS ESCALATES (observable signs; agreed response; who to contact) _________________________________________________________________________ DURATION AND REVIEW (how long this applies; review date; when to reassess) _________________________________________________________________________ Follow-up care in place: [ ] yes [ ] scheduled, first date: ____________ Consent / authorization documented in the clinical record: [ ] yes Signature: ______________________________ Date signed: _________________ Name / credentials: _____________________________________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample return-to-school letter Scenario: an adolescent returning to high school after a structured treatment episode. The letter shows a dated attendance plan and two supports while disclosing the minimum the school needs. All details are fictional. Return-to-School Letter. From: L. Navarro, PhD, licensed psychologist · To: Ms. R. Patel, student services coordinator, Winfield High School · Date: 08/21/2026 · Re: J.M., grade 10 Dear Ms. Patel, I am J.M.'s treating psychologist. With J.M.'s and her parents' consent, I am writing to support her return to school on Monday 08/24/2026 after an absence for health care. I last met with J.M. on 08/14/2026 and 08/20/2026, and this letter draws on those visits and on the discharge recommendations from her recent treatment program. Attendance plan: half days (mornings) from 08/24/2026 through 09/04/2026, then full days from 09/08/2026 if the gradual return goes as expected. If a step needs to change, the family or the school can contact me and I will update the plan in writing. Current functioning: J.M. is managing coursework and social contact well in shorter blocks, tires by early afternoon, and concentrates best with brief breaks. She is ready for regular classwork on this schedule; large make-up loads and timed testing are the main strains at present. Requested supports: (1) permission to take a short break in the counseling office when she shows or signals she needs one, without a public explanation in class; (2) make-up work spread over the four weeks to 09/18/2026 rather than due at once, with deadlines set through her counselor. If distress escalates: J.M. will go to the counseling office, and the school should contact her parents at the numbers on file. Please direct questions about this plan to me; the family has consented to the school contacting my office. Follow-up and review: J.M. continues in weekly care with me. I will review this plan with the family by 09/25/2026, or sooner if the school or family raises a concern, and will send an update if the recommendations change. Sincerely, L. Navarro, PhD, licensed psychologist (license PSY 24815) Cedar Grove Child & Adolescent Psychology · (555) 013-4482 · signed 08/21/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works It never names a diagnosis, a medication, or the reason for the absence: the school receives function and a plan, which is everything it needs to act, and the disclosure stays minimal as copies circulate. The attendance plan is dated at every step (half days to 09/04, full days from 09/08, review by 09/25), so attendance staff can enter it without interpreting anything. Both supports are operational: who does what, where, and through whom, instead of a general request for flexibility. The escalation line gives an observable trigger, an agreed response, and a contact path without reproducing the clinical safety plan. The opinion is bounded: contact dates state its basis, consent is on record, follow-up care is confirmed, and a review date keeps a temporary plan from hardening into a permanent record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Privacy does the heavy lifting here, because the letter's defining act is disclosure to a third party. In the US, a HIPAA-covered clinician generally needs a valid authorization under 45 CFR 164.508 before sending health information to a school; the serious-threat exception exists for genuine emergencies, not routine return letters. Who signs that authorization is its own question: HIPAA generally treats parents as an unemancipated minor's personal representative, but state minor-consent laws can shift control to the adolescent, and in Ontario a capable patient aged 16 or older ordinarily controls the disclosure even when a parent requests the letter, under PHIPA and the provincial privacy commissioner's guidance; the minor consent page covers the framework. Australia's Privacy Act points the same direction through APP 6, and a small practice is not exempt merely because it is small: providing a health service and holding health information keeps it covered. Everywhere, PHIPA's standard is worth adopting as practice even where it is not law: disclose no more than is reasonably necessary, and prefer the least identifying information that serves the purpose. Keep session-process material out entirely; separately maintained psychotherapy notes carry their own authorization rules, and a school letter is the last place their content belongs. Two boundary facts keep the rest in proportion. First, the same physical letter lives under two privacy regimes at once: your copy stays part of your clinical record under HIPAA and state or provincial health law, while the copy a US elementary or secondary school files ordinarily becomes an education record under FERPA, governed by education rules and readable by school officials on that side of the line, as the 2019 joint HHS and Education guidance explains. That is the practical argument for writing every line as if the whole school community might one day handle it. Second, the letter has no payer status: no CMS manual, provincial plan, or MBS item recognizes a return-to-school letter as its own covered service, so the writing time is not separately payable by default, and the treatment record, not the letter, is what any audit of the underlying care examines. Retention follows the clinical record's rules, and for minors those run long: Ontario psychologists keep the record until ten years after the client turns 18 or after last contact, whichever is later; New South Wales and Victoria keep records made when the client was under 18 until age 25; New York keeps a minor's record until at least one year after age 21. The letter is a convention; the consent behind it and the record around it are law. ↑ Back to the template and downloads ## Common return-to-school letter errors auditors flag Start with an honest audit fact: no OIG, CMS, provincial payer, or MBS audit program isolates the return-to-school letter, and no denial code exists for a missing or inadequate one. The audit weight sits behind the letter, in the billed care and in the disclosure itself. The OIG's national psychotherapy audit (A-09-21-03021, 2023) found that only 84 of 216 sampled Medicare psychotherapy days met all requirements and estimated $580 million in improper payments , and a 2020 OIG audit of a single psychology group found 111 of 120 sampled psychotherapy claims noncompliant: if the treatment episode behind your letter is billed, those are the records that must stand up. On the disclosure side, Ontario College complaint decisions have directed clinicians to use detailed written consent or to document verbal consent properly after consent records fell short, which is exactly the failure a school letter exposes. The BastionGPT Clinical Advisory Board sees the same errors most often in return-to-school letter reviews: Safety guarantees. "Poses no risk to self or others" and "cleared as safe" assert more than a time-limited assessment can establish; state the functional opinion, its basis and date, the conditions, and the reassessment triggers instead. Diagnosis where function belongs. No reviewed regulation requires a diagnosis in this letter, Ontario's human rights guidance discourages demanding one, and the school's copy becomes an education record; leading with a label discloses the most sensitive fact to the widest audience for the least gain. Consent assumed rather than documented. A parent's phone request is not an authorization, a capable Ontario 16-year-old ordinarily controls the disclosure personally, and regulators have faulted clinicians for consent they could not evidence. The safety plan pasted in. The full clinical safety plan hands warning-sign detail and history to everyone who ever opens the file; the letter needs observable signs, an agreed response, and a contact, and the restricted plan stays restricted. An open-ended plan. No duration, review date, or reassessment condition, so a two-week half-day schedule quietly becomes the student's permanent profile. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on return-to-school letters. Draft a complete return-to-school letter from a few bullets, a dictation, or your discharge recommendations, with the return date, attendance plan, functional limitations, and supports in order. Run a minimum-necessary pass before the letter leaves your desk: flag the diagnosis mentions, treatment details, and identifiers the school does not need, with functional language suggested in their place. Rewrite the plan in plain language for the student and family, so everyone leaves with the same return date, the same supports, and the same contact path. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a return-to-school letter be? Under one page: most run 100 to 300 words. No US, Canadian, or Australian rule sets a length, a format, or a required form for this letter, and brevity is a privacy feature as much as a courtesy: the school's copy typically becomes part of the student's education record, so every extra clinical detail travels further than you intend. Say what the school must act on, then stop. Is a return-to-school letter legally required after a mental health absence? No national US, Canadian, or Australian instrument requires one. Where a school demands a letter, that is local policy, not national law, and US hospital discharge-planning rules under 42 CFR 482.43 govern the care transition, not school correspondence. A blanket demand can deserve scrutiny when it delays a student with a disability or insists on a diagnosis. The letter is a convention; the consent that authorizes sending it is the legal requirement. Should I write that the student is "cleared" or "safe to return"? No. No regulation reviewed for this page defines a mental health "clearance," and a promise that a student is safe or poses no risk asserts more than a time-limited assessment can establish. Write a bounded opinion instead: the dates and basis of your assessment, the attendance the student can manage under stated conditions, and the signs that should prompt reassessment. The school keeps its own responsibility for attendance and operational decisions. Does the letter need to include the diagnosis? Usually no. Section 504 and IDEA require the school to evaluate from multiple sources rather than accept or demand a diagnosis letter, and Ontario's human rights guidance says a diagnosis is generally unnecessary where accommodation needs can be described functionally. Include it only when it is authorized and materially needed for the stated purpose; medication instructions, where a school must administer or monitor, usually belong in separate school medication paperwork. Who consents to the letter when the student is a minor? It depends on the jurisdiction, and "the parent signs" is not a universal answer. HIPAA generally treats parents as an unemancipated minor's personal representative, but state minor-consent laws can give the adolescent control of information about care they consented to themselves; in Ontario, a capable patient aged 16 or older ordinarily controls the disclosure; Australian practice turns on capacity and the Privacy Act. Settle who consents before drafting, and document it: see the minor consent template. Can I bill for the time spent writing the letter? There is no automatic payment entitlement. No Medicare benefit, Canadian provincial plan policy, or Australian MBS item reviewed recognizes a return-to-school letter as its own covered service; whether letter-writing time is payable depends on your payer contract and on the covered service actually furnished. Whatever the billing answer, the letter never substitutes for the treatment record: audits of psychotherapy claims examine the notes, time support, and treatment plan behind the care. What is the difference between a return-to-school letter and a Section 504 clinician statement? Timeframe and decision-maker. The return letter manages an immediate, time-limited transition: a date, an attendance plan, temporary supports, a review point. A Section 504 clinician statement feeds the school's own evaluation of whether a student needs formal, ongoing accommodations, decided by a knowledgeable school group under 34 CFR 104.35; special-education questions run through the IEP process . If the supports your letter recommends still matter a term later, the 504 route is usually the right next step. What happens to the letter once the school has it, and how long do I keep mine? The copy a US elementary or secondary school maintains ordinarily becomes a FERPA education record, handled under education rules rather than HIPAA; your copy stays part of the clinical record. Retention follows the record, and minors' records run long: Ontario psychologists keep the record until ten years after the client turns 18 or after last contact, whichever is later; New South Wales and Victoria keep records made in childhood until age 25; New York keeps a minor's record until at least one year past age 21. Keep the letter alongside the consent that authorized it. Can BastionGPT draft a return-to-school letter from my notes? Yes. Give it a few bullets, a dictation, or your discharge recommendations and it drafts the letter with the return date, attendance plan, functional limitations, and supports in order, then runs a minimum-necessary check for details the school does not need and can produce a plain-language version for the family. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates School & Teacher Collateral Note Section 504 Clinician Statement IEP Psychoeducational Input Return-to-Work & Accommodation Letter Consent for Treatment of a Minor Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/section-504-statement All note templates # Section 504 Clinician Statement: Template & Sample Letter Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A Section 504 clinician statement is a letter from a treating clinician that describes a student's health condition, its functional impact on learning, and accommodation suggestions for the school's Section 504 team. The team considers it as one source of evidence when deciding eligibility and accommodations under the Rehabilitation Act. Most run 300 to 600 words, about one page. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating therapists, psychologists, counselors, psychiatrists, pediatricians; any licensed clinician within scope Audience The school's Section 504 team; parents and guardians; the student's education record Typical length 300 to 600 words · 20 to 30 minutes by hand (clinical team estimate) Format family Disability-accommodation support letter (compare: return-to-school letter, IEP input) When it's used When a family asks you to support a school's Section 504 evaluation or the review of an existing 504 plan Standards context No federal form or required fields exist; 34 CFR 104.35 governs how the school must weigh the letter, not how you write it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a Section 504 clinician statement? A Section 504 clinician statement is a letter or short report from a treating clinician to a school's Section 504 team, describing a student's health condition, the concrete ways it limits learning and school participation, and accommodations worth considering. Section 504 itself is the civil-rights provision of the Rehabilitation Act of 1973 (Public Law 93-112) that bars federally funded schools from discriminating on the basis of disability; the Department of Education's implementing regulations at 34 CFR Part 104 date to 1980. Neither the statute nor the regulations created a clinician-letter form: the standardized thing is the school's evaluation process, not your correspondence. Families and schools call the same document a 504 letter, a 504 support letter, a therapist letter for a 504 plan, or simply a school accommodation letter. The load-bearing distinction: your statement is not the 504 plan, not the evaluation, and not the eligibility decision. Under 34 CFR 104.35 , the school must draw evaluation information from a variety of sources, document and carefully consider all of it, and have the decision made by a group knowledgeable about the student, the data, and the placement options. Your letter is one source that group must consider; it decides nothing by itself. The Department of Education is equally clear in the other direction: no medical diagnosis or clinician letter is a prerequisite, and a school may determine disability without either. That double status, valuable evidence but never a required or deciding document, shapes everything about how the letter should be written. For the neighboring document that feeds special education under IDEA rather than access under Section 504, see IEP input . ## Who uses Section 504 clinician statements and when Any treating clinician may be asked for one: therapists and counselors treating anxiety or depression, psychologists, psychiatrists managing medication, pediatricians. Federal law names no required profession; what you may state is bounded by your state scope of practice, and prelicensed clinicians follow their board's co-signature rules. The request usually arrives when a family starts a 504 evaluation, when an existing plan comes up for review, or when new symptoms change what the student needs. Reach for this letter when the school needs focused information about the functional impact of a condition you already treat. When the question is a learning disorder, cognitive testing, or eligibility for specially designed instruction, a psychoeducational evaluation or ADHD evaluation report is the right instrument, and input into an IDEA process is its own document ( IEP input ). A student re-entering after a treatment episode needs a return-to-school letter , which answers a different question and does not establish 504 eligibility. ## Section 504 clinician statement structure: what goes in each section Addressee, student, and purpose. Date the letter, address it to the school's Section 504 coordinator or team, identify the student with a second identifier (date of birth or student ID), and state the purpose in one sentence: information offered for the team's Section 504 evaluation or plan review, written at the family's request with consent on file. Pitfall: leaving the purpose implicit. A letter that never says what process it serves gets read as a fitness certification or a risk clearance, questions you were not asked and may not have assessed. Clinician identification and treating relationship. Your name, degree, license number and state, and practice contact details, then the treating relationship: how long you have seen the student, at what frequency, and the date of the most recent session or evaluation. Pitfall: omitting the relationship and recency. The team needs to know this is a treating opinion rather than an independent evaluation, and how current it is; an undated basis reads as stale evidence. Condition and functional limitations. Name the condition if the release covers it and it helps the team; then do the real work: the specific, observable ways the condition limits the student. Slowed processing under time pressure, difficulty initiating written tasks, physical symptoms that make staying in the classroom hard. Pitfall: the bare-diagnosis letter. "J. has anxiety; please provide accommodations" gives the team nothing to evaluate: the OCR resource guide directs schools to weigh functional and educational information, and a diagnosis alone ordinarily does not identify what a student needs. Impact on major life activities. Connect the limitations to the activities Section 504 recognizes: concentrating, thinking, reading, communicating, sleeping, attendance, class participation. Note episodic patterns and variability, such as symptoms that flare during timed assessment or mornings after high-symptom nights. Pitfall: naming symptoms without the nexus. The legal question is substantial limitation of a major life activity; a letter that never touches that frame leaves the team to build the bridge alone. Sources, basis, and limits. Say what rests on your own sessions, what on rating scales with dates and scores, and what on parent or teacher report. State plainly what you have not done: no school observation, no cognitive or academic testing. Pitfall: presenting secondhand school facts as your own findings. Attribution is what makes the rest of the letter credible; a team that catches one unattributed claim discounts the document. Accommodation suggestions tied to function. For each suggestion, name the barrier it answers: because time pressure escalates symptoms, reduced-distraction testing or time flexibility; because symptoms spike without an exit, a discreet break pass. Frame them as clinical suggestions for the team to weigh. Pitfall: prescribing. Under 34 CFR 104.35(c) the knowledgeable school group decides services and placement; "the school must provide 100% extended time" overclaims your role and invites the team to discount the letter. Consent, signature, and date. Reference the signed release naming this school and this purpose, then sign with credentials and date. Pitfall: attaching records to strengthen the letter. Psychotherapy notes carry heightened HIPAA protection and need their own specific authorization; a purpose-limited summary discloses less and persuades more. ### Blank template (copy and adapt) SECTION 504 CLINICIAN STATEMENT Date: __________ Purpose: [ ] initial 504 evaluation [ ] plan review To: Section 504 coordinator / team, ___________________________ (school) Student: ____________________ DOB: __________ Grade: _________________ CLINICIAN AND TREATING RELATIONSHIP Name / credentials / license no. and state: ____________________________ Practice contact: ______________________________________________________ Treating since: ________ Frequency: ____________ Last seen: __________ CONDITION (a diagnosis is not federally required; include if authorized and clinically useful) ________________________________________________________________________ FUNCTIONAL LIMITATIONS (concrete and school-relevant: task initiation, processing under time pressure, remaining in class) ________________________________________________________________________ ________________________________________________________________________ IMPACT ON MAJOR LIFE ACTIVITIES (concentrating, thinking, reading, sleeping, attendance, participation; note any episodic pattern) ________________________________________________________________________ ________________________________________________________________________ BASIS OF OPINION ([ ] sessions [ ] rating scales, dated [ ] parent report [ ] school input [ ] records) and limits (e.g. no school observation, no cognitive testing) ________________________________________________________________________ ACCOMMODATION SUGGESTIONS, EACH TIED TO A STATED BARRIER (the school's knowledgeable group decides; frame as clinical suggestions) 1. _____________________________________________________________________ 2. _____________________________________________________________________ 3. _____________________________________________________________________ CONSENT: release on file naming this school, signed ________ by ____________________ (guardian) Scope: _______________________________ Signature / credentials: ____________________ Date signed: ____________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample Section 504 clinician statement Scenario: a licensed psychologist supports a school's Section 504 evaluation for a 14-year-old client with generalized anxiety disorder. All details are fictional. Section 504 Clinician Statement. From: R. Vasquez, PhD, Licensed Psychologist, Cedar Park Psychology · To: Section 504 Coordinator, Lakeview High School · Date: 08/05/2026 · Re: J.H., DOB 03/22/2012, Grade 9 Dear members of the Section 504 team, Purpose and consent: I am writing at the request of J.H.'s parents, under a release signed 07/29/2026 naming Lakeview High School, to provide information for the team's Section 504 evaluation. I have treated J.H. in weekly cognitive behavioral therapy since February 2026, 22 sessions to date, most recently on 08/03/2026, for generalized anxiety disorder. Functional limitations: During periods of heightened anxiety, J.H. shows slowed information processing, difficulty initiating written tasks, and physical symptoms (racing heart, nausea) that make remaining in the classroom difficult. By her and her parents' report, these effects are most likely during timed tests and unstructured transitions. GAD-7 scores across the last three months have ranged from 12 to 16, most recently 13 on 08/03/2026. Impact on major life activities: In my clinical judgment these symptoms substantially limit concentrating and thinking during sustained academic tasks, and they affect attendance: her parents report six late arrivals since April on mornings following high-symptom nights. The pattern is episodic; on low-symptom days her functioning is unremarkable, which is typical of the condition rather than evidence against impairment. Basis and limits: My opinion rests on 22 clinical sessions, monthly GAD-7 administrations, and parent interviews. I have not observed J.H. at school and have not conducted cognitive or academic testing; the team's own data will speak to classroom performance. Suggestions for the team's consideration: Because symptom escalation impairs concentration under strict time pressure, consider reduced-distraction testing and time flexibility sufficient to permit brief regulation breaks. Because physical symptoms escalate when she feels unable to leave, consider a discreet break pass with an agreed re-entry routine. Because late arrivals follow high-symptom nights, consider an attendance arrangement that lets her make up missed morning work without penalty. These are clinical suggestions tied to the barriers I observe; the team is best placed to decide which supports fit. I am glad to answer questions with the family's continued consent. Sincerely, R. Vasquez, PhD, Licensed Psychologist (license PSY 8841) Cedar Park Psychology · (555) 013-7702 · signed 08/05/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The purpose sentence names the process (a Section 504 evaluation), the requester, and the consent on file, so the team knows exactly what the document is and that it can lawfully use it. Functional limitations are concrete and school-shaped (initiating written tasks, remaining in class, timed assessment), which a team can act on where a bare diagnosis gives it nothing to evaluate. The major-life-activities paragraph does the 504-specific work, tying symptoms to concentrating, thinking, and attendance, with an episodic pattern the team can verify against its own records. Sources are attributed and limits stated (no school observation, no testing), which marks the letter as treating evidence rather than advocacy and tells the team where its own data must fill in. Each accommodation is paired with the barrier it answers and framed as a suggestion, respecting the group's decision role under 34 CFR 104.35 while giving it something concrete to adopt. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A 504 statement changes legal hands the moment it arrives. In your chart it is part of the clinical record, governed by HIPAA and state law; once the school maintains it, it generally becomes an education record under FERPA and falls outside HIPAA's definition of protected health information, per the HHS and Department of Education joint guidance . Parents hold FERPA rights to inspect it, so write every line as if the family will read it, because they can. Send it only under a signed authorization naming the school and the purpose, and keep psychotherapy notes out: they carry heightened HIPAA protection and their own authorization rules, and a purpose-limited functional summary both protects the client and persuades better. Retention attaches to your record, not to the letter as a class: no single federal period governs clinical records, and the archived APA record-keeping guideline's seven years after last service for adults, longer for minors, is convention, with state licensing and payer rules controlling. Keep the letter, the request, the authorization, and what you relied on in the chart. Ongoing contact with teachers around the plan is its own documentation event: record each call in a school collateral note . Sort the rules by force before you write. LAW governs the school's process: evaluation drawing on multiple sources, a decision by a knowledgeable group, periodic reevaluation, and procedural safeguards for parents (34 CFR 104.35 and 104.36). What law does not reach is your letter: no federal regulation prescribes a template, requires a diagnosis or DSM code, designates an authoring profession, sets a deadline, an expiration date, or an annual recertification, or gives your recommendations binding effect. District forms requesting those things are local policy, worth accommodating but not federal law, and the widely repeated 60-day evaluation clock belongs to IDEA, not Section 504. PAYER POLICY decides nothing here: no general insurance mandate covers preparing school correspondence, so handle the fee through your informed-consent agreement. One point in the family's favor is worth knowing: OCR's 2016 resource guide says that if the district itself decides a medical assessment is necessary to complete its evaluation, the district must offer one at no cost to the parent. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common Section 504 clinician statement errors auditors flag No payer or federal auditor reviews 504 clinician statements as a billed document class, and this page will not invent a denial rate for one: the accountability trail runs through the school side. The Department of Education's Office for Civil Rights enforces the process, and its complaint files show where letters fail: in OCR matter 04-25-1604 , involving Wilkes County Schools, the dispute included a district position that necessary medical documentation was missing even though the record contained physician information: the recurring failure is not bad letters so much as letters lost or mischaracterized inside the evaluation. The nearest federal audit dollars sit in school-based claiming: a 2024 HHS OIG audit estimated that $551.4 million of roughly $590 million in Pennsylvania school-based Medicaid claims was improperly claimed for documentation-support reasons. And the population is large: Zirkel and Gullo's analysis of the Department of Education's 2021-22 Civil Rights Data Collection puts 504-only enrollment at 3.93% of US students. The BastionGPT Clinical Advisory Board sees the same errors most often in Section 504 clinician statement reviews: A diagnosis where the function should be. "Has anxiety, please accommodate" with no functional explanation: schools must weigh how the impairment affects the student, and official guidance treats a diagnosis alone as ordinarily insufficient to identify services. Prescribing instead of suggesting. The letter approves a 504 plan or orders named accommodations. Eligibility and services belong to the knowledgeable school group under 34 CFR 104.35(c); overclaiming invites the team to discount everything else the letter says. Unattributed sources. Parent report, student report, teacher emails, and the clinician's own observation blur into one clinical voice, making conclusions look more certain than the basis supports. Overdisclosure. Psychotherapy notes or the full chart attached to strengthen the request. Heightened HIPAA protections apply to psychotherapy notes, and a purpose-limited functional summary was the right instrument. The recycled template. Boilerplate findings never actually assessed, no date, no signature, no treating-relationship statement: nothing the school can authenticate, and a professional-accuracy problem if the letter is ever challenged. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on Section 504 clinician statements. Draft a complete statement from a few bullets or a dictation: condition, functional limitations, major-life-activity impact, and accommodation suggestions each tied to a barrier. Translate clinical language into the plain, school-relevant functional statements a 504 team can act on, without diluting the clinical substance. Check the draft for overdisclosure before you send it: flag diagnosis detail, treatment content, and identifiers beyond what the letter's purpose needs. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a Section 504 clinician statement be? Most run 300 to 600 words, about one page. No federal rule sets a length, requires letterhead, or prescribes a template: 34 CFR Part 104 regulates the school's evaluation process and never mentions clinician correspondence. A focused page stating the condition, the functional limitations, the major-life-activity impact, and function-linked suggestions serves the team better than a long clinical summary, and it discloses less. Does a student need a diagnosis from a clinician to get a 504 plan? No. The Department of Education states that nothing in Section 504 requires a medical diagnosis as a precondition to eligibility, and a school may determine disability without medical documentation. A diagnosis you do provide must be considered, but by itself it ordinarily does not identify the services a student needs. And if the district decides a medical assessment is necessary to complete its evaluation, OCR says it must offer one at no cost to the parent rather than sending the family to buy one. Can I write the 504 letter as a therapist, or must it come from a physician? Federal Section 504 law designates no profession. What you may state, a diagnosis or a testing conclusion, is bounded by your state scope-of-practice rules, and prelicensed clinicians follow their board's and employer's co-signature requirements. What matters to the team is that the author is licensed, competent to make the statements offered, and clear about the treating relationship behind them. Can I tell the school which accommodations it must provide, and can it ignore my letter? Neither. Under 34 CFR 104.35 the school must document and carefully consider information from a variety of sources, so it cannot lawfully ignore your statement; but eligibility and services are decided by a group knowledgeable about the student and the data, so it need not adopt your recommendations either. Function-linked suggestions framed as clinical opinion are the form of influence the process actually allows, and the form teams take most seriously. Do I need a release before sending a statement to the school? Ordinarily yes: a signed authorization naming the school and the purpose, before anything travels. Once the school maintains the letter, it generally becomes an education record under FERPA rather than HIPAA protected health information, which means parents can inspect it and school-side handling follows education law ( HHS and Department of Education joint guidance ). Keep psychotherapy notes out entirely: they carry heightened HIPAA protection and need their own specific authorization. What is the difference between a 504 clinician statement and IEP input? The statute and the stakes. A 504 statement supports access accommodations under the Rehabilitation Act's civil-rights framework. IEP input feeds an IDEA process that requires a full initial evaluation, on a 60-day federal default timeline, and can end in specially designed instruction rather than access changes alone. If instruction itself may need to change, the IDEA pathway is the right target; if the question is a learning disorder or test scores, a psychoeducational evaluation is the right instrument. How recent does the letter have to be, and does it expire? No federal rule gives it a shelf life, requires annual recertification, or sets a renewal date. The school's own duty is periodic reevaluation, and districts that review plans annually do so as local practice, not federal mandate. Judge currency clinically: the stability of the condition, the time since you last saw the student, and any change in functioning. Districts and testing agencies may impose their own recency rules; treat those as local policy and say so in the letter if you are writing against one. Can I bill insurance for writing a 504 letter? Usually not, and no general mandate says a plan must pay. Letter preparation is administrative correspondence outside most benefit contracts, so attempted claims fail on ordinary grounds, a noncovered service (CARC 96) or unsupported necessity, not on any 504-specific rule. The clean path is a documented fee in your informed-consent agreement, quoted to the family before you write. Whatever you charge, keep the letter, the request, and the authorization in the chart. Can BastionGPT draft a Section 504 clinician statement? Yes. Give it a few bullets: the condition, what you observe in session, what parents report about school, and the barriers you see. It drafts the statement with functional limitations and accommodation rationale connected, in plain language a 504 team can act on, and it can check a finished draft for overdisclosure before you send it. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates IEP Psychoeducational Input Psychoeducational Evaluation Report ADHD Evaluation Report Return-to-School Letter School & Teacher Collateral Note CELF-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/iep-input All note templates # IEP Psychoeducational Input: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 IEP psychoeducational input is a treating psychologist's or other outside clinician's written report to a school team: findings, educationally relevant functional effects, and recommendations offered for the IEP process. Clinicians send it, with a signed authorization, before evaluations, eligibility determinations, and IEP reviews. A short input letter runs 300 to 900 words; a full private evaluation runs far longer. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating psychologists, therapists, counselors; independent evaluators submitting private evaluations Audience The school IEP team, school psychologist, parents; the clinician's own record on the sending side Typical length 300 to 900 words · 30 to 60 minutes by hand (clinical team estimate) Format family Clinical correspondence / collateral report (compare: psychoeducational evaluation, 504 statement) When it's used Before school evaluations, eligibility determinations, IEP development, and IEP reviews Standards context No law prescribes the letter; IDEA prescribes what the school must do with it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is IEP psychoeducational input? IEP psychoeducational input is the written contribution a clinician outside the school system makes to a student's special-education process: a report, letter, or addendum describing clinically observed symptoms, diagnoses where relevant, evaluation findings, the effects on learning, and possible supports for the IEP team to weigh. Nobody standardized it. Unlike the IEP itself or a school eligibility report, this document has no promulgating authority, official template, or prescribed field list; it grew up as the practical bridge between clinical care and school decision-making, which is why practice knows it under a dozen names: IEP support letter, IEP input letter, therapist letter for school, treating-provider statement, functional-impact letter, private or outside evaluation. This page covers the outside clinician's document, not the school's IEP. Two boundary lines do the most work. First, an outside report is not automatically an independent educational evaluation. Under 34 CFR 300.502 , an IEE is an evaluation conducted by a qualified examiner who is not employed by the district, and agency criteria, public-funding rules, and hearing rights attach to that defined category; a treatment summary does not become an IEE because a psychologist signed it. What a private report earns instead is consideration: the team must consider a privately obtained evaluation that meets agency criteria and must review evaluations and information the parents provide, but "must consider" never means "must implement," and the district still owes its own sufficiently comprehensive evaluation using multiple tools. Second, input for an IEP is not the same document as support for a Section 504 plan: an IDEA IEP delivers specially designed instruction, a 504 plan delivers equal-access accommodations, and the two arise under different statutes with different eligibility tests. When your document is a full evaluation, the psychoeducational evaluation report covers that genre; when the question is access, see the Section 504 clinician statement . ## Who uses IEP psychoeducational input and when Treating psychologists, therapists, and counselors of school-age clients write IEP input whenever the school process needs what treatment already knows: before an initial evaluation, for an eligibility determination, ahead of an annual review, or when a reevaluation is due. Independent evaluators submit the same kind of addendum when a family commissions a private evaluation and wants it in front of the team. Choose the format by what you actually did. When you conducted standardized testing and are integrating cognitive, academic, and behavioral data, the full psychoeducational evaluation report applies, and instrument write-ups such as the WISC-V report feed it. When the issue is equal access rather than specially designed instruction, the US vehicle is a Section 504 clinician statement . When the school wants observations of the student in context, that is a school and teacher collateral note , and for children below school age the developmental assessment covers the evaluation side. Renaming a short treatment letter a psychoeducational evaluation does not make it one; the team will read it against what it claims to be. ## IEP psychoeducational input structure: what goes in each section Header, recipient, and purpose. Date, student identifiers (name or initials, date of birth, grade), your name, degree, license, and contact details, the recipient (the IEP team or a named district contact), and one purpose sentence saying what the document is: treatment-provider input, an excerpt of a private evaluation, or a full report. Say what it is not, too. Pitfall: a letter with no purpose line. A treatment summary that never states its scope gets read as a comprehensive educational evaluation, then tested against criteria it was never meant to meet. Authorization and treatment relationship. The signed authorization's date and scope, who authorized release, and the basis of your knowledge: how long you have worked with the student, in what role, at what frequency. Pitfall: sending protected health information to a school without a valid authorization. HIPAA generally requires one, and once the school maintains your report it ordinarily becomes a FERPA education record with different access and redisclosure rules. Sources of information. Sessions, parent and teacher interviews, rating scales, records reviewed, and any direct testing, with dates and the language of assessment. Then keep three evidentiary levels separate throughout: observed fact, collateral information, professional opinion. Pitfall: secondhand reports written as direct findings. "Completes half of independent work (teacher interview)" is defensible; the same sentence without the informant is not. Findings and index scores. When you tested, report each instrument with dates, index or composite scores with percentiles, and a validity statement; when you did not test, say so plainly. Index scores and interpretation travel; item content and protocols do not. Pitfall: attaching record forms or quoting test items. Score summaries serve the team; item-level content compromises test security and adds nothing an IEP team can use. Educationally relevant functional effects. The section the team actually acts on: translate findings into attendance, concentration, emotional regulation, transitions, work completion, peer interaction, and access to instruction, naming the informant for each. Pitfall: stopping at the diagnosis. IDEA eligibility turns on educational need established through multiple sources, and a diagnosis alone does not determine eligibility or services. Recommendations for team consideration. Specific supports, each tied to a finding, framed as input for the team to weigh, plus the data you would want reviewed at the next annual review. Pitfall: prescribing exact minutes, placement, or a one-to-one aide. Programming decisions depend on classroom data and team analysis a treating clinician does not hold, and IDEA assigns them to the team. Scope, limitations, and signature. What you did not do (no classroom observation, no academic testing, no records beyond those listed), your availability for the meeting, and a dated signature with credentials, plus a supervisor co-signature where your jurisdiction requires one for supervised practice. Pitfall: an unscoped, unsigned letter. The scope statement keeps a solid treatment letter from being dismissed as a failed evaluation, and signature rules still bind the clinical record behind it. ### Blank template (copy and adapt) IEP PSYCHOEDUCATIONAL INPUT (OUTSIDE CLINICIAN) Date: __________ To (IEP team / district contact): ____________________ Student: ____________________ DOB: __________ Grade: __________________ From: ______________________ Degree / license: _________________________ Practice + contact: _____________________________________________________ PURPOSE (what this document is, and is not) [ ] Treatment-provider input [ ] Private evaluation excerpt [ ] Other _________________________________________________________________________ AUTHORIZATION AND TREATMENT RELATIONSHIP Signed authorization dated: __________ Scope: __________________________ Role and duration (treating / evaluating, since when, frequency): _________________________________________________________________________ SOURCES OF INFORMATION (sessions, interviews, scales, records, testing; mark each finding: observed fact / collateral report / opinion) _________________________________________________________________________ FINDINGS AND INDEX SCORES (instrument, date, index score, percentile; validity statement; index scores only, no item content) _________________________________________________________________________ _________________________________________________________________________ EDUCATIONALLY RELEVANT FUNCTIONAL EFFECTS (attendance, concentration, regulation, transitions, work completion, access to instruction) _________________________________________________________________________ _________________________________________________________________________ RECOMMENDATIONS FOR TEAM CONSIDERATION (each tied to a finding) 1. ______________________________________________________________________ 2. ______________________________________________________________________ 3. ______________________________________________________________________ SCOPE AND LIMITATIONS (what was not done; e.g. no classroom observation) _________________________________________________________________________ Available for the IEP meeting: [ ] in person [ ] by phone [ ] no Signature / credentials: ______________________ Date signed: ___________ Supervisor co-signature (if required): __________________________________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample IEP psychoeducational input Scenario: an outside psychologist who completed a private evaluation submits an input addendum before a fourth grader's IEP eligibility meeting. All details are fictional. IEP Psychoeducational Input (Addendum). From: L. Navarro, PhD, Licensed Psychologist · To: The IEP team, Cedar Grove Elementary (Ridgeview District) · Date: 08/05/2026 · Re: J.K., DOB 03/14/2017, Grade 4 Purpose and scope: This addendum summarizes the private psychoeducational evaluation I completed on 07/28/2026 and is submitted at his parents' request for the team's consideration at the eligibility meeting scheduled for 08/19/2026. It is input to the team's process: it is not the district's evaluation, and I have not observed J.K. in his classroom. Authorization and relationship: J.K.'s parents signed an authorization on 08/03/2026 covering release of this summary and, on request, the full report. I evaluated J.K. across two testing sessions in July 2026 and one parent feedback session; I am an outside evaluator, not his treating therapist. Sources: Review of report cards and spring universal reading screening results for grades 2 and 3 (provided by the family), parent and teacher interviews, standardized parent and teacher rating scales, and individually administered, nationally normed cognitive and academic achievement batteries. All assessment was conducted in English, J.K.'s primary language. Index scores (July 2026): Index / composite Standard score Percentile Range General cognitive ability 102 55 Average Verbal comprehension 110 75 High average Working memory 85 16 Low average Processing speed 83 13 Low average Reading composite 84 14 Below average Mathematics composite 98 45 Average J.K. engaged readily across both sessions, and results are considered valid estimates of current functioning. Index scores only are reported here; no test item content is included. Eligibility-relevant findings: Overall cognitive ability is average, with relative weaknesses in working memory (16th percentile) and processing speed (13th percentile). Reading achievement is below average (14th percentile) against average mathematics. Parent and teacher rating scales converge on clinically significant inattention. Per teacher interview, J.K. loses multi-step directions, completes roughly half of independent work, and avoids grade-level text. In my opinion these findings meet DSM-5-TR criteria for ADHD, combined presentation, and indicate a specific weakness in reading; whether J.K. meets IDEA eligibility criteria, under Other Health Impairment, Specific Learning Disability, or neither, is the team's determination. Recommendations for the team's consideration: (1) If found eligible, a structured, evidence-based reading intervention with progress monitoring; screening results have been flat since grade 2, which suggests core instruction alone has not closed the gap. (2) Classroom supports aimed at working memory and processing speed: directions chunked and repeated back, written task lists, reduced copying, and extended time on multi-step assignments. (3) Review of reading fluency and work-completion data at the first annual review, so the team can judge response to whatever it puts in place. I am available to join the 08/19/2026 meeting by phone at the family's invitation, and the full report is available to the team under the same authorization. L. Navarro, PhD, Licensed Psychologist (license PSY 23841) Northgate Psychological Services · (555) 013-4821 · signed 08/05/2026 This sample is fictional and for educational purposes. It does not describe a real patient. ↑ Back to the template and downloads ### Why this sample works The purpose line says what the document is and is not, so the team reads it as private-evaluation input instead of testing it against IEE or district-evaluation criteria it never claimed to meet. Scores appear as indexes with percentiles and a validity statement, no item content, which gives the team usable data without compromising test security. Every finding is translated into a classroom effect with its informant named (multi-step directions, work completion, grade-level text), the educational-need language IDEA teams act on. Recommendations are tied to findings and framed for the team's consideration, and eligibility is expressly left to the team, matching who decides under IDEA. Authorization, scope limits, meeting availability, and a dated signature are all on the page, closing the privacy and process questions before anyone has to ask. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations An IEP input document lives on both sides of a legal boundary, and the rules change as it crosses. In your chart it is clinical correspondence: part of the record the client or parents can request, and protected health information that generally needs a valid, signed authorization before it travels to a school. Once the school receives and maintains it, the same pages ordinarily become an education record under FERPA rather than a HIPAA medical record, with FERPA's own access, amendment, and redisclosure rules; the Departments of Education and Health and Human Services describe this handoff in joint guidance first issued in 2008 and revised in December 2019. Retention splits the same way: HIPAA's six-year rule (45 CFR 164.530(j)) covers Privacy Rule documentation such as the authorization itself, not the chart, so state law and your licensing board govern how long you keep the clinical record, while state and district schedules govern the school's copy, with FERPA adding only that a requested record cannot be destroyed while an access request is pending. Disclose to the minimum the educational purpose needs: functional effects can go to the team where a full treatment history would be excessive. Keep the three layers of authority separate, because template sites blur them. What is LAW binds the school, not your letter: the team must review evaluations and information provided by the parents ( 34 CFR 300.305 ), must consider a privately obtained evaluation that meets agency criteria (34 CFR 300.502(c)), and must evaluate with multiple tools across all areas of suspected disability, never a single measure as the sole criterion ( 34 CFR 300.304 ). Nothing in IDEA prescribes your letter's fields, length, letterhead, validity period, or delivery deadline, and the 60-day initial-evaluation clock runs from parental consent, not from the date your letter arrives. Even the "written input before the meeting" rule (34 CFR 300.321(e)(2)) applies to excused school-team members, not to outside clinicians. What is PAYER POLICY concerns the services around the letter: no benefit pays for an "IEP letter" as such, and if you bill psychotherapy or testing in connection with school questions, the record must support that covered service, including medical necessity, time, and the rendering provider; CMS's testing billing guidance expects the record to show why testing was needed for diagnosis or treatment planning, so testing performed solely for a school-administrative purpose may not meet a medical benefit. Everything else on this page, the field list included, is CONVENTION: prudent practice, enforceable only where a licensing standard, agency IEE criterion, or contract adopts it. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common IEP psychoeducational input errors auditors flag No OIG, CERT, or Medicare contractor audit series targets IEP letters, and that absence is the finding: the letter is usually not the claim. Audit exposure sits in the clinical services billed around it. In HHS OIG's audit of On-Site Psychological Services (2020), 111 of 120 sampled psychotherapy claims failed Medicare requirements, every one of the 111 with a deficient treatment plan, 109 records carried pasted images of clinician signatures, and OIG put the estimated overpayment at $3.3 million. In its Grand Desert Psychiatric Services audit , OIG recommended recovery of an estimated $421,272 over inadequate time documentation, incident-to failures, and misidentified rendering providers, the exact record gaps a bare "spoke with school" entry creates. The contrast case matters too: a 2025 OIG audit of a Florida mental health center found only one incorrect procedure code in 100 sampled claim lines, so the risk is concrete, not presumed. The BastionGPT Clinical Advisory Board sees the same errors most often in IEP psychoeducational input reviews: Sold as an IEE when it is not one. Under IDEA, an independent educational evaluation is an evaluation by a qualified examiner not employed by the district, with agency criteria attached; labeling a treatment summary an IEE invites the district to reject it against standards it was never written to meet. Diagnosis presented as eligibility. The team must use multiple tools and may not rest on any single measure, so a letter that ends at the diagnostic label leaves the educational-need work undone and reads as a conclusion without evidence. Secondhand report written as direct finding. Parent-described classroom behavior recorded as if observed; keep observed fact, collateral information, and professional opinion labeled, or the strongest sentence in the letter becomes its least defensible. School work billed as a clinical service. Calls, meeting attendance, and letter writing relabeled as psychotherapy or family therapy without meeting the payer's definition of the service; time, rendering provider, and proof the service occurred are precisely what the OIG audits above recovered on. Signature failures compound it: CMS's July 2025 signature guidance tells reviewers they may deny claims that lack signed, dated documentation from the responsible clinician. Prescribing the program. Exact minutes, placement, or a one-to-one aide directed from the clinic chair exceeds the treating clinician's data and role; recommendations framed for team consideration, tied to findings, are the ones that survive the meeting. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on IEP psychoeducational input letters. Turn a full private evaluation or a treatment record into a focused input addendum for the IEP team: index scores, functional effects, and recommendations, with test item content left out. Rewrite findings in the plain, classroom-level language school teams act on, and trim disclosure to what the authorization and the educational purpose actually cover. Check the document before you sign: purpose stated, authorization documented, observed fact separated from collateral report, each recommendation tied to a finding, scope limits and signature in place. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Does the IEP team have to follow my recommendations? No. IDEA requires the team to review evaluations and information the parents provide and to consider a privately obtained evaluation that meets agency criteria, but consideration is the whole obligation: the team decides eligibility, services, and placement, and it may not base them on any single measure. Your influence comes from translation: findings tied to classroom function give the team something it can act on, and the parent or the district can invite you to the meeting as a person with knowledge or special expertise. Is my report an independent educational evaluation (IEE)? Not automatically. Under 34 CFR 300.502 an IEE is an evaluation conducted by a qualified examiner who is not employed by the district, and agency criteria, public-funding rules, and hearing rights attach to that category. A treatment summary or accommodation letter is not an IEE because a psychologist signed it; it is information the team reviews. If you did conduct a full evaluation, the psychoeducational evaluation report page covers that genre. How long should IEP input from a psychologist be? No law sets a length, field list, letterhead rule, validity period, or delivery deadline for the clinician's document. A focused input letter or addendum typically runs 300 to 900 words; a full private evaluation runs far longer. Length matters less than three things: a purpose line, functional effects with informants named, and recommendations tied to findings. IDEA's 60-day initial-evaluation clock runs from parental consent, not from the date your letter arrives. Do I need a parent's authorization before sending anything to the school? Generally yes. A HIPAA-covered practice needs a valid, signed authorization before disclosing protected health information to a school unless a specific exception applies. Once the school maintains your report, it ordinarily becomes an education record under FERPA rather than a HIPAA record, with different access, amendment, and redisclosure rules; the Departments of Education and HHS describe the handoff in joint guidance revised in December 2019. Disclose the minimum the educational purpose needs: functional effects can go to the team where a full treatment history would be excessive. Can I bill insurance for writing an IEP letter or attending the meeting? There is no universal stand-alone benefit for either. The work may be bundled into a covered service, payable under a specific contract, or self-pay by advance agreement. What you cannot do is relabel it: if you bill psychotherapy or testing around school questions, the record must support that covered service, including medical necessity, time, and the rendering provider, and OIG psychotherapy audits recover millions on exactly those elements. CMS testing guidance adds that testing performed solely for a school-administrative purpose may not meet a medical benefit. Should I disclose the diagnosis, or are functional effects enough? No IDEA rule requires a treating clinician to disclose a diagnosis, and disclosure follows the authorization and the minimum necessary for the purpose. Functional effects are the load-bearing content either way: IDEA eligibility turns on educational need shown through multiple sources, so a diagnosis alone neither establishes nor substitutes for it. Many defensible letters describe concentration, regulation, attendance, and access to instruction without a diagnostic label. What is the difference between IEP input and a Section 504 statement? They feed different laws. An IEP delivers specially designed instruction under IDEA; a 504 plan delivers equal-access accommodations under Section 504 of the Rehabilitation Act, with different eligibility standards and procedures. At least one currently ranking template page describes a 504 plan as a type of IEP, which is incorrect. If the question in front of the school is access rather than specialized instruction, write a Section 504 clinician statement instead. How long is my report valid, and how long must everyone keep it? No reviewed IDEA provision assigns an expiry date to an outside clinician's report; a school may reasonably weigh recency and changed functioning, and a testing program or agency may impose its own recency rule as policy. Retention splits by holder. Your copy follows state law and licensing rules, not HIPAA's six-year rule, which covers Privacy Rule documentation such as authorizations rather than the chart. The school's copy follows state and district schedules; FERPA adds only that a record cannot be destroyed while an access request is pending. Can BastionGPT draft IEP input from my evaluation or session notes? Yes. Give it your scores, session bullets, or a full report, and it drafts the addendum with the purpose line up front, findings separated from collateral report, and recommendations tied to findings, then trims disclosure to what the authorization covers. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Psychoeducational Evaluation Report Section 504 Clinician Statement Child & Adolescent Developmental Assessment School & Teacher Collateral Note WISC-V Report Write-Up CELF-5 Report Write-Up GFTA-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/medical-source-statement All note templates # Disability Evaluation & Medical Source Statement: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A medical source statement is a treating clinician's written opinion describing what a patient can still do despite mental impairment, rated across four work-related domains: understanding and memory, concentration and pace, social interaction, and adaptation. Clinicians complete one when a disability claim, most often Social Security, needs functional evidence from the treating source. Most combine a structured rating form with narrative comments, 2 to 6 pages. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating psychologists, psychiatrists, therapists, and other licensed clinicians Audience SSA disability examiners and judges, private disability insurers, claimant attorneys Typical length 600 to 1,200 words (form plus narrative) · 30 to 60 minutes by hand (clinical team estimate) Format family Structured functional-opinion form or narrative report (compare: capacity evaluation, FMLA certification, psychological evaluation report) When it's used When a disability claim needs treating-source functional evidence, most often SSDI/SSI and private long-term disability Standards context Opinion content is defined in SSA regulation; no law mandates a specific form, and HA-1152 is an appeals-stage tool On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a medical source statement? A medical source statement (MSS) is a treating clinician's written opinion about what a patient can still do despite mental impairment, and about impairment-related limits in the work-related mental activities the regulation names: understanding, remembering, and carrying out instructions; concentration, persistence, and pace; interacting with supervisors, coworkers, and the public; and adapting to workplace conditions and pressures (20 CFR 404.1513(a)(2)). The best-known version is SSA Form HA-1152, Medical Source Statement of Ability to Do Work-Related Activities (Mental), introduced by SSA's hearings office in 2002 for the appeals stage. You will also hear mental RFC, MRFC, mental impairment questionnaire, treating-source statement, and, from private disability insurers, attending physician statement. One disambiguation saves most of the confusion: three different documents share the "mental RFC" nickname. A treating-source MSS is evidence you supply. Form SSA-4734-F4-SUP is an internal assessment completed by agency consultants, a "prior administrative medical finding," not a clinician's form. And the residual functional capacity determination itself is the adjudicator's finding, which no form decides. Since the rules changed for claims filed on or after March 27, 2017, a treating clinician's opinion gets no controlling weight; SSA weighs every opinion for supportability and consistency (20 CFR 404.1520c). An MSS is also not a full psychological evaluation report : the evaluation establishes diagnoses and test findings, while the MSS translates them into sustained work functioning, and unlike a capacity evaluation , which answers one specific legal decision, it addresses functioning across a workweek and a year. ## Who uses medical source statements and when Treating psychologists, psychiatrists, therapists, and counselors complete these statements when a patient applies for SSDI or SSI, usually at the request of the claimant's representative, and the same functional-opinion genre appears in private long-term disability claims as the attending physician statement and in state and veterans programs under other names. Pick the tool by the legal question. Job-protected leave runs through an FMLA certification ; an employer's present-capacity question about a specific job is a fitness-for-duty evaluation ; a benefits adjudicator asking about sustained work-related functioning over 12 months or more is asking for a medical source statement. Complete one when the chart already holds the evidence: dated scores, observed behavior, and a treatment relationship long enough to support longitudinal statements. ## Medical source statement structure: what goes in each section Claimant, clinician, and treatment relationship. Identify the claimant and claim, then establish yourself as the source: name, degree, license, specialty, and how long and how often you have treated. Under 20 CFR 404.1502 a licensed psychologist is an acceptable medical source, and other licensed clinicians acting within scope are medical sources whose opinions SSA evaluates. Pitfall: an unsigned or undated statement. SSA rejects proxy signatures, rubber stamps, and "dictated but not read" in the exams it purchases, and a reviewer discounts a treating-source statement with the same defects. Diagnoses, symptoms, treatment, and response. State each diagnosis with its ICD-10 code, the clinical findings behind it, medication and therapy history, response, side effects, and prognosis. This is the "other medical evidence" that makes the ratings believable. Pitfall: letting the diagnosis argue the case. The regulatory question is what the claimant can still do; a diagnosis or a high score, standing alone, answers nothing about sustained work activity. Understanding, remembering, and carrying out instructions. The first of the four work-related mental domains in 20 CFR 404.1513(a)(2) : rate simple and complex instructions separately, address judgment, and attach the observed evidence for each rating. Pitfall: a checked box with a blank comments field. Under 20 CFR 404.1520c(c)(1), an opinion is persuasive in proportion to its supporting evidence and explanation; an unexplained "marked" carries little weight. Concentration, persistence, and pace. Describe initiating, sustaining, sequencing, and completing tasks over a normal schedule, tolerance for ordinary pace, and recovery after interruption. Pitfall: treating off-task percentages and monthly-absence counts as required fields. They are conventions of attorney and vendor questionnaires, not elements of the regulation or the current HA-1152; give a number only when your records support one. Social interaction. Address supervisors, coworkers, and public contact: accepting instructions, responding to criticism, managing conflict, and behavioral regulation, with the evidence for each. Pitfall: generalizing from the office. Cooperative behavior in a structured, supportive hour does not by itself predict a workweek; say whether the in-session presentation is representative and why. Adaptation and workplace pressures. Cover response to change, ordinary work stress, decision-making, safety awareness, and independent planning. Pitfall: opining "disabled" or "unable to work." Disability is the adjudicator's conclusion; SSA instructs its own examiners to describe work abilities instead, and an ultimate-issue opinion gives the reviewer nothing to adopt. Duration, fluctuation, and prognosis. Date the onset you can support, the duration already observed, the episodic pattern (frequency, length, recovery time, good-day and bad-day variation), and expected persistence, which matters because the statutory test looks at 12 months or longer. Pitfall: unsupported retroactive onset dates; tie every date to a record. Supporting explanation and records reviewed. Link each limitation to findings, dated scores, observed behavior, treatment response, and collateral, and identify the records the opinion rests on. Pitfall: leaving a conflict with the longitudinal record unexplained. Consistency is one of the two most important regulatory factors, and an unaddressed contradiction is the standard reason a treating opinion is found unpersuasive. ### Blank template (copy and adapt) MENTAL MEDICAL SOURCE STATEMENT (TREATING SOURCE) Claimant: [initials] Claim/file #: _________ Date: _________ Clinician: [name, degree, license #, specialty] Treatment relationship: first visit ____ last visit ____ Frequency and total visits: ____________________________________ Diagnoses (ICD-10): ____________________________________________ Clinical findings, treatment, and response: [medications, psychotherapy, dated measure scores, side effects] WORK-RELATED MENTAL ACTIVITIES (rate and give the evidence) 1) Understand, remember, and carry out instructions: [simple vs complex; judgment; evidence] 2) Concentration, persistence, and pace: [sustaining tasks over a normal schedule; evidence] 3) Social interaction (supervisors, coworkers, public): [evidence] 4) Adaptation and workplace pressures (change, stress, decisions): [evidence] Duration and course: [onset support, observed duration, episodic pattern, expected persistence, prognosis] What the claimant can still do: ________________________________ Basis: [records reviewed, testing, collateral] Dates: _________ Note: the agency decides disability; describe function, not the legal conclusion. Signature/credentials: __________________ Date signed: ________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample medical source statement Scenario: a treating psychologist completes a mental medical source statement for an SSDI claim at the reconsideration stage, at the request of the claimant's representative. Diagnoses are major depressive disorder and generalized anxiety disorder; treatment has run 14 months. All details are fictional. Mental medical source statement · Program: SSDI, reconsideration stage · Prepared: 08/14/2026 at the request of the claimant's representative Claimant: D.K., 41. Source: A. Whitmore, PhD, Licensed Psychologist (license on file). Treatment relationship: weekly to biweekly individual psychotherapy since 06/2025; 43 sessions through 08/2026. Records reviewed: my complete treatment record 06/2025 to 08/2026, PHQ-9 and GAD-7 logs, the primary care medication list, and a collateral interview with the claimant's spouse on 07/30/2026. Diagnoses, treatment, and response: Major depressive disorder, recurrent, moderate (F33.1) and generalized anxiety disorder (F41.1), diagnosed at intake 06/2025. Treatment: weekly cognitive behavioral therapy with behavioral activation; sertraline prescribed by primary care since 09/2025 with partial response. PHQ-9: 21 (09/2025), 18 (01/2026), 16 (07/2026). GAD-7: 15 (07/2026). Attendance is consistent for afternoon appointments; three morning appointments were missed and rescheduled, documented as symptom-related. 1. Understanding, remembering, and carrying out instructions: D.K. understands and carries out simple one- and two-step instructions without support. Multi-step instructions require written prompts and re-explanation: in 9 of the last 12 sessions, three-step homework rationales had to be repeated (chart-documented), and D.K. maintains a written checklist for routine household sequences. 2. Concentration, persistence, and pace: Observed sustained task focus in session runs 15 to 25 minutes before redirection is needed. D.K. reports starting and abandoning household tasks most days, corroborated by the spouse's collateral report. Pace slows markedly under time pressure; a timed worksheet exercise on 06/18/2026 was abandoned at the halfway point with visible distress. 3. Social interaction: Cooperative and appropriate in the treatment setting. Tolerates brief, structured interactions such as store checkouts. After two extended family gatherings in 07/2026, D.K. experienced documented panic symptoms and two to three days of avoidance. Responds to perceived criticism with withdrawal lasting days, reported consistently across the treatment record. 4. Adaptation and workplace pressures: Routine changes, such as a rescheduled appointment or an unfamiliar building, produce documented anxiety spikes and avoidance. Decision-making under pressure deteriorates: D.K. defers routine decisions to the spouse during symptom exacerbations. No unsafe behavior observed; safety awareness is intact. Duration, fluctuation, and prognosis: Symptoms at or near this severity are documented since intake, 14 months. Exacerbations lasting 2 to 4 days occur two to three times monthly, during which D.K. does not leave home and household functioning depends on the spouse. With continued treatment I expect meaningful but incomplete improvement over the next 12 months; the limitations described have already lasted more than 12 months and are expected to persist. What the claimant can still do: Manage self-care and routine household finances with reminders; sustain simple, familiar tasks in low-demand settings for short periods; interact appropriately in brief, structured contacts. Basis and limits: The findings above are drawn from 43 documented sessions, dated measure scores, and collateral, each cited by date in the record. The treatment setting is structured and supportive and may overstate workplace functioning; the longitudinal record is the better guide. Whether D.K. meets the statutory definition of disability is a determination reserved to the agency. Signature: A. Whitmore, PhD, Licensed Psychologist · Date signed: 08/14/2026 This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or claim. ↑ Back to the template and downloads ### Why this sample works Every rating carries its evidence. Each domain statement cites observed behavior, dated scores, or session counts, which is what the supportability factor in 20 CFR 404.1520c weighs most. The in-session paradox is addressed head on. The statement explains why cooperative presentation in a structured hour may not predict a workweek, closing the consistency gap reviewers use to discount treating opinions. Function runs in both directions. Retained abilities appear alongside limitations, answering the regulatory question, what the claimant can still do, instead of arguing a conclusion. Duration is dated arithmetic. Onset documentation, 14 months observed, a quantified episodic pattern, and expected persistence past 12 months map onto the statutory duration requirement. The legal conclusion is left to the agency, explicitly. SSA's own examiner guidance treats that as the more useful contribution, and the signature and date close the authentication gaps reviewers flag. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations A medical source statement joins the claim file and the patient's designated record set: the patient can request it, opposing reviewers will read it, and it can resurface in later claims, hearings, or litigation, so write it as a document you can defend line by line from the chart. Two ownership points surprise clinicians. First, an opinion form completed at an attorney's request is not shielded as attorney work product: 20 CFR 404.1513 excludes the lawyer's legal analysis from the evidence but expressly keeps medical opinions and factual medical evidence in it. Second, retention follows the chart, and HIPAA sets no chart-retention period : state law and payer contracts govern, with APA record-keeping guidance suggesting at least seven years for adult records as professional convention. Keep the statement, the request that prompted it, and the fee agreement together in the record. Sort the demands on this document into three piles before drafting. The four work-related mental domains and the "what the claimant can still do" frame are law, the definition in 20 CFR 404.1513(a)(2), and so is the weighing rule for claims filed on or after March 27, 2017, which makes supportability and consistency decisive. HA-1152 is agency practice for the appeals stage, policy rather than statute, and no rule makes it the exclusive format. Off-task percentages, monthly absence counts, and five-point severity scales are conventions of attorney and vendor questionnaires: useful when your records support them, safely left blank when they do not. The content is the requirement; the form is a convenience. A psychological evaluation report supplies the diagnostic and testing foundation that the statement's ratings then translate into sustained work functioning, and a capacity evaluation answers a different question entirely, one specific legal decision rather than work capacity over time. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ## Common medical source statement errors auditors flag The closest federal audit trail concerns the consultative examinations SSA purchases, the adjudicated neighbor of the treating-source statement, and its findings map directly onto the same failure modes. In the SSA Office of the Inspector General's July 2024 review of consultative examination oversight , 11 of 52 state Disability Determination Services oversight reports (21.2%) were missing required elements, and DDSs removed 47 CE providers from approved panels in fiscal year 2022, 44 of them for cause, with report-quality problems and late reports among the leading reasons. SSA's own report-review procedure (POMS DI 22510.020) checks completeness, internal consistency, documentation of evidence reviewed, and the examiner's actual signature. The BastionGPT Clinical Advisory Board sees the same errors most often in medical source statement reviews: Checked severity with no explanation. A "marked" or "extreme" rating with a blank comments field fails the supportability factor in 20 CFR 404.1520c(c)(1); the rating that persuades carries the finding, score, or observed behavior behind it. A conflict with the longitudinal record left unaddressed. Consistency is the other most-important factor: an opinion that contradicts months of "stable, engaged" progress notes without explaining why in-session presentation does not generalize to a workweek is easy to find unpersuasive. Symptoms and diagnosis substituted for function. The regulatory definition asks what the claimant can still do; a diagnosis list or a severity score alone does not answer it, and Australia's Table 5 makes the same point by requiring corroborated, sustained-activity evidence. The ultimate-issue opinion. "Disabled" and "unable to work" are adjudicative conclusions. SSA instructs its own examiners to describe work-related abilities instead, and a statement that stops at the legal conclusion gives the reviewer nothing to adopt. Signature, dating, and completeness failures. Missing pages, unanswered requested items, no list of records relied on, or a proxy signature: each is a documented return reason in SSA's review procedure, and each undermines a treating-source statement the same way. ### Medical source statements in the US, Canada, and Australia Aspect United States Canada Australia Status Opinion content is defined in law (20 CFR 404.1513(a)(2)); no statute mandates a specific form, and HA-1152 is an appeals-stage collection No national mental-RFC form; CPP disability report content is set by law (CPP Regulations s. 68), with the principal report completed by a physician or nurse practitioner No national mental-RFC form; DSP impairment is assessed under the legislated 2023 Impairment Tables, and the SA478 evidence checklist is expressly optional Terminology Medical source statement, mental RFC, mental impairment questionnaire; attending physician statement (private insurers) CPP disability medical report (ISP-2519), Disability Tax Credit certification, attending physician statement (insurers) DSP medical evidence, SA478 checklist, NDIS functional capacity report What changes The weighing rule turns on filing date: claims filed on or after March 27, 2017 get no treating-source controlling weight, and supportability plus consistency decide; SSA and private-insurer workflows differ Physicians and NPs complete the principal CPP report, and a narrative may substitute for the form; psychologists provide supporting functional evidence and can certify DTC mental functions Table 5 requires a medical practitioner's diagnosis, plus registered-psychologist evidence when the diagnosis is not from a psychiatrist; corroboration is statutory, and self-report alone is insufficient Retention No HIPAA chart-retention period; state law and payer contracts govern, with APA guidance suggesting 7+ years for adults as convention Provincial; Ontario's psychology standard expects 10 years after the last professional contact for adult records (effective July 2024) State and territory law; NSW requires 7 years after the last adult service and until age 25 for records of minors The four functional domains travel well as clinical content; the legal machinery around them does not. A US-style mental RFC questionnaire answers no Canadian or Australian statutory test, so match the document to the program before adapting it. For Australian clinicians writing functional evidence for support planning rather than income benefits, the NDIS functional capacity assessment page covers that distinct document family. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on medical source statements. Draft a complete mental medical source statement from your chart summary, a dictation, or a few bullets, organized around the four work-related domains with the supporting evidence attached to each rating. Pull the functional evidence out of months of notes: dated measure scores, observed behavior, attendance patterns, and treatment response, ready to cite under each domain. Check a finished statement before you sign: every rating supported, duration and expected persistence stated, no ultimate-issue conclusions, signature and date in place. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How long should a mental medical source statement be? No regulation sets a length. Most treating-source statements run 2 to 6 pages: a structured form rating the four work-related mental domains plus narrative comments, or a narrative letter of roughly 600 to 1,200 words. SSA's paperwork estimate for its own HA-1152 form assumes 15 minutes per response, but that is a Paperwork Reduction Act assumption, not a measured completion time. Plan 30 to 60 minutes to do the supporting explanations well (clinical team estimate); the explanation, not the length, is what carries weight. Do I have to use SSA Form HA-1152? No. HA-1152, Medical Source Statement of Ability to Do Work-Related Activities (Mental), entered use in 2002 as a tool for administrative law judges and the Appeals Council, and no regulation makes it the exclusive format at any stage. SSA accepts a narrative letter, an attorney's questionnaire, or your own template, as long as the content addresses what the claimant can still do and the work-related mental limitations defined in 20 CFR 404.1513(a)(2) . The familiar shape persists because reviewers can read it quickly. Can an LCSW, LPC, or LMFT sign a medical source statement? Often, yes. For current claims, any licensed clinician acting within scope of practice is a "medical source" whose opinion SSA will evaluate. Licensed psychologists, like physicians, are also "acceptable medical sources," the narrower category that matters for establishing a medically determinable impairment. So a therapist's functional opinion is usable evidence, but the file still needs an acceptable medical source behind the diagnosis. Read the request's wording, too: some private insurers and programs direct their forms to specific professions. Does a treating clinician's opinion carry extra weight with SSA? Not for claims filed on or after March 27, 2017. Under 20 CFR 404.1520c , no medical opinion receives controlling or predetermined weight; supportability and consistency are the two most important factors. Claims filed before that date keep the older treating-source rule. The practical consequence: the explanation attached to each rating, and its fit with the longitudinal record, now do the work the treatment relationship used to do. Should I write that my client is disabled or unable to work? No. Disability is a conclusion the adjudicator reaches; SSA instructs its own consultative examiners to describe work-related abilities and not to opine on the statutory conclusion, and the same logic protects a treating-source statement. The same discipline applies to fields you cannot support: off-task percentages and monthly-absence estimates appear on many attorney questionnaires, but no regulation requires them and they are not on the current HA-1152. Complete them when your records support a number; leave them blank when they do not. How is this different from an FMLA certification or a fitness-for-duty evaluation? The legal question differs. An FMLA certification supports job-protected leave: a serious health condition, incapacity, and a leave schedule, usually over weeks or months. A fitness-for-duty evaluation answers an employer's present-capacity question about one specific job. A medical source statement answers a benefits question: sustained work-related functioning despite impairment, usually across 12 months or longer. Reusing one document for another purpose imports the wrong legal test, so adapt rather than forward. Can I charge the patient for completing the form? There is no universal rule. In the US, state law, payer contracts, and your financial agreement govern; billers' forums show wide variation and no uniform coverage for standalone form completion, so set the fee policy before the request arrives. Service Canada pays up to CAD 85 toward the CPP disability medical report, with any excess potentially the patient's responsibility under the program's published process. Services Australia treats evidence gathered during a clinically necessary consultation as billable under the ordinary consultation framework, which does not make a standalone administrative report reimbursable. How does disability evidence work in Canada and Australia? Neither country uses the US mental-RFC model. Canada's CPP disability benefit requires a medical report whose content is set by regulation (CPP Regulations s. 68), completed under current instructions by a physician or nurse practitioner; psychologists supply supporting functional evidence and can certify the mental-functions route of the Disability Tax Credit. Australia's Disability Support Pension is assessed under legislated Impairment Tables: Table 5 requires diagnosis by a medical practitioner, with a registered psychologist's evidence required when the diagnosis is not from a psychiatrist, and the SA478 evidence checklist is expressly optional. For NDIS support planning rather than income benefits, see the NDIS functional capacity assessment page. Can BastionGPT draft a medical source statement? Yes. Give it your chart summary, measure log, and the request or form, and it drafts the statement around the four work-related domains with the supporting evidence attached to each rating, flags severity language your records do not support, and keeps the legal conclusion out. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Capacity & Competency Evaluation Fitness-for-Duty Evaluation FMLA & Leave Certification Support Note Psychological Evaluation Report NDIS Functional Capacity Assessment Report GAF Score Documentation WHODAS 2.0 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/fmla-certification All note templates # FMLA & Leave Certification Support Note: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 An FMLA certification support note is the documentation a mental health clinician prepares when an employer requests medical certification of a serious health condition: the completed WH-380-E entries plus the chart note that backs them. Clinicians treating employees of covered US employers write it within the employee's 15-day return window. Form entries plus the supporting note typically run 150 to 400 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinical psychologists, clinical social workers, psychiatrists; counselors and MFTs when accepted under 29 CFR 825.125 Audience The employer's HR or leave administrator, often a third-party leave vendor; never the Department of Labor or a payer Typical length 150 to 400 words · 20 to 45 minutes by hand (clinical team estimate) Format family Employment-leave certification (compare: return-to-work letter, medical source statement, fitness-for-duty evaluation) When it's used When a covered employer requests certification for FMLA leave; the employee has 15 calendar days to return it Standards context Requiring certification is law (29 CFR 825.305); the WH-380-E form itself is optional, and 29 CFR 825.306 caps what an employer may ask On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is an FMLA certification support note? A serious-health-condition certification is the statement a health care provider supplies so a covered employer can decide whether an eligible employee's absence is protected by the Family and Medical Leave Act, the federal leave law enacted February 5, 1993. The Department of Labor's Wage and Hour Division publishes optional forms for it: WH-380-E for the employee's own condition and WH-380-F for care of a spouse, parent, or child. Clinicians and HR teams also call it FMLA paperwork, FMLA medical certification, provider certification, or simply the 380-E. The support note is the chart-side companion: the documentation you keep that backs every entry you make on the form. Two boundaries carry the weight. First, the certification requirement is law while the form is not: 29 U.S.C. 2613 and 29 CFR 825.305 let an employer require certification, 29 CFR 825.306 fixes what it may ask, and DOL describes its own forms as optional, so a complete and sufficient provider letter can do the same work. Second, the completed certification goes to the employer or its leave administrator, never to the Department of Labor and never to a health insurer: it is employment-leave evidence, not a claim, and no payer medical-necessity rule decides whether it is sufficient. Keep the neighbors distinct as well: a return-to-work letter speaks to reinstatement and accommodation, and a disability medical source statement speaks to sustained work capacity for an adjudicator. ## Who uses FMLA certification support notes and when Any clinician treating working adults meets this request: outpatient therapists, psychologists, psychiatric prescribers, and group practices where a third-party leave vendor sends the employer's packet. It covers continuous leave for an episode of care and intermittent leave for recurrent conditions, and it can even be completed from Canada or Australia for a US-covered employee, because 29 CFR 825.307(f) requires the employer to accept a sufficient certification from a provider authorized to practice where treatment occurs, subject to an English translation. Reach for the neighbors when the question changes: safe return and modified duties belong in a fitness-for-duty evaluation or a return-to-work letter , and sustained incapacity for a benefits adjudicator belongs in a medical source statement . ## FMLA certification support note structure: what goes in each section Provider identification and qualifying status. Open with who you are: name, address, phone and fax, credential, and type of practice or specialty, the identifying content 29 CFR 825.306 lists first. Then settle the question the form cannot ask: whether you qualify under 29 CFR 825.125. Clinical psychologists and clinical social workers are expressly named; licensed professional counselors and marriage and family therapists are not named categories and usually qualify, when they do, as a provider the employer or its group health plan accepts. Pitfall: signing as "therapist." The title does not establish qualifying status, and a certification from a non-qualifying signer can fail even when every clinical entry is right. Condition onset and probable duration. Give the approximate date the condition began and its probable duration, and keep two clocks apart: how long the underlying condition has existed or will persist, and how long the present incapacity or leave need should last. For a recurrent condition, the condition is measured in years while an episode is measured in days. Pitfall: one duration entry doing both jobs. In Hansler v. Lehigh Valley Hospital Network (3d Cir. 2015), a certification listing only a one-month duration muddied the chronic-condition theory the leave request depended on. Appropriate medical facts. 29 CFR 825.306 asks for "appropriate medical facts" supporting the need for leave: functional consequences, treatment needs, hospitalization when relevant. A diagnosis is one possible fact, not a universal federal requirement, and the Department of Labor cautions providers to limit answers to the condition supporting the leave. Pitfall: pasting chart content. The regulation bounds what an employer may ask, and nothing in it requires progress notes, psychotherapy notes, or a diagnostic history. Essential job functions the employee cannot perform. Tie the opinion to functions the employer identified or the job description describes: sustained concentration on safety-critical tasks, scheduled attendance, customer-facing composure. Pitfall: a bare "unable to work." When only some duties are affected, an unanchored global statement is less useful to the reader and easier to challenge. Leave schedule and estimates. For continuous leave, give best-estimate beginning and ending dates. For intermittent or reduced-schedule leave, state why it is medically necessary and estimate treatment dates, episode frequency, and episode duration; the regulations permit estimates and do not demand certainty. Pitfall: "as needed." An unbounded answer where a clinical estimate was possible can be treated as vague or nonresponsive, and it sends the employee into the deficiency-cure cycle. Signature, routing, and the disclosure log. Sign with credentials and date it, then return the form to the client or the employer's designated process, never to the Department of Labor, and log what left the office: the completed form, and nothing else without a separate authorization . Pitfall: skipping the log. When HR calls to clarify, the record of what was released and who may receive contact is the difference between a routine call and an improvised disclosure. ### Blank template (copy and adapt) FMLA & LEAVE CERTIFICATION SUPPORT NOTE (WH-380-E COMPANION) Worksheet mirroring the WH-380-E elements. The official DOL form, or the employer's own form, is the document you return; this note stays in your chart. PROVIDER AND QUALIFYING STATUS Clinician / credential / license #: ____________________________ Practice type or specialty: ____________________________________ Address and phone / fax: _______________________________________ Qualifying route (29 CFR 825.125): [ ] clinical psychologist [ ] clinical social worker [ ] other provider accepted by the employer or group health plan: _________________________________ CONDITION AND DURATION Approximate date condition began: ______________________________ Probable duration of the condition: ____________________________ Duration of the PRESENT incapacity or leave need: ______________ MEDICAL FACTS SUPPORTING LEAVE (diagnosis optional) Facts limited to what supports this leave: functional impact, treatment needs, episode pattern: ______________________________ ________________________________________________________________ ESSENTIAL JOB FUNCTIONS AFFECTED Function(s) the employee cannot perform, from the employer's list or job description: _______________________________________ LEAVE SCHEDULE (best estimates, never "as needed") [ ] Continuous: from ___________ to ___________ [ ] Intermittent / reduced schedule, medically necessary Episode frequency: _____ per [ ] week [ ] month Typical episode duration: _____ [ ] hours [ ] days Planned treatment schedule: ____________________________________ ROUTING AND DISCLOSURE LOG Returned to (never the DOL): [ ] client [ ] employer process Released: [ ] form only [ ] other (list): _____________________ Authorization on file for clarification contact: [ ] yes Clinician signature / credentials: _____________ Date: ________ Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample FMLA certification support note Scenario: a clinical social worker completes the employer's WH-380-E for intermittent leave during recurrent depressive episodes, and charts the support note behind it. All details are fictional. FMLA certification support note. Cedar Row Counseling LLC · Client: D.K. · Note date: 08/05/2026 · Clinician: A. Reyes, LCSW Request and qualifying status (08/03/2026): Client brought the employer's certification packet, a WH-380-E issued by HR on 08/01/2026 with a return date of 08/16/2026. Signer: A. Reyes, LCSW, state license on file, outpatient psychotherapy practice; clinical social workers are an expressly recognized provider category for FMLA certification. Client signed an authorization on 08/03/2026 permitting clarification contact with the employer's leave administrator only. Condition and duration: In weekly treatment at this practice since 02/2025 for recurrent depressive episodes, first documented by history in 2019. The underlying condition is chronic and expected to persist beyond six months; this certification covers 08/03/2026 through 02/03/2027. Each episode, by contrast, is short: the two clocks are recorded separately on the form. Medical facts and functional impact: During an episode the client shows marked early-morning impairment, loss of sustained concentration, and withdrawal from task demands; in the 07/2026 episode the client could not complete assigned reconciliation work or attend two consecutive scheduled shifts. Affected essential functions, per the employer's job description: sustained concentration on financial reconciliation and reliable scheduled attendance. Form entries were limited to these facts and the treatment plan; a diagnosis was not required to make the certification sufficient and was not entered. Leave schedule estimates: Intermittent leave is medically necessary. Best estimate from the pattern documented since 02/2025: one to two episodes per month, each lasting two to three days. Planned treatment: weekly 50-minute psychotherapy, Wednesdays, through at least 02/2027, plus prescriber medication reviews approximately every eight weeks. Between episodes the client performs all essential functions. Routing and disclosure log: Completed WH-380-E returned to the client on 08/05/2026 for submission to HR; nothing sent to the Department of Labor. Released: the completed form only; no progress notes, no psychotherapy notes, no chart export. A copy of the completed form is retained in the record with this note. This sample is fictional and for educational purposes. It does not describe a real client, clinician, or employer. ↑ Back to the template and downloads ### Why this sample works The estimates are bounded. One to two episodes per month, two to three days each: a responsive best estimate under 29 CFR 825.306, where an unbounded "as needed" invites a deficiency notice. The two durations are kept apart. The chronic condition runs in years, the episode in days, the distinction the Hansler case shows a certification can live or die on. Functional impact is anchored to named job functions. Reconciliation work and scheduled attendance, from the employer's own description, not a bare "unable to work." Disclosure stays at certification scope. The form went back; the chart, the diagnosis, and the psychotherapy notes did not, matching what the regulation lets an employer ask. Routing and authorization are logged. Who received the form, what was released, and who may call for clarification are all on record before HR ever phones. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The regulation that authorizes the request also caps it. Once a certification is complete and sufficient, employer follow-up narrows to two actions, authentication and clarification, and the messenger is restricted: under 29 CFR 825.307 a human-resources professional, a leave administrator, a management official, or a health care provider may contact you, and the employee's direct supervisor may never make that contact. HIPAA still governs the content of any answer, so keep a signed release of information on file before discussing anything of substance, and remember that no FMLA provision entitles the employer to progress notes, psychotherapy notes, or the chart. If the employer doubts the opinion, the remedy is structural rather than interrogative: a second opinion at the employer's expense, then a jointly approved third opinion that is final and binding, with copies available to the employee on request. The deadlines belong to the employment process, not to you. The employer generally must request certification within five business days of the leave notice, the employee has 15 calendar days to return it, and an incomplete or insufficient certification earns a written deficiency notice and ordinarily seven calendar days to cure; federal law sets no clinician turnaround time and no clinician fee schedule, so completion charges follow state law, professional standards, and practice policy, and payers commonly treat form completion as noncovered. Recertification ordinarily may not be sought more often than every 30 days in connection with an absence. Retention splits the same way: the employer must keep FMLA records, certifications included, for three years under 29 CFR 825.500, while no FMLA rule sets a clinician retention period, so your licensing board and health-record rules govern the chart. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. ↑ Back to the template and downloads ## Common FMLA certification errors employers and leave administrators flag The scale of the process comes from the Department of Labor's 2018 FMLA surveys, released in 2020 from 4,470 employee and 2,206 worksite interviews: 56% of US employees were FMLA-eligible, 15% had taken leave for a potentially qualifying reason in the prior year, and 32% of qualifying-reason leaves were intermittent, up from 24% in 2012, with intermittent leave-takers absent on an average of seven separate occasions. Intermittent schedules are exactly where estimate quality decides sufficiency, and the leading appellate case, Hansler v. Lehigh Valley Hospital Network (3d Cir. 2015), turned on an employer denying leave over an insufficient certification without the required chance to cure it. The BastionGPT Clinical Advisory Board sees the same errors most often in FMLA certification reviews: "As needed" where an estimate was possible. Intermittent certification calls for a best estimate of episode frequency and duration; an unbounded answer can be treated as vague or nonresponsive under 29 CFR 825.305 and triggers the cure cycle instead of an approval. A duration entry that defeats the leave theory. Writing the episode length where the condition duration belongs, the trap Hansler documents: distinguish the flare that resolves in days from the underlying condition expected to persist. A signer who does not satisfy 29 CFR 825.125. Counselor and MFT licenses are not named categories in the federal provider definition; when the route is employer or group-plan acceptance, confirm it before signing rather than after a challenge. Over-disclosure. Sending progress notes, psychotherapy notes, or a diagnostic history when the regulation bounds the employer to certification content; the fix for an insufficiency notice is answering the identified deficiency, not opening the chart. Misrouting the form. Certifications go back to the patient or the employer's designated process; the Department of Labor neither receives nor approves them, and a misdirected form burns the employee's 15-day window. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on FMLA and leave certification support notes. Draft the support note from chart bullets or dictation: onset, medical facts, functional impact, and bounded frequency and duration estimates, ready to transfer onto the WH-380-E or the employer's form. Check a completed draft against the 29 CFR 825.306 elements before you sign: a missing onset date, a missing duration, an unanchored "unable to work," or an "as needed" schedule. Flag over-disclosure before the form leaves the office: diagnosis details, treatment specifics, and chart content beyond what the certification needs. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Can a therapist sign FMLA paperwork, or does it have to be a psychiatrist? A psychiatrist is not required. The federal provider definition at 29 CFR 825.125 expressly recognizes clinical psychologists and clinical social workers authorized to practice under state law. Licensed professional counselors, mental health counselors, and marriage and family therapists are not named categories: they may still qualify through another route, most often as a provider from whom the employer or its group health plan accepts certification, and they must be authorized to diagnose and treat under applicable law. Confirm the route before signing, because a certification from a non-qualifying signer can fail on that ground alone. Do I have to put the client's diagnosis on the form? Not invariably. The federal standard at 29 CFR 825.306 is "appropriate medical facts" sufficient to support the leave: incapacity, functional impact, treatment needs, duration, and the expected absence pattern can establish sufficiency without a diagnosis. The Department of Labor cautions providers to limit answers to the condition supporting the leave, and other law may restrict disclosure. Include the diagnosis when it is clinically appropriate, authorized, and needed to make the certification sufficient. Can I write "as needed" for unpredictable episodes? Give a bounded estimate instead. Intermittent-leave certification calls for the provider's best estimate of episode frequency and typical duration, such as one to two episodes per month lasting two to three days each. Certainty is not expected, and unexpected variation later supports recertification, but an unbounded "as needed" where an estimate was possible can be treated as vague or nonresponsive, sending the employee into the written-deficiency and seven-day cure cycle. Can the client's supervisor call me about the certification? No. Under 29 CFR 825.307 the employee's direct supervisor may never contact you about a certification. Permitted contact comes from a human-resources professional, a leave administrator, a management official, or a health care provider, only for authentication or clarification, and generally only after the employee has had the chance to cure a deficiency. HIPAA governs what you may say, so keep a signed release of information on file before answering questions of substance. What happens if the employer says the certification is insufficient? The employer must identify the deficiency in writing and ordinarily allow seven calendar days to cure it; in Hansler v. Lehigh Valley Hospital Network (3d Cir. 2015), a denial issued without that opportunity did not stand. Answer the identified deficiency rather than rewriting unrelated sections or releasing records. The cure process addresses missing or vague information; it is not a mechanism to change an honestly adverse clinical opinion. Is the WH-380-E form mandatory, and does its printed OMB date matter? The form is optional; the certification is what the law requires when properly requested. The Department of Labor describes its forms as optional and accepts the required information in any complete and sufficient format, including an employer's own form or a provider letter, though an employer form may not demand more than 29 CFR 825.306 permits. The OMB control date printed on the PDF is a paperwork-clearance date, not an expiration of anyone's FMLA rights: the 2026 renewal went to OMB as an extension without change. Is an FMLA certification the same as a disability evaluation or an accommodation letter? No, and mixing them applies the wrong legal test. The FMLA certification asks whether protected-leave criteria are met. A disability medical source statement addresses sustained work-related capacities for a benefits adjudicator. An accommodation letter under the ADA addresses limitations and workplace modifications while the employee keeps working, closer to the return-to-work letter . Fitness-for-duty at reinstatement is governed separately by 29 CFR 825.312 and belongs in a fitness-for-duty evaluation . One client may legitimately need several of these documents. Can I charge for completing FMLA paperwork, and how long must it be kept? FMLA sets no provider fee schedule, does not require insurers to cover form completion, and does not require clinicians to work free: charging follows state law, professional standards, contracts, and practice policy, and payers commonly treat completion as noncovered. On retention, the federal three-year rule at 29 CFR 825.500 binds the employer's FMLA records, not the treating clinician; no FMLA rule sets a clinician period, so your licensing board and health-record rules govern how long the form and support note stay in the chart. Can BastionGPT draft FMLA certification support notes? Yes. Give it chart bullets, a dictation, or the employer's blank form fields, and it drafts the onset, medical facts, functional statements, and bounded frequency and duration estimates for your review, then checks the draft for the elements 29 CFR 825.306 lists before you sign. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Return-to-Work & Accommodation Letter Disability Evaluation & Medical Source Statement Fitness-for-Duty Evaluation Release of Information (ROI) Authorization Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/court-report All note templates # Court Report & Treatment Progress Update: What It Includes, With Sample Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A court report or treatment progress update is a limited report a treating clinician sends to a court, probation officer, or attorney with a valid legal basis, summarizing attendance, engagement, progress toward treatment goals, and recommendations for continuing care. It stays within the treatment role and is not a forensic evaluation. Most run 1 to 3 pages, about 400 to 900 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Treating therapists, counselors, and psychologists; supervisees per state rules and the supervision agreement Audience Judges, probation and drug-court teams, child-protection workers, attorneys; a copy stays in the clinical record Typical length 400 to 900 words · 30 to 60 minutes by hand (clinical team estimate) Format family Limited-scope treatment report (compare: forensic evaluation report, subpoena response) When it's used When a court, probation officer, or attorney requests a treatment status update and a lawful disclosure basis exists Standards context No law mandates the format; HIPAA, 42 CFR Part 2, and the scope of the order or authorization govern the content On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a court report or treatment progress update? A court report, in the treating clinician's sense, is a limited-purpose report that summarizes information generated in treatment for a court, probation officer, drug-court team, child-protection agency, or attorney: dates of service, attendance, goals, interventions, observed change, barriers, and recommendations for continuing care. No statute, payer, or standards body created a universal format. The closest lineage is the Association of Family and Conciliation Courts' Guidelines for Court-Involved Therapy, approved in 2010, which address role clarity, consent, records, and the limits on a treating clinician's opinions, and which AFCC expressly states are not mandatory practice. Clinicians and courts also call the document a treatment progress report, treatment status report, court letter, therapy progress letter, attendance verification, or completion letter. Some agencies say compliance letter, a name that misleads unless the clinician was actually asked and authorized to verify defined conditions. Two boundaries define the document. First, the author is a treating clinician, not a forensic evaluator. A forensic evaluation is an independent assessment built to answer a legal question, with methods, collateral sources, and neutrality chosen for that purpose; a treatment relationship supplies none of that, so a progress update reports treatment facts and treatment opinions and leaves the psycholegal questions alone. Second, a progress summary is not a HIPAA psychotherapy note. The federal definition at 45 CFR 164.501 excludes summaries of diagnosis, functional status, the treatment plan, symptoms, prognosis, and progress, along with session times and modalities. The letter is still protected health information, so it needs a valid disclosure basis, but the special psychotherapy-notes authorization category is narrower than the phrase therapy notes suggests. ## Who uses court reports and when Outpatient therapists, counselors, and psychologists write these updates whenever treatment intersects with supervision: probation-mandated therapy, drug courts, family court and dependency matters, diversion programs, and reentry planning. Probation officers and drug-court teams usually want attendance and engagement on a schedule the order sets; attorneys request updates to support a hearing; child-protection workers ask for status ahead of case reviews. Reach for this document when the request is for current treatment status from the treating clinician. When the request is for the chart itself, that is a subpoena response and records production , and when the court needs an independent opinion on custody, competency, or risk, that calls for a forensic evaluation or capacity evaluation by an evaluator who is not the treating clinician. ## Court report structure: what goes in each section Recipient, legal basis, and reporting period. Name the recipient, the case or docket identifier, the reporting period, and the exact basis for disclosure: a signed authorization on file, a court order whose scope you have read, or a subpoena that satisfies the conditions at 45 CFR 164.512(e). If any record comes from a federally assisted substance use disorder program, confirm 42 CFR Part 2 consent or a Part 2 court order before writing a word. Pitfall: treating a court order to attend treatment as blanket permission to disclose; the order authorizes only what sits inside its scope. Author identity and treatment role. State your name, credentials, license, practice contact, when treatment began, and your role: treating clinician. If services are delivered under supervision, say so accurately. Pitfall: implying appointment as an independent evaluator; the report's authority comes from the treatment role, and overstating it invites cross-examination the record cannot support. Attendance and participation. Give first and most recent service dates, sessions attended out of sessions scheduled, missed or late appointments when relevant, and current status: active, completed, or interrupted. Pitfall: letting attendance stand in for progress or legal compliance; attendance establishes attendance, and compliance usually involves conditions outside the clinician's knowledge. Treatment goals and interventions. State goals at the level the authorized recipient needs and name the modality and frequency. Diagnosis belongs here only when the request reaches it, it is relevant, and disclosure is lawful. Pitfall: reflexively including the diagnosis; no general rule requires one in every court update, and privacy law favors the minimum necessary for the stated purpose. Progress, engagement, and risk-relevant information. Describe change against the stated goals, and label every statement by source: directly documented fact, information the client reported, or professional opinion within the treatment role. Ontario's CRPO makes that three-way labeling an explicit report-writing standard, and it is strong drafting anywhere. Include risk-relevant information only as the authorized purpose requires, and separate current observation from history and allegation. Pitfall: conclusory labels such as fully compliant or low risk; an unlabeled conclusion reads as advocacy and collapses under questioning. Recommendations within the treatment role. Recommend treatment: continuation, frequency, level of care, referrals, supports. If the request asks for more, state the limit in the letter itself, in one sentence, and offer what the treatment role can support. Pitfall: ultimate-issue opinions; custody, parenting time, sentencing, and termination of supervision are the court's questions, and a treating clinician's letter is not the instrument for them. Signature, date, and the file copy. Sign with credentials and date, and keep the request, the authorization or order, the final letter, and the transmission record in the chart. Pitfall: an unprovable disclosure; when the release is questioned later, the retained request and basis are the evidence that the letter was lawful. ### Blank template (copy and adapt) COURT REPORT / TREATMENT PROGRESS UPDATE Prepare only with a valid legal basis. Scope: treatment role only; this is not a forensic evaluation. To: [recipient and agency] Date: Re: [client initials / court ID] Case #: Reporting period: __________ to __________ Basis for disclosure: [ ] signed authorization dated ______ [ ] court order (scope reviewed) [ ] other lawful basis 42 CFR Part 2 records involved: [ ] no [ ] yes (Part 2 consent or Part 2 court order confirmed before release) AUTHOR AND ROLE Name, credentials, license #: _____________________________ Treating clinician since: ________ Supervised: [ ] no [ ] yes Role statement: treating clinician in this matter; not an independent forensic evaluator. ATTENDANCE AND PARTICIPATION First session: ________ Most recent session: ________ Attended ____ of ____ scheduled Missed / late: __________ Current status: [ ] active [ ] completed [ ] interrupted GOALS AND INTERVENTIONS (within the authorized scope) Goals addressed: __________________________________________ Modality and frequency: ___________________________________ PROGRESS AND ENGAGEMENT (label each statement: documented fact / client report / clinical opinion) ___________________________________________________________ ___________________________________________________________ Risk-relevant information, only as the purpose requires: ___________________________________________________________ RECOMMENDATIONS (treatment needs only) ___________________________________________________________ Opinions outside the treatment role are declined; the legal determination rests with the court. Signature and credentials: ______________ Date: __________ File copy kept with the request and authorization/order. Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample court report and treatment progress update Scenario: a probation officer requests a quarterly progress update on a client attending court-referred individual therapy; a signed authorization is on file. All details are fictional. Treatment progress update. Riverbend Counseling Associates · Client: D.K. · Case: 26-4471 · Reporting period: 05/12/2026 to 08/06/2026 · Sent: 08/07/2026 To and from: Officer J. Alvarez, Adult Probation. From L. Warren, LCSW, treating therapist, Riverbend Counseling Associates. Basis for disclosure: D.K.'s signed authorization dated 05/12/2026 permits release of attendance, participation, a progress summary, and treatment recommendations to the probation department for supervision purposes. This update stays within that scope. No 42 CFR Part 2 program records are involved. Role: I am D.K.'s treating therapist. I have not been appointed to perform a forensic evaluation in this matter, and nothing in this letter is an independent forensic opinion. Attendance and participation (documented fact): D.K. began weekly individual therapy on 05/12/2026 under a court-referred treatment condition and remains in active treatment. D.K. attended 11 of 12 scheduled sessions through 08/06/2026; the one missed session, on 06/24/2026, was called in ahead of time and made up within the same week. Sessions run 50 minutes. Goals and interventions: Treatment addresses two goals set with D.K. at intake: strengthening emotion regulation in high-conflict situations, and building structured problem-solving for work and family stressors. The modality is weekly individual cognitive behavioral therapy, with a brief self-report measure completed at the start of each session to track change. One scheduling barrier, a rotating work shift, was addressed in July with a standing alternate appointment time. Progress and engagement: Documented fact: D.K. arrived on time to all attended sessions and completed written between-session practice for 9 of the last 10. Client report: D.K. describes using the skills practiced in session during two recent disagreements and reports that both ended without escalation. Clinical opinion, within the treatment role: D.K. is engaged, and progress toward both goals is steady; continued weekly sessions are clinically indicated to consolidate gains. No safety concerns were identified during this reporting period; safety is reviewed as part of routine care. Recommendations and scope: I recommend continued weekly individual therapy through the next review period, stepping down to every other week when goal progress supports it. The department's request form also asks for an opinion on early termination of supervision. That question is outside my role as treating clinician, and I respectfully decline to offer an opinion on it; if the court requires one, an independent evaluation is the appropriate instrument. I can provide the next update on the department's schedule while D.K.'s authorization remains in effect. Signature: L. Warren, LCSW · 08/07/2026 · The request, the authorization, and a copy of this letter are retained in the clinical record. This sample is fictional and for educational purposes. It does not describe a real client, clinician, or case. ↑ Back to the template and downloads ### Why this sample works The legal basis leads. The authorization, its date, and its scope are stated before any clinical content, so the disclosure is provably lawful. The role is named and bounded. Treating clinician, not forensic evaluator, stated in the letter itself, which is the boundary complaints and cross-examinations test first. Attendance is fact, not verdict. Sessions attended out of sessions scheduled, with dates; nothing is labeled compliant, and the legal question stays with the court. Every statement carries its source. Documented fact, client report, and clinical opinion are labeled separately, the three-way split Ontario's CRPO writes into its report standard. The out-of-scope question is declined in writing. One sentence declines the early-termination opinion and names the right instrument, and the retained file copy proves the whole exchange. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The letter is protected health information, and the lawful-basis question comes before the writing question. A signed authorization is the clean path. A court order permits disclosure only within its scope, and a subpoena without an order is not the same thing: under 45 CFR 164.512(e) it requires satisfactory assurances about notice to the client or a protective order, which is why the safe reflex on receiving one is the subpoena response workflow rather than a quick letter. Substance use disorder records from a federally assisted program carry the separate 42 CFR Part 2 overlay, under which an ordinary subpoena or a generic release is not enough; the 2024 Part 2 final rule reached its mandatory compliance date on February 16, 2026. A mandatory-reporting duty or serious-threat exception is a separate disclosure pathway with its own documentation, covered on the mandated reporting page, not a reason to widen a routine progress update. The letter also joins the clinical record: Ontario's CRPO advises discussing the requested content with the client and offering a chance to review the letter before it goes out, which is sound practice anywhere. Almost everything about the format is convention. No US federal law sets a deadline or a template for a treating clinician's court letter; the controlling deadline usually comes from the court order, the probation contract, or the hearing date, and Ontario's CRPO, one of the few regulators to address report writing directly, expects delivery within 30 days of a request. The letter is not payer documentation either: no claim adjustment code exists for it because it is not a claim, Medicare audits test the underlying session notes, treatment plan, time, and signatures, and a polished court letter cures none of that. Writing time is generally billed to the client or requesting party by written agreement rather than to a health plan. Keep the letter, the request, and the authorization or order in the record under your jurisdiction's retention rules, which vary widely: New York's professional-conduct rule, as one state example, expects six years, and for minors until at least one year past age 21. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ## Common court report errors auditors flag The audit record here concerns the treatment documentation underneath the letter, and it is unforgiving. In a 2020 HHS OIG audit of one psychotherapy practice, On-Site Psychological Services, 111 of 120 sampled claims had treatment plans that never documented improvement or a reasonable expectation of improvement, 109 of 120 treatment notes carried pasted digital images of clinician signatures, and OIG estimated at least $3.3 million in overpayments. A 2022 OIG audit of a New York City provider found all 100 sampled beneficiary days noncompliant, 96 of them lacking the required plan signature, with about $1.1 million in estimated overpayments. A court letter asserting progress that the contemporaneous record cannot show is the same failure with a letterhead. The BastionGPT Clinical Advisory Board sees the same errors most often in court report reviews: Scope creep past the legal basis. The order or authorization covers attendance and participation; the letter narrates diagnosis, history, and third parties. Disclosure is lawful only within the basis it rests on. Attendance dressed up as compliance. Fully compliant asserts a legal conclusion involving conditions the clinician cannot see. Report sessions attended out of sessions scheduled and leave the arithmetic of compliance to the court. Unlabeled sources. Fact, client report, and opinion run together in one paragraph, so the strongest sentence inherits the weakest sourcing. Label all three, every time. Forensic opinions from a treatment chair. Custody, parenting time, sentencing, and termination of supervision are evaluator and court questions; offering them from a treatment relationship exceeds the role and invites a board complaint. A letter the chart cannot back. Retrospective global claims of progress with no contemporaneous notes, goals, or measures behind them, the exact gap the OIG audits keep finding in the underlying records. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on court reports and treatment progress updates. Draft a progress update from your attendance log and session notes, with documented fact, client report, and clinical opinion labeled as you go. Check a finished letter before it is sent: legal basis stated, content inside the order or authorization's scope, no opinions beyond the treatment role, signature and date in place. Flag over-disclosure before it happens: diagnosis, third-party information, or substance use content the authorization does not reach. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What should a treatment progress letter for court include? Six things cover most requests: the recipient and the legal basis for disclosure; your identity, credentials, and treatment role; attendance and participation with dates and counts; goals and interventions at the level the request needs; progress with each statement labeled as documented fact, client report, or clinical opinion; and recommendations limited to treatment. Most run 1 to 3 pages, about 400 to 900 words. No law prescribes the format; the order or authorization prescribes the scope. Does a court order for treatment mean confidentiality is gone? No. An order to attend treatment is not an order to disclose everything the therapist knows. A court order permits disclosure only within its scope, and a subpoena without an order triggers extra conditions under HIPAA at 45 CFR 164.512(e), such as satisfactory assurances about notice to the client or a protective order. Read the order, disclose what it reaches, and use a signed authorization to cover routine reporting; for records demands, see the subpoena response page. Can I recommend custody, unsupervised visits, or a lighter sentence? Usually not from treatment alone. The AFCC court-involved therapy guidelines confine a treating clinician's opinions to the treatment role: functioning, progress, coping, treatment needs. Custody, parenting time, sentencing, and probation termination are psycholegal questions that call for an independent forensic evaluation with methods and collateral sources a therapy relationship does not provide. The defensible move is one sentence declining the question and naming the right instrument. How is a court report different from a forensic evaluation? Source and purpose. A court report summarizes what treatment generated: attendance, goals, response, and recommendations for care, written by the treating clinician. A forensic evaluation is an independent assessment built to answer a legal question, with an evaluee rather than a patient, adversarial scrutiny in mind, and collateral records and instruments chosen for that purpose. One clinician should not hold both roles in the same case. Do I need a signed release if treatment is court-ordered? Get one whenever possible. It makes the disclosure basis unambiguous, defines the scope, and lets the client know what leaves the office. Where the order itself authorizes specific reporting, disclosure can rest on the order, confined to its scope. The release of information page covers the authorization elements; either way, name the basis in the letter itself. What about substance use disorder treatment records? They can carry protection beyond HIPAA. Records from a federally assisted substance use disorder program fall under 42 CFR Part 2, and a subpoena or a generic release alone is not sufficient: disclosure needs Part 2-compliant consent or a qualifying Part 2 court order, and the 2024 final rule's mandatory compliance date was February 16, 2026. Confirm Part 2 status before drug-court or probation reporting, and keep the consent with the file copy. Can I charge for a court letter, and how fast must I send it? There is no US federal deadline or fee schedule for a treating clinician's court letter; the timeline usually comes from the order, the probation contract, or the hearing date, and writing time is typically billed to the client or requesting party by written agreement, not to a health plan. Ontario's CRPO, which regulates report writing directly, permits a reasonable preparation fee with an estimate given first and expects delivery within 30 days of a request. How long do we keep the letter and the request? Keep the final letter, the request, and the authorization or order in the clinical record for as long as your jurisdiction requires, and expect wide variation: New York's professional-conduct rule expects six years, and for minors until at least one year past age 21, while Ontario's CRPO expects at least ten years and New South Wales runs seven years for adults or until age 25 for childhood records. HIPAA's six-year rule covers compliance documentation such as policies, not a national chart-retention period. Can BastionGPT draft a court progress update? Yes. Give it the request, the authorized scope, and your attendance and session data, and it drafts a progress update with facts, client reports, and opinions labeled, recommendations kept to treatment, and a scope statement declining questions outside the role. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Forensic Psychological Evaluation Report Subpoena Response & Records Affidavit Mandated Reporting & Duty-to-Protect Documentation Release of Information (ROI) Authorization Capacity & Competency Evaluation GAF Score Documentation SASSI Documentation PCL-R Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/subpoena-response All note templates # Subpoena Response & Records Affidavit: Definition, Template & Example Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 A subpoena response and records affidavit is the packet a therapy practice assembles when legal process demands client records: an intake log, a written objection or privilege assertion where grounds exist, a custodian affidavit authenticating any copies produced, and a production letter. Records custodians and clinicians use it, with counsel, whenever a subpoena arrives without a court order or client authorization. A typical packet runs 400 to 900 words. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Records custodians and practice managers with the treating clinician; the practice's attorney reviews; a custodian or another qualified person signs the affidavit Audience The issuing attorney or agency, the court, the client and their counsel, your own compliance log Typical length 400 to 900 words across the packet · 40 to 90 minutes by hand (clinical team estimate) Format family Legal response packet (compare: records request packet, release of information, court report) When it's used When compulsory legal process seeks client records without client authorization; deadlines come from the instrument and the governing court rule Standards context Objection deadlines, privilege procedure, and affidavit predicates are law (Rule 45, FRE 902(11), 45 CFR 164.512(e), state statutes); the intake log and cover letter are convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is a subpoena response and records affidavit? A subpoena response and records affidavit is not one standardized document. It is the packet a practice assembles when compulsory legal process demands client records, and it can contain up to four legally distinct documents: a written objection or notice of objection, a privilege assertion that identifies withheld material without revealing it, a custodian affidavit or certification establishing that produced copies are authentic business records, and a production or nonproduction letter. No professional association or agency invented the format. It merges three older legal families: the subpoena duces tecum of compulsory process, the business-record evidence foundation codified when the Federal Rules of Evidence were enacted in 1975, and the confidentiality and privilege objections that decide whether otherwise relevant records may be withheld. Clinicians and lawyers also call the pieces a response to subpoena duces tecum, custodian-of-records affidavit, business-records affidavit, records certification, notice of objection, or motion to quash. The load-bearing distinction runs through the middle of the packet: the objection and the privilege assertion govern whether records may be disclosed at all, while the affidavit governs only whether copies can be authenticated as evidence, and signing one says nothing about the other. Just as load-bearing: a subpoena signed by an attorney is not a court order. Under the HIPAA judicial-proceedings rule at 45 CFR 164.512(e), a court order permits disclosure of exactly what it authorizes, while a subpoena alone permits disclosure only after satisfactory assurance that the client was notified, or a qualified protective order. Federal common law adds a psychotherapist-patient privilege, recognized in Jaffee v. Redmond, 518 U.S. 1 (1996), and state statutes are often stricter. A client asking for their own chart is a different workflow entirely, the records request , and a third party holding a signed authorization belongs in the release of information lane. ## Who uses subpoena response packets and when Every practice that keeps records is eventually served: solo therapists, group practices, community clinics, and hospital records departments, most often in family, personal-injury, criminal, and administrative matters. The work divides three ways: a records custodian or practice manager runs intake and signs the affidavit, since federal evidence law accepts a custodian or another qualified person rather than the treating clinician; the clinician supplies the privilege and psychotherapy-notes analysis; and the practice's attorney reviews anything contested. Reach for this packet when process arrives without client authorization. When an attorney simply writes with a signed authorization, use the release of information workflow; when the client requests their own chart, the records request clock and fee rules apply; and when a court wants observations or opinions from treatment rather than copies, that is a court report with its own scope boundaries. ## Subpoena response structure: what goes in each document Intake and triage log. Record service the day it happens: date and method of service, who received it, the court and case caption, the issuing attorney or agency, the compliance or return date, the records and date range demanded, whether testimony is also commanded, and whether the client is a party. This entry is how you later prove which deadline applied and why you chose the response path you did. Pitfall: assuming a 30-day clock. The 30-day period belongs to a client's own records request ; a subpoena's deadline is whatever the instrument and the governing rule say, and in federal civil cases written objections are due before the earlier of the compliance date or 14 days after service. Authority screen. Classify what actually arrived before anything else happens: a judge-signed court order, an attorney-issued subpoena, or a demand accompanied by a client authorization. Under 45 CFR 164.512(e), a subpoena without an order permits disclosure only after satisfactory assurance of reasonable efforts to notify the client, or a qualified protective order; an order permits only the information it expressly authorizes. Pitfall: treating an attorney subpoena as a court order. The two sit in different legal categories, and collapsing them produces either an unlawful release or demands for conditions that do not exist. Scope and privilege review. Map the demand against the record: which categories are in scope, whether separately kept psychotherapy notes are implicated, whether 42 CFR Part 2 substance-use-disorder material is present, and which grounds exist: privilege, statutory confidentiality, overbreadth, undue burden, defective service. Pitfall: treating everything a therapist wrote as a psychotherapy note. The HIPAA definition is narrow and turns on separate storage; misclassification leads either to overproducing separately kept analysis notes or to wrongly refusing ordinary treatment records. Written objection. Where grounds exist, serve written objections before the deadline: identify the categories objected to, state the grounds, and request the relief sought, such as quashing, modification, in camera review, or a protective order. Pitfall: a blanket refusal, or silence. In federal practice a late objection risks waiving the ordinary grounds, and ignoring the instrument invites contempt exposure; a client's instruction not to release anything supports an objection, it does not cancel compulsory process. Privilege assertion. Where records are withheld on privilege grounds, expressly claim the privilege and describe the nature of the withheld material without revealing the protected content, the pattern Rule 45(e)(2) and Rule 26(b)(5) require in federal cases. The federal psychotherapist-patient privilege recognized in Jaffee v. Redmond covers confidential communications in psychotherapy with licensed psychiatrists, psychologists, and licensed social workers. Pitfall: writing that all records are HIPAA privileged. HIPAA is a privacy regulation, not a privilege; privilege comes from evidence or substantive law, and the assertion must name the actual source. Custodian affidavit. The authentication document: the signer's name, role, and basis for knowing the recordkeeping system, identification of the records by category and date range, page or file count, the copying method, and the business-record predicates: made at or near the time by or from a person with knowledge, kept in the course of a regularly conducted activity, made as a regular practice. A custodian or another qualified person may sign under Federal Rule of Evidence 902(11); the treating clinician is not required, and no universal notarization rule exists. Pitfall: certifying too much. The affidavit authenticates recordkeeping; it does not vouch that treatment was appropriate, the diagnosis correct, or the charges reasonable, and drifting into opinion converts it into a different document nobody requested. Production letter and close-out. When authority is confirmed, transmit with a letter stating what is produced, the date and method, page or media count, withheld categories and their basis, and any agreed narrowing of scope; when no responsive records exist, say so, a statement California's custodian-affidavit statute expressly contemplates. Close the log and retain the complete packet. Pitfall: producing the lifetime file. An order-based disclosure permits only what the order expressly authorizes, and a defined-period demand answered with the whole chart exceeds the legal permission. ### Blank template (copy and adapt) SUBPOENA RESPONSE AND RECORDS AFFIDAVIT PACKET Up to four documents, not one: objection, privilege assertion, custodian affidavit, production letter. Deadline: the instrument plus the governing rule (federal civil: the earlier of the compliance date or 14 days after service). Confirm authority BEFORE any release. INTAKE AND TRIAGE LOG Date served: __________ Method: __________ Received by: ______ Court / case caption and number: _______________________________ Issuing attorney or agency: ____________________________________ Compliance/return date: _________ Objection deadline: _________ Records and date range demanded: _______________________________ Testimony also commanded: [ ] yes [ ] no Client a party: [ ] yes [ ] no AUTHORITY SCREEN (complete before anything is released) [ ] Judge-signed court order (produce ONLY what it authorizes) [ ] Attorney subpoena + satisfactory assurance of client notice or qualified protective order (45 CFR 164.512(e)) [ ] Signed client authorization covering this disclosure [ ] Part 2 records demanded: court order AND subpoena both present [ ] None of the above: object before the deadline, request an authorization or protective order, consult counsel Client / client's counsel contacted: ________ Practice attorney: ______ SCOPE AND PRIVILEGE REVIEW Categories in scope: ___________________________________________ Psychotherapy notes kept separately: [ ] yes [ ] no (not produced without specific authority) Privilege / confidentiality grounds asserted: __________________ Other grounds (overbreadth, burden, service): __________________ CUSTODIAN AFFIDAVIT (authenticates recordkeeping, not opinions) Signer, role, basis of knowledge: ______________________________ Records identified (categories, dates, page count): ____________ [ ] Made at or near the time by a person with knowledge [ ] Kept in the course of a regularly conducted activity [ ] Made as a regular practice [ ] True and complete copies No-record or partial-record statement, if any: _________________ Execution form per the governing rule (declaration or sworn): __ PRODUCTION / NONPRODUCTION LETTER AND CLOSE-OUT Produced on: ________ Method: ________ Pages/media: __________ Withheld categories and basis: _________________________________ Retain the full packet with the record and any legal hold. Download the template (PDF) Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt. ## Sample subpoena response packet Scenario: a two-clinician private practice is served with an attorney-issued subpoena for a former client's complete record, with no court order and no client authorization. The packet shows the triage, the response letter, and the unsigned affidavit skeleton. All details are fictional. Subpoena response packet. Fernhill Counseling Group LLC · Client: L.P. · Served: 08/03/2026 · Objections served: 08/12/2026 Intake log (opened 08/03/2026): Subpoena duces tecum served by process server at 2:10 pm on C. Whitfield, office manager. Civil action pending in federal district court, Case No. 26-cv-0412; issuing counsel R. Malone, attorney for the defendant. Demands all mental health records, notes, and communications concerning former client L.P., no date range stated. Compliance date on the instrument: 08/24/2026. No court order attached, no client authorization on file, no documentation of satisfactory assurance enclosed. Client is a party to the action. Objection deadline calculated: 08/17/2026, the earlier of 14 days after service and the compliance date. Routed to the practice's attorney the same day. Authority screen (completed 08/04/2026): Judge-signed order: no. Satisfactory assurance of client notice or qualified protective order: none received. Client authorization covering this disclosure: none on file. Part 2 substance-use-disorder records: none held for this client. Psychotherapy notes: kept separately from the chart; excluded from any production absent specific authority. Determination: no current authority to disclose; produce nothing; do not confirm in any response whether a treatment relationship exists; contact the client's counsel through the practice attorney. Response letter to issuing counsel (sent 08/12/2026, with written objections): The letter states that the practice received the subpoena on August 3, 2026; that federal and state law restrict disclosure of mental health information; and that the subpoena is not accompanied by a court order, a qualified protective order, or documentation of satisfactory assurance under 45 CFR 164.512(e). It states the practice can neither produce records nor confirm whether responsive records exist unless it receives a signed client authorization, a qualified protective order, or an order of the court. The enclosed written objections, served within the Rule 45(d)(2)(B) window, object to the demand as unlimited in time and scope and as seeking material protected by the psychotherapist-patient privilege, described by category only. The letter closes by committing to prompt compliance with a lawful instrument of appropriate scope. Custodian affidavit skeleton (prepared 08/12/2026, unsigned, held pending authority): Designated signer C. Whitfield, records custodian, familiar with the practice's recordkeeping system. The draft identifies records by category and date range once scope is fixed, and recites: entries made at or near the time of each service by a person with knowledge; records kept in the course of the practice's regularly conducted activity; making such records a regular practice; copies produced are true copies; a no-record or partial-record statement to be added if any demanded category is empty. The draft certifies recordkeeping only and offers no opinion on treatment, diagnosis, causation, or charges. It will be executed in whatever form the governing rule requires, declaration or sworn affidavit, only when production is authorized. Outcome (logged 08/12/2026): Objections served on day 9 of the 14-day window. Nothing produced; the existence of records neither confirmed nor denied to the issuing party. Awaiting authorization, qualified protective order, or court order. Packet, service documentation, and correspondence retained with the underlying record under legal hold. Logged by C. Whitfield; practice attorney copied. This sample is fictional and for educational purposes. It does not describe a real client, practice, attorney, or case, and it is not legal advice. ↑ Back to the template and downloads ### Why this sample works Nothing moved before authority. The screen separates a judge-signed order from an attorney subpoena, the exact line 45 CFR 164.512(e) draws, and the determination is recorded before any drafting starts. The objection beat the deadline. Served on day 9 of the 14-day federal window, which preserves grounds that silence or lateness would risk waiving. The letter avoids over-disclosure. It neither confirms a treatment relationship nor attaches records, and it names the three instruments that would permit production. Privilege is asserted without revealing content. The objections name the privilege and describe the withheld material by category only, the pattern federal privilege procedure requires. The affidavit stays in its lane. Recordkeeping predicates only, no clinical opinion, and it stays unsigned until production is actually authorized. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The risk in this workflow runs opposite to most documentation risk: the expensive failure is disclosing too much, too early, to the wrong instrument. HIPAA permits disclosure in judicial proceedings; it does not compel it, and the source of compulsion is the subpoena, order, or statute itself. When a court order is the authority, 45 CFR 164.512(e) permits only the information the order expressly authorizes, so a lifetime chart produced against a defined-period order exceeds the legal permission on its face. Separately kept psychotherapy notes ordinarily need their own psychotherapy notes authorization even when ordinary records may flow, and substance-use-disorder records under 42 CFR Part 2 follow a dual-instrument rule: an authorizing court order does not compel production without a subpoena or other mandate, and a subpoena alone does not replace the required order. State law layers on top and is often stricter: Illinois requires specified warning language on subpoenas seeking protected mental health records, Washington requires advance notice before compulsory discovery of health care information, and California prescribes the custodian affidavit's content, including statements when no or only some responsive records exist. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary. No payer ever adjudicates this packet: there is no claim and no denial code for a defective subpoena response, so nothing downstream will catch an error the way a billing office catches a rejected claim. The consequences are legal and professional instead: waived objections, contempt exposure, costs, liability for unlawful disclosure, and board complaints. Retention splits the same way it does for access documentation: the HIPAA six-year rule at 45 CFR 164.530(j) covers Privacy Rule compliance documentation, not the chart, while the clinical record follows state and board rules, at least six years in New York with an extended rule for minors, and the prudent convention is to keep the complete legal-response packet, service evidence, correspondence, and an index of anything produced at least as long as the underlying record and any active legal hold. The intake log and cover letter are convention; the objection deadline, the privilege procedure, and the affidavit predicates are law. When the court wants observations or opinions from treatment rather than copies, step out of this packet and into the court report , and route a later request to correct an error in the record through the record amendment process. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ## Common subpoena response errors courts and counsel flag This document type has no payer audit trail: a targeted review of OIG mental health reports, CMS CERT materials, and remittance-code lists found no denial code and no published national error rate for custodian affidavits answering therapy-record subpoenas, because the packet never enters claim adjudication. A 2023 HHS Office of Inspector General audit did estimate $580 million in improper Medicare psychotherapy payments, but that number measures billing documentation, not subpoena handling, and importing it here would be false precision. The documented failure modes live in the rules themselves: in federal practice an objection served after the earlier of the compliance date or 14 days risks waiving the ordinary grounds, and in Australian family proceedings unexplained noncompliance can end in an arrest warrant or a costs order. The BastionGPT Clinical Advisory Board sees the same errors most often in subpoena response reviews: Treating an attorney subpoena as a court order. The two are different legal categories under 45 CFR 164.512(e); collapsing them produces either an unlawful release or demands for conditions that do not exist. Producing more than the instrument authorizes. An order-based disclosure is limited to the information the order expressly authorizes; answering a defined-period demand with the whole chart is the over-disclosure this workflow exists to prevent. Misclassifying psychotherapy notes, in both directions. The HIPAA category turns on a narrow definition and separate storage; getting it wrong either sends protected analysis notes out the door or wrongly withholds ordinary treatment records such as diagnoses, treatment plans, and progress summaries. An affidavit that proves too little or vouches too much. A bare statement that the copies are true records misses the business-record predicates of timing, knowledge, regular course, and regular practice; drifting into clinical accuracy, necessity, or causation turns a custodial certification into an opinion document nobody requested. Checking for only one instrument on Part 2 records. A subpoena alone does not replace the required Part 2 court authorization, and an authorizing order alone does not compel production without a subpoena or other mandate; a workflow that looks for just one fails in both directions. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on subpoena response and records affidavit packets. Extract a served subpoena into a triage log: caption, issuing party, service date and method, compliance date, records demanded, and the calculated objection deadline. Draft the packet documents for attorney review: the response letter seeking an authorization, qualified protective order, or court order, the written objection, and a custodian affidavit skeleton carrying the business-record predicates. Check a response before anything is released: authority confirmed, scope matched to the instrument, psychotherapy notes and Part 2 material flagged, and deadlines verified. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Can I ignore a subpoena if my client tells me not to release anything? No. A client's instruction supports a privilege or confidentiality objection, but it does not cancel compulsory process. The lawful paths are written objections, a motion to quash or modify, a protective-order request, or attendance at the return date, each inside the instrument's deadline. Ignoring a subpoena risks waived objections, contempt exposure, and costs; in Australian family proceedings, unexplained noncompliance can lead to an arrest warrant. Is a subpoena from a lawyer the same as a court order? No, and the distinction drives the whole response. Under the HIPAA judicial-proceedings rule, a court or administrative order permits disclosure of exactly what it authorizes. An attorney-issued subpoena without an order permits disclosure only after satisfactory assurance that the client was notified with a chance to object, or a qualified protective order. State law can prohibit release even then, so the safe sequence is authority first, production second. Do I have to produce psychotherapy notes in response to a subpoena? Only material that actually meets the definition gets the special handling. HIPAA's psychotherapy-notes category covers a clinician's separately stored session-analysis notes and excludes diagnoses, treatment plans, medication records, session times, test results, functional status, prognosis, and progress summaries. Separately kept notes ordinarily need their own psychotherapy notes authorization or equivalent legal authority; everything commingled with the chart is ordinary record, whatever it is labeled. Does the treating therapist have to sign the records affidavit? Usually not. Federal Rule of Evidence 902(11) accepts a certification by the custodian or another qualified person who knows the recordkeeping system, and no federal evidence rule requires notarization. Some state statutes and prescribed forms are narrower about who signs and how, California's custodian-affidavit statute being the detailed example, so check the governing rule before choosing the signer. What deadline applies to a subpoena for therapy records? The one on the instrument, read with the governing rule, and it is usually shorter than people assume. In federal civil cases, written objections are due before the earlier of the stated compliance date or 14 days after service. The 30-calendar-day clock many practices know belongs to a client's own records request under access law; it has nothing to do with subpoenas. What if the demand touches substance use disorder records under 42 CFR Part 2? Part 2 runs on a dual-instrument rule. Without patient consent, disclosure in a legal proceeding takes both an authorizing court order meeting Part 2's standards and a subpoena or other legally sufficient mandate. An order alone does not compel production, and a subpoena alone does not substitute for the order, so a workflow that checks for just one instrument fails in both directions. If I certify records as the custodian, am I vouching that everything in them is true? No. A business-records certification establishes how records were made and kept: near-contemporaneous entries by people with knowledge, in the regular course of a regularly conducted activity, as a regular practice, with true copies produced. It does not certify that treatment was appropriate, a diagnosis correct, or future care needed, and the other side can still challenge trustworthiness. When a court wants clinical observations or opinions, that is a court report , a different document with different rules. How long do we keep the subpoena response packet? No national rule sets a retention period for the packet itself. The prudent convention is to keep the complete response file, the instrument, service evidence, objections, correspondence, and an index of anything produced, at least as long as the underlying clinical record and any active legal hold. Chart retention itself is state and board law: at least six years in New York with an extended rule for minors, ten years past age 18 or last contact for Ontario psychologists, and seven years or to age 25 in New South Wales. Can BastionGPT draft a subpoena response? Yes, as the drafting layer under your attorney's judgment. Give it the instrument's facts and it produces the triage log, the response letter seeking an authorization, qualified protective order, or court order, the written objection, and an unsigned custodian affidavit skeleton for counsel to review. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Related templates Records Request & Chart Copy Packet Release of Information (ROI) Authorization Psychotherapy Notes Authorization Court Report & Treatment Progress Update Record Amendment & Correction Response Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/phq-9 All note templates # PHQ-9 Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The PHQ-9 is a nine-item, public-domain depression severity measure scored 0 to 27, validated by Kroenke, Spitzer, and Williams in 2001. Primary care and behavioral health teams use it for depression screening and measurement-based care. This page covers how to document and interpret PHQ-9 results in the chart, including item 9 follow-up, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report; any clinician or trained staff can administer and score, no license or qualification level required Audience Treating clinicians and care teams, collaborative care registries, quality programs (MIPS, HEDIS), payers and auditors Typical length 2 to 6 chart lines, plus item-9 follow-up when triggered · patient completion under 3 minutes Format family Self-report depression severity measure (9 items, scored 0 to 27) When it's used Depression screening, symptom monitoring and treatment-to-target, collaborative care registries, perinatal and adolescent pathways Standards context Public domain per the PHQ Screeners terms; no US, Canadian, or Australian authority mandates it by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PHQ-9? The Patient Health Questionnaire-9 (PHQ-9) is a nine-item depression severity measure validated by Kurt Kroenke, Robert Spitzer, and Janet Williams in 2001, derived from the clinician-administered PRIME-MD and developed with an educational grant from Pfizer. Patients rate nine symptoms over the past two weeks from 0 to 3, for a total of 0 to 27. It is in the public domain: the official PHQ Screeners terms state that "no permission is required to reproduce, translate, display or distribute" the PHQ family, which covers EHR builds, patient portals, apps, and printed forms at no cost. The family matters for charting: the PHQ-2 is the two-item gateway screen, the PHQ-8 drops the self-harm item for research use (total 0 to 24), and the PHQ-9 Modified for Teens and the broader adolescent PHQ-A are related but not interchangeable forms, so the chart should name the exact version. The load-bearing distinction for documentation: the PHQ-9 measures symptom severity, it does not make a diagnosis. A total of 10 or higher is a positive screen with pooled sensitivity and specificity near 85% against semistructured diagnostic interviews, and the published bands (minimal, mild, moderate, moderately severe, severe) describe reported symptom burden, not established major depressive disorder; the diagnosis comes from a diagnostic evaluation that weighs duration, impairment, and the bipolar, substance, medical, and grief alternatives. Two published scoring methods also coexist: the dominant continuous total, and a legacy DSM-IV-era symptom-count algorithm that should be labeled provisional if your organization still uses it, because meta-analysis finds it less sensitive than the cutoff of 10. And item 9 is a safety flag, not a suicide risk assessment : a nonzero response starts a direct clinical inquiry, and a zero does not end one. ## Who uses PHQ-9 documentation and when Primary care teams document PHQ-9 results inside screening workflows driven by USPSTF recommendations and quality measures; therapists and psychiatric prescribers trend it for measurement-based care and treatment-to-target; care managers run registries on it in collaborative care ; and perinatal and adolescent pathways adapt it with population-specific forms. It pairs naturally with the GAD-7 when anxiety runs alongside depression, and it anchors the depression column of an outcome measure note . Neighbors matter at the boundaries: the licensed BDI-II serves settings that want a longer proprietary self-report, the EPDS owns most perinatal pathways, and Australian Better Access workflows usually reach for the K10 first, so a PHQ-9 entry there should say why a depression-specific, two-week measure fits the clinical question. ## How to document PHQ-9 results in the chart No statute prescribes a PHQ-9 note format. The elements below are the convention that survives review because each one maps to something a future clinician, a quality measure, or an auditor will need. Most results fit in two to six lines; the note grows only when item 9 is positive. Each element carries the pitfall that most often undermines it. Instrument identity, mode, and language. Name the exact form (PHQ-9, PHQ-2, PHQ-8, PHQ-9 Modified for Teens), the administration mode (paper, portal, tablet, interview, telephone), the language, and any interpreter or staff assistance. Pitfall: "PHQ positive" with no form named. A PHQ-8 total runs 0 to 24, teen variants carry supplemental questions outside the total, and PHQ-A and PHQ-9 Modified for Teens are different instruments, so an unnamed form leaves the number uninterpretable and the quality extract empty. Total over 27 with the published band. Write the score as a fraction and its severity descriptor: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 19 moderately severe, 20 to 27 severe. Pitfall: writing "14/30". The unscored functional item is not part of the total, and a denominator error signals to reviewers that the scorer does not know the instrument. Item 9, separately, every time. Record the 0 to 3 response even when it is zero. A nonzero response gets a same-visit direct inquiry: passive death wishes versus active ideation, plan, intent, preparatory behavior, history, and protective factors, then a documented risk formulation and a proportionate disposition, with a safety plan where indicated. Pitfall: treating the checkbox as the assessment. The instrument's own manual states that final determination of self-harm risk requires a clinical interview, and item 9 mixes thoughts of death with thoughts of self-harm. The functional-interference item, unscored. Record the patient's answer (not difficult through extremely difficult) beside the total, never inside it. Pitfall: dropping it. It is the instrument's only direct read on impairment, which is what medical necessity and treatment decisions actually turn on. Prior score, with change in points and percent. Name the baseline score and date, then give both numbers: points answer the measurement-error question, percent answers the response question. Pitfall: the five-point reflex. A 2026 individual-participant meta-analysis puts the general-practice threshold for change beyond measurement error near six points, response still means a 50% reduction, and remission means below 5; the three constructs deserve three separate sentences. An interpretation sentence. Say what the result is (a severity screen in clinical context) and what it is not (a stand-alone diagnosis); after the interview, "consistent with" is the defensible verb. Pitfall: promoting a band to a diagnosis. "Moderate depressive symptom severity" is a finding; "moderate depression" is a diagnostic claim the questionnaire cannot carry alone. Plan linkage and the next measurement. State what continues, changes, or is deferred because of the result, and when the next administration is due, tied to treatment phase rather than habit. Pitfall: an orphaned score. A number with no plan sentence documents administration, not care, and it is the first thing collaborative care reviewers and payer auditors flag. ### Blank template (copy and adapt) PHQ-9 DOCUMENTATION BLOCK Form: [PHQ-9 / PHQ-2 / PHQ-8 / PHQ-9 Modified for Teens] Language: [ ] Mode: [paper / portal / tablet / interview] Date: [ ] Completed by: [patient self-report / staff-assisted] Total: [ ]/27 Severity: [minimal / mild / moderate / moderately severe / severe] Item 9: [0-3] If nonzero, same-visit inquiry: [passive vs active ideation; plan / intent / preparatory behavior; history; protective factors] Risk formulation + disposition: [ ] Safety plan: [updated / n/a] Functional interference (unscored): [not difficult ... extremely] Prior score / date: [ ] Change: [points] ([percent]) Construct: [beyond measurement error / response >=50% / remission <5 / within error band] Interpretation: [severity result in clinical context; screen, not a stand-alone diagnosis] Plan linkage: [continued / changed / deferred, and why] Next measurement: [interval + reason] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. ## Sample PHQ-9 documentation (fictional) Scenario: a psychotherapy follow-up four weeks into treatment. The total has improved but not to response, and item 9 is newly positive at 1, the combination that gets documented badly most often: the improvement invites a short note, and the item-9 response demands a full one. All details are fictional. Patient: R.T., 34 · Visit: Psychotherapy follow-up, week 4 · Clinician: M. Idris, LCSW · Note date: 08/12/2026 Measure: PHQ-9 (English, patient portal, self-completed 08/12/2026): 14/27, moderate depressive symptom severity. Item 9 = 1. Functional interference reported as very difficult. Prior PHQ-9: 20/27 on 07/15/2026 (paper, in office). Change from baseline: Six-point reduction (30%) over four weeks. The change is at the current general-practice threshold for exceeding measurement error, does not meet the 50% response definition, and is not remission. Trajectory is consistent with early improvement. Item-9 follow-up (same visit): Reviewed directly today. R.T. describes passing thoughts of being "better off not waking up" on the worst mornings, most recently three days ago. She denies current active suicidal ideation, plan, intent, or preparatory behavior, and has no history of suicide attempt or self-harm. Protective factors include treatment engagement, her partner's support, and stated future plans. Acute risk assessed as low, with reasoning discussed; chronic elevation associated with recurrent depressive symptoms noted. Safety plan reviewed and updated with her today; crisis and after-hours contacts confirmed, including 988. Telephone check-in scheduled within 48 hours. Interpretation: Today's result is a severity measurement within an established episode of major depressive disorder, single episode, moderate, diagnosed by clinical interview at intake. The score monitors severity; it did not establish the diagnosis. Plan linkage: Continue weekly CBT with a behavioral activation focus. Prescriber review already scheduled for next week will use today's score and item-9 follow-up; coordination message sent today. Next PHQ-9 in two weeks, per active-treatment cadence. Depression follow-up was provided and documented at today's qualifying encounter for the practice's screening measure. This sample is fictional and for educational purposes. It does not describe a real patient or record, and the scores are invented for illustration and correspond to no real person. ↑ Back to the template and downloads ### Why this sample works It names the exact form, language, mode, and date, so the score is comparable at the next administration and extractable for quality reporting. The total sits over its 27-point denominator with the published band, and the unscored functional item stays separate from the arithmetic. Item 9 gets a documented same-visit inquiry, a risk formulation with reasoning, and a proportionate disposition instead of a checkbox or a reflex emergency referral. Change is stated in points and percent against a named baseline, and each measurement construct (error band, response, remission) is claimed separately. The score links to a plan and a dated next measurement, and the diagnosis stays anchored to the clinical interview rather than the questionnaire. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations No law in the US, Canada, or Australia mandates the PHQ-9 by name, at any interval, or at any score threshold; the operative rules are PAYER POLICY and CONVENTION, and they run on different clocks. The USPSTF recommendations for depression screening (adults, 2023; adolescents 12 to 18, 2022) are Grade B convention: they name no instrument and specify no interval. MIPS Quality ID #134 is payer policy with teeth: screening with an age-appropriate standardized tool on the encounter date or up to 14 days before, once per performance period, with a documented follow-up plan for a positive screen. In 2026 the MIPS CQM version lets that documentation land up to two calendar days after the encounter while the Medicare Part B claims version requires it the same date, and a suicide-risk assessment alone does not count as the depression follow-up. A documented refusal is an allowed exception; a declined screen entered as a zero is a falsified record. Quality ID #370 , depression remission at twelve months, starts a fixed clock at the first qualifying score above 9 with a depression diagnosis and looks for a score below 5 at 12 months plus or minus 60 days; a later, higher score does not restart the clock, so preserve the index date, form, and score. HEDIS runs three more clocks: follow-up within 30 days of a positive screen, a PHQ-9 result in each four-month assessment period for members with depression, and response or remission at 4 to 8 months. Collaborative care billing (99492 through 99494) requires validated rating scales and a registry as a billing element , which the PHQ-9 usually satisfies. For 96127 there is no stable national unit limit: CMS updates its edit files quarterly and commercial daily-limit tables differ, so verify the current plan-specific rule instead of repeating a number. In accredited settings, Joint Commission NPSG 15.01.01 adds the accreditation layer: a positive suicide screen, which an item-9 endorsement can trigger, must lead to an evidence-based assessment covering ideation, plan, intent, behaviors, and risk and protective factors, with the risk level, justification, and mitigation plan documented; item 9 alone does not complete that assessment. The national postures genuinely diverge, so a cross-border template cannot assume the US pattern. Canada's task force (2025) recommends against routine questionnaire screening of asymptomatic adults, while CANMAT supports targeted case-finding (PHQ-2 at 2 or above, then PHQ-9 at 10 or above) and measurement every 2 to 4 weeks during active treatment with less frequent checks in maintenance; a Canadian chart should show assessment of symptomatic patients rather than universal screening justified by US guidance. Australia's Better Access guidance (March 2026) requires an outcome measure in a mental health treatment plan unless clinically inappropriate but prescribes no tool; the K10 is the incumbent by convention, and the MBS plan-item structure changed on 1 November 2025, so date any billing statement. On measurement itself: official translations may be reproduced freely but are not all independently validated, so record the language and any assistance used; and respect the instrument's error band when interpreting small movements, which the 2026 meta-analytic estimate puts near six points in general practice. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The PHQ-9 and the PHQ instrument family are in the public domain; the official PHQ Screeners site states that no permission is required to reproduce, translate, display, or distribute them. PRIME-MD is a trademark of Pfizer Inc. BastionGPT is not affiliated with, or endorsed by, Pfizer, the instrument authors, or the PHQ Screeners project. This page reproduces no proprietary test items, norms, or scoring materials from any licensed instrument. ↑ Back to the template and downloads ## Common PHQ-9 documentation errors reviewers flag The accountability data here are unusually specific. Validated against an electronic C-SSRS criterion, item 9 carried a positive predictive value of 28.6% (Na et al., 2018: 41.1% of 841 patients were item-9 positive while 13.4% met the risk criterion), and the misses run the other way too: in a large health-system cohort , 39% of suicide attempts and 36% of suicide deaths within 30 days followed an item-9 response of zero (Simon et al.), and across 447,245 VA assessments, 71.6% of subsequent suicides occurred among patients who had answered "not at all" (Louzon et al., 2016). Meanwhile the 2026 individual-participant meta-analysis put the PHQ-9's minimal detectable change near six points, not the five that habit assumes. The BastionGPT Clinical Advisory Board sees the same errors most often in PHQ-9 documentation reviews: Severity bands promoted to diagnoses. "PHQ-9 = 12, moderate depression" reads as a diagnostic statement. The defensible phrase is moderate depressive symptom severity, with the diagnosis carried by the interview; the CMS screening measure itself expects treatment decisions to follow adequate diagnostic evaluation, not the score alone. Item 9 treated as a verdict in either direction. A nonzero response routed straight to the emergency department without an assessment, or a zero response closing a risk inquiry the interview should have kept open. Both patterns ignore the same evidence: the item overflags (positive predictive value 28.6% against a structured criterion) and underdetects (most near-term deaths followed a zero). Document the inquiry, the formulation, and why the disposition is proportionate. The form left ambiguous. "PHQ score 14" without naming PHQ-9 versus PHQ-8 versus PHQ-9 Modified for Teens, the language, or the mode. The denominators differ, the teen supplements sit outside the total, and structured quality extraction fails when the instrument name is missing, which quietly zeroes out measures the organization reports. Change math without its constructs. A five-point drop labeled real change when the 2026 general-practice threshold for exceeding measurement error sits nearer six points, a percent response with no named baseline, or two same-day administrations read as clinical change on an instrument that asks about two weeks. Points, percent, response, and remission each answer a different question; the note should say which one it is answering. Scores invented, prorated, or overwritten. A declined screen entered as zero (which erases the allowed refusal exception and falsifies the record), missing items silently prorated into a "validated" total with no named method, or a patient's corrected answer replacing the original with no audit trail. Document what actually happened: offered, declined, incomplete, or amended, and by whom. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Paste the responses, or just the total, item 9, and the prior score, and get a documentation-ready score block: severity band, change in points and percent against the named baseline, and an item-9 follow-up narrative scaffold when the response is nonzero. Cross-check a finished note for the gaps reviewers flag: a total that disagrees with its band, a positive item 9 with no documented follow-up, a missing form name or baseline, or a functional item folded into the total. Summarize a serial score history into response, remission, and measurement-error language for treatment reviews and collaborative care registries. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do PHQ-9 scores actually mean? The total runs 0 to 27, and the published bands are 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 19 moderately severe, and 20 to 27 severe. Those are severity descriptors for reported symptoms over the past two weeks, not diagnoses. A score of 10 or higher is the conventional positive-screen threshold, with pooled sensitivity and specificity near 85% in the largest meta-analysis . In a low-prevalence clinic, many positives will not turn out to be major depression after interview, which is why chart language should stay at "positive screen, moderate severity" until the diagnostic evaluation speaks. What counts as meaningful change between two PHQ-9 scores? Three thresholds answer three different questions. Change beyond measurement error: the 2026 individual-participant meta-analysis estimated the minimal detectable change at 5.72 points in general practice, so about six points is the cautious threshold, replacing the older five-point habit. Response: a reduction of at least 50% from a named baseline. Remission: a total below 5, which still says nothing about residual individual symptoms until you review item-level responses. Write the points, the percent, and the construct you are claiming; "improved" with no baseline is the version auditors cannot use. How do I document a positive item 9 without sending everyone to the ED? Record the response, then document a same-visit direct inquiry: passive death wishes versus active ideation, plan, intent, preparatory behavior, past attempts, and protective factors. Close with a risk formulation covering acute and chronic risk with your reasoning, a proportionate disposition, and the follow-up interval. A structured tool can organize the inquiry (the suicide risk assessment page covers it) and a safety plan documents the mitigation. Emergency referral follows the assessment when the assessment supports it; no authority reviewed for this page makes it automatic for every nonzero response. Does item 9 equal to zero mean my patient is safe? No. It is one data point with measured limits: in a large health-system cohort , 39% of suicide attempts and 36% of suicide deaths within 30 days followed a zero, and across 447,245 VA assessments, 71.6% of subsequent suicides occurred among patients who had answered "not at all" (Louzon et al., 2016). Record the zero, and let history, behavior, collateral information, and the interview keep their full weight. A zero never overrides contrary clinical evidence, and a note should never cite it as the reason an inquiry stopped. How often should the PHQ-9 be repeated? No authority requires it at every visit. CANMAT describes every 2 to 4 weeks as useful during active treatment, with less frequent measurement in maintenance. MIPS #134 requires screening once per performance period; HEDIS looks for a result in each four-month period for members with depression and checks response or remission at 4 to 8 months; collaborative care programs re-measure on their registry cadence. Pick the cadence from the treatment phase and the program you actually report to, then write it into the plan. Do we need permission to put the PHQ-9 in our EHR or forms? No. The instrument family is in the public domain, and the official PHQ Screeners terms state that no permission is needed to reproduce, translate, display, or distribute it. That covers EHR builds, portals, apps, and paper forms at no cost. Two cautions travel with the freedom: a reworded or restructured version is no longer the validated instrument, and the licensing world next door is different. The BDI-II (Pearson) and the EPDS (Royal College of Psychiatrists) carry reproduction restrictions, so the free-to-copy habit must not migrate to them. What do I document when a patient declines the screen? That it was offered, that it was declined, and any reason the patient wants recorded, then the assessment you did instead through conversation and observation. Never enter a zero or a "negative screen" for a form that was not completed. For MIPS #134, a documented patient refusal is an allowed denominator exception, so the honest note is also the one the measure accepts. Is the PHQ-9 the same as the PHQ-2, PHQ-8, or the teen versions? They are one family with different jobs, and the chart should name the exact form. The PHQ-2 is the two-item gateway (0 to 6) that triggers a full PHQ-9; the PHQ-8 drops item 9 for research settings (0 to 24, no safety flag); the PHQ-9 Modified for Teens adds youth-specific supplemental questions that sit outside the total; and the broader PHQ-A is a substantially modified adolescent instrument, not a synonym for the teen PHQ-9. Totals from different forms are not interchangeable, and "PHQ positive" without a form name is the ambiguity that starts most downstream errors. Can BastionGPT draft my PHQ-9 documentation? Yes. Paste the responses, or just the total, item 9, and the prior score, and it drafts the documentation block: severity band, change in points and percent against the named baseline, an item-9 follow-up narrative scaffold when the response is nonzero, and the plan-linkage sentence, ready for your review. It can also check a finished note for the classics: a total that disagrees with its band, a positive item 9 with no documented follow-up, or a missing form name. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: PHQ Screeners, official screener site and instruction manual : public-domain status, version family, severity bands, the unscored functional item, and the requirement that self-harm risk be determined by clinical interview. Kroenke, Spitzer & Williams, 2001, Journal of General Internal Medicine, the PHQ-9 validation study : development, original 88%/88% operating characteristics at the cutoff of 10, and reliability. Negeri et al., 2021, BMJ, updated individual-participant-data meta-analysis (100 studies, 44,503 participants): pooled sensitivity and specificity of 0.85 at the cutoff of 10; Levis et al., 2019, BMJ, the first IPD meta-analysis . Manea et al., 2015, meta-analysis of the PHQ-9 diagnostic algorithm : the legacy symptom-count method is less sensitive than the total-score cutoff. Na et al., 2018, Journal of Affective Disorders, item 9 validated against the eC-SSRS : sensitivity 87.6%, specificity 66.1%, positive predictive value 28.6%, negative predictive value 97.2%. Simon et al., large health-system cohort of PHQ administrations : graded attempt risk by item-9 response and the share of near-term attempts and deaths that followed a zero response; Louzon et al., 2016, Psychiatric Services 67(5), the VA cohort of 447,245 assessments. 2026 BMJ individual-participant meta-analysis : PHQ-9 minimal detectable change (MDC95) of 5.72 points in general practice and 6.48 in inpatient settings. CMS Quality Payment Program, 2026 specifications: Quality ID #134 (MIPS CQM) , #134 (Medicare Part B claims) , and #370 (Depression Remission at Twelve Months) . NCQA, HEDIS depression measures : screening follow-up, PHQ-9 monitoring, and remission-or-response windows. CMS, behavioral health integration billing guidance (validated rating scales in collaborative care) and the medically unlikely edits program (quarterly updates; no stable national 96127 unit limit). Joint Commission, standards interpretation FAQs on suicide risk ( screening and assessment expectations ): what an evidence-based assessment after a positive screen must address. CANMAT 2023 depression guideline update, Canadian Journal of Psychiatry : two-stage screening pathway, measurement cadence, and response and remission definitions; Canadian Task Force on Preventive Health Care, 2025 depression screening update . Australian Government Department of Health, Disability and Ageing, Better Access treatment-plan guidance (March 2026): outcome-measure requirement with clinician tool choice. USPSTF, adult depression screening recommendation (2023) and child and adolescent recommendation (2022); LOINC, 44261-6 , the PHQ-9 total-score code used in structured extraction. ## Related templates GAD-7 Score Documentation C-SSRS Documentation Outcome Measure Review Note Suicide Risk Assessment Safety Plan Collaborative Care (BHI/CoCM) Note BDI-II Score Documentation EPDS Score Documentation DASS-21 Score Documentation PSC-17 Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation MADRS Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/gad-7 All note templates # GAD-7 Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The GAD-7 is a seven-item anxiety severity measure scored 0 to 21, developed by Spitzer, Kroenke, Williams, and Löwe in 2006 and free to reproduce under the official PHQ Screeners terms. Primary care and behavioral health teams use it for anxiety screening and symptom monitoring. This page covers how to document and interpret GAD-7 results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report; any clinician or trained staff can administer and score, no license or qualification level required Audience Treating clinicians and care teams, collaborative care registries, accreditation surveyors, payers and auditors Typical length 2 to 5 chart lines, often alongside a paired PHQ-9 · patient completion under 2 minutes Format family Self-report anxiety severity measure (7 items, scored 0 to 21) When it's used Anxiety screening, symptom monitoring during treatment, collaborative care registries, paired depression and anxiety measurement Standards context Free to reproduce without permission under the PHQ Screeners terms; no US, Canadian, or Australian authority mandates it by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the GAD-7? The Generalized Anxiety Disorder 7-item scale (GAD-7) is a seven-item anxiety severity measure developed by Robert Spitzer, Kurt Kroenke, Janet Williams, and Bernd Löwe, published in 2006 from a validation study across 15 US primary-care clinics, with development supported by a Pfizer educational grant. Patients rate seven symptoms over the past two weeks from 0 to 3, for a total of 0 to 21, with four published bands: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, and 15 to 21 severe. The first two items double as the GAD-2 gateway screen (0 to 6, positive at 3 or above). The official PHQ Screeners terms permit reproduction, translation, display, and distribution without permission or fee, and the developers describe the instrument family as public domain, which covers EHR builds, portals, apps, and printed forms; roughly 90 translations are distributed the same way, though translation availability is not the same thing as language-specific validation. The load-bearing distinction for documentation is that the GAD-7 measures two weeks of symptoms while the GAD diagnosis requires about six months of excessive, hard-to-control worry: in the original validation, only 67% of patients scoring 10 or above reported symptoms lasting six months, so roughly a third of positive screens fail the duration element alone. The scale is also deliberately transdiagnostic in practice: at the cutoff of 10 it flagged panic disorder, social anxiety disorder, and PTSD with lower sensitivity than GAD itself, which makes a high score a useful anxiety-distress signal and a weak classifier. And it contains no self-harm question, so it can never stand in for a risk screen . Its natural partner is the PHQ-9 , documented once each with a single integrated interpretation. ## Who uses GAD-7 documentation and when Primary care teams document GAD-7 results inside anxiety screening workflows driven by USPSTF recommendations; therapists and psychiatric prescribers trend it for measurement-based care; care managers track it in collaborative care registries, where the billing codes require a validated scale without naming one; and accredited behavioral health organizations use it to satisfy standardized outcome-monitoring expectations. It anchors the anxiety column of an outcome measure note and usually travels with the PHQ-9. Neighbors matter at the boundaries: the Beck Anxiety Inventory is a licensed Pearson instrument with a heavier somatic emphasis, PROMIS anxiety measures carry their own reproduction terms, and Australian Better Access workflows name the K10 and DASS-21 as their example outcome tools, so a GAD-7 entry there rests on the clinician-discretion clause and should say why an anxiety-specific, two-week measure fits. ## How to document GAD-7 results in the chart No statute prescribes a GAD-7 note format. The elements below are the convention that survives review because each maps to something a future clinician, an accreditation surveyor, or an auditor will need. Most results fit in a few lines beside the paired PHQ-9. Each element carries the pitfall that most often undermines it. Instrument and administration context. Name the form (GAD-7 or the two-item GAD-2), the mode (paper, portal, tablet, telephone, interview), the language, and any assistance. Pitfall: treating administration detail as noise. A randomized 2026 study shifted scores 2.63 points, enough to cross a band, by changing instructional emphasis alone, and a paper-versus-smartphone comparison found only moderate agreement between modes. Total over 21 with the published band. Write the fraction and the band: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 21 severe. Pitfall: importing the PHQ-9's five bands onto this four-band scale, or calling 15 moderate; 15 begins the severe band, an error live on at least one currently ranking documentation page. The functional-difficulty item, separately. Record the unscored response about work, home, and relationships beside the total, never inside it. Pitfall: an eighth scored item. The validated range is 0 to 21, and the functional answer is the score's impairment context, not its arithmetic. Trajectory, in points and percent. Name the baseline score and date, then both numbers. Pitfall: one universal change threshold. The published minimal clinically important difference is about 4 points from a chronic-depression sample, reliable change has been estimated near 6, the 2026 MBHR2 quality measure defines response as a 25% reduction at 4 to 8 months, and research conventions use 50%; name the construct you are claiming. An interpretation sentence. State severity, concordance with the interview and function, and, for a prospective GAD diagnosis, the duration bridge: the scale asks about two weeks, the diagnosis requires about six months. Pitfall: "GAD-7 = 12, generalized anxiety disorder" charted from the score. In the original validation, a third of patients above the screening threshold did not report six months of symptoms. What the score did not assess. When the picture includes panic attacks, trauma symptoms, social fear, or safety concerns, say how they were assessed, because the GAD-7 under-detects those disorders and contains no self-harm item; risk gets its own documentation in a risk screening note . Pitfall: a low total charted as anxiety ruled out. Pooled meta-analytic sensitivity at the cutoff of 10 is about 0.74, and lower for panic, social anxiety, and PTSD. Plan linkage and the next measurement. State what continues, changes, or is deferred because of the result, and when the next administration is due. Pitfall: citing an every-session rule as the instrument's requirement. Every-session administration is a program convention (Ontario Structured Psychotherapy runs one), not a steward rule; accreditation standards leave the interval to the organization. ### Blank template (copy and adapt) GAD-7 DOCUMENTATION BLOCK Form: [GAD-7 / GAD-2] Language: [ ] Mode: [paper / portal / tablet / interview] Date: [ ] Total: [ ]/21 Band: [minimal 0-4 / mild 5-9 / moderate 10-14 / severe 15-21] Functional difficulty (unscored): [not difficult ... extremely] Prior score / date: [ ] Change: [points] ([percent]) Construct: [MCID ~4 / reliable change ~6 / response per named convention / remission <5 (local convention)] Paired measures this visit: [PHQ-9 __/27 or n/a] Interpretation: [severity + concordance with interview and function; for new GAD dx: duration bridge, two weeks measured vs six months required] Not assessed by this scale: [panic / trauma / social anxiety / safety, and how each was addressed] Plan linkage: [continued / changed / deferred, and why] Next measurement: [interval + reason] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. ## Sample GAD-7 documentation (fictional) Scenario: an intake where the GAD-7 is elevated, the paired PHQ-9 is not, and the clinician needs the note to carry the duration bridge from a two-week screen toward a six-month diagnosis. This is the score documentation only, not the full intake. All details are fictional. Patient: R.M., 29 · Visit: Intake assessment · Clinician: D. Whitfield, PhD, Licensed Psychologist · Note date: 08/13/2026 Measures: GAD-7 (English, patient portal, self-completed 08/13/2026): 14/21, moderate anxiety symptoms. Functional difficulty reported as very difficult. PHQ-9 same visit: 5/27, mild range, item 9 = 0. No prior administrations. Duration and context (interview): R.M. describes excessive, difficult-to-control worry across work performance, finances, and family health on most days for approximately nine months, with initial insomnia and concentration interference. The two-week score and the interview history are concordant, and the duration element the questionnaire cannot measure is met by history. Interpretation: The GAD-7 result is a positive severity screen consistent with generalized anxiety disorder, which was diagnosed today by clinical interview covering duration, impairment, and panic, trauma, social-anxiety, substance, medical, and mood alternatives. The score did not establish the diagnosis. Discrete panic episodes denied; no trauma-related symptoms reported. Safety screening was completed and documented separately in the intake risk section. Plan linkage: Today's 14/21 is the measurement baseline. Begin weekly CBT targeting worry and intolerance of uncertainty with behavioral sleep strategies; re-administer the GAD-7 in four weeks with the PHQ-9 alongside, and interpret change against the baseline in points and percent with the convention named. Result reviewed with R.M., who found the severity band consistent with her experience. This sample is fictional and for educational purposes. It does not describe a real patient or record, and the scores are invented for illustration and correspond to no real person. ↑ Back to the template and downloads ### Why this sample works It names the instrument, mode, language, and date, so later administrations are comparable and administration drift cannot masquerade as clinical change. The total sits over 21 with the correct four-band descriptor, and the unscored functional item stays out of the arithmetic. The duration bridge is explicit: the note says what the two-week screen showed and how the six-month diagnostic element was established by interview. The paired PHQ-9 appears once with a single integrated interpretation instead of a duplicated narrative under each instrument. The score links to a baseline, a plan, and a dated next measurement, and safety is documented in its own section rather than inferred from an instrument that never asks about it. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations No law in the US, Canada, or Australia mandates the GAD-7 by name; the operative expectations are PAYER POLICY and CONVENTION, and for anxiety they are thinner than clinicians assume. The USPSTF recommends anxiety screening for adults 19 through 64 (Grade B, 2023) with an insufficient-evidence statement for 65 and older, and for ages 8 to 18 (Grade B, 2022), naming no instrument and no interval. From there the US quality infrastructure is asymmetric: as of 2026 there is no national HEDIS or MIPS anxiety-screening measure comparable to the depression set, and the anxiety-specific reporting that does exist lives in a registry: MBHR2, Anxiety Response at 6 Months , takes adults with an anxiety diagnosis and an index GAD-7 of at least 8 and defines response as a 25% reduction at 4 to 8 months, a definition that matches neither the 50% research convention nor any severity band, which is exactly why the note should name the convention it is using. Collaborative care billing (99492 through 99494) requires validated rating scales and a registry without naming an instrument. In accredited behavioral health organizations, Joint Commission standard CTS.03.01.09 (guidance updated May 2026) expects a standardized instrument administered at multiple intervals with results feeding goals and treatment decisions, and leaves the instrument and cadence to the organization. For 96127 there is no national unit rule: state Medicaid programs and commercial plans publish materially different limits, so verify the current plan-specific policy and date the claim. Cross-border, the labels matter. In Canada, Ontario Health's anxiety-disorders quality standard calls for validated severity measures and regular monitoring without mandating a tool, and Ontario Structured Psychotherapy directs every-session GAD-7 administration inside that program, a workflow rule that should never be cited as a national requirement. In Australia, Better Access requires an outcome tool in a mental health treatment plan unless clinically inappropriate and names the K10 and DASS-21 as examples; the GAD-7 fits through the clinician-discretion clause, and the plan-review cadence rules (generally not more often than every three months) govern plan reviews, not how often a brief measure may be administered. On measurement integrity: keep the standardized wording and instructions, because a 2026 randomized study moved scores 2.63 points, enough to cross a severity band, by changing instructional emphasis alone; record the mode, because paper and smartphone administrations agreed only moderately in a randomized crossover; and record the language, because the roughly 90 distributed translations are not all independently validated. The GAD-7 may be used, reproduced, translated, displayed, and distributed without permission or fee under the official PHQ Screeners terms, and its developers describe the PHQ and GAD instruments as public domain; one instrument registry currently labels it copyrighted by Pfizer, so this page states the operative permission rather than a formal dedication. BastionGPT is not affiliated with, or endorsed by, Pfizer or the instrument authors. This page reproduces no proprietary test items, norms, or scoring materials from any licensed instrument. ↑ Back to the template and downloads ## Common GAD-7 documentation errors reviewers flag The evidence base is specific about where this instrument gets over-read. Pooled meta-analytic sensitivity at the standard cutoff of 10 is about 0.74, not the original study's 0.89 ; in the original validation itself, only 67% of patients at 10 or above reported the six months of symptoms the GAD diagnosis requires; and a 2026 randomized study shifted scores 2.63 points with an instruction-emphasis change and no treatment at all. Even the severity bands get mangled: one currently ranking documentation page labels a score of 15 moderate when 15 begins the severe band. The BastionGPT Clinical Advisory Board sees the same errors most often in GAD-7 documentation reviews: Bands imported or invented. The PHQ-9's five bands applied to this four-band, 0 to 21 scale, a 15 called moderate, or "GAD-7 positive" with no cutoff named when 8, 10, and program-specific thresholds all circulate. Write the fraction, the band, and, where a cutoff matters, which one. A screen promoted to a diagnosis. "GAD-7 = 12, GAD" charted from the score alone. The scale measures two weeks; the diagnosis requires about six months of excessive, hard-to-control worry plus impairment and differentials, and a third of above-threshold patients in the validation sample failed the duration element alone. The interview carries the diagnosis; the score carries severity. Change claimed without a construct. A 4-point drop labeled clinically meaningful (that estimate came from a chronic-depression sample), a 6-point reliable-change threshold and a 25% quality-measure definition and a 50% research convention used interchangeably, or "improved" with no baseline. Points, percent, and the named convention belong in the note. A low score treated as anxiety ruled out. With pooled sensitivity near 0.74 at the cutoff, and lower for panic disorder, social anxiety disorder, and PTSD, a low total cannot close the question when the presentation argues otherwise, and it says nothing about risk because the instrument has no self-harm item. Administration drift read as clinical change. Reworded items, coached responses, a switch from paper to phone, or an unrecorded translation, then a serial comparison across the change. Preserve standardized wording, record mode and language every time, and flag any administration change when interpreting the trend. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Paste the responses, or just the total and the prior score, and get a documentation-ready block: band, trajectory in points and percent, the paired PHQ-9 integrated once, and the duration-bridge sentence when a new GAD diagnosis is on the table. Cross-check a finished note for the gaps reviewers flag: a band that disagrees with its total, a diagnosis resting on the score alone, a functional item folded into the arithmetic, or a change claim with no named construct. Summarize a serial score history against the convention you name (reliable change, percent response, or a registry definition) for treatment reviews and collaborative care registries. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do GAD-7 scores actually mean? The total runs 0 to 21 across seven items, and the four published bands are 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, and 15 to 21 severe. Those describe reported symptom frequency over the past two weeks, not a diagnosis. For scale: in a large general-population sample, only about 5% of adults scored 10 or above and about 1% scored 15 or above, so a moderate or severe band is genuinely uncommon background noise. A score of 10 or higher is the conventional threshold for further evaluation, and what follows it should be an interview, not a diagnosis line. Is a GAD-7 of 10 or more a GAD diagnosis? No. The mismatch is structural: the scale asks about two weeks, while the GAD diagnosis requires excessive, difficult-to-control worry on most days for about six months, plus impairment and the differential work a questionnaire cannot do. In the original validation, only 67% of patients scoring 10 or above reported six months of symptoms, so roughly a third of positive screens failed the duration element alone. The defensible chart language is "positive severity screen, consistent with GAD once the interview establishes duration and rules out the alternatives," with the diagnostic evaluation carrying the conclusion. What counts as meaningful change between two GAD-7 scores? Name the construct, because several circulate and they disagree. The published minimal clinically important difference is about 4 points, estimated in a chronic-depression sample; reliable change, the threshold for exceeding measurement error, has been estimated near 6 points; the 2026 MBHR2 quality measure defines response as a 25% reduction at 4 to 8 months from an index score of at least 8; and research programs commonly use a 50% reduction, with remission below 5 as a local convention. A note that says "16 to 8, down 8 points, a 50% reduction, meeting this clinic's response convention" survives review; "improved" does not. Can a low GAD-7 rule out an anxiety disorder? No. Pooled meta-analytic sensitivity at the cutoff of 10 is about 0.74 , and the follow-on validation found it lower still for the neighbors: 0.74 for panic disorder, 0.72 for social anxiety disorder, and 0.66 for PTSD. A patient with disabling panic attacks or trauma symptoms can sit comfortably under every GAD-7 threshold. When the presentation argues with the score, document both and let the interview decide; and remember the instrument has no self-harm item, so safety is always assessed and documented separately. How often should the GAD-7 be repeated? No steward or national authority sets a cadence. The USPSTF found no evidence for an optimal screening interval; Joint Commission outcome-monitoring expectations leave the interval to the organization; and every-session administration is a program rule where it exists (Ontario Structured Psychotherapy runs one), not an instrument rule. A defensible pattern ties frequency to treatment phase: more often during active treatment changes, less often in maintenance, always at decision points, written into the plan with the next date. Billing follows its own rules: collaborative care requires systematic measurement, and 96127 unit limits are payer-specific. Do we need permission to use the GAD-7 in our EHR or forms? No. The official PHQ Screeners terms permit reproduction, translation, display, and distribution without permission or fee, and the developers describe the family as public domain (one instrument registry still carries a "copyrighted by Pfizer" label, which does not change the operative permission). Two cautions: a reworded or restructured version is no longer the validated instrument, and the neighbors are licensed differently. The Beck Anxiety Inventory is a paid Pearson instrument, and PROMIS measures carry their own reproduction terms, so the free-to-copy habit stops at the GAD-7's edges. How do I document the GAD-7 and PHQ-9 from the same visit? Each score once, then one integrated interpretation: "PHQ-9 = 11, previously 15; GAD-7 = 8, previously 13; both domains improving, residual sleep disruption tracks the depressive symptoms; plan below." The two scales correlate around 0.75 while still measuring separable constructs, so duplicated narratives under separate headings add length without information. Where a service wants a single combined trend, the published PHQ-ADS composite sums both instruments into a 0 to 48 measure with cutpoints of 10, 20, and 30, which is a documented option, not a requirement. The PHQ-9 documentation page covers the depression-side specifics, including item 9. Does the GAD-7 screen for suicide risk? No. None of the seven items asks about self-harm, death, or suicide, so a GAD-7 of any value says nothing about risk in either direction, and a note should never cite it as a risk screen. Safety gets its own assessment and its own documentation: a risk screening note for the screening layer and a suicide risk assessment when concern exists. When the GAD-7 travels with the PHQ-9, that instrument's item 9 provides a safety flag, with its own documented follow-up rules. Can BastionGPT draft my GAD-7 documentation? Yes. Paste the responses, or just the total and prior score, and it drafts the documentation block: the band, trajectory in points and percent, the paired PHQ-9 integrated once, the duration-bridge sentence when a new GAD diagnosis is in play, and the plan linkage, ready for your review. It can also check a finished note for a band that disagrees with its total, a diagnosis carried by the score alone, or a change claim with no named construct. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: PHQ Screeners, official screener site and terms of use : no-permission reproduction rights, the clinician-verification caveat, and translation distribution; ePROVIDE, GAD-7 registry entry (the coexisting copyright label). Spitzer, Kroenke, Williams & Löwe, 2006, Archives of Internal Medicine, the GAD-7 development and validation study : scoring, bands, operating characteristics, the duration finding, and functional-impairment associations. Kroenke et al., 2007, Annals of Internal Medicine, anxiety disorders in primary care : cross-disorder sensitivity and specificity at the cutoff of 10. Plummer et al., 2016, General Hospital Psychiatry, screening meta-analysis : pooled sensitivity 0.74 and specificity 0.83 at the cutoff of 10; Toussaint et al., 2020, Journal of Affective Disorders (minimal clinically important difference of about 4 points); Bischoff et al., 2020 (reliable change near 6 points). Young et al., 2026, JAMA Network Open, randomized instructional-emphasis study : a 2.63-point between-group score shift from wording emphasis alone; Hasegawa et al., 2023, smartphone-versus-paper crossover (moderate between-mode agreement). USPSTF, anxiety screening in adults (2023) and in children and adolescents (2022). APA-affiliated Mental and Behavioral Health Registry, MBHR2, Anxiety Response at 6 Months (2026 specification): index GAD-7 of at least 8, response as a 25% reduction at 4 to 8 months. NCQA, HEDIS depression measures : the depression-side quality infrastructure with no anxiety-screening equivalent as of 2026. CMS, behavioral health integration billing guidance : validated rating scales and registry requirements for 99492 through 99494. Joint Commission, standards interpretation on outcome measurement (CTS.03.01.09) , guidance updated May 2026. Ontario Health, anxiety disorders quality standard ; Ontario Structured Psychotherapy program materials, GAD-7 administration guidance (every-session program convention). Australian Government Department of Health, Disability and Ageing, Better Access professional fact sheet (March 2026): outcome-tool requirement with K10 and DASS-21 as named examples. ## Related templates PHQ-9 Score Documentation C-SSRS Documentation Outcome Measure Review Note General Risk Screening Note Collaborative Care (BHI/CoCM) Note Better Access Mental Health Treatment Plan Referral DASS-21 Score Documentation PSC-17 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/c-ssrs All note templates # C-SSRS Documentation: Versions, Categories & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The C-SSRS (Columbia-Suicide Severity Rating Scale) is a widely adopted instrument that classifies suicidal ideation and behavior into defined categories. Hospitals, clinics, schools, and research programs use it to standardize screening and the response that follows. This page covers how to document C-SSRS administration and results in the chart, with a fictional, structure-only sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Anyone may administer the healthcare versions per the steward; clinicians own interpretation, assessment, and follow-up Audience Treating teams and consultants, accreditation surveyors and tracers, quality and risk managers, attorneys reviewing records Typical length 4 to 10 chart lines for the screen, plus the assessment a positive result triggers · screener completion 1 to 3 minutes Format family Structured ideation and behavior classification; category results, not a summed score When it's used Suicide risk screening at intake and transitions, ED triage, inpatient monitoring, VA Risk ID, clinical trial safety monitoring Standards context Free for healthcare use and EHR embedding; copyright held by RFMH; major regulators require a validated process, not this tool by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the C-SSRS? The Columbia-Suicide Severity Rating Scale (C-SSRS), also distributed as the Columbia Protocol, is a structured instrument for identifying and classifying suicidal ideation and behavior, developed by Kelly Posner and colleagues for an NIMH-funded adolescent treatment study and first validated at scale in 2011. It is disseminated by the Columbia Lighthouse Project under Columbia University auspices, and its copyright is held by the Research Foundation for Mental Hygiene, Inc. The licensing is bifurcated and widely misunderstood: healthcare and community use is free, including creating setting-specific triage steps and embedding the tool in an EHR, while pharmaceutical-funded clinical trials require a per-protocol license, with the electronic eC-SSRS routed through Clario. Training splits the same way: the steward recommends but does not require training for clinical use, while research raters must train and re-train every two years. The version family is the first documentation trap, because materially different instruments share the name: brief screeners (Recent and Since Last Contact), full scales (Lifetime/Recent for baselines, Since Last Visit for follow-up), pediatric and very-young-child adaptations, a frequent-monitoring screener for shift-based settings, an inpatient-discharge screener, and self-report versions. "C-SSRS completed" identifies none of them. The second trap is the output: the instrument classifies ideation categories and behavior categories with recency, and it produces no summed total; the steward's own materials put "score" in quotation marks and map affirmative answers to example triage tiers. A defensible note therefore records the category-level result, the local triage decision, and the clinician's overall risk formulation as three different things, because they are. ## Who uses C-SSRS documentation and when Hospitals and behavioral health organizations document C-SSRS results inside accreditation-driven screening workflows; emergency departments and inpatient units use the setting-specific screeners at triage, during stays, and at discharge; the Veterans Health Administration mandates the screener internally through its Risk ID strategy, with a comprehensive evaluation required the same day in ambulatory care and within 24 hours in inpatient and emergency settings; schools, corrections, and crisis lines use the community versions; and clinical trials use licensed versions for prospective safety monitoring under FDA draft guidance. The documentation always sits inside a larger chain: a positive screen feeds a suicide risk assessment , mitigation lands in a safety plan , acute encounters generate a crisis note , and the generic screening-layer architecture lives in a risk screening note . This page owns the instrument-specific layer: version, categories, and the record that connects them to action. ## How to document C-SSRS results in the chart No regulator prescribes a C-SSRS note format, but surveyors trace the record as a chain: covered patient, validated tool administered as designed, categorical result, response to a positive, assessment, overall risk with rationale, mitigation, and follow-up. The elements below make that chain visible. Each carries the pitfall that most often breaks it. Exact version and administration circumstances. Name the instrument and version (Screener Recent, full Lifetime/Recent, Since Last Contact, and so on), the timeframe it covers, the mode, the date and time, the administrator's role, and any language, privacy, or participation limits. Pitfall: "C-SSRS done." Different versions measure different timeframes with different items, and the record cannot be traced, compared, or defended without the version. The category-level result. Record the highest ideation category identified, the behavior category or "none identified," the relevant recency, and whether the result is positive or negative under the named local protocol. Pitfall: a bare number or color. "C-SSRS score 4" is ambiguous (item, category, intensity subtotal, or local tier?), and "red" without the underlying category and timeframe documents the wall chart, not the patient. Branching fidelity. Note that all applicable branches were completed, and record anything that limited participation. Pitfall: partial administration read as negative. The instrument's branching is part of its validity, and accreditation guidance warns against altering validated wording or skipping the flow. The screen-to-assessment chain. A positive screen leads to an evidence-based assessment covering ideation, plan, intent, behavior, risk factors, and protective factors, documented in a suicide risk assessment the same visit. Pitfall: the screen standing in for the assessment. Screening organizations that stop at the screener are among the most-cited survey findings, and a negative screen never closes an inquiry the clinical picture keeps open. Overall risk level, with the reasoning. Where policy requires a categorical level, document it under the organization's rubric with the clinical rationale: current state against baseline, dynamic and historical factors, protective factors that are specific and currently accessible, collateral information, and foreseeable changes. Pitfall: the screener category pasted in as the overall level. The instrument result, the local triage tier, and the clinician's formulation are three outputs, and reviewers expect the third to carry reasoning, not a color. Actions completed, not recommended. Document the safety plan work, means-safety counseling at a non-method-specific level, monitoring status, consultation and its outcome, confirmed referral or handoff with a named receiver, disposition, and who owns the next contact. Pitfall: a positive result with no visible response, or a reflex disposition. No authority makes hospitalization or one-to-one observation automatic; the steward's own experience is that 1 to 2% of people screened need more acute care, and the record should show why the chosen response is proportionate. Reassessment cadence and triggers. Record when re-screening is due under written policy and what changes trigger immediate reassessment. Pitfall: invented universal intervals. "Every encounter" is a Zero Suicide convention, annual screening is a VHA rule inside VA care, and the frequent-monitoring screener exists precisely because cadence is a local policy decision. ### Blank template (copy and adapt) C-SSRS DOCUMENTATION BLOCK Version: [Screener Recent / Since Last Contact / full Lifetime-Recent / Since Last Visit / other] Timeframe: [ ] Date/time: [ ] Administered by (role): [ ] Mode: [ ] All applicable branches completed: [yes / limits noted] RESULT (category level, never method detail) Highest ideation category: [ ] Behavior category: [ / none] Recency: [ ] Result under [named local protocol]: [pos / neg] IF POSITIVE -> same-visit evidence-based assessment [ideation, plan, intent, behavior, risk factors, protective factors; documented in the risk assessment note] OVERALL RISK (clinician formulation, org rubric) Level: [ ] Rationale: [state vs baseline; dynamic + historical factors; protective factors and their current accessibility; collateral; foreseeable changes] ACTIONS COMPLETED [safety plan updated / means-safety counseling / monitoring status / consultation + outcome / confirmed handoff to (name)] Disposition + why proportionate: [ ] FOLLOW-UP Next contact owner: [ ] Reassessment due: [per policy] Immediate-reassessment triggers: [status change / new ideation / new behavior / care transition] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. ## Sample C-SSRS documentation (fictional) Scenario: an outpatient visit where the brief screener is positive for active ideation without intent, the clinician completes the assessment the same visit, and care continues safely outpatient. The sample shows the documentation structure only: it names categories and actions, and deliberately contains no method, means, or item content. All details are fictional. Patient: J.L., 41 · Visit: Outpatient follow-up · Clinician: A. Reyes, LPC · Note date: 08/11/2026 Screen: C-SSRS Screener Basic, Recent version, administered verbally in private at 10:20 by this clinician; all applicable branches completed. Highest ideation category: active suicidal ideation without intent. No suicidal-behavior category identified in the screener's timeframe. Result is positive under the clinic's written screening protocol. Same-visit assessment: An evidence-based suicide risk assessment was completed this visit and is documented in today's risk assessment note: current and historical ideation and behavior, intent, mental status, dynamic and historical risk factors, protective factors and their current accessibility, and collateral information from J.L.'s spouse with consent. No prior suicidal behavior by history or record review. Overall risk and rationale: Elevated from J.L.'s baseline but managed outpatient under the clinic's formulation rubric. Reasoning: ideation is recent and active but without intent, plan, or preparatory behavior; presentation is engaged and future-oriented; protective factors (children at home, treatment alliance, involved spouse) are current and accessible; the identified stressor is acute and being addressed. The screener category informed but did not determine this formulation. Actions completed today: Collaborative safety plan updated with J.L.; means-safety counseling provided and arrangements confirmed; supervising psychiatrist consulted, concurred with outpatient management; spouse engaged with J.L.'s consent; crisis and after-hours contacts reviewed, including 988. Disposition: continue outpatient care with increased contact, proportionate to the assessment above. Follow-up: This clinician owns next contact: telephone check-in within 48 hours and appointment in 5 days. Re-screen at the policy interval, and immediately on any status change, new ideation, new behavior, or transition of care. This sample is fictional and for educational purposes. It does not describe a real patient or record, shows documentation structure only, and intentionally contains no method or means content. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). ↑ Back to the template and downloads ### Why this sample works The exact version, timeframe, mode, administrator, and branching fidelity are named, so the result is traceable and comparable at the next administration. The result is recorded at category level with its recency and its status under a named local protocol, never as a bare number or color. The chain is visible: positive screen, same-visit evidence-based assessment, overall formulation with reasoning, and actions that were completed rather than recommended. The overall risk level is the clinician's formulation with stated rationale, not the screener category pasted forward, and the disposition is justified as proportionate. Follow-up has an owner, a date, and named reassessment triggers, which is exactly the sequence a surveyor tracer or a reviewing attorney will walk. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The C-SSRS is widely adopted, and almost nothing mandates it by name: that distinction shapes defensible documentation. Joint Commission accreditation requires a validated suicide screening tool and names the C-SSRS only as an example, and its 2026 reorganization split the standard's location: hospitals and critical access hospitals now work under NPG.08.01.01 while behavioral health care programs remain under NPSG.15.01.01, with closely aligned elements: screening from age 12 for patients evaluated or treated primarily for behavioral health conditions, an evidence-based assessment after a positive screen covering ideation, plan, intent, behavior, risk factors, and protective factors, a documented overall risk level and mitigation plan, written policy on reassessment and monitoring, discharge counseling and follow-up, and quality monitoring. CMS is expressly tool-neutral: its March 2026 interpretive guidance states it does not endorse or require a particular instrument, while the underlying regulations require care in a safe setting and psychiatric records that justify the degree and intensity of treatment. Inside VA care, PAYER POLICY is specific: the Risk ID strategy uses the C-SSRS Screener with annual universal screening, and a positive screen requires the comprehensive suicide risk evaluation the same day in ambulatory care and within 24 hours in inpatient and emergency settings. LAW appears mostly at the state layer: California requires general acute-care hospitals to screen patients 12 and older under written policies, and a dozen states mandate suicide-specific training for defined professions; none of these makes the C-SSRS itself compulsory. Zero Suicide's every-encounter screening is a CONVENTION worth naming as one in policy documents. Canada and Australia run the same closed loop with different labels, and one genuine divergence. Accreditation Canada applies Health Standards Organization's suicide prevention program standard as a required practice (the current edition uses Required Safety Practice terminology; older materials say Required Organizational Practice), covering routine screening, assessment, safety planning, transitions, and postvention, all tool-neutral. Australia's NSQHS Comprehensive Care actions 5.31 and 5.32 require systems that identify and respond to suicide and self-harm risk, and the Commission's June 2026 fact sheet , which supersedes the old AS18/14 advisory, explicitly calls screening without resulting action and carrying forward old risk ratings without reassessment low-value care. The divergence: US accreditation requires a documented overall risk level, while NSW Health policy discourages high/medium/low stratification as the basis for care decisions, citing evidence that most people categorized high-risk never die by suicide and about half of deaths arise from lower-risk groups. A defensible chart works in both worlds the same way: record the category-level result, then carry the weight in the formulation and the actions rather than in a tier label. On the instrument itself, keep the validated wording and branching intact, record refusals as refusals rather than negatives, and remember the prediction limits: screening organizes detection and response, and no instrument accurately predicts individual outcomes. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The C-SSRS is disseminated by the Columbia Lighthouse Project under Columbia University auspices; its copyright is held by the Research Foundation for Mental Hygiene, Inc. Healthcare and community use, including EHR embedding, is free under the steward's terms, and clinical-trial use requires a per-protocol license. BastionGPT is not affiliated with, or endorsed by, the Columbia Lighthouse Project, Columbia University, or the Research Foundation for Mental Hygiene. This page reproduces no instrument items, stimuli, or scoring materials. ↑ Back to the template and downloads ## Common C-SSRS documentation errors reviewers flag The evidence sets the tone: this instrument classifies and organizes, it does not predict. In a cohort of 18,684 psychiatric-emergency patients, the screener's one-week discrimination ran near 56.5% sensitivity and 75.6% specificity ; across the risk-categorization literature, roughly 95% of patients categorized high-risk do not die by suicide while about half of deaths arise from lower-risk groups ; and the first C-SSRS-specific meta-analysis (2026) found meaningful associations with later nonfatal attempts but could not estimate fatal-outcome prediction at all. The steward's own experience is that 1 to 2% of people screened need more acute care. The BastionGPT Clinical Advisory Board sees the same errors most often in C-SSRS documentation reviews: A number or a color charted as the result. "C-SSRS score 4" could be an item, a maximum ideation category, an intensity subtotal, or a local tier; "red" names the wall chart. Record the highest ideation category, the behavior category or none, the recency, and the result under the named local protocol. The screen charted as the assessment. Screening organizations that stop at the screener are a recurring accreditation survey finding. A positive screen leads to a same-visit evidence-based assessment with its own documentation; a single process counts as both only when it genuinely covers ideation, plan, intent, behavior, risk factors, and protective factors. The screener category pasted in as the overall risk level. The required overall level is the clinician's formulation with reasoning: state versus baseline, dynamic and historical factors, protective factors that are current and accessible, collateral, and foreseeable changes. A tier with no rationale reads as a checkbox in a tracer and in a courtroom. A positive screen treated as an automatic disposition. No steward, accreditor, or regulator makes hospitalization or one-to-one observation the default response, and reflex dispositions cut both ways: over-restriction the record cannot justify, or a discharge with no visible reasoning. Document why the chosen response is proportionate. Refusals and negatives over-read. A refusal is not a negative screen: document the version offered, the refusal, the alternative sources used, the limitations, and the plan. And a negative screen never closes an inquiry that history, behavior, collateral, or presentation keeps open; charting should never cite it as the reason assessment stopped. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the category-level facts (version, highest ideation category, behavior category, recency, protocol result) and it drafts the documentation chain for your review: screen block, assessment scaffold, formulation-with-rationale prompts, and completed-actions and follow-up sentences. Cross-check a finished note for the gaps surveyors trace: a positive screen with no same-visit assessment, an overall level with no reasoning, a missing version or timeframe, or follow-up with no named owner. Draft the handoff summary: version and timeframe, category-level result, formulation status, safety plan status, monitoring, and the trigger for the next reassessment, ready to confirm with the receiving clinician. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Is there such a thing as a C-SSRS score? Not in the PHQ-9 sense. The instrument classifies the highest ideation category reached and any behavior category with its recency; research protocols may code categories numerically and total a separate intensity component, but there is no validated summed clinical score, and the steward's own materials put "score" in quotation marks before mapping answers to example triage tiers. A chart entry like "C-SSRS score 4" is ambiguous four different ways. Document the category, the timeframe, and the result under your named local protocol instead. Which C-SSRS version do I document at intake versus follow-up? Use and name the version that matches the clinical job: the full Lifetime/Recent scale when a comprehensive baseline history is needed, the brief Screener (Recent) for front-line identification, Since Last Contact or Since Last Visit forms after a prior administration, the frequent-monitoring screener in shift-based inpatient workflows, the inpatient-discharge screener at that transition, and pediatric or very-young-child adaptations where age requires them. The setting-specific triage layouts change the suggested next steps, not the questions. Whatever you choose, the note records the exact version and timeframe, because results from different versions are not interchangeable. Does a positive C-SSRS mean the patient must be hospitalized? No. No steward, accreditor, or regulator reviewed for this page makes admission, involuntary treatment, or one-to-one observation the automatic consequence of a positive screen; the steward's published experience is that 1 to 2% of people screened need more acute care. A positive result triggers the locally defined response, normally a same-visit evidence-based assessment , and disposition follows the full formulation, legal criteria, available supports, and the least restrictive safe option, with the reasoning documented. What reviewers flag is not the disposition itself but a disposition with no visible rationale. Is the C-SSRS required by law or by Joint Commission? Generally no authority requires it by name. Joint Commission requires a validated screening tool and lists the C-SSRS as an example; since 2026, hospitals work under NPG.08.01.01 while behavioral health programs remain under NPSG.15.01.01. CMS states expressly that it does not endorse or require a particular tool. Canada's accreditation requirement and Australia's NSQHS framework are tool-neutral. The clear exception is internal: the VA's Risk ID strategy mandates the C-SSRS Screener within VHA care. So the defensible description is widely adopted to satisfy validated-process requirements, and the defensible chart names the local policy the result was interpreted under. Does the C-SSRS screener replace a suicide risk assessment? No, and collapsing the two is the most-cited survey finding in this territory. The screener detects and classifies; the assessment integrates ideation, plan, intent, behavior, risk factors, protective factors, collateral, and clinical judgment into a formulation with a documented overall risk level and mitigation plan. One process can satisfy both roles only when it genuinely covers all the assessment domains. A negative screen does not end the obligation either: when history, behavior, or presentation raises concern, the assessment proceeds anyway, and the note says why. How often does the C-SSRS need to be repeated? There is no universal interval. Joint Commission requires each organization's written policy to define reassessment frequency and triggers (status change, new ideation, new behavior); Zero Suicide's every-encounter screening is a convention its adopters choose; the VA's annual universal screening is internal VHA policy; and the steward publishes a frequent-monitoring screener precisely for settings that re-screen by shift or day under local rules. Document the cadence your policy sets, the next due date, and the triggers, and re-screen immediately at material transitions rather than carrying an old result forward; Australian guidance now explicitly calls stale carried-forward risk ratings low-value care. Do we need a license or training to use the C-SSRS? For healthcare and community use, no license and no fee: the steward's terms allow use, setting-specific triage steps, and EHR embedding without permission, and training is recommended rather than required, with free modules in about 30 languages. The boundaries: pharmaceutical-funded clinical trials require a per-protocol license (copyright sits with the Research Foundation for Mental Hygiene, and the electronic eC-SSRS routes through Clario), research raters must train with re-training every two years, and your own organization or accreditor may mandate training as a competency requirement even though the steward does not. What do I document when a patient declines the screen? A refusal is not a negative screen, and the chart should never convert one into the other. Document the exact version offered, that the patient declined or could not participate, any stated reason, the steps taken to make participation easier, the sources used instead (observation, records, collateral), the limits created by the missing self-report, the clinical assessment and safety actions taken anyway, and the reassessment plan. When concern remains despite refusal, follow local law and policy, and record the reasoning; the liability literature is consistent that over-reliance on a single declined self-report is the pattern that fails review. Can BastionGPT draft my C-SSRS documentation? Yes, for the documentation layer. Give it the category-level facts and it drafts the screen block, the assessment scaffold, the formulation-with-rationale prompts, and the completed-actions and follow-up sentences for your review; it can also check a finished note for a positive screen with no assessment, a level with no reasoning, or a missing version. The clinical decisions (risk formulation, disposition, safety planning) remain yours, and the tool is built to keep them that way. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Columbia Lighthouse Project: about the scale , healthcare and community permissions , risk identification , pharmaceutical licensing , and research training requirements . Joint Commission, suicide prevention portal and standards-interpretation FAQs on reassessment and screening and assessment : the 2026 NPG.08.01.01 / NPSG.15.01.01 split and the element chain. CMS, State Operations Manual Appendix A transmittal (March 2026): tool-neutral screening strategy; 42 CFR 482.13 and 482.61 . Zero Suicide toolkit, screening and assessment and risk formulation : screening for care planning, not prediction. VA/DoD, 2024 suicide risk clinical practice guideline ; VHA, Directive 1160.07 (Risk ID context). California Health & Safety Code, section 1259.6 (hospital screening policies); AFSP, June 2026 state training-law inventory . Health Standards Organization, suicide prevention program standard (Accreditation Canada required practice, tool-neutral). Australian Commission on Safety and Quality in Health Care, Comprehensive Care Standard and the June 2026 risk screening and assessment fact sheet ; NSW Health, PD2022_043 . Posner et al., 2011, American Journal of Psychiatry, initial C-SSRS validation ; Bjureberg et al., 2022, screener performance in 18,684 psychiatric-emergency patients ; Brown et al., 2020, mixed psychometric performance in high-risk ED patients . Daray et al., 2026, first C-SSRS-specific meta-analysis (association with nonfatal attempts; no fatal-outcome estimate; steward-affiliated coauthors). Franklin et al., 2017, Psychological Bulletin, 50 years of risk-factor prediction research ; Large et al., meta-analysis of risk categorization after self-harm . Giddens, Sheehan & Sheehan, 2014, Innovations in Clinical Neuroscience, critique of the classification structure (authors market a competing instrument, relevant context); Simon and Obegi in the Journal of the American Academy of Psychiatry and the Law, probable standards of care for suicide risk assessment . ## Related templates PHQ-9 Score Documentation Suicide Risk Assessment Safety Plan General Risk Screening Note Crisis Psychotherapy Note Self-Harm (NSSI) Assessment MADRS Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/moca All note templates # MoCA Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The MoCA (Montreal Cognitive Assessment) is a 30-point cognitive screening instrument developed by Ziad Nasreddine and validated in 2005, built to detect mild cognitive impairment. Primary care, neurology, and geriatrics teams use it at wellness visits and in memory workups. This page covers how to document and interpret MoCA results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any healthcare professional for total-score screening; certification required to interpret subscores and domains (policy dated December 2025) Audience Referring and treating clinicians, memory clinics and neuropsychologists, payers and care programs, families receiving results Typical length 3 to 8 chart lines · administration about 10 minutes Format family Clinician-administered cognitive screen (30 points; adapted versions scored out of 22 and 15) When it's used Medicare annual wellness visits, memory complaints and MCI workups, delirium recovery tracking, pre-treatment cognitive baselines Standards context Accessed via mocacognition.com; certification optional for total-score screening since December 2025; no authority mandates it by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MoCA? The Montreal Cognitive Assessment (MoCA) is a roughly ten-minute, 30-point cognitive screening instrument developed by neurologist Ziad Nasreddine and validated in 2005 as a tool sensitive to mild cognitive impairment. It is a trademarked instrument accessed through mocacognition.com , and its rules have moved: mandatory universal training, announced in 2019, gave way in a policy dated December 2025 to a tiered model in which certification is optional for healthcare professionals who use only the total score for screening, triage, or referral, and required for anyone interpreting individual tasks, subscores, or domain-specific impairments, with exemptions for neuropsychologists and clinicians with a year of postdoctoral cognitive-assessment training. Training is free for students, faculty, academic researchers, and full-time employees of publicly operated healthcare institutions; the official site no longer publishes a standard paid price for everyone else. One caution the site itself creates: older Terms and Permissions pages still carry universal-certification language, so institutions relying on the exemption should keep a copy of the dated FAQ or obtain written confirmation. The load-bearing documentation facts are arithmetic. The Full versions (8.1, with alternate forms 8.2 and 8.3 for repeat testing) are scored out of 30, with one education correction: add one point for 12 or fewer completed years of education, only when the total is below 30, and chart the raw and adjusted values separately. The version family changes the denominator: MoCA-Basic serves low-literacy populations at /30, the Blind/Telephone version is scored out of 22, and the five-minute version out of 15, and the official FAQ states plainly that converting a /22 result to /30 is not validated. And the famous cutoff of 26 is a sensitivity-oriented threshold from the 2005 validation sample, not a verdict: in a diverse population-based sample, 66% of adults scored below it, and meta-analytic work supports lower cutoffs near 23 in many settings. A MoCA result is a screen that feeds a fuller evaluation ; it diagnoses nothing by itself. ## Who uses MoCA documentation and when Primary care teams document MoCA results at Medicare annual wellness visits, where the law requires detection of cognitive impairment but names no instrument; clinicians billing cognitive assessment and care plan services fold it into a much larger documentation set (independent historian, function, safety, staging instruments, care plan); memory clinics use it as the brief baseline before neuropsychological evaluation and treatment pathways, where payer rules sometimes accept a MoCA range as an eligibility gate; hospital teams track it through delirium recovery, with the during-delirium caveat written in; and capacity and driving questions cite it only as collateral evidence, never as the determinant, alongside a capacity evaluation . Neighbors matter: the MMSE is a licensed Pearson-era incumbent that is less sensitive to MCI, SLUMS and Mini-Cog are free alternatives, RUDAS serves culturally and linguistically diverse populations, and the RBANS is the purchased next step when domain profiling with repeatable forms is needed. ## How to document MoCA results in the chart No statute prescribes a MoCA note format. The elements below are the convention that keeps a screening number interpretable months later, when a different clinician compares against it. Each carries the pitfall that most often undermines it. Clinical state and purpose. Say why the screen was done and the patient's state: routine screening, a complaint, a baseline before treatment, or recovery tracking, and whether delirium, acute illness, or major fatigue was present. Pitfall: a score obtained during delirium charted without that context, then treated as the premorbid baseline forever after. Exact version, language, mode, and denominator. Name the form (Full v8.1, 8.2, or 8.3, Basic, Blind/Telephone, five-minute, hearing-adapted, digital), the language, the mode (in person, telephone, videoconference), and the denominator. Pitfall: "MoCA 18" from a /22 telephone version read against the /30 cutoff. The official FAQ states that converting /22 results, or scores from partially administered Full versions, to /30 is not validated. Raw and adjusted scores, both. Record the raw total, the completed years of education, and whether the one-point Full-version correction was applied (12 or fewer years, only when the total is below 30). Pitfall: one merged number. A raw 24 with correction and an uncorrected 25 look identical downstream, and serial comparison quietly breaks; correction applied at 30/30 or at every education level is simply wrong. Testing conditions. Note hearing aids, corrective lenses, motor limits, literacy, interpreter use, interruptions, and anything that deviated from standard administration. Pitfall: conditions omitted. A later three-point difference cannot be interpreted when nobody recorded that the baseline was done without hearing aids in a noisy room. Domain observations, only when qualified, as patterns. Under the December 2025 policy, interpreting subscores and domains requires certification or an exemption; where qualified, chart relative patterns ("relative weakness in new learning and executive control") rather than localization. Pitfall: mapping a missed task to an etiology. Large registry analyses find domain scores substantially less reliable than the total. Interpretation with the cutoff in context. Below 26 means below the original sensitivity-oriented validation threshold, a flag for further evaluation; it is not "abnormal, has dementia." Meta-analytic work supports cutoffs near 23 in many settings, and a high score does not rule impairment out. Pitfall: verdict language. In one diverse population sample, 66% scored below 26; among cognitively healthy adults aged 70 to 99, 46% did. Function, plan, and serial comparability. Tie the score to daily function and collateral report, state the plan (reversible-cause workup, monitoring, referral), and set up the next administration: same language and mode, an alternate form at intervals of about three months or less, and the exact form recorded. Pitfall: over-reading small changes. Published reliable-change estimates run from roughly two to four points depending on cohort and interval, so a one- or two-point movement is usually measurement noise unless function moves with it. ### Blank template (copy and adapt) MoCA DOCUMENTATION BLOCK Reason / clinical state: [screening / complaint / baseline / recovery tracking; delirium or acute illness present?] Version: [Full v8.1 / 8.2 / 8.3 / Basic / Blind-Telephone / 5-minute / other] Language: [ ] Mode: [ ] Date: [ ] Administered by: [ ] Raw score: [ ]/[30 / 22 / 15] Education: [ ] completed years Correction: [+1 applied / not applied / n/a] Adjusted: [ ]/30 Conditions: [hearing aids / lenses / motor limits / interpreter / interruptions / deviations] Domain observations (certified or exempt only): [relative patterns, not localization] Interpretation: [screening result vs applicable threshold in this population; not a stand-alone diagnosis] Function / collateral: [IADLs, medication and finance management, informant report] Plan: [reversible workup / monitoring / referral / none] Next administration: [interval; alternate form if under ~3 months; same language and mode] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. ## Sample MoCA documentation (fictional) Scenario: an annual wellness visit with a borderline result and a collateral concern, the everyday case where the difference between a defensible note and "MoCA 24" matters most. All details are fictional. Patient: H.B., 74 · Visit: Annual wellness visit · Clinician: S. Okafor, MD (family medicine) · Note date: 08/10/2026 Context: Cognitive screening at the annual wellness visit. H.B. reports occasional word-finding pauses; her husband, present with consent, describes two medication-refill errors in the past three months. No delirium, acute illness, or sleep deprivation today. Measure: MoCA Full v8.1, English, in person, administered by this clinician: raw 24/30. Education 12 completed years; one-point Full-version correction applied; adjusted 25/30. Hearing aids worn, reading glasses used, no interruptions; standard administration throughout. Total-score screening use only; no domain interpretation is charted. Interpretation: The adjusted score falls below the original validation threshold of 26, a screening flag rather than a diagnosis, and population studies show many cognitively healthy older adults score in this range. Taken with the collateral medication-management concern, the result supports further evaluation; it does not establish mild cognitive impairment or dementia. Function and plan: Basic daily activities intact; instrumental activities notable for the refill errors. Plan: medication reconciliation with a simplified regimen, laboratory review of reversible contributors, mood and sleep screening, and a repeat administration in about 12 months using alternate form 8.2 under the same conditions, sooner if function changes. Referral for fuller cognitive evaluation will follow if the concern persists or the trajectory declines. Results discussed with H.B. and her husband in plain language: areas worth watching, nothing diagnosed today. This sample is fictional and for educational purposes. It does not describe a real patient or record, and the scores are invented for illustration and correspond to no real person. ↑ Back to the template and downloads ### Why this sample works The clinical state is on the record (no delirium, standard conditions), so the score can serve as a real baseline later. Version, language, mode, and administrator are named, and raw and education-adjusted scores are preserved separately with the education years. The interpretation places the number against the cutoff's actual meaning: a sensitivity-oriented screening threshold, not a diagnosis, with the population caveat stated. The score is tied to function and collateral report, which is what turns a borderline number into a clinical decision. The plan sets up a comparable serial measurement: alternate form named, same conditions, a defined interval, and an escalation path if function moves. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations In the US, the rules around cognitive screening are mostly tool-neutral, and the note should reflect which layer it is serving. The Medicare annual wellness visit regulation ( 42 CFR 410.15 ) is LAW that requires detection of cognitive impairment through direct observation and patient and informant report; it names no instrument, so "Medicare requires a MoCA" is false, while a MoCA can document the detection work. The cognitive assessment and care plan service (99483) is PAYER POLICY that expects far more than a score: an independent historian, function and decision-making ability, behavioral symptoms, safety including driving, caregiver needs, staging instruments, and a written care plan; "MoCA 20/30" alone supports nothing, and some local contractor articles name a short MoCA as one acceptable documentation route, which is a local rule rather than a national mandate. CMS's GUIDE dementia model runs its tiering on the CDR, FAST, and Zarit instruments, not the MoCA. Treatment gates deserve their own label: Medicare's national anti-amyloid coverage framework sets no numerical MoCA threshold, while some commercial policies (for example, 2026 UnitedHealthcare criteria ) accept a MoCA of 17 to 30 as one eligibility route; those ranges are payer gates, not clinical cutoffs or validated staging. And for the two highest-stakes questions, the score is collateral evidence only: capacity is decision-specific and functional, documented in a capacity evaluation , and driving-fitness guidance (NHTSA and the AGS in the US) treats cognitive screens as triggers for fuller evaluation, with state reporting laws that turn on clinical judgment rather than any score. Canada and Australia run the same logic with their own anchors. The Canadian Medical Association's 2025 driver's guide states that a cognitive screen generally cannot determine fitness to drive by itself except in severe impairment, provincial reporting duties (mandatory in some provinces, discretionary in others) turn on the clinician's overall judgment, and provincial drug plans historically key coverage to MMSE-family scores rather than the MoCA (British Columbia's donepezil criteria use the SMMSE), while Canada's August 2026 national lecanemab recommendation permits clinically accepted alternatives such as the MoCA without setting a numerical threshold. In Australia, Austroads' fitness-to-drive framework sets no MoCA cutoff, the June 2026 aged-care assessment materials name GPCOG and KICA pathways rather than the MoCA, RUDAS remains the culturally responsive alternative for diverse and low-education populations, and current PBS dementia-drug criteria publish no MoCA threshold. Across all three countries, the measurement cautions are the same: nearly 100 translations exist but availability is not validation, remote administrations agree only moderately with in-person testing at the individual level, the digital product line is in transition (MoCA Duo is being retired for MoCA Solo during 2026, and the public XpressO self-test is an awareness tool, not a clinical administration), so the chart names the exact product, language, and mode every time. MoCA is a registered trademark of TEST MOCA INC., and the instrument is distributed through mocacognition.com under its permission terms, which prohibit republishing the form; this page links to the official source and hosts nothing. BastionGPT is not affiliated with, or endorsed by, MoCA Cognition, TEST MOCA INC., or the instrument's developers. This page reproduces no test items, stimuli, forms, scoring keys, or normative tables. ↑ Back to the template and downloads ## Common MoCA documentation errors reviewers flag The over-identification data are the backdrop for most of these. In a diverse population-based sample of 2,653 adults, 66% scored below the conventional cutoff of 26 ; among cognitively healthy adults aged 70 to 99, 46% did ; a Cochrane review found more than 40% of people without dementia screen positive at that threshold; and meta-analytic work supports cutoffs near 23 in many settings. Meanwhile the instrument's own rules moved in December 2025, and most template pages have not noticed. The BastionGPT Clinical Advisory Board sees the same errors most often in MoCA documentation reviews: A bare number. "MoCA 24" with no version, denominator, language, mode, education years, or correction status. With /30, /22, and /15 forms in circulation and an education point that applies only below 30, the bare number is uninterpretable and unserviceable for serial comparison. The 26 cutoff applied as a verdict. "Below 26 = impaired" over-identifies in older, less-educated, and culturally diverse populations, and "26 or above = normal" ignores ceiling effects in highly educated patients. Chart the score against the threshold appropriate to the population, as a flag for evaluation, never as the conclusion. The education correction misapplied. Added at every education level, added to a 30/30, silently baked into a single reported number, or inconsistently applied across serial tests so a one-point "change" is really a bookkeeping artifact. Raw score, education years, and correction status are three separate chart facts. Unvalidated conversions and severity bands. A /22 telephone result prorated to /30 (the official FAQ says this is not validated), a partial administration scored as if complete, or score ranges charted as dementia stages when the FAQ itself concedes parts of the banding were set arbitrarily; staging is functional, not a screen arithmetic. Serial changes over-read. A one- or two-point movement charted as progression or recovery when published reliable-change estimates run roughly two to four points, practice effects inflate early retests, alternate forms differ slightly in difficulty, and state factors (illness, mood, sleep, hearing) move scores. Real change usually brings function along with it; the note should say whether it did. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the score facts (version, raw total, education years, conditions, prior results) and it drafts the documentation block: raw and adjusted values stated separately, the interpretation caveat in place, and the function and plan sentences scaffolded for your review. Cross-check a finished note for the gaps reviewers flag: a denominator mismatch, a correction applied at 30/30 or at the wrong education level, verdict language around the cutoff, or a serial comparison across changed conditions. Summarize a serial score history with measurement-error framing, alternate-form tracking, and the functional corroboration a reviewer will look for. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does a MoCA score actually mean? The Full versions total 30 points, and the familiar threshold of 26 comes from the 2005 validation study , where it caught 90% of mild cognitive impairment and 100% of mild Alzheimer disease in a clinically enriched sample. In broader populations it over-identifies: 66% of a diverse community sample and 46% of cognitively healthy adults aged 70 to 99 scored below 26, and meta-analyses support cutoffs near 23 in many settings. So a below-threshold score is a flag for evaluation of reversible causes, function, and, where indicated, fuller testing; it is not a diagnosis, and a high score does not rule impairment out. Do I need MoCA certification to administer it? Under the policy dated December 2025 on the official MoCA FAQ : no, when a healthcare professional uses only the total score for screening, triage, or referral; yes, when anyone interprets individual tasks, subscores, or domain-specific impairment, with exemptions for neuropsychologists and clinicians with a year of postdoctoral cognitive-assessment training. Training is free for students, faculty, academic researchers, and employees of publicly operated healthcare institutions. Two cautions: older pages on the same site still carry the 2019-era universal-mandate language, so keep the dated FAQ or written confirmation, and your employer may require training regardless. How does the education correction work? For the Full versions: add one point when the patient has 12 or fewer completed years of formal education (counted after kindergarten), only when the total is below 30, and never above 30. Chart the raw score, the education years, and the correction status as three separate facts: "raw 24/30; 12 years education; +1 applied; adjusted 25/30." A single merged number breaks serial comparison, because nobody can later tell whether a one-point difference is clinical or bookkeeping. And the correction does not fully remove education effects: research finds lower-education patients remain more likely to be misclassified even after the point is applied, which belongs in the interpretation, not in improvised extra points. How do I document telephone, blind, or short MoCA versions? By their own names and denominators. The Blind/Telephone version is scored out of 22, the five-minute version out of 15, and MoCA-Basic (for low-literacy populations) out of 30 as its own adapted instrument. The official FAQ states that converting a /22 result to /30, or prorating a Full administration with omitted visual items, is not validated, so write "MoCA Blind/Telephone 18/22," never "MoCA 18" or "equivalent to 25/30." If a data system forces a /30 field, record the native result in text and label any system-generated conversion as unvalidated. Telephone adaptations in the literature vary in content and maxima, which is exactly why the note names the specific version. Does a MoCA below 26 mean my patient has dementia? No. A dementia diagnosis requires evidence of cognitive decline plus functional impairment established clinically; a screening score supplies neither by itself, and the official MoCA materials say a low result does not establish a diagnosis while portions of the published severity bands were set arbitrarily. The defensible note reads: below the applicable threshold, screening result, reversible contributors reviewed (medications, mood, sleep, sensory, metabolic), function and collateral documented, and a plan: monitoring, repeat testing under comparable conditions, or referral for neuropsychological evaluation when the stakes or the uncertainty warrant it. How much MoCA change between visits is real? More than most notes assume. Published reliable-change estimates run from roughly two points in some cohorts to about four in others, practice effects inflate early retests, and the alternate forms (8.2 and 8.3), built for repeat testing at intervals of about three months or less, are close but not perfectly matched in difficulty. So a one- or two-point movement is usually within measurement error, and even a three- or four-point change deserves a check: same form lineage, language, mode, sensory aids, and clinical state? The strongest evidence of real change is concordance, a score moving in the same direction as function and collateral report, and that is what the note should show. Can a MoCA score determine capacity or driving fitness? No, in every jurisdiction this page covers. Capacity is decision-specific: the conclusion rests on the patient's demonstrated ability to understand, appreciate, reason about, and communicate a specific choice, documented in a capacity evaluation , with the MoCA as collateral cognitive evidence only. Driving guidance runs the same way: US clinician guidance, the Canadian Medical Association's 2025 driver's guide, and Australia's Austroads framework all treat cognitive screens as triggers for fuller assessment (collateral history, occupational-therapy or on-road evaluation), not as pass/fail gates, and state and provincial reporting laws turn on clinical judgment about the condition, not on a score. When should I use the MoCA versus the MMSE, SLUMS, or RUDAS? By purpose, population, and licensing. The MoCA is more sensitive to mild cognitive impairment than the MMSE (which is a licensed, purchased instrument with its own documentation rules) at the cost of more false positives in low-prevalence and low-education settings. SLUMS is a free alternative for older adults, Mini-Cog is the very brief triage option, and RUDAS was built for culturally and linguistically diverse and low-education populations, where the MoCA's conventional threshold misfires most. When the question outgrows any screen (domain profiling, validity assessment, high-stakes decisions), the next steps are an RBANS or a full neuropsychological evaluation. Never convert scores between instruments; chart each by name. Can BastionGPT draft my MoCA documentation? Yes. Give it the facts (version, raw total, education years, conditions, prior scores, the functional picture) and it drafts the documentation block: raw and adjusted values separated, the screening-not-diagnosis caveat in place, and the plan and serial-comparison sentences scaffolded for your review. It can also check a finished note for a denominator mismatch, a misapplied education correction, verdict language around the cutoff, or a serial comparison across changed conditions. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: MoCA Cognition, official FAQ (December 2025 certification policy, education correction, version denominators, unvalidated conversions, severity-band caveat), training and certification , permissions , and disclaimer . Nasreddine et al., 2005, Journal of the American Geriatrics Society, the MoCA validation study : the 26 threshold and its original operating characteristics. Islam et al., 2023, meta-analysis of MoCA cutoffs (sensitivity-specificity tradeoffs at 23 to 25); a 2024 amnestic-MCI meta-analysis, 55 studies (median research cutoff below 24). Rossetti et al., 2011, Neurology, population-based MoCA norms (66% below 26); Malek-Ahmadi et al., 2015, healthy older adults (46% below 26); Davis et al., 2021, Cochrane review (false-positive rates at the 26 threshold). White et al., 2022, education effects after the one-point correction ; Stimmel et al., 2024, language and population effects on classification . Ratcliffe et al., 2024, NACC norms and reliable-change analysis (18,410 participants; domain-score reliability); the 2024 Memory Index Score replication, Archives of Clinical Neuropsychology . Kopecek et al., 2016 and Krishnan et al., 2017, reliable-change studies ( one- and multi-year cohorts); Cooley et al., 2015, practice effects ; Lebedeva et al., 2016, alternate-form difficulty. Medicare annual wellness visit regulation, 42 CFR 410.15 ; CMS, cognitive assessment and care plan services ; CMS GUIDE model, participation framework . CMS, national anti-amyloid coverage framework ; UnitedHealthcare, 2026 lecanemab policy (MoCA 17 to 30 as one eligibility route). NHTSA and the American Geriatrics Society, clinician's guide to older drivers ; Canadian Medical Association, 2025 driver's guide ; Austroads, Assessing Fitness to Drive . CDA-AMC, lecanemab reimbursement recommendation (MoCA accepted as an alternative scale, no numerical threshold); British Columbia PharmaCare, donepezil coverage criteria (SMMSE-based). Australian Government, aged care assessment manual (June 2026; GPCOG and KICA pathways); NSW Agency for Clinical Innovation, cognitive screening tools guidance (RUDAS). ## Related templates MMSE Score Documentation Mental Status Exam (MSE) Note Capacity & Competency Evaluation Neuropsychological Evaluation Report RBANS Report Write-Up Mini-Cog Score Documentation SLUMS Score Documentation Clock Drawing Test Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mmse All note templates # MMSE Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The MMSE (Mini-Mental State Examination) is a 30-point cognitive screen published by Folstein, Folstein, and McHugh in 1975, now a licensed instrument sold per form by its publisher. Hospitals, clinics, and drug-coverage programs still key decisions to it. This page covers how to document and interpret MMSE results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Credentialed healthcare professionals using purchased or licensed forms (publisher qualification level S) Audience Treating and referring clinicians, drug-coverage programs in Canada and Australia, courts and capacity evaluators, families Typical length 3 to 8 chart lines · administration 10 to 15 minutes Format family Clinician-administered cognitive screen (30 points; second-edition versions score 16, 30, and 90) When it's used Cognitive screening and serial dementia tracking, drug-subsidy documentation in Canada and Australia, inpatient cognitive baselines Standards context Licensed PAR instrument sold per form; Canadian and Australian drug plans still key coverage to its bands; no US authority mandates it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MMSE? The Mini-Mental State Examination (MMSE) is a 30-point, clinician-administered cognitive screen published by Marshal Folstein, Susan Folstein, and Paul McHugh in 1975. The fact that shapes everything else about using it today is legal, not clinical: after circulating freely for about 25 years, its intellectual property was registered to MiniMental LLC in 2000, and in 2001 Psychological Assessment Resources (PAR) received the exclusive worldwide license. Enforcement followed, including the 2011 withdrawal of the free Sweet 16 screen after an infringement demand. Under the publisher's current terms, forms are purchased consumables (US list prices in August 2026: $2.50 per original-MMSE form, with second-edition forms priced per version), a purchased copy does not authorize photocopying, and reproducing prompts, stimuli, or scoring rules in an EHR template, an improvised translation, or administration "from memory" all fall outside the stated permissions. What is ordinary clinical documentation: recording the edition, version, score, date, limitations, and interpretation in the chart. The scoring facts are just as easy to get wrong. The original and the second edition's Standard Version total 30 points, and the publisher equated them so longitudinal data carry across; the second edition's Brief Version totals 16 and its Expanded Version 90, so "MMSE-2 score" means nothing without the version, and each version ships in Blue and Red alternate forms. The famous cutoff of 24 is a convention built on a 23/24 boundary from early validation work, and it bends with demographics: in the largest US population study, the median score for adults with zero to four years of schooling was 22, which is why a low score in a low-education patient can be population-typical while a "normal" 27 can represent real decline in a highly educated one. The MMSE is also one letter away from the mental status exam , the descriptive clinical examination it is routinely confused with, and one comparison away from the MoCA , which detects mild impairment the MMSE misses. It is a screen; it diagnoses nothing alone. ## Who uses MMSE documentation and when Hospital and clinic teams document MMSE results as cognitive baselines and serial dementia measurements; the payer reality is where the score still carries hard weight. Canadian provincial drug plans key cholinesterase-inhibitor coverage to score bands (Ontario and British Columbia use 10 to 26, with BC on the Standardised MMSE plus a staging scale, while Alberta's 2026 criteria accept the MMSE or SLUMS, RUDAS, or interRAI alternatives). Australia's PBS requires a baseline of 10 or higher for subsidized cholinesterase inhibitors (10 to 14 for memantine) with specialist confirmation, runs a formal low-score pathway for patients whose scores reflect language, education, cultural, sensory, or dysphasia barriers rather than disease severity, and judges continuation on a documented clinically meaningful response, having abolished its old two-point-improvement rule in 2013. In the US, no Medicare rule names it: wellness-visit cognition detection and capacity and delirium workflows are all tool-neutral, and delirium pathways use delirium-specific instruments. When the question outgrows a screen, the next steps are an RBANS or a full neuropsychological evaluation . ## How to document MMSE results in the chart No statute prescribes an MMSE note format. The elements below are what keeps a licensed, versioned, convention-laden screen interpretable and defensible over time. Each carries the pitfall that most often undermines it. Edition, version, and form. Name the instrument precisely: original MMSE or MMSE-2, the version (Brief, Standard, Expanded), the Blue or Red form where applicable, and the authorized language edition. Pitfall: "MMSE 22." The three second-edition versions total 16, 30, and 90 points, only the Standard Version is score-equated to the original, and an unnamed version makes the number and every later comparison ambiguous. The attention method, for legacy administrations. The original instrument permits two alternative attention tasks, and research shows the serial-subtraction and reverse-spelling approaches are not interchangeable: the arithmetic version runs harder and both are education-sensitive. Pitfall: a method switch between administrations charted as cognitive change. Record which method was used, every time. Raw total over its denominator, with demographic context. Write the fraction and the interpretive context: completed education, literacy where material, language and interpreter use, and sensory or motor limits with aids in place. Pitfall: reading a low score in a low-education patient as disease. In the major US population norms, the median score with zero to four years of schooling was 22; education, language, and culture move totals without any pathology. Validity and completeness. State whether administration was complete, cooperative, and standard; document refusals, interruptions, and untestable components with reasons. Pitfall: silent zeros and improvised prorating. No generally authorized rule converts an incomplete administration into a standard 30-point total; "administration incomplete; standard total not interpretable" is the defensible entry. Domain pattern and function. Describe the broad pattern in your own words (orientation, registration, attention, delayed retrieval, language, construction) and tie it to function and collateral report. Pitfall: treating identical totals as identical patients. The total is a composite; very different cognitive patterns produce the same number, and function is what separates them. Interpretation against the conventional threshold. Below 24 means below the conventional screening threshold, a flag for evaluation. Pitfall: verdict language. The 23/24 boundary comes from early validation work whose authors said the MMSE alone cannot diagnose, and ceiling effects mean highly educated patients with real impairment routinely score "normal": in one highly educated cohort, the standard cutoff caught barely more than half of dementia cases. Serial comparison discipline. Compare only like with like: same edition, version, language, attention method, sensory aids, and clinical state, on alternate forms where available, with the interval stated. Pitfall: over-reading small movements. Reliable-change estimates run from roughly two to four points in some cohorts to five to seven in others, while average Alzheimer-type decline runs about three points a year, so a two-point drop is usually noise unless function moves with it. ### Blank template (copy and adapt) MMSE DOCUMENTATION BLOCK Edition: [original MMSE / MMSE-2] Version: [Brief 16 / Standard 30 / Expanded 90] Form: [Blue / Red / n/a] Language (authorized edition): [ ] Interpreter: [ ] Attention method (legacy original): [serial subtraction / reverse spelling] Date: [ ] Setting: [ ] Administered by: [ ] Raw total: [ ]/[30 / 16 / 90] Context: [education years; literacy; sensory aids in place; motor limits; acute illness or delirium risk] Validity: [complete and standard / incomplete: which parts and why; refusals documented, not scored as errors] Domain pattern (own words): [orientation / registration / attention / delayed retrieval / language / construction] Function and collateral: [ADLs, IADLs, informant report] Interpretation: [screening result vs conventional threshold in demographic context; not a diagnosis] Prior: [date, edition/version/form, method, score] Change: [points; reliable-change and condition-comparability check] Plan: [workup / referral / reassessment interval] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces any MMSE test content. ## Sample MMSE documentation (fictional) Scenario: an outpatient serial comparison, the use case where version, form, and method discipline earn their keep. This entry documents results and interpretation only; it reproduces no test content. All details are fictional. Patient: W.K., 79 · Visit: Memory clinic follow-up · Clinician: P. Lindqvist, MD (geriatrics) · Note date: 08/11/2026 Measure: MMSE-2 Standard Version, Red form, English (authorized edition), administered in clinic by this clinician: 21/30. Education 14 years. Hearing aids and reading glasses in place; no acute illness; administration complete and standard. Purchased protocol used; this note records results only. Prior: MMSE-2 Standard Version, Blue form, English, this clinic, 06/09/2025: 25/30, same conditions and aids. Interval 14 months; alternate forms used to limit practice effects; versions are score-equated, so the totals are comparable. Pattern and function: Current performance was relatively weaker in orientation and delayed retrieval, with language and construction comparatively preserved. W.K.'s son reports new help with bills and two missed appointments since spring; medication management now supervised. Interpretation and plan: The four-point decline may exceed short-term measurement variability in some cohorts, and the interpretation rests principally on its concordance with the reported functional decline rather than the number alone. The result is a screening measurement, not a diagnosis or a stage. Plan: laboratory review of reversible contributors, medication reconciliation, structured functional assessment, and discussion of fuller neuropsychological evaluation; reassess with the Blue form in 12 months under the same conditions, sooner on functional change. This sample is fictional and for educational purposes. It does not describe a real patient or record, reproduces no MMSE test content, and the scores are invented for illustration and correspond to no real person. ↑ Back to the template and downloads ### Why this sample works Edition, version, form, and language are named at both time points, so the comparison is between comparable measurements rather than between two ambiguous numbers. Alternate forms, matched conditions, and the stated interval are on the record, which is what lets a four-point change carry meaning. The demographic and sensory context is documented, and the domain pattern is described in the clinician's own words without any test content. The change is interpreted through reliable-change caution and anchored to concordant functional decline, not the arithmetic alone. The note stays inside the licensing lines: results, interpretation, and plan in the chart, purchased protocols for administration, nothing reproduced. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The MMSE's compliance layer is unusual for a bedside screen: the operative questions are publisher licensing first and payer policy second. Under PAR's stated terms , authorized protocols are purchased consumables; a purchased copy does not authorize photocopying, downloading a "free" online copy, reproducing any part of the instrument in an EHR template, modifying, reformatting, or translating it, and the publisher's enforcement history is concrete (the free Sweet 16 screen was withdrawn in 2011 after an infringement demand). Academic commentary has long questioned how far copyright can reach into a clinical method, and no court judgment on the merits was located, but a dispute in the literature is not a permission slip: the low-risk operational course is purchased or licensed materials, with the distinction that matters for records kept clean. Documenting the edition, version, form, score, date, limitations, and interpretation is ordinary clinical charting; embedding prompts, stimuli, or scoring keys in a template is reproduction. In the US, the decision contexts are tool-neutral: the Medicare annual wellness visit requires detection of cognitive impairment by observation and report without naming an instrument, the cognitive assessment and care plan service (99483) requires a far larger documentation set than any score, delirium pathways call for delirium-specific instruments rather than an MMSE threshold, and capacity is decision-specific: the Department of Justice's capacity guide says no single test result should drive the determination, and the American Bar Association's guidance for lawyers puts it more bluntly: "Thou shalt not covet the mini-mental status exam." A score can inform a capacity evaluation ; it can never be one. Where the score still gates money, precision matters. In Canada, Ontario's drug benefit requires an MMSE of 10 to 26 to start and continue cholinesterase-inhibitor coverage; British Columbia keys donepezil to a Standardised MMSE of 10 to 26 plus Global Deterioration Scale stage 4 to 6 (and note that SMMSE means the Standardised MMSE, a different thing from the Severe MMSE used below the ordinary floor); Alberta's April 2026 criteria accept an MMSE of 10 to 26 or, notably, SLUMS, RUDAS, or interRAI Cognitive Performance Scale ranges instead, a payer formally accommodating instrument bias; Manitoba dropped the score gate for donepezil and galantamine back in 2018. Australia's PBS instrument requires a baseline MMSE or Standardised MMSE of at least 10 for cholinesterase inhibitors (10 to 14 for memantine) with specialist confirmation, runs an explicit low-score pathway when a score of 9 or less reflects language, education, Indigenous cultural factors, intellectual disability, sensory impairment, or disproportionate dysphasia rather than disease severity (with a clinician's-impression assessment instead), and since 2013 judges continuation on a documented, six-monthly, clinically meaningful response, having abolished the old two-point-improvement rule its own advisory committee found unreliable. Two final measurement cautions: a 2025 central review of 10,203 administrations flagged administration problems in 26.8% and scoring problems in 27%, so standardized administration is a live quality issue, not a formality; and published MoCA crosswalks are estimates for context, never substitutes for an administered score a payer requires. MMSE is a registered trademark of Psychological Assessment Resources, Inc.; MMSE materials are copyright MiniMental LLC. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common MMSE documentation errors reviewers flag The quality data are blunt: a 2025 central review of 10,203 MMSE administrations flagged administration problems in 26.8% and scoring problems in 27.0%. Interpretation has its own traps: in a highly educated cohort the standard cutoff caught barely more than half of dementia cases, while the major US population norms put the median score for adults with zero to four years of schooling at 22, inside the range routinely labeled impaired. The BastionGPT Clinical Advisory Board sees the same errors most often in MMSE documentation reviews: Edition, version, form, and method omitted. "MMSE 22" could be an original administration with either of two non-equivalent attention methods, or a second-edition Brief (out of 16), Standard (out of 30), or Expanded (out of 90) version on either alternate form. Every later comparison, payer submission, and audit inherits the ambiguity. Below 24 charted as dementia. The 23/24 boundary is a screening convention from early validation work whose own authors said the instrument cannot diagnose alone; education and age move the number, ceiling effects hide impairment in educated patients, and severity bands repeated across websites are heuristics with inconsistent edges, not stages. Licensing errors written into the workflow. Photocopied protocols, "free download" forms, improvised translations, or test content built into an EHR template all sit outside the publisher's stated permissions. Chart results and interpretation freely; administer from purchased or licensed materials. Incomplete administrations scored as complete. Refusals and untestable components silently scored as errors, or missing items prorated into a standard total with no authorized rule. Document what was not done and why, and report the administration as incomplete rather than inventing a comparable number. Serial changes over-read or manufactured. A two-point movement called progression when reliable-change estimates run roughly two to seven points depending on cohort; an attention-method or version switch mid-series masquerading as decline; or a crosswalk-estimated score standing in for the administered score a drug plan requires. Compare like with like, and let function corroborate. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (edition, version, form, score, education and language context, prior results) and it drafts the documentation block: denominator explicit, demographic caveats in place, and the screen-not-diagnosis interpretation scaffolded for your review. Cross-check a finished note for the gaps reviewers flag: a missing version or attention method, a verdict hung on the 24 threshold, an incomplete administration scored as complete, or a serial comparison across changed conditions. Draft payer-facing summaries that carry exactly what the plan needs: the administered score with its version and date, the band it must sit in, and the response documentation for continuation, with nothing converted or estimated. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does an MMSE score actually mean? It is a 30-point screening composite, and the familiar cutoff of 24 is a convention built on a 23/24 boundary from early validation work, where the authors themselves said the instrument cannot diagnose alone. Demographics move the number: in the largest US population norms, the median score was 29 with nine or more years of schooling and 22 with zero to four years, and scores drift down with age. A Cochrane review found reasonable but threshold-dependent accuracy in older community populations. So the defensible reading is "below the conventional screening threshold, in this demographic context, warranting evaluation," never a diagnosis or a stage. Is the MMSE free to use? Can we photocopy the form? No and no, under the publisher's stated terms. The instrument circulated freely for decades, but its rights were consolidated in 2000 and exclusively licensed to PAR in 2001, and enforcement is real (the free Sweet 16 screen was withdrawn in 2011 after an infringement demand). Authorized protocols are purchased consumables (US list price for the original ran $2.50 per form in August 2026), and the publisher's position is that purchase does not authorize photocopying, free-download copies, modification, or improvised translation. Legal commentators have questioned the copyright's reach, but no merits ruling settles it, so the low-risk course for a clinic is purchased or licensed materials, or a free alternative instrument chosen deliberately. Can we build the MMSE into our EHR? Not the test itself, without a license. The distinction that keeps records clean: entering the edition, version, form, score, date, limitations, and interpretation is ordinary result documentation and is fine; building the prompts, stimuli, response rules, or scoring key into an EHR smart form is reproduction and reformatting of protected content, which the publisher says requires written permission, as does electronic administration. Scanned completed protocols sit in between: store them only where the organization is licensed to possess them, with restricted access and audit controls. When in doubt, chart the result and keep the instrument on purchased paper or a licensed electronic platform. Does a score below 24 mean dementia? No. Dementia requires clinically established decline plus functional impairment, and the score supplies neither alone. The error runs both directions: a low score in a patient with limited schooling, a different first language, or sensory barriers can be population-typical (the low-education median in US norms sits at 22), while ceiling effects mean a highly educated patient with real impairment can score "normal": in one highly educated cohort the standard cutoff caught barely more than half of dementia cases. The defensible note pairs the number with demographic context, function, and collateral report, and routes genuine concern to fuller evaluation, starting with reversible contributors. How much serial MMSE change is real? More than the folklore assumes. Published reliable-change estimates run from roughly two to four points in some cohorts to five to seven in others, while average Alzheimer-type decline runs about three points a year, so a small drop is frequently measurement noise. Before interpreting any change, check comparability: same edition and version, same attention method (the two legacy alternatives are not psychometrically equivalent, and a mid-series switch can manufacture "decline"), same language, aids, and clinical state, ideally on alternate forms. Then let function arbitrate: a score falling alongside new help with medications and finances means something a bare number does not. What are the MMSE-2 versions and do they change the score? The 2010 second edition comes in three versions with different totals: Brief (16 points, about five minutes), Standard (30 points, score-equated to the original so longitudinal data carry across), and Expanded (90 points, built to remove the ceiling for subtler impairment), each in Blue and Red alternate forms with published equivalence data. The documentation consequences: never enter a Brief or Expanded total into a classic 30-point severity habit, always name the version and form, and when switching from the original instrument, the Standard Version is the one designed to preserve your serial record. Do drug plans still require MMSE scores? In Canada and Australia, yes; in the US, no national rule does. Ontario requires 10 to 26 to start and continue cholinesterase-inhibitor coverage; British Columbia uses a Standardised MMSE of 10 to 26 plus a staging scale; Alberta's 2026 criteria accept the MMSE or SLUMS, RUDAS, or interRAI ranges; Manitoba dropped its gate in 2018. Australia's PBS requires a baseline of at least 10 (10 to 14 for memantine), runs a formal low-score pathway when barriers other than disease severity depress the score, and judges continuation on a documented clinically meaningful response, the two-point-improvement rule having been abolished in 2013. For payer submissions, document the administered score with its version and date; a crosswalk estimate never substitutes. How is the MMSE different from the MSE, and from the MoCA? Two different confusions. The mental status exam (MSE) is the descriptive clinical examination of appearance, mood, thought, perception, insight, and judgment; the MMSE is a scored, licensed cognitive screen; a chart can contain either or both, and the shared abbreviation causes real errors. The MoCA is the neighboring screen: markedly more sensitive to mild cognitive impairment (in its validation cohort it detected 90% of MCI against the MMSE's 18%), free to use under its own terms, with its own certification rules. Published crosswalks let you relate MMSE and MoCA scores as estimates, with the original scores always preserved, and no conversion satisfies a payer that names the MMSE. Can BastionGPT draft my MMSE documentation? Yes. Give it the facts (edition, version, form, score, education and language context, prior results, the functional picture) and it drafts the documentation block with the denominator explicit, demographic caveats in place, and the screen-not-diagnosis interpretation scaffolded for your review; it never needs, and should never be given, any test content. It can also check a finished note for a missing version or attention method, a verdict hung on the 24 threshold, or a serial comparison across changed conditions. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: PAR, MMSE product page , MMSE-2 product page (versions, totals, alternate forms, equivalence data, pricing captured August 2026), and intellectual-property position ; ePROVIDE, MMSE registry entry (copyright string, translations). Folstein, Folstein & McHugh, 1975, Journal of Psychiatric Research, the original publication ; Anthony et al., 1982, the 23/24 boundary validation (education-linked false positives; no diagnosis from the score alone). Newman & Feldman, 2011, New England Journal of Medicine, and Feldman, Newman & Johnston, 2013, the copyright chronology ; UCSF, the Sweet 16 withdrawal ; Badgett et al., 2026, citation trajectories of restrictive versus permissive instruments . Crum et al., 1993, JAMA, population-based MMSE norms by age and education (18,056 adults). Creavin et al., 2016, Cochrane review of MMSE accuracy ; Tsoi et al., 2015, JAMA Internal Medicine, cognitive-test meta-analysis . O'Bryant et al., 2008, ceiling effects in highly educated cohorts ; a 2025 near-ceiling item-response analysis, measurement precision at the top of the scale . Ganguli et al., 1990, non-equivalence of the legacy attention alternatives ; Hawkins et al., 2011, systematic total-score differences between methods . Hensel et al., 2007, reliable-change limits in older adults ; a 2026 rare-dementia study, thresholds for meaningful MMSE change ; a 2025 central review of 10,203 administrations, administration and scoring quality . Nasreddine et al., 2005, the MoCA validation (MCI sensitivity comparison); Fasnacht et al., 2023, MoCA-MMSE conversion study (crosswalks as estimates). CMS, annual wellness visit guidance and cognitive assessment and care plan services (tool-neutral); US Department of Justice, capacity resource guide ; American Bar Association, guidance on capacity and cognitive screens . Ontario drug benefit, limited-use criteria ; British Columbia PharmaCare, donepezil criteria ; Alberta, special-authorization criteria (April 2026 multi-instrument menu). Australia, PBS listing instrument (baseline thresholds, low-score pathway, continuation standard) and PBAC, review minutes on the abolished two-point rule . ## Related templates MoCA Score Documentation Mental Status Exam (MSE) Note Capacity & Competency Evaluation RBANS Report Write-Up Neuropsychological Evaluation Report Mini-Cog Score Documentation SLUMS Score Documentation Clock Drawing Test Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pcl-5 All note templates # PCL-5 Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The PCL-5 (PTSD Checklist for DSM-5) is a free, public-domain, 20-item self-report measure of PTSD symptoms scored 0 to 80, stewarded by the VA's National Center for PTSD. Trauma therapists, VA and DoD clinicians, and primary care teams use it for screening and treatment monitoring. This page covers how to document and interpret PCL-5 results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report; administered and interpreted by qualified health professionals per the steward's terms Audience Trauma therapists and prescribers, VA, DoD, and TRICARE programs, compensation and workers' compensation reviewers Typical length 3 to 7 chart lines per administration · patient completion 5 to 10 minutes Format family Self-report PTSD symptom measure (20 items, scored 0 to 80, DSM-5 clusters B through E) When it's used Follow-up to a positive PC-PTSD-5 screen, baseline and session monitoring in trauma-focused therapy, program outcome reporting Standards context Public domain from the National Center for PTSD; cutoffs are setting-dependent (31 to 33 = probable PTSD); no official severity bands exist On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PCL-5? The PTSD Checklist for DSM-5 (PCL-5) is a 20-item self-report measure of the DSM-5 PTSD symptoms, developed in 2013 by Weathers, Litz, Keane, Palmieri, Marx, and Schnurr at the VA's National Center for PTSD , which remains its steward. The official page states the measure is "in the public domain and not copyrighted" and intends it for qualified health professionals and researchers; the forms are free downloads with no license. Version identity is the first documentation fact: there are two time frames (the default past-month form and an official past-week form dated February 2024 for session-by-session monitoring, now supported by a 2026 longitudinal validation, which retires the older guidance that only the monthly form was validated) and three Criterion A packages (without Criterion A when the index trauma is already established, with a brief Criterion A check, or with the LEC-5 and an extended assessment). A staff-confirmed Spanish translation exists but is expressly unvalidated in published research. The DSM-IV predecessors (PCL-C, PCL-M, PCL-S) scored 17 items from 17 to 85, and no authorized arithmetic converts old totals to the 0-to-80 scale. The load-bearing distinctions: each item is rated 0 to 4 for a total of 0 to 80, with cluster totals (intrusion, avoidance, negative cognitions and mood, arousal) that are descriptive only, because the steward publishes no cluster cutoffs and, notably, no severity bands at all; every mild/moderate/severe table on the internet is someone's invention. Provisional interpretation runs by either of two methods: a total-score cutoff, where 31 to 33 indicates probable PTSD across many samples but validated optima span roughly 22 to 49 by setting and purpose, or the DSM-aligned symptom-count pattern (items rated 2 or higher counting as endorsed: at least one intrusion, one avoidance, two cognition/mood, and two arousal symptoms). Either way the result is provisional: diagnosis requires the full criteria, and the structured-interview gold standard is the CAPS-5, inside a fuller trauma assessment . ## Who uses PCL-5 documentation and when Trauma therapists document it as the backbone of measurement-based care: the Cognitive Processing Therapy manual prescribes session-level administration and review across its roughly 12 sessions, Prolonged Exposure materials support measurement during treatment and at pre- and post-treatment, and VHA policy layers its own collect-share-act expectation of a valid outcome measure at every or nearly every encounter, with the PC-PTSD-5 as the upstream primary-care screen. TRICARE's current behavioral-health manual sets a different clock (baseline, roughly 60-day intervals, and discharge), and compensation systems read the documentation without being driven by it: VA disability ratings turn on occupational and social impairment established through a qualified examination, Canadian workers' compensation presumptions still require a DSM diagnosis from a psychiatrist or psychologist, and Australian DVA claims ordinarily require psychiatric evidence. The score feeds an outcome measure note and the surrounding psychotherapy progress note ; the diagnostic work lives in the trauma assessment . ## How to document PCL-5 results in the chart No statute prescribes a PCL-5 note format. The elements below keep a serial trauma measure interpretable across a course of treatment and defensible when a program, payer, or claims reviewer reads it later. Each carries the pitfall that most often undermines it. Version, time frame, and Criterion A package. Name the form: past-month or past-week, and whether Criterion A was assessed here (brief form or LEC-5) or established elsewhere. Pitfall: an unlabeled series. Weekly and monthly totals are not interchangeable observations, and a weekly score charted against a monthly baseline manufactures a change the data never showed. The index event and its source. State that the same index trauma was used as in prior administrations, and where exposure was established (interview, brief form, LEC-5 package, prior records). The chart needs the minimum clinically necessary description, not a trauma narrative in every note. Pitfall: index drift. A series that quietly switches events measures two different things under one heading. Completeness and the total over 80. State that all 20 items were completed, then the total, with cluster totals (intrusion 0 to 20, avoidance 0 to 8, cognitions and mood 0 to 28, arousal 0 to 24) as descriptive observations when useful. Pitfall: silent proration. No official rule converts an incomplete PCL-5 into a standard total, and the steward publishes no cluster cutoffs to be "positive" on. The interpretation method, named. If a cutoff is used, say which and why ("exceeds the clinic's screening threshold of 31, selected to favor detection"); if the symptom-count method, say so. Keep the language provisional. Pitfall: "PCL-5 = 36, patient has PTSD." The steward's own framing is probable PTSD, validated optimal cutoffs span roughly 22 to 49 by setting, and there are no official severity bands to quote. Change, with the right constructs. Report the raw change and direction, then the construct: the steward suggests a 10-point reduction as a response indicator, sample-derived reliable-change estimates run from about 9 to 12 points in primary care to 15 to 18 in veteran treatment samples, and a final score below 28 is the current clinically-significant-change indicator (the underlying study wrote 28 or below; the VA page writes below 28; quote the one you use). Pitfall: the five-point rule. That convention belongs to the DSM-IV predecessor and does not transfer. Context, especially for a rise. Before interpreting an increase, verify comparability (same form, time frame, index event, completeness) and record context: anniversaries, trauma-processing sessions, new stressors, sleep, substances, safety. Pitfall: one spike charted as treatment failure or exaggeration. A temporary rise during exposure work is common, and a rise during a compensation claim is not evidence of over-reporting. Plan linkage and the cadence authority. State what the result changed and when the next administration is due, naming whose cadence governs: the CPT manual's session-level rule, VHA measurement-based-care policy, a TRICARE interval, or clinic protocol. Pitfall: conflated cadences. Treatment-manual procedure, health-system policy, and payer contract are three different authorities, and notes that blur them fail exactly the reviews they were written for. ### Blank template (copy and adapt) PCL-5 DOCUMENTATION BLOCK Form: [past-month / past-week] Criterion A: [not assessed here / brief form / LEC-5 extended / established previously] Index event: [same as prior administrations? y/n] Date: [ ] Completed: [all 20 items / incomplete: n missing, not totaled] Total: [ ]/80 Clusters (descriptive): B [ ]/20 C [ ]/8 D [ ]/28 E [ ]/24 Interpretation method: [cutoff (value + purpose) / DSM-aligned symptom count] Result: [provisional language only] Prior comparable score / date: [ ] Change: [points, direction] Construct: [10-pt response indicator / sample reliable-change estimate / final score vs 28 indicator] Context: [anniversary / exposure work / stressors / sleep / substances / safety, as relevant] Plan linkage: [continue / intensify / consult / reassess + why] Next administration: [date + governing cadence: manual / policy / payer / clinic] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. ## Sample PCL-5 documentation (fictional) Scenario: the final session of a 12-session course of Cognitive Processing Therapy, where the note has to reconcile a weekly monitoring series, a mid-course spike, and the directly comparable monthly pre/post pair. All details are fictional. Patient: D.S., 38 · Visit: CPT session 12 (final) · Clinician: K. Aronov, PsyD · Note date: 08/12/2026 Measure: PCL-5, past-month form, without Criterion A (index event established at intake assessment and unchanged throughout), all 20 items completed: 22/80 today. Clusters, descriptively: intrusion 6/20, avoidance 2/8, cognitions and mood 8/28, arousal 6/24. Baseline past-month score at intake (same index event): 52/80. Weekly series (past-week form, sessions 2 through 11): 47, 44, 46, 41, 38, 35, 31, 29, 26, 24. The session-4 rise from 44 to 46 coincided with the trauma anniversary and the start of written trauma-account work; safety, adherence, and the index event were reviewed at the time and documented, and the trajectory resumed downward. Weekly scores are recorded as their own series and were not compared against the monthly baseline. Interpretation: The directly comparable monthly pair shows a 30-point reduction, well beyond the steward's suggested 10-point response indicator, and the final score of 22 is below the current clinically-significant-change indicator of 28. These are self-report findings: today's result supports response and is not by itself a diagnostic reassessment. D.S. reports returned driving, improved sleep, and resumed family activities, consistent with the score trajectory. Plan linkage: Results reviewed collaboratively. Transition to relapse-prevention spacing with a booster session in one month; residual guilt-related cognitions identified for the relapse-prevention plan. Next PCL-5 (past-month form, same index event) at the booster visit, per clinic protocol. Diagnostic status to be revisited clinically at that visit rather than inferred from today's score. This sample is fictional and for educational purposes. It does not describe a real patient or record, and the scores are invented for illustration and correspond to no real person. ↑ Back to the template and downloads ### Why this sample works The form, time frame, Criterion A handling, index event, and completeness are all named, so every number in the series is traceable and comparable. Weekly and monthly scores are kept as separate series, and the response math runs only on the directly comparable monthly pair. The mid-course rise is documented with its context and the verification done at the time, instead of being smoothed over or over-read. Change is claimed with named constructs: the 10-point response indicator and the below-28 clinically-significant-change indicator, not folklore thresholds. The score links to a plan and a governed next administration, and diagnostic status stays with clinical reassessment rather than the questionnaire. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations In the US, the authorities around PCL-5 documentation are layered, and notes should say which layer they are serving. VHA measurement-based-care policy expects a valid patient-reported outcome measure collected every or nearly every encounter, reviewed, shared with the Veteran, and acted on, with specialty PTSD programs required to chart that collect-share-act cycle; the upstream VHA screen is the PC-PTSD-5 , not the PCL-5. Disability compensation runs on different rails entirely: the diagnosis must conform to DSM-5 under 38 CFR 4.125 , the examination lives on the relevant disability benefits questionnaire completed by a qualified examiner, ratings are assigned under the mental-disorders formula in 38 CFR 4.130 by occupational and social impairment, not by any questionnaire total, and while the CAPS-5 is the diagnostic gold standard, no authority reviewed for this page makes it mandatory in every compensation examination. TRICARE's current behavioral-health manual specifies PCL-5 collection at baseline, roughly 60-day intervals, and discharge in covered settings, a contract cadence distinct from both the CPT manual's session-level procedure and VHA policy. Brief-assessment billing under 96127 follows payer-specific unit rules. And in claims contexts, restraint cuts both ways: the recently developed PCL-5 symptom-validity indices are preliminary research tools with modest sensitivity, so neither a high total nor an increase during a pending claim is evidence of exaggeration, and a note should never say otherwise. Canada and Australia separate the questionnaire from the legal conclusion even more explicitly. Veterans Affairs Canada requires the PTSD diagnosis itself from specified regulated professionals with comprehensive documentation, and the provincial workers' compensation presumptions (Ontario's first-responder statute and its counterparts, with Saskatchewan's broadened presumption effective May 2025) shift part of the causation analysis while still requiring a DSM diagnosis from a psychiatrist or psychologist, timely claims, and clinical documentation; a chart should keep four propositions distinct: the PCL-5 result, the clinician's diagnosis, the qualifying exposure, and the legal decision. In Australia, DVA guidance (amended July 2026) requires the evidence file to address all DSM-5 criteria and ordinarily requires a psychiatrist's report for PTSD, its trauma-program outcome measurement runs at intake, discharge, and follow-ups, and Open Arms guidance states plainly that the PCL-5 is not a primary diagnostic tool; Better Access outcome measurement is the clinician's choice of tool, with the K10 the incumbent. Across all three countries the measurement rules are the same: keep the index event and time frame stable and labeled, treat translations as unvalidated until shown otherwise (including the steward's own Spanish form), report incomplete administrations as incomplete rather than prorated, and let the trauma assessment carry the diagnostic weight. The PCL-5 was developed by staff of the VA's National Center for PTSD and is in the public domain per the steward's published terms, which intend it for qualified health professionals and researchers. BastionGPT is not affiliated with, or endorsed by, the National Center for PTSD or the US Department of Veterans Affairs. This page reproduces no instrument items and no CAPS-5 interview content. ↑ Back to the template and downloads ## Common PCL-5 documentation errors reviewers flag The numbers behind these errors are specific. Across the systematic-review literature, validated optimal cutoffs span roughly 22 to 49 depending on setting , with the familiar 31-to-33 band accounting for only 11 of 30 candidate thresholds in one review; sample-derived reliable-change estimates run from about 9 to 12 points in primary care to 15 to 18 in veteran treatment samples , against a steward-suggested 10-point response indicator; and the steward publishes no severity bands at all. The BastionGPT Clinical Advisory Board sees the same errors most often in PCL-5 documentation reviews: "33 or above equals PTSD." The steward's language is probable PTSD across a 31-to-33 range with the cutoff chosen by purpose, and validated optima run far wider by population. The defensible entry names the threshold and its purpose, uses provisional language, and routes diagnosis to the clinical assessment, with the CAPS-5 as the interview gold standard. Invented severity bands charted as official. No National Center severity ranges exist, and cluster totals have no official cutoffs either. "PCL-5 52, severe PTSD per VA" cites a standard that was never published; "52/80, a high raw symptom burden well above the clinic's screening threshold" says what is actually known. DSM-IV change rules imported. The five-point response convention belongs to the 17-item predecessor. Current guidance suggests 10 points as a response indicator, reliable change is sample-dependent, and the below-28 clinically-significant-change indicator carries a wording nuance (the source study used 28 or below; the VA page says below 28) that matters exactly at 28. Name the construct and the authority. Versions and series pooled. A past-week score compared against a past-month baseline, an index event that drifts mid-series, legacy PCL totals (scored 17 to 85) trended against PCL-5 totals, or an incomplete administration silently prorated: each produces a change the data never measured. Label every administration and keep series separate. Rises over-read. A mid-treatment spike charted as failure, non-adherence, or exaggeration without checking comparability and context (anniversaries, trauma-processing work, sleep, stressors), or a compensation-context elevation treated as invalid by default when the embedded validity indices remain preliminary research tools. Document the verification, the context, and the plan; let the trajectory and an adequate dose carry the treatment decision. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Paste the score facts (form, total, clusters, prior scores and dates) and get a documentation-ready block: version and index-event lines in place, change stated with named constructs, and provisional interpretation language that survives review. Cross-check a finished note for the gaps reviewers flag: a weekly score compared against a monthly baseline, a diagnosis hung on a cutoff, an invented severity band, or a change claim with no named construct. Summarize a full treatment series for program reporting or case review: baseline and post pairs, the weekly trajectory with contextualized spikes, and response status against the steward's indicators. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do PCL-5 scores actually mean? Twenty items rated 0 to 4 sum to a total of 0 to 80, with cluster totals (intrusion, avoidance, negative cognitions and mood, arousal) available as descriptive detail. The steward's guidance puts probable PTSD at a total of 31 to 33 across many samples, with the cutoff deliberately chosen by purpose: lower to screen, higher to minimize false positives, and validated optima span roughly 22 to 49 by setting. What the number is not: a diagnosis, a severity stage (no official bands exist), or a rating driver. It is a self-reported symptom burden that feeds clinical assessment and tracks change. Does a PCL-5 of 33 or more mean my patient has PTSD? No. The steward's own language is probable PTSD, and a provisional result by either method (total cutoff, or the DSM-aligned symptom count requiring endorsed items across all four clusters) still leaves Criterion A, duration, impairment, exclusions, and differential diagnosis to the clinician. The structured-interview gold standard is the CAPS-5, and the diagnostic work belongs in a trauma assessment . The defensible chart line: "consistent with a provisional PTSD symptom pattern; diagnostic confirmation by clinical assessment," never "PCL-5 = 36, patient has PTSD." Past-month or past-week: which PCL-5 do I use? Both are official downloads with different jobs. The past-month form is the default for intake, provisional interpretation, and comparable pre/post pairs; the past-week form (an official VA form since February 2024, supported by a 2026 longitudinal validation) exists for closely spaced session monitoring. The old guidance that only the monthly form was validated is stale. The rules that keep a series honest: label the time frame on every administration, keep the index event stable, and never chart a weekly score as change from a monthly baseline; run the response math on directly comparable pairs. How much PCL-5 change counts as real improvement? Name the construct, because they differ. The steward suggests a 10-point reduction as a response indicator, while noting the evidence is limited. Sample-derived reliable-change estimates run from about 9 to 12 points in primary care to 15 to 18 in male veteran treatment samples. A final score below 28 is the current clinically-significant-change indicator, with a wording nuance worth quoting precisely: the source study used 28 or below, the VA page says below 28, and a patient at exactly 28 lands differently under each. The five-point rule belongs to the DSM-IV predecessor and should not appear in PCL-5 notes; and response, reliable change, and remission are different claims. How often should the PCL-5 be given during CPT or PE? Name the authority, because three different clocks exist. The CPT manual prescribes session-level administration and review across the protocol, with formal reviews at the midpoint and end. Prolonged Exposure materials support standardized measurement during treatment and at pre- and post-treatment without a universal every-session PCL-5 rule of their own. On top of the manuals sit policies: VHA measurement-based care expects a valid measure at every or nearly every encounter, and TRICARE's current manual specifies baseline, roughly 60-day, and discharge collection in covered settings. The note should say which cadence governs and keep the weekly monitoring series separate from the comparable pre/post pair. My patient's score went up mid-treatment. What do I document? The rise, the verification, the context, and the plan. First check comparability: same form and time frame, same index event, all items completed. Then record context: trauma anniversaries, the start of exposure or written-account work, new stressors, sleep, substances, and safety, with any risk change assessed on its own track. A temporary increase during trauma processing is common and is not, by itself, treatment failure, non-adherence, or exaggeration; what matters is the trajectory over an adequate dose. Persistent non-response is documented as exactly that, with protocol review, patient discussion, consultation, and a considered treatment decision, not a verdict from one administration. Can the PCL-5 alone support a disability or compensation claim? No, in any of the three countries. US VA ratings are assigned by occupational and social impairment under the mental-disorders formula, from a qualified examination with a DSM-5-conforming diagnosis; the score neither diagnoses nor sets the percentage, and the CAPS-5, while the gold standard, is not mandated in every examination. Canadian workers' compensation presumptions still require a DSM diagnosis from a psychiatrist or psychologist plus timing and occupation rules, and Veterans Affairs Canada requires diagnosis by specified professionals. Australian DVA claims ordinarily require a psychiatrist's report. In every system, the PCL-5 is supporting symptom evidence, and the chart should keep the score, the diagnosis, the exposure, and the legal decision as four separate facts. Do we need permission to use the PCL-5, and what about translations? No permission or fee: the steward states the measure is in the public domain and not copyrighted, intended for qualified health professionals and researchers, with the forms as free downloads from the National Center for PTSD . The boundaries: the steward's Spanish translation is staff-confirmed but expressly unvalidated in published research and should be documented as such; the CAPS-5 is obtained through the National Center's own process and its interview content is not for reproduction; and the child and adolescent instruments (CPSS-5, UCLA Reaction Index, CAPS-CA-5) are the right family for minors, with their own terms, rather than adult PCL-5 cutoffs applied downward. Can BastionGPT draft my PCL-5 documentation? Yes. Paste the score facts (form, total, clusters, prior scores with dates, the clinical context) and it drafts the documentation block: version and index-event lines in place, change stated with named constructs against the steward's indicators, provisional interpretation language, and the plan linkage, ready for your review. It can also check a finished note for a weekly-versus-monthly comparison, a diagnosis hung on a cutoff, or an invented severity band, and summarize a full series for case review or program reporting. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: National Center for PTSD, PCL-5 measure page (public-domain status, versions and Criterion A packages, cutoff guidance, the 10-point response indicator and below-28 change indicator) and PC-PTSD-5 screen page . Blevins et al., 2015, Journal of Traumatic Stress, development and initial psychometrics ; Bovin et al., 2016, veteran validation (the 31-to-33 range and its moderate diagnostic agreement). Forkus et al., 2023, systematic review of PCL-5 psychometrics, VA-hosted article : cutoff spread across studies and structural findings; Marx et al., 2022 (reliable change of 15 to 18 points in veteran samples; clinically significant change at 28 or below); Blanchard et al., 2023 (9 to 12 points in primary care). Ashbaugh et al., 2016, English and French Canadian validation ; Pettrich et al., 2025 and van der Meer et al., 2024 (German and Dutch validations illustrating setting-dependent optimal cutoffs). National Center for PTSD, DSM-5 criteria page, DSM-5-TR status (no adult PTSD criterion change requiring a PCL revision). OHSU clinician handout, PCL-5 reference : the explicit note that no severity cutoff bands have been established. 38 CFR 4.125, diagnosis conforming to DSM-5 ; the mental-disorders rating formula at 38 CFR 4.130 (ratings by occupational and social impairment); VA Review PTSD disability benefits questionnaire and adjudication materials (examiner qualifications; CAPS-5 not mandated in every examination). VHA Directive 1160.05 (measurement-based care; collect, share, act) and VHA Directive 1160.03 (specialty PTSD programs); TRICARE behavioral-health manual, current T-5 edition (baseline, approximately 60-day, and discharge collection). Ontario, Supporting Ontario's First Responders Act (2016) , and the provincial presumption family it typifies (DSM diagnosis by psychiatrist or psychologist required); Veterans Affairs Canada PTSD entitlement guideline (reviewed January 2025). Open Arms (Australian DVA), treating PTSD guidance (PCL-5 not a primary diagnostic tool; CAPS-5 gold standard); Australian DVA compensation guidance amended July 2026 (psychiatric evidence requirements). Schroeder & Bieu, 2024, and the 2026 active-duty cross-validation of embedded PCL-5 symptom-validity indices (preliminary tools with modest sensitivity); Crowe et al., 2025 (measurement invariance by race supported, partial by gender). Darnell et al., 2026 (weekly-version longitudinal psychometrics) and the official past-week PCL-5 form, February 2024. ## Related templates Trauma Assessment Outcome Measure Review Note Psychotherapy Progress Note PHQ-9 Score Documentation GAD-7 Score Documentation CAPS-5 Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/ciwa-ar All note templates # CIWA-Ar Documentation: Scoring, Protocols & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The CIWA-Ar (Clinical Institute Withdrawal Assessment for Alcohol, revised) is a 10-item alcohol-withdrawal severity scale published by Sullivan and colleagues in 1989 and scored 0 to 67. Hospital, emergency, and detoxification teams use it for serial monitoring under symptom-triggered and fixed-schedule protocols. This page covers how to document CIWA-Ar assessments defensibly, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Nurse-administered in most settings under protocol; combines observation with substantial patient self-report Audience Hospital and detox nursing teams, hospitalists, addiction medicine, ED and ICU clinicians, order-set committees, auditors Typical length One flowsheet row per assessment plus a decision note · administration about 2 to 5 minutes Format family Nurse-administered withdrawal scale (10 items: nine scored 0 to 7, one 0 to 4, total 67) When it's used Serial severity monitoring after alcohol withdrawal is diagnosed, symptom-triggered medication orders, escalation, protocol discontinuation Standards context The 1989 appendix grants free use of the scale; every threshold, interval, and stop rule is a guideline or order-set convention On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CIWA-Ar? The Clinical Institute Withdrawal Assessment for Alcohol, revised (CIWA-Ar) is a 10-item, clinician-administered measure of current alcohol-withdrawal severity, published by Sullivan, Sykora, Schneiderman, Naranjo, and Sellers in the British Journal of Addiction in 1989. Nine items are scored 0 to 7 and one, orientation and clouding of sensorium, is scored 0 to 4, so the maximum total is 67. The design history explains what the number can and cannot say. The authors selected a ten-item subset that efficiently reproduced the older 15-item CIWA-A total and validated it against that predecessor scale, not against seizures, delirium, ICU transfer, or death. Seizure was considered as an item and rejected (two events among 137 source observations, with timing unsuited to repeated scoring), blood pressure was tested and excluded, and pulse was recorded on the form but never scored: vital signs live outside the total by design. Two distinctions carry the documentation weight. First, licensing: the 1989 appendix states the scale "is not copyrighted and may be used freely," which is why it can sit in order sets and EHRs without a license, a permission the copyrighted CIWA-AD derivative and most adjacent instruments do not share. Second, the score is the instrument and everything else is a protocol. Severity bands, medication thresholds, reassessment intervals, and stop rules come from guidelines and local order sets, and they legitimately differ. The scale also assumes a patient who can participate: it measures apparent severity after alcohol withdrawal has been clinically identified, and it does not diagnose withdrawal (that lives in the substance use assessment ), predict who will later seize or develop delirium, or function in delirious, intubated, or non-communicative patients. Its opioid neighbor, the COWS , is a different instrument for a different syndrome; the scores are never comparable or interchangeable. ## Who uses CIWA-Ar documentation and when Hospital medicine and detoxification units run serial CIWA-Ar scores under symptom-triggered or fixed-schedule order sets, and nurses administer and chart most assessments, which raises the documentation stakes: under symptom-triggered orders the recorded score directly triggers medication. Emergency departments use it while the withdrawal diagnosis and disposition take shape. ICUs are where it is most often misapplied: ASAM recommends against CIWA-Ar for alcohol-withdrawal delirium because the scale depends on patient report, pointing instead to CAM-ICU, RASS, and MINDS-type instruments, and an ICU cohort study found CIWA-Ar documented in 56% of alcohol-withdrawal patients against 94% for RASS. In Canada, Alberta Health Services publishes score-linked intervals and notification triggers, while the 2023 CRISM national guideline builds risk assessment around PAWSS and clinical parameters rather than any CIWA number. In Australia the instrument itself must be verified before charting: Western Australia operationalizes a CIWA-Ar chart, while Queensland leans on the locally developed Alcohol Withdrawal Scale, so an Australian withdrawal chart may not be a CIWA-Ar at all. Medication decisions land in the medication management note , deterioration escalates through a crisis note , and the withdrawal course is summarized in the discharge summary . ## How to document CIWA-Ar results in the chart No statute prescribes a CIWA-Ar note format, and no law or accreditor sets a CIWA-Ar threshold or interval. What survives review is a record that makes the decision trace auditable: a valid assessment, the ordered criterion, the medication or the documented hold, the reassessment, and the response. Each element below carries the pitfall that most often undermines it. A time-stamped total with its component values. Record the actual bedside-assessment time (label late entries; the EHR entry time is not the assessment time), the total over 67, and all ten component values, ideally in a flowsheet, under the name CIWA-Ar with the active order set identified. Pitfall: a bare total. In a 2025 handoff study, unassessable components made a valid total impossible to calculate in 34.2% of paired assessments, and inter-rater correlation was 0.68; the components and observations are what make the number defensible. The two validity gates. Chart documented recent alcohol use (and how it is known) and that the patient was alert, communicative, and able to understand the questions, with interpreter details when language matters. Pitfall: skipping them. In a Mayo Clinic Proceedings review only 48% of patients on symptom-triggered therapy met both criteria, and a later sample found 20% with no documented recent alcohol use and 14% unable to communicate when the protocol was ordered. Separate objective observations. Blood pressure, pulse, respiratory rate, oxygen saturation, temperature, and sedation or arousability are recorded alongside, never inside, the score: the developers tested vital signs and left them out of the total. Pitfall: a falling total standing in for safety. Oversedation and respiratory depression develop outside the score, and ASAM requires monitoring them separately. Confounders and attribution. Say which component may be unreliable and why: pain, delirium from another cause, head injury, postictal states, baseline tremor, psychiatric symptoms, intoxication, or a language barrier. Pitfall: zero-filling unanswerable items or nudging a component toward a desired dose. A subtotal with silent zeros is not a valid CIWA-Ar; chart "not interpretable as a valid total" with the reason and escalate instead. The named protocol and its exact criterion. Name the order set with its version or date and paraphrase the band actually met. Hospitals legitimately trigger at 8, 10, or 12, or use a non-CIWA pathway; ASAM offers bands only as examples and states that severity classification is ultimately clinician judgment. Pitfall: "CIWA above 8 requires benzodiazepines" charted as a universal rule. No US, Canadian, or Australian statute or regulation sets any CIWA-Ar threshold. The medication or the documented hold. Drug, dose, route, and exact administration time tied to the criterion met; or the hold parameter, contraindication, sedation, diagnostic concern, refusal, or provider instruction, with who was notified, when, and the response. Pitfall: "Ativan given for CIWA," or a score-linked dose silently not given. Both break the trace between score and action. Reassessment against the ordered interval. When the reassessment was due, when it actually happened, the repeat total or the alternative assessment, the response, and adverse effects. Intervals are protocol property: ASAM describes every 1 to 4 hours initially for moderate, severe, or treated withdrawal, and Alberta, Western Australia, and New South Wales each publish different score-linked ladders. Pitfall: "serial CIWA per protocol" with no clock times, or an interval remembered from a different hospital. Escalation, discontinuation, and handoff. Notification triggers with times, persons, and orders; the exact local stop criterion (Western Australia, for example, requires four consecutive totals of 8 or less plus at least 48 hours since the last drink); any switch to CAM-ICU, RASS, or a MINDS-type instrument with the reason; and a handoff carrying the last valid total, trend, last dose and cumulative exposure, next due time, and unresolved confounders. Pitfall: quietly stopping assessments, or continuing totals after the patient can no longer participate. ### Blank template (copy and adapt) CIWA-AR DOCUMENTATION BLOCK Date: [ ] Actual assessment time: [ ] (label late entries) Setting: [ ] Assessor: [ ] Instrument: CIWA-Ar Order set: [name + version/date] Validity gates: [recent alcohol use + how known] [alert, communicative, understands questions: y/n] [interpreter service + language, if used] Total: [ ]/67 Components: [all 10 values in the flowsheet] Separate observations: [BP / pulse / RR / SpO2 / temp / sedation-arousability] Confounders + attribution: [affected items, why, action] Criterion: [ordered band, paraphrased] Met: [y/n] Medication: [drug, dose, route, exact time] OR Hold / variance: [reason; provider notified + time + response] Reassessment: [due -> done; repeat total; response; adverse effects] Escalation: [trigger, person notified, time, orders received] Stop / transition: [local stop criterion met, or switched to: instrument + reason] Handoff: [last valid total + trend, last dose + cumulative, next due, notification triggers] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale itself. ## Sample CIWA-Ar documentation (fictional) Scenario: uncomplicated moderate withdrawal on a symptom-triggered order set, from first assessment through step-down and handoff. The order set, bands, and doses are invented to illustrate documentation structure; they are not treatment instructions. All details are fictional. Patient: M.H., 48 · Setting: Medical unit, symptom-triggered order set · Clinician: R. Osei, RN / L. Grant, MD · Note date: 08/13/2026 08:00, assessment: Last drink approximately 14 hours ago per patient, corroborated by spouse. Alert, oriented, conversant in English, answers internally consistent. CIWA-Ar 12/67: tremor 3, paroxysmal sweats 2, anxiety 3, nausea 1, headache 2, agitation 1; perceptual items and disorientation 0; all ten component values in the flowsheet. Recorded separately: BP 148/92, pulse 104, RR 16, SpO2 98% on room air, afebrile, no sedation. The unit's Alcohol Withdrawal Order Set (revision April 2026, fictional) specifies lorazepam 2 mg PO for a valid total of 10 to 15: criterion met, dose administered 08:05, provider aware. Reassessment due in one hour per the order set. 09:05, reassessment (60 minutes post-dose): Easily arousable, no respiratory depression, RR 15. CIWA-Ar 6/67 (tremor 2, sweats 1, anxiety 2, headache 1). Below the ordered medication bands; no dose indicated. Baseline essential tremor considered and excluded: no tremor history per patient and chart. 13:05: CIWA-Ar 4/67, patient comfortable, tolerating oral fluids. Interval extended to every four hours per the order set's step-down rule; vital signs and sedation checks continue on the unit schedule. 19:00, handoff: Scores 12, 6, 4 today; last lorazepam 2 mg PO at 08:05 (cumulative 2 mg); next assessment due 21:05; both validity gates intact all shift; no active confounders. Notify provider for a total of 20 or higher, seizure, new disorientation or hallucinations, unstable vital signs, or oversedation. This sample is fictional and for educational purposes. It does not describe a real patient or record; the order set, bands, doses, and scores are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works Both validity gates are on the record: recent alcohol use with how it is known, and an alert, communicative patient, the pair the misuse studies most often find missing. Every assessment carries an actual bedside time, a total over 67, and component values, so the trajectory can be audited against the dose clock. Vital signs, arousability, and respiratory status are charted separately from the total, because the score contains none of them by design. The medication is tied to a named, dated order set and its paraphrased band, never to "CIWA protocol" in the abstract, and the confounder check (baseline tremor) is written down rather than assumed. The handoff carries the trend, cumulative exposure, next due time, and explicit notification triggers, which is where reviewers stop tracing. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: under 42 CFR 482.24 , hospital records must be complete, dated, timed, and authenticated, and standing orders and protocols must be adopted and periodically reviewed through medical, nursing, and pharmacy leadership (LAW), but nothing in it names a CIWA-Ar cutoff or interval. Federal nursing conditions require medication administration to follow practitioner orders and hospital policy (LAW), with scope details left to state nurse-practice acts. No accreditor-specific CIWA-Ar rule was verified: accreditation exposure runs through order-set currency, staff competency, administration per orders, reassessment records, and deterioration response (ACCREDITATION REQUIREMENT), so a specific threshold should never be labeled as accreditor-required without a citable current standard. No standalone CIWA-Ar CPT code or national payer threshold exists; the score supports severity and medical-necessity documentation but is not itself billable (PAYER POLICY). ASAM's 2020 guideline prefers symptom-triggered treatment when trained staff can reassess reliably and describes monitoring every 1 to 4 hours initially, relaxing to every 4 to 8 after sustained stability (CONVENTION). The oversight record is concrete: a 2025 VA Office of Inspector General inspection found inaccurate and delayed scoring and protocol-inconsistent medication administration in a fatal case, a 2024 national review found local-guidance gaps across 30 systems, and a settled federal suit alleged scores assigned without the underlying assessments (an allegation resolved by settlement, not an adjudicated finding). Canada and Australia run the same pattern with different anchors. The 2023 CRISM guideline recommends PAWSS plus clinical parameters for severe-withdrawal risk and non-benzodiazepine options for lower-risk patients (CONVENTION); Alberta Health Services publishes a protocol using every-four-hour scoring for totals of 0 to 9, hourly scoring at 10 or higher until three consecutive totals are below 10, provider notification after two scores of 20 or higher, and reconsideration of the protocol after five days (Alberta CONVENTION, not national law); and British Columbia's revised nursing documentation standards, effective April 2026, require timely, chronological records with actual times and labeled late entries (PROFESSIONAL REGULATION). Australia has no single national workflow: New South Wales guidance, mandatory in NSW Health facilities, describes roughly four-to-six-hourly assessment below 10, two-to-four-hourly from 10 to 20, and hourly above 20; Western Australia's chart uses four-hourly for 0 to 8, two-hourly for 9 to 14, and hourly at 15 or higher, with discontinuation after four consecutive totals of 8 or less and at least 48 hours since the last drink; Queensland leans on the Alcohol Withdrawal Scale, whose validation was never formally published. A 2025 New South Wales inquest into a custody death (an AWS case, not CIWA-Ar) recommended that the withdrawal-monitoring form itself carry observation-frequency guidance and a field for an individualized plan, documentation-design evidence any jurisdiction can learn from. The psychometrics counsel the same humility as the law. The 1989 validation reproduced the predecessor CIWA-A total rather than predicting outcomes; a 2025 handoff study found unassessable components blocking a valid total in 34.2% of paired assessments, with inter-rater correlation of 0.68; a 2019 analysis of 479 acutely ill and injured inpatients did not support the scale's reliability or validity in those populations; and an ICU cohort showed CIWA-Ar documented in only 56% of alcohol-withdrawal patients. A 1988 study associated totals above 15 with higher risk of severe withdrawal (relative risk 3.72), an association, not a validated treatment cutoff. On rights: the 1989 appendix grants free use of the scale itself; that permission does not extend to the journal article's typesetting, commercial repackagings, or derivatives, and the CIWA-AD adaptation is copyrighted. BastionGPT is not affiliated with the instrument authors or the journal. This page describes the instrument in original prose and reproduces no item text or scoring anchors. ↑ Back to the template and downloads ## Common CIWA-Ar documentation errors reviewers flag The misuse literature is unusually quantified. In a Mayo Clinic Proceedings review, only 48% of patients on symptom-triggered therapy met both validity criteria (recent alcohol use and intact verbal communication); a later general-hospital sample found 20% with no documented recent alcohol use, 57% with one or zero documented withdrawal risk factors, 14% unable to communicate, and benzodiazepine-associated adverse events in 15%. A 2025 handoff study could not compute a valid total in 34.2% of paired assessments. And a 2025 VA Office of Inspector General inspection connected a death to exactly these records: inaccurate and delayed scoring, and medication administration inconsistent with the facility's own CIWA-Ar protocol. The BastionGPT Clinical Advisory Board sees the same errors most often in CIWA-Ar documentation reviews: Thresholds charted as law. "CIWA above 8 requires benzodiazepines" cites a rule that does not exist. ASAM's bands (below 10 mild, 10 to 18 moderate, 19 and higher severe) are offered as examples with the explicit caveat that classification is ultimately clinician judgment, and institutions legitimately trigger at 8, 10, 12, or run non-CIWA pathways. Name the order set, its version, and the band actually met. Validity gates skipped. No documented recent alcohol use, or totals produced on patients who could not participate. The instrument was validated in communicative detoxification patients; the misuse literature describes protocolized use in complex inpatients as the equivalent of off-label application, which is exactly what the chart should surface rather than bury. Zero-filled or manufactured totals. Silently scoring unanswerable subjective items as zero, carrying ratings forward without a fresh assessment, or adjusting a component to reach or avoid a medication band. Chart "not interpretable as a valid CIWA-Ar total" with the reason; withdrawal-monitoring litigation has included allegations of scores charted without the assessment behind them. Bare totals with nothing around them. The total contains no blood pressure, pulse, respiratory rate, oxygen saturation, or sedation by design, so a falling score is not evidence of safe treatment; chart the separate observations and the component values that explain material ratings. Assessment time and entry time conflated. Late entries unlabeled, the EHR timestamp standing in for the bedside time, or reassessment intervals recalled from a different hospital instead of the ordered one. The wrong instrument continued. Totals produced on delirious, intubated, or postictal patients instead of a documented switch, with the reason, to CAM-ICU, RASS, or a MINDS-type scale, which is what ASAM recommends once patient report fails. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the assessment facts (times, totals and components, the validity gates, separate vitals and sedation, the named order set and band, doses and times) and it drafts the time-stamped documentation trace: assessment blocks, criterion reasoning, holds with notification, serial response, and the handoff, ready for your review. Cross-check a finished note for the gaps reviewers flag: a threshold charted as a requirement, a bare total with no components, missing validity gates, zero-filled items, or a falling score standing in for sedation and respiratory checks. Draft the escalation or handoff summary: last valid total and trend, doses against the clock with cumulative exposure, confounders, next assessment due, and explicit notification triggers, ready to confirm with the receiving team. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do CIWA-Ar scores actually mean? Ten items: nine scored 0 to 7 and one (orientation and clouding of sensorium) scored 0 to 4, for a maximum of 67. The total estimates the apparent severity of alcohol withdrawal at one moment, after the diagnosis has been made clinically. The bands attached to it are conventions that vary by source: ASAM's examples are below 10 mild, 10 to 18 moderate, and 19 or higher severe, offered with the explicit caveat that classification is ultimately clinician judgment. The score does not diagnose withdrawal, predict who will later seize or develop delirium, or contain any vital sign; it was validated by reproducing the older CIWA-A total, not by predicting outcomes. Cite whose bands you are using, or better, the order set actually in force. What CIWA-Ar score requires medication? None, by law: no US, Canadian, or Australian statute or regulation sets a CIWA-Ar medication threshold. The instrument's own appendix says only that patients scoring below 10 do not usually need additional withdrawal medication. Everything firmer is a protocol convention: one hospital's order set starts at 8, another at 10 or 12, and some units run non-CIWA pathways entirely. The randomized evidence (median 9 versus 68 hours of treatment and 100 versus 425 mg of chlordiazepoxide in the 1994 trial; 39% versus 100% of patients medicated in the 2002 trial) supports symptom-triggered dosing in communicative, selected patients, not automatic medication from any particular number. Chart the named order set, its version, and the band actually met. How often should CIWA-Ar be repeated? The instrument sets no interval; the order or protocol does. ASAM describes every 1 to 4 hours initially for moderate, severe, or pharmacologically treated withdrawal, relaxing to every 4 to 8 hours after sustained stability. Alberta's protocol uses every four hours for totals of 0 to 9 and hourly at 10 or higher until three consecutive totals fall below 10; Western Australia uses four-hourly for 0 to 8, two-hourly for 9 to 14, and hourly at 15 or higher; New South Wales describes roughly four-to-six-hourly below 10, two-to-four-hourly from 10 to 20, and hourly above 20. Chart the interval actually ordered, the actual reassessment times, and the reason for any delay, never a ladder remembered from a different hospital. Do you wake a sleeping patient for a CIWA-Ar? Local policy decides, and the chart should say what happened either way. Sleep is not a substitute for assessing the ten domains, and unanswered subjective items should never be silently scored zero; some services (Western Australian nursing education, for example) instruct staff to wake the patient for a scheduled assessment, which is a local rule rather than a property of the instrument. Whatever the policy, document arousability, respiratory status and sedation, what was genuinely observable while the patient slept, and why a full assessment was or was not completed, with the provider instruction when one was given. What if the score does not match the patient? Treat the mismatch as clinical data and never adjust a component to fix it. Verify the history and timing, repeat the assessment when appropriate, name the components affected by confounders (pain, delirium from another cause, head injury, baseline tremor, psychiatric symptoms, intoxication, language barriers), and use collateral information or an interpreter. If a score-linked instruction conflicts with what you see, a drowsy patient the number says to medicate, follow the ordered hold parameters, notify the prescriber, and document who was notified, when, the response, and the revised plan. Chart observable inconsistency neutrally; "patient exaggerating" without objective support is the phrasing that fails review. When is CIWA-Ar invalid, and what replaces it? The scale needs two validity gates: documented recent alcohol use and a patient who can communicate and understand the questions. A Mayo Clinic Proceedings review found only 48% of patients on symptom-triggered therapy met both. It is unreliable or invalid in delirium, intubation, deep sedation, postictal states, significant cognitive impairment, and unmanaged language barriers, and ASAM specifically recommends against CIWA-Ar for alcohol-withdrawal delirium because it depends on patient report, pointing to CAM-ICU, RASS, the Delirium Detection Score, or MINDS-type instruments instead. Document the switch explicitly: the last valid total, why the scale became invalid, the replacement monitoring plan, and the provider order behind it, rather than a bare "CIWA inaccurate." How do I document stopping the protocol? Against the local stop criterion, because no universal one exists. Western Australia requires at least 48 hours since the last drink plus four consecutive totals of 8 or less; Alberta scores hourly until three consecutive totals fall below 10 and asks whether the protocol is still needed after five days; ASAM allows reduced monitoring after sustained stability. Chart the exact criterion met, the last valid score and trend, cumulative sedative exposure, the order or protocol authority for stopping, what monitoring continues, and the handoff. A short symptom-triggered benzodiazepine course does not automatically require a prolonged taper; that is an individualized prescriber decision, documented as such. Is the CIWA-Ar copyrighted, and can we put it in our EHR? The 1989 appendix states the scale "is not copyrighted and may be used freely," which is why it can be embedded in order sets and EHRs without a license, and it is the exception among withdrawal instruments. The permission covers the scale itself, not the journal article's typesetting or commentary, not commercial repackagings, and not derivatives: the CIWA-AD adaptation is copyrighted, and adjacent instruments (PAWSS, the Australian AWS, SEWS, COWS) carry their own rights and should be described, not reproduced. A locally modified form should be identified by organization and version rather than treated as psychometrically identical to the original. Can BastionGPT draft my CIWA-Ar documentation? Yes. Give it the assessment facts (times, totals and components, the validity gates, separate vitals and sedation, the named order set and band, doses and times, confounders) and it drafts the full trace: assessment blocks, criterion reasoning, hold documentation with notification, serial response, escalation, and the handoff, ready for your review. It can also check a finished note for thresholds charted as requirements, missing validity gates, zero-filled items, bare totals, and missing clock times. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Sullivan JT, Sykora K, Schneiderman J, Naranjo CA, Sellers EM, 1989, British Journal of Addiction, the CIWA-Ar publication (scale architecture, the free-use appendix statement, and the design decisions on seizure, blood pressure, and pulse); the federally hosted reproduction . Shaw JM and colleagues, 1981, Journal of Clinical Psychopharmacology, the CIWA-A predecessor ; Reoux JP and Oreskovich MR, 2006, American Journal on Addictions (CIWA-AD agreement: 0.45 points mean difference, within one point 82.6% of the time). ASAM, 2020, Clinical Practice Guideline on Alcohol Withdrawal Management (band examples with the clinician-judgment caveat, monitoring conventions, the symptom-triggered preference, and the recommendation against CIWA-Ar in withdrawal delirium). Saitz R and colleagues, 1994, JAMA, the symptom-triggered randomized trial (median 9 versus 68 hours; 100 versus 425 mg chlordiazepoxide); Daeppen JB and colleagues, 2002, Archives of Internal Medicine (39% versus 100% of patients medicated). Hecksel KA and colleagues, 2008, Mayo Clinic Proceedings, inappropriate symptom-triggered therapy (48% met both validity criteria); Eloma AS and colleagues, 2018, American Journal of Drug and Alcohol Abuse (20% no documented recent alcohol use, 14% unable to communicate, 15% benzodiazepine-associated adverse events). Kane and colleagues, 2025, prospective paired-handoff study (valid totals blocked by unassessable components in 34.2% of CIWA-Ar assessments; inter-rater correlation 0.68); a 2019 psychometric analysis of 479 trauma and internal-medicine inpatients; Steel and colleagues, 2021, Addiction Science and Clinical Practice (ICU documentation 56% for CIWA-Ar versus 94% for RASS). VA Office of Inspector General, 2025 healthcare inspection (inaccurate and delayed scoring and protocol-inconsistent medication administration in a fatal case) and the 2024 national review of alcohol-withdrawal guidance across 30 systems. Reporting on the Estate of Murray settlement (the allegation of scores assigned without the underlying assessments; resolved by settlement, not adjudicated). Alberta Health Services CIWA-Ar protocol (score-linked intervals, notification after two scores of 20 or higher, and the five-day review); British Columbia College of Nurses and Midwives documentation standards, effective April 2026. NSW Health, clinical guidance appendices (CIWA-Ar and AWS) ; the 2025 Visvanathan inquest recommendation (withdrawal-form redesign); Western Australia, the state CIWA-Ar chart (intervals and the discontinuation rule). CRISM, 2023, Canadian guideline for high-risk drinking and alcohol use disorder (PAWSS-based risk assessment and non-benzodiazepine options for lower-risk patients). Foy A, March S, Drinkwater V, 1988, Alcoholism: Clinical and Experimental Research (totals above 15 and severe-withdrawal risk, relative risk 3.72); Medscape clinical reference, the CIWA-AD copyright distinction . ## Related templates COWS Protocol Documentation Substance Use Assessment Crisis Psychotherapy Note Discharge Summary CAGE Questionnaire Documentation AUDIT-C Score Documentation RASS Score Documentation CAM-ICU Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cows All note templates # COWS Documentation: Scoring, Thresholds & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The COWS (Clinical Opiate Withdrawal Scale) is an 11-item, clinician-administered opioid-withdrawal measure published by Wesson and Ling in 2003 and scored 0 to 48. Emergency, hospital, addiction-medicine, and opioid treatment program teams use it around buprenorphine initiation and withdrawal monitoring. This page covers how to document COWS assessments defensibly, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinician-administered (nurses and prescribers); combines measured, observed, and patient-reported findings Audience Induction and observation teams, addiction consultants, ED and OTP programs, auditors and payers reviewing MOUD records Typical length One flowsheet row per assessment plus a decision note · administration about 2 to 5 minutes Format family Clinician-administered withdrawal scale (11 unevenly weighted items, scored 0 to 48) When it's used Buprenorphine initiation decisions, serial withdrawal monitoring, precipitated-withdrawal recognition, methadone-to-buprenorphine transfers Standards context Express clinical-copy permission on the published version; not public domain; no law or regulation sets a COWS threshold On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the COWS? The Clinical Opiate Withdrawal Scale (COWS) is an 11-item, clinician-administered measure of apparent opioid-withdrawal severity, published by Donald Wesson and Walter Ling in the Journal of Psychoactive Drugs in 2003. The arithmetic matters more than most pages admit: the eleven item maxima sum to 48, not the 0-to-47 range some secondary papers repeat, and the items are unevenly weighted (four domains reach 5 points, one of them scored only 0, 3, or 5), so the total is an ordinal composite rather than an equal-interval scale. The published severity bands are 5 to 12 mild, 13 to 24 moderate, 25 to 36 moderately severe, and above 36 severe, which means severe starts at 37, a score of 36 is still moderately severe, and 0 to 4 sits below the published bands with no named category to invent. The developers derived those bands from clinical expertise, not statistics, and later validation work confirmed exactly that, so the bands describe severity and were never validated as treatment cutoffs. Two more distinctions carry the documentation weight. First, licensing: the version distributed through federal resources carries express permission to be copied and used clinically, but that is not public-domain status; current instrument-registry records list a distributor with licensing conditions for some uses (funded research, commercial products, electronic implementations), so copy the authorized form for care and get rights clarification before republishing or embedding it in software. Second, scope: a COWS total estimates withdrawal severity at one moment. It does not diagnose opioid use disorder (that lives in the substance use assessment ), establish which opioid was used or when, prove tolerance, guarantee any medication decision, or rule precipitated withdrawal in or out. And its alcohol-withdrawal neighbor, the CIWA-Ar , is a different instrument for a different syndrome; the scores are never comparable or interchangeable. ## Who uses COWS documentation and when Emergency departments document COWS around ED-initiated buprenorphine, where the dominant pathway lineage starts at a score of 8 with objective signs; hospital and addiction-consult teams run serial scores through observed initiations, high-dose protocols, and low-dose overlap starts that deliberately use no threshold at all; opioid treatment programs fold it into individualized assessments, noting that the 2024 federal OTP rules require no COWS score for methadone or buprenorphine admission; and office-based and telehealth prescribers use it in observed and home-initiation workflows, where what was clinician-observed versus patient-reported must stay distinct. In Canada, provincial pathways set their own criteria (British Columbia's traditional pathway uses a score of at least 12; Ontario's ED toolkit uses at least 13), and Australia's national guideline and state rules run their own thresholds. The decision trace lands in the medication management note , acute deterioration escalates through a crisis note , and relapse-prevention planning picks up at discharge in a relapse prevention plan . ## How to document COWS results in the chart No statute prescribes a COWS note format, and no law sets a COWS threshold. What survives review is a record that makes the decision trace auditable: what was observed, when, under which named protocol, what was done, and what happened next. Each element below carries the pitfall that most often undermines it. A time-stamped total with its component values. Record the exact clock time, the total over 48, and every component value, ideally in a flowsheet. Pitfall: a bare total. The items are unevenly weighted, two identical totals can reflect different clinical pictures, and inter-rater agreement on the severity category was only moderate in ED testing, so the components are what make the number defensible. The objective substrate, separated from report. Preserve what was measured and observed: the actual pulse, visible sweating, pupil appearance in the prevailing light, observed tremor and restlessness, piloerection, tearing. The instrument scores only findings apparently attributable to withdrawal. Pitfall: confounders scored in. Exertion, ambient heat, stimulant effects, chronic pain, anxiety, allergy, and other withdrawal syndromes inflate totals, and the chart should say what was attributed away and why. Last-use history, with its uncertainty. Substance and suspected substance, route, usual pattern, exact or approximate last use, and exposure to methadone, extended-release opioids, fentanyl, sedatives, or alcohol. Pitfall: charting an opioid-free interval as fact when the history is approximate; with fentanyl in the supply, elapsed-time rules are unreliable and the uncertainty itself belongs in the note. The named protocol and its exact criterion. Name the pathway with its version or date, and paraphrase its criterion, including any objective-sign requirement. Authoritative pathways legitimately span a score of at least 6, at least 8 with objective signs, at least 12, and at least 13, and low-dose overlap protocols use no threshold gate at all, which the note should state plainly when that pathway is chosen. Pitfall: "COWS 12 required before buprenorphine" charted as law. No US, Canadian, or Australian statute or regulation sets any COWS threshold. The decision, dose, and exact times. Initiate, defer, switch pathways, observe, or escalate, with the rationale; then medication, dose, route, and administration time, followed by serial scores with elapsed times. Pitfall: "serial COWS per protocol" with no clock times. The trajectory only means something against the dose clock. Response, including any suspected precipitated withdrawal. Document worsening as a timed change from the pre-dose baseline: the numerical rise, elapsed minutes, new objective findings, vital signs and sedation, the differential (spontaneous progression, inadequate dosing, stimulants, acute illness), treatment, and response. Pitfall: a verdict from one number. Published definitions differ (a rapid rise above 5 with objective signs is one common description; others use at least 6 within an hour), so "suspected precipitated withdrawal" with the delta and reasoning is the defensible phrasing. Disposition and the safety net. Observation status, naloxone provision, discharge medication or bridge, confirmed or attempted linkage, return precautions, and who owns follow-up. Pitfall: an induction note that ends at the dose. Reviewers trace the record to the handoff, not to the medication order. ### Blank template (copy and adapt) COWS DOCUMENTATION BLOCK Date / exact time: [ ] Setting: [ ] Assessor: [ ] Mode: [in person / video; phone report is not an observed COWS] Total: [ ]/48 Components: [all 11 values in the flowsheet] Objective findings: [measured pulse / sweating / pupils / tremor / restlessness / piloerection / tearing, as observed] Attribution notes: [confounders considered and excluded] Last use: [substance; route; time +/- uncertainty; fentanyl, methadone, XR-opioid, sedative, alcohol exposure] Safety: [arousal, respiratory status, vitals, intoxication] Protocol: [name + version/date] Criterion: [paraphrase, incl. objective-sign requirement / low-dose: no threshold gate] Criterion met: [y/n] Decision + rationale: [ ] Medication: [drug, dose, route, exact time] Serial scores: [time -> total, elapsed from dose] Response: [improved / progressing / suspected precipitated withdrawal: delta, minutes, objective signs, differential] Interventions + response: [ ] Disposition: [observation / admit / discharge meds / naloxone / follow-up owner + date / return precautions] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale itself. ## Sample COWS documentation (fictional) Scenario: an observed ED buprenorphine initiation where the first assessment does not meet the local criterion and the second does. The protocol, threshold, and doses are invented to illustrate documentation structure; they are not treatment instructions. All details are fictional. Patient: T.R., 31 · Setting: Emergency department, observed initiation · Clinician: E. Vance, RN / J. Malik, MD · Note date: 08/13/2026 09:10, first assessment: Reports last use of intranasal opioid believed to contain fentanyl approximately 20 hours ago (time approximate); denies methadone, extended-release opioids, alcohol, or sedatives. COWS 10/48, all 11 components in the flowsheet; objective findings limited to mildly enlarged pupils and facial moisture. Alert, respiratory rate 16, oxygen saturation 98% on room air. The ED's Observed Buprenorphine Pathway (revision March 2026, fictional) requires a COWS of at least 12 with at least one objective sign: criterion not met. Medication deferred; repeat assessment ordered; symptoms and plan reviewed with T.R. 10:00, second assessment: COWS 13/48 (components in flowsheet). Observed: facial sweating, enlarged pupils, frequent position changes, visible hand tremor, tearing. No sedation or intoxication; no alternate acute diagnosis identified. Criterion met. Buprenorphine/naloxone 4 mg sublingual administered 10:08 per the pathway. Serial response: 11:10 (62 minutes post-dose): COWS 8, partial relief, tremor and facial sweating resolved, no abrupt worsening; second 4 mg dose at 11:18 per the same order set. 13:20: COWS 3, comfortable, ambulatory, no sedation or respiratory change at any point. Disposition: Discharged with the pathway's bridge prescription, naloxone kit dispensed and teach-back completed, clinic appointment tomorrow 09:00 (confirmed by the ED navigator, named in the chart), and return precautions reviewed, including abrupt worsening after any dose. Follow-up owner: bridge clinic; ED to be notified of any bounce-back. This sample is fictional and for educational purposes. It does not describe a real patient or record; the protocol, threshold, doses, and scores are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works Every assessment carries an exact clock time, a total over 48, and flowsheet components, so the trajectory can be audited against the dose times. The objective findings are named concretely and separated from reported symptoms, which is what makes the totals defensible given the instrument's known rater variability. The last-use history is recorded with its uncertainty, including the fentanyl suspicion that shapes the pathway choice and monitoring. The threshold is attributed to a named, dated local pathway and paraphrased exactly, including the objective-sign requirement, never stated as a legal rule. The record runs through to disposition: naloxone, a confirmed follow-up owner, and return precautions, which is where reviewers stop tracing. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations The US regulatory frame around COWS documentation changed almost completely between 2023 and 2026, and none of it created a COWS requirement. The MAT Act provision of the Consolidated Appropriations Act, 2023 eliminated the X-waiver and its patient caps (LAW), leaving ordinary DEA registration, state law, and the MATE Act's one-time eight-hour training attestation; the 2024 revision of 42 CFR Part 8 (effective April 2024, compliance October 2024) rewrote OTP admission, assessment, and take-home rules and requires individualized methadone dosing that considers withdrawal severity, yet never names COWS or any score for admission or medication (LAW); and telemedicine runs on two coexisting federal pathways as of August 2026: the finalized buprenorphine telemedicine rule and the broader fourth temporary extension of COVID-era flexibilities , which is scheduled to sunset December 31, 2026, while the special-registration framework remains a proposal, not law (all dated facts worth re-verifying before year-end). PDMP duties are state law with state-specific timing and exceptions, and payer rules are their own layer: Medicare's OTP bundles and office-based OUD codes require documentation supporting the billed service, and no national payer policy reviewed makes a COWS value a condition of payment, so a threshold should never be charted as a payer requirement unless the active plan policy actually says it. Canada and Australia run the same pattern: national guidance stays threshold-neutral while named pathways set conventions. Canada's CRISM 2024 national guideline expressly leaves induction protocols out of scope; British Columbia's BCCSU pathway requires moderate withdrawal (a COWS of at least 12 in its published criterion) for traditional initiation while its low-dose pathway uses no gate, and Ontario's META ED toolkit uses at least 13. Australia's national guideline dates to April 2014 (a fact worth displaying, since it predates the fentanyl-era literature), describes a methadone-to-buprenorphine transfer pathway at a score of at least 13, and sits atop state-based LAW for authorization and records, with New South Wales' new opioid treatment standards taking effect November 5, 2026. On the measurement itself, the psychometrics counsel humility: the validation study centered on mild withdrawal, ED inter-rater agreement on the severity category was moderate at best with only the pulse item reaching substantial agreement, a one-factor model did not fit in a 516-participant analysis, and no minimal clinically important difference has ever been established, all of which is why the components, times, and observations, not the bare total, carry the record. If a translated version is used, name it; listed translations are not automatically validated. The COWS was published by Wesson and Ling (2003) in the Journal of Psychoactive Drugs. The version distributed through federal clinical resources carries express permission to be copied and used clinically; that permission is not public-domain status, and current instrument-registry records list Mapi Research Trust as distributor with licensing conditions for some uses, including commercial and electronic implementations. BastionGPT is not affiliated with the instrument authors, the journal's publisher, or Mapi Research Trust. This page describes the instrument in original prose and reproduces no scale text, anchors, or scoring materials. ↑ Back to the template and downloads ## Common COWS documentation errors reviewers flag The numbers behind these errors deserve more respect than they get. In paired ED ratings, agreement on the published severity category was 67.5% with a weighted kappa of 0.55 , and only the pulse item reached substantial agreement; the validation study's mean peak score was 7.6, meaning the instrument was validated chiefly for mild withdrawal; no minimal clinically important difference has ever been established; and published precipitated-withdrawal incidence spans roughly 0.76% in a 1,200-patient ED trial to 11.5% among hospitalized patients using fentanyl , depending on setting and definition. The BastionGPT Clinical Advisory Board sees the same errors most often in COWS documentation reviews: Bands misread and ranges garbled. A 36 charted as severe (severe starts above 36), a category invented for scores of 0 to 4 (they sit below the published bands), a 0-to-47 range repeated from secondary papers (the item maxima sum to 48), or the expert-derived severity bands treated as validated induction cutoffs, which they never were. Thresholds charted as law. "COWS 12 required before buprenorphine" cites a rule that does not exist. Authoritative pathways span at least 6, at least 8 with objective signs, at least 12, and at least 13, low-dose overlap uses no gate at all, and federal OTP regulations require no COWS for methadone or buprenorphine. Name the protocol, its version or date, and its exact criterion. Bare totals with no components or attribution. Two identical totals can come from different clinical pictures, confounders (heat, exertion, stimulants, pain, allergy, another withdrawal syndrome) inflate scores, and the inter-rater data show why an unexplained number cannot authenticate itself. Chart the component values, the objective findings, and what was attributed away. Precipitated-withdrawal verdicts from a single score. A high post-dose total does not distinguish precipitated withdrawal from spontaneous progression, inadequate dosing, stimulant effects, or acute illness. Document the timed delta from the pre-dose baseline, the new objective findings, the differential, treatment, and response, and phrase it as suspected: published definitions genuinely differ. Missing clocks and misattributed modes. "Serial COWS per protocol" with no times, doses without administration times, or a telephone symptom report charted as an observed COWS. Audio cannot observe pupils, tremor, sweating, or piloerection; chart phone contacts as patient-reported updates or use an explicitly adapted protocol. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the assessment facts (times, totals, components, last-use history, the named protocol and criterion, doses and times) and it drafts the time-stamped documentation trace: assessment blocks, decision rationale, serial response, and disposition, ready for your review. Cross-check a finished note for the gaps reviewers flag: a threshold charted as a requirement, a bare total with no components, missing clock times, or a precipitated-withdrawal verdict with no delta, differential, or timed trajectory. Draft the handoff or transfer summary: last-use history with its uncertainty, score trajectory against dose times, response, and the follow-up owner, ready to confirm with the receiving team. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do COWS scores actually mean? Eleven unevenly weighted items sum to a total of 0 to 48 (some secondary papers say 47, but the published item maxima add to 48). The developers' bands are 5 to 12 mild, 13 to 24 moderate, 25 to 36 moderately severe, and above 36 severe, so severe starts at 37, and 0 to 4 simply sits below the published bands. Two caveats keep the number honest: the bands came from the authors' clinical expertise rather than statistical derivation, and the score describes apparent withdrawal severity at one moment. It is not a diagnosis of opioid use disorder, evidence of what was used or when, or a validated treatment cutoff. What COWS score is required before starting buprenorphine? None, by law: no US, Canadian, or Australian statute or regulation sets a COWS threshold. What exists are protocol conventions that genuinely differ: common ED pathways start at 8 with objective signs, a 2021 federal quick-start guide suggested above 12 with a higher bar when fentanyl is suspected, British Columbia's traditional pathway uses at least 12, Ontario's ED toolkit uses at least 13, one Australian state pathway uses 6 with objective signs, and low-dose overlap protocols deliberately use no threshold at all. The defensible chart names the active protocol with its version or date and paraphrases its exact criterion; "COWS 12 required" with no source is the phrasing that fails review. How do I document suspected precipitated withdrawal? As a timed delta, not a verdict. Record the pre-dose score with its components and time, the dose and administration time, the onset of worsening, the repeat score with the numerical rise and elapsed minutes, the new objective findings (not just reported distress), vitals and sedation, the differential (spontaneous progression, inadequate initial dosing, stimulant effects, acute illness), the treatment with times, and the response. Published definitions differ: one common description is a rapid rise of more than 5 with objective worsening, another uses at least 6 within an hour, so "suspected buprenorphine-precipitated withdrawal, nine-point rise with new objective findings 25 minutes after the dose" is defensible where "COWS 18, precipitated withdrawal" is not. Is a COWS required before methadone? No. The 2024 revision of the federal opioid treatment program regulations (42 CFR Part 8) requires individualized methadone dosing that considers withdrawal severity, opioid exposure, other substances, and treatment history, and it requires documentation supporting admission and treatment, but it never names the COWS or any score for admission or dosing. A program may adopt COWS locally to structure that assessment, which makes it an organizational convention worth following and citing as such. For methadone-to-buprenorphine transfers the scale is more directly relevant, and Australia's national guidance describes a transfer pathway using a score of at least 13, again a dated protocol convention rather than a universal rule. How often should COWS be repeated? The instrument sets no interval; the pathway does. Published and institutional examples range from reassessment 30 to 60 minutes after a dose to one-to-three-hour intervals scaled to severity, and fentanyl-era guidance leans toward more frequent checks without naming one mandatory schedule. What matters for the record is the actual clock time of every assessment and dose, so the trajectory reads against elapsed time, plus a stated reason when a reassessment was delayed, skipped, or accelerated. "Serial COWS per protocol" with no times is the version that cannot be audited. Can I do a COWS by phone or video? By video, partially: a clinician can observe some findings (restlessness, yawning, tremor, sometimes tearing) and should document the modality and what was genuinely observable. By phone, no: audio cannot observe pupils, sweating, piloerection, tremor, or motor behavior, so a telephone contact should be charted as a patient-reported withdrawal update or run under an explicitly adapted protocol, never as a complete clinician-observed COWS. For home initiations, the note documents the selection rationale, the written protocol and its criterion, teach-back, naloxone, the emergency plan, and how each follow-up was actually conducted (phone report, video observation, or in-person exam). A structured self-report scale is the honest tool for patient-scored home monitoring. Is the COWS objective, and can patients game it? It is deliberately mixed: direct measurement (pulse), observation (pupils, sweating, tremor, restlessness, piloerection), patient report (aches, anxiety, gastrointestinal symptoms), and blended items. That design makes it clinically complete and also explains the reliability data: in paired ED ratings, only the pulse item reached substantial agreement. The scoring rule that keeps it defensible is that higher ratings in most domains require some observable correlate, so document the objective substrate behind every nonzero item. When report and observation diverge, chart both neutrally: what was reported, what was and was not observed, the confounders, and how the discrepancy affected the decision; no published evidence supports labeling a patient as exaggerating from discordant findings alone. Can we copy the COWS into our EHR, app, or website? Copying the authorized version for clinical use is expressly permitted: the federally distributed form states it may be copied and used clinically. That permission is narrower than people assume. It is not public-domain status: current instrument-registry records list a distributor (Mapi Research Trust) with licensing conditions for some uses, including funded research, commercial products, and electronic implementations, and registry metadata about the copyright genuinely conflict, which is itself a reason for caution. The safe pattern: use authorized copies in care, chart results freely, link to an authorized source rather than republishing the scale on a public site, and get rights clarification before embedding the instrument in software or a paid product. Can BastionGPT draft my COWS documentation? Yes. Give it the assessment facts (times, totals and components, objective findings, last-use history with its uncertainty, the named protocol and criterion, doses and times) and it drafts the full decision trace: assessment blocks, criterion-met reasoning, serial response, suspected-precipitated-withdrawal documentation with the timed delta and differential, and the disposition with follow-up ownership, ready for your review. It can also check a finished note for thresholds charted as requirements, missing components or clock times, and verdict phrasing. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Wesson & Ling, 2003, Journal of Psychoactive Drugs, the COWS publication ; the federally distributed instrument copy (the clinical-copy permission statement and the published severity bands). Tompkins et al., 2009, Drug and Alcohol Dependence, the controlled validation study : concurrent validity against CINA (r = 0.85), internal consistency 0.78, a mean peak score of 7.6, and the note that the severity bands were expert-derived. Tomaszewski et al., 2021, JACEP Open, ED inter-rater reliability : 67.5% category agreement, weighted kappa 0.55, and only the pulse item reaching substantial agreement. D'Onofrio et al., 2023, JAMA Network Open, precipitated withdrawal in 1,200 ED initiations (0.76%); Thakrar et al., 2024, JAMA Network Open, hospitalized patients using fentanyl (11.5%); Varshneya et al., 2022, Journal of Addiction Medicine, withdrawal risk by time since fentanyl . CADTH clinical review, no established minimal clinically important difference for the COWS . ASAM, National Practice Guideline (2020 focused update) and its 2023 clinical considerations on buprenorphine initiation in the fentanyl era; SAMHSA, TIP 63 (2021 revision) . SAMHSA, 42 CFR Part 8 final rule resources (2024): OTP admission and dosing documentation with no COWS requirement. DEA/HHS telemedicine framework: the buprenorphine telemedicine final rule and the fourth temporary extension through December 31, 2026 . Yale, ED-initiated buprenorphine evidence (the D'Onofrio pathway lineage); CA Bridge, on-shift initiation resources (clinical-judgment initiation without a fixed number). British Columbia, BCCSU opioid use disorder guideline (traditional threshold of at least 12; low-dose pathway without a gate); CRISM 2024 national guideline, CMAJ (2018 original; induction detail out of scope in the 2024 update). Australia: Australian Prescriber on opioid-dependence medicines and SA Health opioid withdrawal management (state threshold examples; the national guideline dates to 2014). Maryland MACS, precipitated withdrawal and induction fact sheet ; Hämmig et al., 2016, the Bernese low-dose method (no threshold gating). ## Related templates CIWA-Ar Protocol Documentation Substance Use Assessment Medication Management Follow-Up Note Crisis Psychotherapy Note DAST-10 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/asrs All note templates # ASRS Documentation: Scoring, Versions & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The ASRS v1.1 (Adult ADHD Self-Report Scale) is a six-question adult ADHD screener developed through the WHO World Mental Health initiative and published by Kessler and colleagues in 2005. Primary care, psychiatry, and telehealth teams use it at intake and to track treatment. It screens; it never diagnoses. This page covers how to document ASRS results defensibly, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report, scored and interpreted by the clinician; Canadian practice adds an observer version when feasible Audience Primary care and psychiatry, ADHD assessment services, telehealth platforms, payers reviewing stimulant requests, accommodation evaluators Typical length A result line plus interpretation in the note · six-question screener takes about 2 to 5 minutes Format family Self-report screening scale (six items; Part B of the full checklist adds 12 interview probes) When it's used Adult ADHD intake screening, pre-assessment triage, treatment-response tracking, observer corroboration in Canadian workflows Standards context The six-question v1.1 screener is free to use; the 18-item checklist and ASRS-5 require permission; no law requires an ASRS On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the ASRS? The World Health Organization Adult ADHD Self-Report Scale version 1.1 is a screening instrument published by Kessler and colleagues in Psychological Medicine in 2005, developed through the WHO World Mental Health survey initiative. The full Symptom Checklist has 18 questions mapping to the adult ADHD symptom set; the six most predictive form Part A, the validated screener, and the remaining 12 form Part B, which supplies interview probes and no score. The authoritative copies live on Harvard Medical School's National Comorbidity Survey site, with copyright held by New York University and the President and Fellows of Harvard College and permissions routed through the NYU team. Harvard's stated position is that use of the six-question v1.1 screener is "free and does not require any formal permission or approval," with a citation, the copyright notice, and an unaltered form expected; the 18-question checklist and the newer ASRS-5 each require a permission request. Free to use is not public domain. The documentation problem is version control, because three scoring systems now coexist. The original v1.1 rule counts how many of the six Part A responses reach their item-specific thresholds, with four or more a positive screen; the thresholds differ by item, so it is wrong to describe the rule as four answers of Often or Very Often. A second, continuous method sums the six frequency responses to a 0 to 24 total with 14 or higher positive and four probability strata, drawn from the 2007 health-plan validation and endorsed as an alternative scoring in a Harvard update dated February 28, 2024. The ASRS-5, published in 2017 for DSM-5, is a different six-item instrument with machine-learned unequal weights, a corrected range of 0 to 25, and a threshold of 14 or higher. All three are screens. None is a severity scale, and none makes the diagnosis: that belongs to the full evaluation documented in an ADHD evaluation report , with normed instruments like the CAARS-2 playing their own distinct role. ## Who uses ASRS documentation and when Primary care and psychiatry use the ASRS at adult intake and as a longitudinal symptom measure, and the United States is the jurisdiction where its paper trail matters most while having the least national guidance: as of August 2026 there is still no released US adult ADHD clinical practice guideline (the first, from APSARD, is anticipated for distribution in 2027), so American clinicians document against DSM-5-TR criteria, payer policies, and clinical convention. Canadian practice follows the CADDRA workflow, which pairs the patient ASRS with an observer version when feasible, a functional-impairment measure, and a diagnostic interview, and states plainly that the toolkit alone cannot make the diagnosis. Australian practice sits on the 2022 AADPA guideline, which prohibits diagnosis from rating scales alone, layered over fast-moving state stimulant-authority reforms. Telehealth platforms lean on the ASRS heavily, which is exactly where screening-quality problems concentrate, and accommodation evaluators encounter it inside multi-method documentation standards that reject internet checklists on their own. Prescribing decisions and their monitoring live in the medication management note ; youth assessment uses age-specific tools covered on the Conners 4 page. ## How to document ASRS results in the chart No statute prescribes an ASRS note format, and no US, Canadian, or Australian law requires an ASRS, positive or otherwise, before diagnosing ADHD or prescribing a stimulant. What survives review is a record that treats the screen as one data source and shows the reasoning around it. Each element below carries the pitfall that most often undermines it. Instrument identity and scoring method. Record the exact instrument (v1.1 Part A alone, Part A plus Part B, or ASRS-5), the language, the reporter (patient, observer, or other), the administration context, and which scoring system produced the number. Pitfall: "ASRS 14" with nothing else. Three scoring systems share 14 as a meaningful boundary: it could be a v1.1 continuous total at threshold, an ASRS-5 weighted total at threshold, or a miscomputed count, and the three are not interchangeable. The result, stated precisely. For the original v1.1 rule, chart how many of the six Part A responses reached their item-specific thresholds and whether the screen was positive, for example "5 of 6 responses reached their item-specific screening thresholds; positive screen." That preserves the result without reproducing the scoring grid. Any continuous 0 to 24 total is identified separately, with its strata treated as probability regions rather than severity grades. Pitfall: rebranding screen counts or strata as mild, moderate, or severe ADHD; no ASRS number grades severity, and no minimal clinically important difference has been established. The clinical meaning, kept modest. A positive screen supports proceeding to a comprehensive assessment and is not independently diagnostic; a negative screen lowers probability but does not exclude the diagnosis, since the six-item sensitivity was 68.7% in the original validation. Pitfall: "ASRS positive, ADHD confirmed." The screen cannot establish onset, pervasiveness, impairment, or the differential. Childhood-onset evidence. Document that relevant difficulties were present before age 12: school records or report-card comments, childhood evaluations, parent or sibling accounts, repeated concrete historical examples, and the limits of each. Pitfall: "patient reports lifelong ADHD" standing alone. State what the patient recalls and what corroborates it, or does not. Cross-setting impairment. Record examples from at least two important settings and separate symptoms from functional consequences, including impairment concealed by heavy compensation. Pitfall: a symptom list with no functional analysis; frequent distractibility without clinically significant impairment does not meet criteria. Collateral, and what was unavailable. Name the records and informants reviewed, whether the informant knew the patient in childhood, material agreements and disagreements, and what could not be obtained and why. Pitfall: treating missing collateral as either disqualifying or ignorable. It lowers or qualifies confidence and shifts weight to other longitudinal evidence; document that reasoning. Differential and comorbidity. Record the plausible alternatives considered: depression, anxiety, bipolar spectrum, trauma, substance use, sleep disorders and restriction, learning disorders, and medical contributors. The false-positive data make this the load-bearing element: in a major-depression sample the v1.1 positive predictive value was 21.4%, and in treatment-seeking substance-use patients it was 0.26. Pitfall: reading the famous 99.5% specificity, a community-sample figure, as if it applied in comorbid clinical populations. Diagnostic status, validity, and the next step. Use an explicit status (screen positive, assessment pending; provisionally diagnosed; criteria met; not established; deferred pending records or sleep and mood stabilization) and document any external incentive or inconsistency neutrally and behaviorally. Pitfall: character labels. "Drug-seeking" and "malingering" are conclusions the ASRS cannot support; it has no embedded validity scale, and simulation studies show the symptom profile is easy to produce. Chart the facts, the missing data, and how they limit confidence. ### Blank template (copy and adapt) ASRS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Reporter: [patient / observer] Instrument: [v1.1 Part A / Part A + B / ASRS-5] Language: [ ] Scoring method: [item-threshold count / 0-24 sum / ASRS-5 weighted] Result: [count or total + positive/negative at stated threshold] Interpretation: [screen result only; not independently diagnostic] Childhood onset (<12): [records, informants, examples + limits] Cross-setting impairment: [setting 1] [setting 2] [functional consequences, compensation] Collateral: [reviewed / attempted / unavailable + why + effect on confidence] Differential considered: [mood, anxiety, sleep, substance, learning, medical + evidence for/against] Validity considerations: [incentives, inconsistencies, stated neutrally] Status: [screen positive, assessment pending / provisional / criteria met / not established / deferred] Next step: [structured interview, records, informant, follow-up] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale itself. ## Sample ASRS documentation (fictional) Scenario: a positive intake screen documented as a screen, with the developmental, cross-setting, collateral, and differential evidence recorded alongside it and the diagnosis left to the structured assessment. All details are fictional. Patient: D.K., 34 · Setting: Outpatient psychiatry, intake · Clinician: A. Reyes, PMHNP · Note date: 08/12/2026 Screening: Patient completed the English ASRS v1.1 Symptom Checklist today, self-report, referring to the past six months. Part A: 5 of 6 responses reached their item-specific screening thresholds; positive screen, original scoring method. Part B was reviewed as interview probes only; no Part B score exists. Result interpreted as supporting further assessment, not as a diagnosis. History: Longstanding difficulty organizing multistep work, completing administrative tasks, and tracking nonroutine obligations, with impairment at work (missed filing deadlines documented in two performance reviews) and at home (unpaid bills, abandoned projects). Recalls similar problems from elementary school; report cards brought today reference incomplete assignments and needing redirection. Older sister, contacted with consent, independently describes chronic childhood disorganization. Generalized anxiety disorder is active and worsens concentration under deadlines, but the attentional pattern reportedly predates it. Sleep averages seven hours; no current mood episode, psychosis, or substance use disorder identified on screening. Assessment: Positive ASRS v1.1 screen with corroborated childhood onset and cross-setting impairment; anxiety remains a relevant comorbidity and possible contributor. ADHD is not diagnosed today: status is screen positive, structured assessment pending. Plan: Structured DSM-5 adult ADHD interview next visit; remaining school records requested; functional impairment and cardiovascular history to be assessed before any medication discussion. Baseline ASRS retained for treatment-response comparison using the same version and scoring method. This sample is fictional and for educational purposes. It does not describe a real patient or record; the details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The instrument, version, language, reporter, and scoring method are all named, so the result is interpretable years later and comparable at follow-up. The result is phrased as a threshold count without reproducing the scoring grid, and Part B is explicitly treated as probes with no score. The screen stays a screen: the note gives it an explicit status (screen positive, assessment pending) instead of quietly converting it into a diagnosis. Childhood onset and cross-setting impairment each carry named evidence (report cards, performance reviews, a childhood informant), not just the patient's summary. The comorbid anxiety is documented as a possible contributor with the temporal reasoning, which is exactly what the false-positive literature says to do. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the striking fact is what does not exist. There is still no released national adult ADHD clinical practice guideline; APSARD's first-ever US adult guideline was in external review during 2026 with distribution anticipated in 2027 (CONVENTION in progress), leaving DSM-5-TR criteria, the AAFP's multi-visit clinical approach, and payer policy as the operative anchors. Stimulants are Schedule II under the federal schedules (LAW), which governs prescribing and recordkeeping without naming any rating scale. The telemedicine pathway that allows remote stimulant initiation without a prior in-person examination is the fourth temporary extension of COVID-era flexibilities, scheduled to end December 31, 2026, with the special-registration framework still a proposal (LAW, dated). State PDMP checks are their own layer of law and vary: some states require consultation before any Schedule II prescription while others mandate checks only for opioid and benzodiazepine classes, so the PDMP entry is charted separately from the diagnostic reasoning. Payer policy is where scale requirements actually live, and it rarely names the ASRS. Missouri Medicaid's current stimulant criteria require a standardized rating scale only for participants under six, naming Conners and Vanderbilt forms; the 2024 adult prior-authorization form that named an Adult ADHD Self-Report scale has been retired. Most reviewed policies require a documented DSM-5 diagnosis supported by generic standardized rating scales (PAYER POLICY, not law). For accommodations, ETS documentation guidelines require multi-method evaluation and state that no single measure suffices; internet checklists are explicitly insufficient (CONVENTION under the ADA framework). Canada and Australia supply the guidelines the US lacks. CADDRA's 4.1 guideline and eToolkit (CONVENTION) embed the ASRS in a workflow with an observer version when feasible, childhood-informant material where possible, a functional-impairment measure, medical and substance review, and the explicit statement that the toolkit alone cannot make the diagnosis; no Canadian federal rule requires an ASRS. Australia's 2022 AADPA guideline (CONVENTION, NHMRC approval window running to July 2027) requires comprehensive assessment, bars diagnosis from rating scales or observation alone, and expects evidence across settings and reporters. The binding Australian layer is state and territory Schedule 8 stimulant-authority law, which was reformed jurisdiction by jurisdiction through 2025 and 2026: New South Wales opened trained-GP continuation prescribing in September 2025 with GP-led diagnosis expanding through 2026, South Australia opened a trained-GP pathway in February 2026, Queensland and Tasmania broadened GP roles, and the ACT created standing approvals in February 2026 whose GP instrument is a continuation, not initiation, authority (LAW, dated per jurisdiction). None of those laws requires ASRS completion; they regulate prescriber class, authorities, monitoring systems, and co-management. On rights and fidelity: Harvard's distribution page states that use of the six-question v1.1 screener is "free and does not require any formal permission or approval," asking for the 2005 citation, the New York University and Harvard copyright notice, and an unaltered instrument; the 18-question Symptom Checklist and the ASRS-5 each require permission routed through the NYU team, and third-party online versions prove nothing about authorization. The steward lists 20 translations while its background memo notes that no cultural validations of those translations have been conducted, so the chart should name the language used. BastionGPT is not affiliated with the instrument authors, WHO, Harvard, or NYU. This page describes the instrument in original prose and reproduces no item text, response grid, or scoring materials. ↑ Back to the template and downloads ## Common ASRS documentation errors reviewers flag The numbers behind these errors are setting-dependent in a way most pages hide. The famous 99.5% specificity comes from the 2005 community validation; in adults with major depression the v1.1 positive predictive value was 21.4% , in treatment-seeking substance-use patients it was 0.26 , and in two non-treatment-seeking population samples 17% to 26% screened positive , several times the plausible adult prevalence. Simulation studies add that undergraduates can produce convincing ADHD self-report profiles with or without coaching, and the ASRS has no embedded validity scale. The BastionGPT Clinical Advisory Board sees the same errors most often in ASRS documentation reviews: Version and method omitted. "ASRS 14" could be a v1.1 continuous total at its threshold, an ASRS-5 weighted total at its threshold, or an arithmetic mistake; the v1.1 count, the 0 to 24 sum, and the 0 to 25 ASRS-5 are three different systems that happen to share a boundary number. Name the instrument, the scoring method, and the language every time. Screen counts rebranded as severity. Four or more qualifying responses is a case-finding result, and the 0 to 24 strata are probability regions from a health-plan calibration; neither was validated as mild, moderate, or severe ADHD, and no minimal clinically important difference exists for tracking. "ASRS 18 = severe ADHD" adds meaning the instrument does not carry. Part B given a score. Part B supplies interview probes; per the steward's own instructions, no Part B total or diagnostic likelihood is used, it has no cutoff, and it cannot turn a negative Part A positive. Diagnosis charted from the screen. A positive ASRS establishes none of the DSM requirements: onset before age 12, symptoms in two or more settings, functional impairment, or the exclusion of better explanations. Notes that jump from screen to diagnosis omit exactly the elements auditors and accommodation reviewers look for. The community specificity universalized. Quoting 99.5% to a patient or a chart while screening a depressed, anxious, sleep-deprived, or substance-using population imports a number from the wrong setting; in those clinics most positive screens are not ADHD. Validity concerns charted as character. "Drug-seeking" and "malingering" are conclusions, not observations. Document the external incentive, the specific inconsistencies, what is missing, and the effect on confidence; a request for stimulants or accommodations is not itself evidence of deception. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the assessment facts (instrument and version, scoring method, result, onset evidence, settings, collateral, differential, incentives) and it drafts the full screening entry: precise result phrasing, modest interpretation, evidence-by-evidence corroboration, and an explicit status with next steps, ready for your review. Cross-check a finished note for the gaps reviewers flag: a bare "ASRS positive," a missing scoring method, strata charted as severity, a diagnosis resting on the screen alone, or validity language that reads as character judgment. Draft the follow-up comparison: baseline and current results under the same version and method, functional change, medication status at completion, and what the trend does and does not show. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does a positive ASRS actually mean? That further assessment is warranted, and nothing more. Under the original v1.1 rule, a screen is positive when four or more of the six Part A responses reach their item-specific thresholds (the threshold differs by item, so it is not simply four answers of Often or Very Often). The screen cannot establish childhood onset, cross-setting impairment, functional interference, or the differential, all of which the diagnosis requires. A negative screen is not an exclusion either: the six-item sensitivity was 68.7% in the original validation, so roughly three in ten true cases screen negative, and a strong developmental and collateral history justifies continuing the assessment. What is the difference between ASRS v1.1 and ASRS-5? They are different instruments with different arithmetic, and a third scoring system sits between them. The v1.1 original rule counts qualifying Part A responses, 0 to 6, with four or more positive. The v1.1 continuous method sums the six frequency responses to 0 to 24, with 14 or higher positive and four probability strata; a Harvard scoring update dated February 28, 2024 endorsed it as an alternative. The ASRS-5, published in 2017 for DSM-5, uses a different item set with machine-learned unequal weights, a corrected range of 0 to 25 (one response can contribute six points, per the 2019 correction), and a threshold of 14 or higher, and it requires its own permission. The totals are never comparable across systems, which is why the chart names the version and method every time. Is the ASRS free to use, and can we put it in our EHR? The six-question v1.1 screener, yes: Harvard's distribution page states its use is "free and does not require any formal permission or approval," with three conditions worth honoring in an EHR build: cite the 2005 article, retain the New York University and Harvard copyright notice, and do not alter the response options or scoring. That is free use, not public domain. The 18-question Symptom Checklist and the ASRS-5 each require a permission request routed through the NYU team, and a third-party version circulating online proves nothing about authorization. The steward prefers links to its current copies so scoring updates are not stranded in stale reproductions, and its own memo notes the 20 listed translations have had no cultural validations, so name the language used. Can the ASRS measure ADHD severity or treatment response? Not as a severity scale. The screen count is case-finding arithmetic, and the 0 to 24 strata (low negative, high negative, low positive, high positive) are probability regions from a health-plan calibration whose meaning shifts with population prevalence; none of it was validated as mild, moderate, or severe ADHD. For treatment tracking, re-administration is a reasonable convention (Canadian guidance recommends documenting change with scales meaningful to the patient) if the version, scoring method, language, reporter, and recall period are held constant and the score is paired with functional outcomes. No broadly accepted minimal clinically important difference exists; a 2026 analysis found none of 11 randomized trials had reported one. Direction of change is documentable; a point difference is not itself proof of response. Why do so many positive screens turn out not to be ADHD? Because positive predictive value depends on the population, and the ASRS asks about experiences many conditions produce. The 99.5% specificity came from the 2005 community validation design. In adults with major depression, the positive predictive value was 21.4%; in treatment-seeking substance-use patients it was 0.26; in a university sample 37% screened positive while 7.9% received a diagnosis; and in two general-population samples 17% to 26% screened positive, several times any plausible prevalence. Depression, anxiety, sleep restriction, and substance effects all generate attentional complaints the form cannot distinguish. The documentation consequence: a positive screen in a comorbid patient is the beginning of the differential, not the end of it. Is a positive ASRS required before prescribing stimulants? No law in the US, Canada, or Australia requires one. Stimulant law regulates other things: federal Schedule II controls, state PDMP checks, telemedicine rules (the current US flexibility runs through December 31, 2026), and, in Australia, state Schedule 8 authority schemes that specify prescriber class and permits rather than any rating scale. Where a scale requirement exists, it is payer or organizational policy, and it rarely names the ASRS. Missouri Medicaid's current stimulant criteria require a rating scale only for participants under six and name Conners and Vanderbilt forms; most policies ask generically for standardized rating scales supporting a DSM-5 diagnosis. Chart the diagnostic reasoning for the clinical record and the specific payer's requirement when one applies, and never present a payer rule as a legal mandate. What do I document when childhood records are unavailable? The attempts, the reason, the substitutes, and the effect on confidence. Parents may be deceased or estranged and school records destroyed; that is not a diagnostic exclusion. Document what was sought and why it was unavailable; use substitute longitudinal evidence such as siblings, partners, old evaluations, employment history, and repeated concrete historical examples; compare the account with the current functional pattern; and state plainly how the missing corroboration qualifies the conclusion. What fails review is silence: a note that neither sought collateral nor acknowledged its absence, or one that invents reassurance the record does not contain. Can patients fake the ASRS, and how do I chart that risk? The symptom profile is easy to produce: simulation studies found undergraduates could feign ADHD on self-report scales whether or not they were coached, and the ASRS contains no embedded validity scale to detect it. That justifies multimethod assessment, not suspicion of every patient. Chart behavior and evidence rather than character: the external incentive when one exists (a licensing exam, an accommodation deadline), the specific inconsistencies (an abrupt reported onset against clean contemporaneous records), what data are missing, and how those facts limit confidence. Where formal symptom- or performance-validity testing is used, name the measure and its result, and remember that a noncredible result limits interpretation without proving intent. Can BastionGPT draft my ASRS documentation? Yes. Give it the facts (instrument and version, scoring method, result, childhood-onset evidence, settings and impairment, collateral status, differential, any incentives) and it drafts the full screening entry: precise result phrasing without the scoring grid, modest interpretation, corroboration point by point, neutral validity language, and an explicit status with next steps, ready for your review. It can also check a finished note for a bare "ASRS positive," a missing method, strata charted as severity, or a diagnosis resting on the screen alone. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Kessler RC and colleagues, 2005, Psychological Medicine, the ASRS publication (six-item sensitivity 68.7%, specificity 99.5%, kappa 0.76); Kessler and colleagues, 2007, International Journal of Methods in Psychiatric Research (the continuous 0 to 24 calibration and its strata). Harvard Medical School National Comorbidity Survey, the ASRS distribution page (permission language, copyright, fidelity conditions, translations) and the February 28, 2024 scoring update . Üstün B and colleagues, 2017, JAMA Psychiatry, the ASRS-5 (weighted algorithm; general-population sensitivity 91.4%, specificity 96.0%; clinical-sample specificity 74.0%) and the 2019 correction (range 0 to 25). Dunlop BW and colleagues, 2018, Behavioral Sciences, ASRS performance in major depression (sensitivity 60%, specificity 68.6%, positive predictive value 21.4%). van de Glind G and colleagues, 2013, Drug and Alcohol Dependence, the IASP substance-use validation (positive predictive value 0.26, negative predictive value 0.97, n=1,138). Chamberlain SR, Cortese S, Grant JE, 2021, Comprehensive Psychiatry, population screening analysis (17.3% and 26.0% positive-screen proportions; over-identification estimates). Edmundson M and colleagues, 2017, Psychological Assessment (feigning with and without coaching); Quinn CA, 2003, Archives of Clinical Neuropsychology (self-report faked while performance measures resisted). CADDRA, ASRS instructions (Part B as probes with no total) and the Canadian ADHD Practice Guidelines 4.1 . AADPA, 2022, Australian evidence-based ADHD guideline (scales not the sole basis of diagnosis; NHMRC approval window). APSARD, US adult guideline project status ; DEA and HHS, fourth temporary telemedicine extension through December 31, 2026; 21 CFR 1308.12 (Schedule II). Missouri Medicaid, ADHD stimulant PDL criteria (January 2026; rating scales named only for participants under six); ETS, ADHD documentation guidelines (multimethod evaluation; no single measure sufficient). NSW Health, psychostimulant prescribing authority (the state Schedule 8 scheme and 2025 to 2026 GP reforms). ## Related templates ADHD Evaluation Report CAARS-2 Report Write-Up Conners 4 Report Write-Up Medication Management Follow-Up Note Vanderbilt ADHD Scale Documentation T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/bdi-ii All note templates # BDI-II Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The BDI-II (Beck Depression Inventory, second edition) is a 21-item, licensed depression severity measure published by Beck, Steer, and Brown in 1996 and scored 0 to 63. Behavioral health, medical, and evaluation practices use it for baselines, outcome monitoring, and assessment batteries. This page covers how to document BDI-II results defensibly, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report on licensed forms or Q-global; scored and interpreted by a qualified (Level B) clinician Audience Psychologists and therapists, evaluation and forensic practices, bariatric and transplant programs, research teams, auditors Typical length A result line plus interpretation in the note · patient completion about 5 minutes Format family Licensed self-report severity measure (21 items scored 0 to 3, total 0 to 63) When it's used Depression severity baselines, psychotherapy outcome monitoring, evaluation batteries, bariatric and transplant psychosocial assessments Standards context Licensed Pearson instrument (Level B): every administration needs a purchased form or usage; no authority mandates it by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the BDI-II? The Beck Depression Inventory, second edition (BDI-II) is a 21-item self-report measure of depression severity published by Aaron Beck, Robert Steer, and Gregory Brown in 1996 and distributed by Pearson. Each symptom category contributes 0 to 3 points, for a total of 0 to 63, over a two-week window aligned with the DSM-IV criteria of its era (it was not built against DSM-5-TR). The listed age range for the English edition is 13 to 80, completion takes about five minutes, and purchase requires Pearson's Level B qualification. The manual's severity descriptors are totals through 13 minimal, 14 to 19 mild, 20 to 28 moderate, and 29 to 63 severe. Those are severity bands, not diagnoses and not universal case-finding cutoffs: a 2019 diagnostic meta-analysis found study-optimal thresholds ranging from 10 to 25 depending on the population, which is why the defensible chart phrase is "within the manual's moderate severity range," never "moderate major depressive disorder." Two facts frame everything else. First, licensing: the instrument is proprietary and test-secure, so every administration consumes a purchased paper form or a Q-global usage, the "unlimited scoring" subscription covers manually entered data only, and photocopying the form or rebuilding it in a survey platform breaches the license. The rights record lists the instrument copyright to Aaron T. Beck (1996, 1987), with Pearson managing distribution and permissions. Second, age: there is no BDI-III (a dated finding as of August 2026; Pearson's own catalog calls the BDI-II the latest edition), so the foundational norms are three decades old, a limitation regulators and high-stakes reviewers increasingly expect reports to address. The free, nine-item PHQ-9 is the standard comparison: totals correlate near .77, but the severity bands do not map one to one, and the BDI-II classifies more patients as severe, so the two are never spliced into one trajectory. ## Who uses BDI-II documentation and when Psychotherapy practices run serial BDI-II totals for outcome monitoring where continuity with the instrument's large treatment literature matters; evaluation practices fold it into batteries for disability, forensic, bariatric, and transplant questions, where the stakes are highest and the instrument never stands alone. Bariatric programs are the emblematic case: the BDI-II is among the most commonly used depression measures in presurgical psychosocial evaluations, yet no national standard mandates it, and transplant policy likewise requires a psychosocial evaluation without naming any instrument. US disability work inverts the usual assumption: Social Security does not require, endorse, or prefer any depression instrument, and since 2017 its mental-disorders rules rate adult functional limitation from functional evidence rather than test scores (intellectual disorder excepted), so a bare total carries weight only through the narrative around it. Medical settings carry the somatic caveat: illness-driven fatigue, sleep, and appetite changes inflate totals (one multiple sclerosis cohort flagged 54.1% for depression on the BDI-II against 30.7% on its somatic-free sibling), which is why the seven-item BDI-FastScreen exists and why perinatal care prefers the EPDS , built without somatic items. Serial-measurement conventions live in the outcome measure note , battery integration in the psychological evaluation report , and session-level care in the psychotherapy progress note . ## How to document BDI-II results in the chart No law prescribes a BDI-II note format, and no US, Canadian, or Australian authority requires the instrument by name. What survives review is a record that treats the total as one measurement with a named method around it: edition, mode, completion, the suicide item, trajectory, integration, and the licensing trail. Each element below carries the pitfall that most often undermines it. Instrument identity, edition, and mode. Record BDI-II with the language or version, the date, and the administration mode (paper, supervised on-screen, or remote on-screen), with telehealth conditions noted per the publisher's guidance: privacy, interruptions, assistance, technical problems, and whether conditions supported valid interpretation. Pitfall: edition blindness. BDI and BDI-IA scores and cutoffs do not transfer (the 1996 revision changed items and runs a few points higher), so chart the edition at every timepoint and treat any edition change as an instrument transition, not a continuous series. The result, with the manual's descriptor. Chart the total and the band phrase with attribution: "41-year-old completed the English BDI-II; total 24, within the manual's moderate severity range (Beck, Steer, and Brown, 1996)." Pitfall: converting a band into a diagnosis. The band describes reported symptom burden over two weeks; the diagnosis belongs to the clinical interview and differential, and "severe range" is not "severe major depressive disorder." The suicide item, handled separately. Any endorsement of the suicide-related item prompts a timely, separately documented risk assessment (a suicide risk assessment with a safety plan where indicated), regardless of the total. Pitfall: arithmetic in either direction. A minimal or mild total cannot neutralize an endorsed suicide item, and a high total does not establish imminent risk without direct assessment; "item 9 positive, safety plan completed" is usually too thin for a high-risk encounter. Completion and conditions. State whether the administration was complete; if not, record the extent and reason without reproducing item content, and arrange completion or an alternative. Pitfall: prorating. No public publisher rule authorizes averaging or substituting missing responses, Q-global will not report with required responses unresolved, and a partial sum read against the 0 to 63 bands is not a valid total. Trajectory, with method honesty. Record the prior total and date, the interval, and the absolute and percentage change. If you claim reliable or clinically significant change, name the method and inputs; published conventions genuinely differ (a reliable-change threshold near 5 points under one method, 8.46 under another, 10 in other protocols, and a patient-perceived meaningful difference estimated at 17.5% of baseline). Pitfall: "reliable improvement" from a bare point drop, or a study-specific recovery endpoint (13 or lower in one trial convention) charted as a manual rule. Without a named method, the honest entry is the raw and percentage change. Integration and limits. Connect the score with the interview, observed functioning, collateral, and medical status, and state the limits that apply: self-report with no validity scale, somatic inflation in medical illness, language and norm limitations, response-style considerations. Pitfall: silent resolution. A score-interview discrepancy is described and investigated, not resolved automatically in favor of either source, and an unusually high total is a hypothesis about reporting style, never proof of exaggeration. The licensing trail. Keep an organizational record that administrations used purchased forms or Q-global usages under a qualified user, store protocols securely, and keep item content out of the EHR. Pitfall: photocopied forms, survey-platform rebuilds, or test items transcribed into notes; purchase of a kit is not a reproduction permission. ### Blank template (copy and adapt) BDI-II DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Instrument: BDI-II Language/version: [ ] Edition held constant across series: [y/n] Mode: [paper / on-screen / remote on-screen + conditions] Completion: [complete / incomplete: extent + reason + plan] Total: [ ]/63 Manual severity range: [minimal / mild / moderate / severe] (Beck, Steer, and Brown, 1996) Suicide item: [not endorsed / endorsed -> separate risk assessment documented at: ] Trajectory: [prior total + date -> current; absolute + % change; method named, or raw change only] Integration: [interview, function, collateral: consistent / discordant + how investigated] Limits: [somatic confounds, language/norms, response style, nonstandard conditions] Action: [risk, formulation, treatment intensity, referral, follow-up interval] Licensing: [purchased form / Q-global usage; qualified user] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the instrument itself. ## Sample BDI-II documentation (fictional) Scenario: serial outcome monitoring through psychotherapy, including one administration where the suicide item was endorsed and assessed separately, and a change statement that names no threshold it cannot support. All details are fictional. Patient: J.T., 41 · Setting: Outpatient psychotherapy, week 12 review · Clinician: M. Calloway, PhD · Note date: 08/12/2026 Administration: English BDI-II completed today in office on a licensed Q-global on-screen administration; complete, no omitted responses; edition and language constant across the series. Total 16, within the manual's mild severity range (Beck, Steer, and Brown, 1996). Suicide item not endorsed today. Series: Intake (05/20/2026) total 32, manual severe range; the suicide item was endorsed at a low level at intake and was assessed separately the same day: passive thoughts without intent, plan, or preparation, documented in the intake risk assessment with a collaborative safety plan and 988 contact information provided. Week 6 (07/01/2026) total 24, moderate range, suicide item not endorsed. Today 16, mild range. Change statement: The total decreased 16 points from intake, a 50% reduction. The direction and pace are consistent with observed functioning: regular work attendance resumed, morning routine re-established, two social commitments kept weekly. No universal reliable-change threshold is assumed; under the trial convention that requires a 10-point improvement with an endpoint of 13 or lower, today's result would meet the change requirement but not that study's recovery endpoint, reported here for transparency, not as a manual rule. Integration and plan: Score consistent with interview and function; residual symptoms remain clinically significant (sleep, self-critical rumination). Continue weekly CBT with relapse-prevention focus; next BDI-II in four weeks under the same mode and edition; prescriber review shares today's total and series. This sample is fictional and for educational purposes. It does not describe a real patient or record; the totals, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The edition, language, mode, completion status, and licensing basis are all named, and the series explicitly holds them constant, which is what makes the trajectory interpretable. Every total carries the manual's band phrase with attribution, and no band is converted into a diagnosis anywhere in the note. The endorsed suicide item at intake was assessed separately the same day and cross-referenced, and the low total at that visit did not determine the disposition. The change statement gives raw and percentage figures, names the one study convention it cites, and refuses to invent a universal threshold. Integration ties the numbers to observable function, and the plan fixes the next administration's interval and mode, which is what reviewers look for in measurement-based care. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: Social Security does not require, endorse, or prefer any depression instrument, and its testing policy instructs adjudicators not to rely on test results alone; since the 2017 mental-disorders rewrite, adult functional limitation is rated from functional evidence rather than test scores, intellectual disorder excepted (PAYER POLICY). The same policy expects standardized administration by a qualified specialist, recent and appropriate norms, and a narrative connecting scores to function, which is exactly where a 1996-normed instrument needs explicit justification. In forensic work, Federal Rule of Evidence 702 governs the methodology (LAW), and the convention is to pair the BDI-II with symptom-validity measures: it has no embedded validity scale, and studies of elevated totals as over-reporting indicators found sensitivity too limited for stand-alone use, so a high score is a hypothesis, never proof of exaggeration (CONVENTION). Bariatric and transplant programs use it widely, but no national standard mandates it: ASMBS materials list it among recommended instruments without crowning a gold standard, and OPTN living-donor policy requires a psychosocial evaluation without naming any measure (CONVENTION and PAYER POLICY by program). On billing, the relevant distinction is 96127 versus the 96130 series: a current Medicare contractor policy uses a brief Beck depression questionnaire as its example of testing that should not be billed as psychological testing when it can be performed within the clinical interview, and license cost is not a coding criterion (PAYER POLICY). Canada and Australia add professional-regulation and program layers. Ontario's College of Psychologists and Behaviour Analysts requires registrants to protect test security and copyright, to avoid outdated or obsolete tests or document a reasoned departure, and to record assessment activity begun but not completed with the reason, direct authority for charting an incomplete BDI-II rather than prorating one (LAW, professional regulation); the Canadian Psychological Association's 2019 position paper found no federal or provincial oversight of the psychological test market, which leaves publisher terms and college standards as the operative controls. The official Canadian French edition differs from the English one (published 1998, listed for ages 18 and older), so a bilingual practice should not silently apply the English 13-to-80 range. In Australia, Better Access guidance expects an outcome measure in a mental health treatment plan unless clinically inappropriate but names the K10 and DASS-21 as examples, not the BDI-II, so the licensed instrument is a discretionary, clinician-funded choice with no MBS item of its own (PAYER POLICY), and registered psychologists remain bound by Board competence and record standards (LAW). If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The psychometrics support disciplined use. Internal consistency runs near .90 to .92 with retest reliability of .73 to .96 across 118 studies; a meta-analysis of 62 samples (20,475 participants) supports a two-factor cognitive and somatic-affective structure while the total remains the defensible interpretive unit; and the somatic factor is exactly where medical illness inflates totals, the reason the seven-item BDI-FastScreen exists and the EPDS omits somatic items for perinatal use. The norm question deserves a balanced sentence in high-stakes reports: the BDI-II remains a current Pearson edition with extensive contemporary validation, but its foundational manual and normative frame date to 1996 (a development sample of roughly 500 outpatients, 91% white, mean age 37), and population studies since, including the VA cohort, show meaningful drift, so normative claims should be justified for the population at hand. On rights: the Beck Depression Inventory, second edition, is a proprietary, Level B Pearson assessment; the rights record lists approximately 90 catalogued translations (availability is not local validation) and displays the instrument copyright as 1996 and 1987 by Aaron T. Beck. BastionGPT is not affiliated with Pearson, Beck Institute, or the instrument authors. This page reproduces no items, response options, forms, scoring keys, or proprietary report content. ↑ Back to the template and downloads ## Common BDI-II documentation errors reviewers flag The numbers behind these errors deserve more respect than they get. Study-optimal case-finding cutoffs ranged from 10 to 25 in the 2019 diagnostic meta-analysis ; a VA study of 152,260 veterans found scores far above the 1996 normative sample (non-depressed subgroup effect size 1.34) and suggested a local cutoff near 27; one multiple sclerosis cohort flagged 54.1% for depression on the BDI-II against 30.7% on the somatic-free short form; and a 2026 review found only 13% of eligible psychotherapy trials used any reliable-change criterion, spread across at least seven calculation methods. The BastionGPT Clinical Advisory Board sees the same errors most often in BDI-II documentation reviews: Bands read as diagnoses or universal cutoffs. "Severe depressive disorder" charted from a total of 31, or 14 treated as a diagnostic threshold. The manual bands describe reported symptom burden; case-finding cutoffs are population-specific; and the diagnosis belongs to the interview and differential. The suicide item folded into the arithmetic. A mild total used to skip follow-up of an endorsed suicide item, or a severe total charted as imminent risk with no direct assessment. The item is reviewed independently of the total, in both directions, with the risk assessment documented separately. Prorated partials. A total computed from fewer than 21 completed categories and read against the 0 to 63 bands. No public publisher rule authorizes proration; chart the administration as incomplete with the extent and reason, and arrange completion or an alternative. Change verdicts without a method. "Reliable improvement" from a nine-point drop, or a study-specific recovery endpoint presented as a manual rule. Published thresholds differ by method and population; name the method and inputs, or report raw and percentage change only. Instruments and editions spliced. PHQ-9 totals converted to BDI-II totals by formula, BDI-IA history read against BDI-II bands, or an edition switch buried mid-series. Correlation is not equivalence; mark every transition and re-baseline. Licensing shortcuts. Photocopied record forms, the instrument rebuilt in a survey platform, protected item text transcribed into the EHR, or remote administration outside the authorized modes. Purchase is not a reproduction permission, and Canadian regulators make test security a professional obligation, not a courtesy. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (edition, language, mode, completion, total, suicide-item status, prior scores and dates, function, limits) and it drafts the full entry: band phrasing with attribution, separate risk-assessment cross-references, an honest change statement, and integration with the interview, ready for your review. Cross-check a finished note for the gaps reviewers flag: a band charted as a diagnosis, a suicide item resolved by arithmetic, a prorated partial, a change verdict with no method, or an edition switch buried mid-series. Draft the outcome summary for a referral or review: the series with modes and intervals, raw and percentage change, functional corroboration, and stated limits, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do BDI-II scores actually mean? Twenty-one symptom categories scored 0 to 3 sum to a total of 0 to 63 over a two-week window. The manual's severity descriptors are totals through 13 minimal, 14 to 19 mild, 20 to 28 moderate, and 29 to 63 severe (Beck, Steer, and Brown, 1996). Those bands describe self-reported symptom burden; they are not diagnoses, treatment mandates, or universal screening cutoffs, and the 2019 diagnostic meta-analysis found optimal case-finding thresholds anywhere from 10 to 25 depending on the population. The defensible chart phrase is "within the manual's moderate severity range," with the diagnosis established separately by clinical interview and differential assessment. Is a specific BDI-II score required for disability, surgery, or treatment? No. Social Security requires no psychological test for depression claims, prefers no instrument, and rates adult functional limitation from functional evidence rather than test scores under its 2017 rules; a BDI-II contributes only through a well-supported narrative tied to function. Bariatric and transplant programs commonly include it in psychosocial evaluations, but no national standard mandates it by name, and payer or program requirements are checked individually rather than assumed. No score band specifies psychotherapy, medication, or level of care; treatment decisions integrate diagnosis, impairment, risk, history, and preference. Where you see a hard number requirement, it is a local program or payer policy, and it should be cited as such in the note. What do I do when the suicide item is endorsed? Assess it directly and document that assessment separately, whatever the total. Any endorsement prompts timely clarification of currency, frequency, intent, plan, preparation, access to means, history, dynamic risks, and protective factors, with consultation, safety planning, and disposition as indicated. The arithmetic works in neither direction: a minimal total cannot neutralize an endorsed item, and a severe total does not establish imminent danger without direct assessment. "Item 9 positive, safety plan completed" is usually too thin for a high-risk encounter; the risk assessment gets its own documentation with reasoning, and the BDI-II note cross-references it. How much BDI-II change counts as real improvement? There is no universal number, and pretending otherwise is the most common charting error. Published conventions include a reliable-change threshold near 5 points under one Jacobson-Truax calculation, 8.46 under another with different inputs, 10 points in other protocols, a trial convention requiring a 10-point improvement plus an endpoint of 13 or lower, and a patient-perceived meaningful difference estimated at 17.5% of baseline (higher, around 32%, in longer-duration treatment-resistant depression). A 2026 review found only 13% of eligible psychotherapy trials used any reliable-change criterion, across at least seven methods. Name the method and inputs when you claim significance; otherwise chart the raw and percentage change and let the functional evidence carry the interpretation. Can I photocopy the BDI-II or build it into our EHR? No. The instrument is proprietary and test-secure: every administration uses a purchased record form or a Q-global usage, the annual "unlimited scoring" subscription covers manually entered data only (on-screen and remote administrations are per-use), and reproduction, translation, platform embedding, and research use are separate permission requests to the publisher. Keep protected item content out of the EHR; chart results and interpretation instead. Canadian colleges make test security and copyright a professional obligation, and Ontario's standards additionally require documenting assessment activity begun but not completed, which is the correct handling for a partial administration. Remote use goes through the authorized modes with the telehealth conditions documented, not through a scanned form in email. Should we use the BDI-II or the PHQ-9? They answer to different economics, not different truths. Totals correlate near .77 overall, responsiveness to change is similar, and neither is more valid because of its price; but severity bands do not map one to one, and the BDI-II classifies more patients as severe. The PHQ-9 is free, shorter, and the default for routine screening and high-volume measurement-based care. The BDI-II earns its licensing cost when continuity with an established BDI-II series or the historical treatment literature matters, when its broader cognitive-affective coverage is clinically useful, or when a protocol or evaluation methodology specifies it. Never convert one total to the other by formula, and never splice the two into one longitudinal graph; migrate with an overlap period and a new baseline. How do I handle medical patients whose symptoms inflate the score? Document the confound; never rewrite the score. Fatigue, sleep, appetite, and concentration items can be driven by illness, treatment, or depression, and the inflation is measurable: one multiple sclerosis cohort flagged 54.1% for depression on the BDI-II against 30.7% on the somatic-free seven-item BDI-FastScreen, and non-depressed pregnant patients score higher specifically on somatic items. The defensible pattern is to chart the relevant medical conditions and treatments, examine the cognitive-affective and somatic patterns qualitatively, consider a measure designed for the setting (the FastScreen in medical populations, the EPDS in perinatal care), and avoid claiming either that every somatic point is depression or that somatic symptoms never are. Are the 1996 norms still defensible in 2026? Defensible, if you address them. The severity bands are criterion-referenced and still apply, and the instrument carries an extensive contemporary validation literature; but the foundational manual and normative frame date to 1996, built on a development sample of roughly 500 outpatients that was 91% white with a mean age of 37, and there is no BDI-III (a dated finding as of August 2026). Population drift is documented: a VA study of 152,260 veterans found scores well above the 1996 sample and suggested a local cutoff near 27. High-stakes reports should say why the normative frame remains appropriate for this examinee and question, and disability policy expects recent, appropriate norms, which makes that sentence load-bearing rather than boilerplate. Is the BDI-II the same test as the BDI-3? No, and the acronym collision confuses search results, reading lists, and occasionally charts. The BDI-II on this page is the Beck Depression Inventory, Second Edition (Pearson): a 21-item self-report questionnaire measuring depression severity in adolescents and adults ages 13 to 80. The BDI-3 is the Battelle Developmental Inventory, Third Edition (Riverside Insights): a comprehensive early-childhood developmental battery, administered by evaluators across five domains, that anchors many Part C early-intervention eligibility decisions. Different publishers, constructs, respondents, ages, and uses; the two share nothing but the letters. If you were looking for the developmental battery, its write-up guide lives on the Battelle (BDI-3) report page ; this page covers the depression questionnaire, where the successor question people usually mean is "is there a BDI-III of the Beck?", and the answer remains no. Can BastionGPT draft my BDI-II documentation? Yes. Give it the facts (edition and language, mode, completion, total, suicide-item status, prior scores with dates, function, medical confounds, licensing basis) and it drafts the full entry: the band phrase with attribution, the separate risk-assessment cross-reference, an honest change statement with raw and percentage figures, integration with the interview, and stated limits, ready for your review. It can also check a finished note for bands charted as diagnoses, arithmetic around the suicide item, prorated partials, method-free change verdicts, and spliced instruments or editions. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Beck AT, Steer RA, Brown GK, 1996, Manual for the Beck Depression Inventory-II, The Psychological Corporation; the current Pearson product listing (architecture, bands, age range, Level B, pricing model). ePROVIDE, the BDI-II rights record (instrument copyright 1996 and 1987 by Aaron T. Beck; approximately 90 catalogued translations); Pearson, permissions and licensing policies (purchase is not a reproduction permission). von Glischinski M and colleagues, 2019, Quality of Life Research, diagnostic meta-analysis (study-optimal cutoffs 10 to 25; pooled sensitivity .86, specificity .78 near 14.5). Huang C and Chen JH, 2015, Assessment (two-factor meta-analysis, 62 samples, 20,475 participants); Wang YP and Gorenstein C, 2013 (118-study reliability review: alpha near .90, retest .73 to .96). Kung S and colleagues, 2013, BDI-II versus PHQ-9 comparison (r = .77 overall; discordant severity assignment); Titov N and colleagues, 2011, responsiveness comparison (similar change sensitivity, fair-to-moderate category agreement). Button KS and colleagues, 2015, Psychological Medicine, patient-perspective MCID (17.5% of baseline; about 32% in treatment-resistant depression); Seggar LB, Lambert MJ, Hansen NB, 2002 (clinically significant change 8.46 under their inputs); Crenshaw and colleagues, 2026, reliable-change usage review (13% of 226 trials; at least seven operationalizations). 2019, Psychological Services, VA renorming study (152,260 veterans; non-depressed effect size 1.34; suggested local cutoff near 27). Social Security Administration, POMS DI 24583.050 (no required or preferred instrument; norm recency and integration expectations); Federal Rules of Evidence, Rule 702 . CGS Administrators, LCD L34353 (a brief Beck depression questionnaire performable within the interview is not separately billable psychological testing). College of Psychologists and Behaviour Analysts of Ontario, Standards of Professional Conduct (test security, outdated-test avoidance, incomplete-service documentation); Canadian Psychological Association, 2019 position paper on test safety . Australian Department of Health, Better Access treatment-plan information (outcome measure expected; K10 and DASS-21 named); Services Australia, MBS mental health billing rules . ASMBS, presurgical psychosocial evaluation recommendations ; OPTN, living-donor psychosocial evaluation policy (evaluation required; no instrument named). ## Related templates PHQ-9 Score Documentation EPDS Score Documentation Outcome Measure Review Note Psychological Evaluation Report Battelle Developmental Inventory (BDI-3) Report Write-Up Geriatric Depression Scale (GDS-15) Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/epds All note templates # EPDS Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The EPDS (Edinburgh Postnatal Depression Scale) is a 10-item perinatal depression screen published by Cox, Holden, and Sagovsky in 1987 and scored 0 to 30 over the past seven days. Obstetric, midwifery, pediatric, and primary care teams use it through pregnancy and the postpartum year. It screens; it never diagnoses. This page covers how to document EPDS results defensibly, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report (may be administered verbally); scored and actioned by the clinician; partners screen as separate respondents Audience OB and midwifery practices, pediatric well-child teams, primary care, perinatal mental health programs, quality reviewers Typical length A result line plus interpretation and follow-up in the note · patient completion under 5 minutes Format family Perinatal self-report screen (10 items scored 0 to 3, total 0 to 30, seven-day window) When it's used Antenatal and postpartum depression screening, well-child-visit caregiver screening, serial rescreening, partner screening Standards context RCPsych copyright, not public domain: own-use photocopying only; distribution or republication needs written permission On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the EPDS? The Edinburgh Postnatal Depression Scale (its exact name; it is not the "Edinburgh Postpartum Depression Scale") is a 10-item self-report screen for perinatal depressive symptoms, published by Cox, Holden, and Sagovsky in the British Journal of Psychiatry in 1987. Each item is scored 0 to 3 for a total of 0 to 30, over the preceding seven days, a shorter window than the two weeks many pages misremember. It was built for pregnancy and the postpartum period and deliberately omits somatic items, because the developers found it too hard for a self-completed scale to separate the physical changes of childbearing from mood disorder. The result is a screen, never a diagnosis: a positive total warrants clinical assessment, and the scale does not detect bipolar disorder or postpartum psychosis at all. Two facts frame the documentation. First, thresholds are tradeoffs, not truths. In the largest individual-participant meta-analysis (58 studies, 15,557 participants), a threshold of 10 or higher ran sensitivity 0.85 with specificity 0.84, 11 or higher gave the best combined balance at 0.81 and 0.88, and 13 or higher reached specificity 0.95 while missing roughly one-third of major-depression cases at sensitivity 0.66, with accuracy broadly similar in pregnancy and postpartum. The chart therefore names the threshold applied and its purpose. Second, copyright: the scale belongs to the Royal College of Psychiatrists and is not public domain. The College's terms allow an individual clinician or researcher to photocopy a complete, unmodified, attributed copy for their own use, and require written permission for distribution to others or republication in any medium, which is why this page describes the instrument without reproducing it and links out rather than embedding a form. The free PHQ-9 is the usual general-purpose comparison; the licensed BDI-II is the instrument whose somatic items make it a poor perinatal fit. ## Who uses EPDS documentation and when Obstetric and midwifery practices screen through pregnancy and postpartum: US convention (ACOG's 2023 guideline) calls for a validated instrument at the initial prenatal visit, later in pregnancy, and at postpartum visits, inside a system that can assess, treat, and follow up. Pediatric teams are the second major user, and the setting with the most documentation traps: AAP recommends caregiver screening at the 1-, 2-, 4-, and 6-month well-child visits, records it in the infant's chart because the screen is performed for the infant's benefit, and advises against opening a maternal chart solely for it. Australia is the jurisdiction that names the EPDS itself: the NHMRC-approved 2023 national guideline recommends universal EPDS screening early and later in pregnancy and again postpartum, and Medicare's obstetric items require a mental-health assessment to be offered and the result, or the decision not to participate, recorded. Canada runs the instructive contrast: the national task force recommends against universal instrument-based screening while the obstetric society and several provinces operationalize risk- and concern-based use, so a Canadian chart names whose guidance it follows. Positive item-10 responses route to a suicide risk assessment , brief triage findings sit in a risk screening note , and serial totals follow the conventions in the outcome measure note . ## How to document EPDS results in the chart No US federal law prescribes an EPDS note format, and no jurisdiction makes a score diagnostic. What survives review is a record with the version, the timing anchor, the threshold and its purpose, a separate item-10 line, a modest interpretation, and a closed follow-up loop. Each element below carries the pitfall that most often undermines it. Instrument, version, and mode. Chart the full 10-item RCPsych EPDS, or name the variant explicitly (an authorized translation with its language, the EPDS-US adaptation, a verbal administration, interpreter assistance, and whether completion was private). Pitfall: unverified forms. A 2003 journal letter documented circulating copies with altered item wording and missing preambles, so confirm the local form matches the original; and a modified or translated version charted as plain "EPDS" hides exactly what a reviewer needs to know. The timing anchor. Record gestational age or weeks postpartum, and at a well-child visit, the infant's age and the respondent's relationship to the infant. Pitfall: a bare score with no perinatal timepoint, which makes serial comparison and guideline mapping impossible; and the seven-day recall window misdescribed as two weeks. The total, with its threshold and purpose. Chart "x/30" plus the threshold applied and why: 10 or higher where sensitivity is the priority, 11 or higher as the pooled best balance, 13 or higher where specificity is intended, per the local program or guideline. Pitfall: "EPDS positive" with no number or threshold, or 13 treated as a diagnosis; at that threshold the pooled data miss about one-third of major-depression cases. Item 10, separately, every time. Record the self-harm item as zero or nonzero on its own line, apart from the total. Any nonzero response gets a same-day, direct risk assessment regardless of the total, with the disposition matched to assessed acuity rather than an automatic transfer, documented in a suicide risk assessment and cross-referenced. Pitfall: arithmetic in either direction (a low total used to skip follow-up, or every nonzero response treated as an emergency), "denies SI" charted against a nonzero item without reconciling the two, and method or means detail written into a screening note. Interpretation kept modest. A positive screen means elevated symptoms warranting assessment; among screen-positive women in the largest screening cohort, 22.6% of those assessed had bipolar disorders, and the scale cannot rule out postpartum psychosis, which is an emergency on clinical signs alone. The items 3 to 5 anxiety cluster is a conversation aid, not an anxiety diagnosis. Pitfall: a negative EPDS charted as ruling out depression, bipolar disorder, psychosis, or risk. Action and the closed loop. Document the same-visit assessment or referral, the warm handoff and its receipt, treatment connection verified at follow-up, the rescreen date, and who owns each open item; US quality measures count follow-up within 30 days of a positive screen. Pitfall: a positive screen that ends at a passive referral order, which quality reviewers and plaintiffs both read the same way. Whose chart, the billing line, and the declined screen. A caregiver screen at a well-child visit is recorded in the infant's chart with the minimum necessary (instrument, result, discussion, action), no maternal chart created for it, proxy visibility explained beforehand, and the caregiver pathway billed under the appropriate code (96161 for a caregiver instrument administered for the infant's benefit; 96127 is for the billed patient's own assessment). A declined screen is charted as offered and declined with any volunteered reason and a re-offer plan. Pitfall: a detailed maternal mental-health narrative in the infant's record, the two codes swapped, or a declined screen entered as a zero or negative result. ### Blank template (copy and adapt) EPDS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Instrument: full 10-item EPDS Version/language: [ ] Mode: [self-completed privately / verbal / interpreter] Timing: [gestational age / weeks postpartum / infant age + respondent relationship, for well-child screens] Total: [ ]/30 Threshold applied + purpose: [>=10 / >=11 / >=13 + local program or guideline] Item 10: [zero / nonzero -> same-day assessment documented at: ] Interpretation: [screen result only; not a diagnosis; bipolar and psychosis not excluded] Action: [assessment / referral + warm handoff / crisis response / rescreen date] Closed loop: [handoff received; treatment connection verified; follow-up owner + date] Chart + billing: [infant chart, minimum necessary (well-child) / own record; bare code noted] Declined: [offered + declined + reason if volunteered + re-offer plan; never entered as a score] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale itself. ## Sample EPDS documentation (fictional) Scenario: a caregiver screen at a two-month well-child visit, positive at the program threshold with a zero item 10, documented in the infant's chart with the minimum necessary and a closed referral loop. All details are fictional. Patient: Infant A.R., 8 weeks (caregiver screen: mother, 29) · Setting: Pediatric clinic, 2-month well-child visit · Clinician: S. Okafor, MD · Note date: 08/12/2026 Screening (infant chart entry): Caregiver postpartum depression screening offered per the 2-month well-child protocol and accepted after explanation that the result is recorded in the infant's chart. Full 10-item English EPDS self-completed privately by the infant's mother, 8 weeks postpartum. Total 12/30; clinic follow-up threshold is 10 or greater: screen positive. Item 10: zero. Result reviewed privately with the mother. Discussion and action: Brief same-visit conversation: she describes low mood, poor sleep independent of the infant's schedule, and reduced enjoyment for about three weeks; no thoughts of self-harm on direct inquiry, consistent with the zero item 10. Screen interpreted as elevated symptoms warranting assessment, not a diagnosis. With her consent, warm handoff completed by phone during the visit to the perinatal behavioral health intake line; intake appointment scheduled 08/19/2026, and she agreed to notification of her own obstetric clinician, sent today. Infant-focused plan: Infant growth, feeding, and development reviewed and normal today; caregiver support and infant-care resources provided. Rescreen at the 4-month visit per protocol, earlier if concerns emerge. Follow-up call by clinic nursing within one week to confirm the intake appointment was kept. Administrative: Caregiver screening documented in the infant's record with the minimum necessary detail; the mother's own assessment and treatment will be documented by her receiving clinicians in her record. Caregiver-instrument administration reported under 96161 for this encounter per payer rules. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, threshold, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, language, mode, private completion, and postpartum timing are all named, and the entry lives in the infant's chart with the minimum necessary detail, exactly as pediatric guidance directs. The total carries its threshold and the threshold's purpose, and the screen is explicitly interpreted as a screen, not a diagnosis. Item 10 gets its own line even at zero, and the direct-inquiry finding is reconciled with it rather than left implicit. The loop actually closes: a warm handoff completed during the visit, a scheduled intake date, consented notification of her own clinician, and a nursing call to verify connection, which is what the 30-day follow-up quality logic looks for. The billing line uses the caregiver-instrument pathway under the infant without importing maternal clinical detail that belongs in her own record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: screening is convention plus a patchwork of real law. USPSTF's 2023 Grade B recommendation covers depression screening in adults including pregnant and postpartum persons (CONVENTION, and a coverage driver), ACOG's 2023 guideline sets the initial-prenatal, later-pregnancy, and postpartum cadence with systems for assessment and follow-up (CONVENTION), and AAP guidance puts caregiver screening at the 1-, 2-, 4-, and 6-month well-child visits, in the infant's chart, billed under the caregiver-instrument pathway (CONVENTION and PAYER POLICY). Several states legislate directly (LAW): New Jersey requires screening around delivery and early postnatal visits; Illinois's Perinatal Mental Health Disorders Prevention and Treatment Act requires clinicians to invite and review a validated questionnaire, with pediatric invitations through the first year and consent before results go to the mother's clinician absent acute danger; California requires prenatal and postpartum clinicians to screen or offer screening; Arkansas requires facilitation of screening in the first six postpartum weeks and, distinctively, requires the record to state when a patient refused, with a liability consequence attached to that documentation; Louisiana conditions pediatric caregiver screening on consent and provides that screening alone creates no caregiver-clinician relationship. On the payer side, 12-month postpartum Medicaid coverage had been adopted by 49 states and DC as of July 2026 (Arkansas the holdout), a 2016 CMS bulletin lets state Medicaid programs cover the caregiver screen as a service to the child under EPSDT, and NCQA's prenatal and postpartum measures count a standardized screen and, for positives, follow-up within 30 days, with the EPDS threshold set at 10 or greater in measure logic, a quality data element, not a diagnosis (PAYER POLICY). Canada and Australia pull in opposite directions, which is itself the documentation point. Canada's national task force conditionally recommended against universal instrument-based perinatal screening in 2022, while the SOGC's 2024 guideline operationalizes validated tools when symptoms, risk factors, or concern are present, and provincial programs (British Columbia, Alberta, Manitoba) run their own cadences (CONVENTION, divergent; name whose guidance the chart follows). Australia's NHMRC-approved 2023 guideline recommends universal EPDS screening early and later in pregnancy and at six to 12 weeks postpartum plus once more in the first year, treats 13 or greater as the further-assessment threshold with 10 to 12 generally repeated within two to four weeks, and requires immediate further assessment on any positive item 10 (CONVENTION); Medicare's obstetric and postnatal items (16590, 16591, 16407, with telehealth equivalents) require a mental-health assessment to be offered and the result, or the decision not to participate, recorded (PAYER POLICY), and New South Wales publishes authorized translations with cautions that lower thresholds may fit some migrant and refugee populations. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The psychometrics and rights reward precision. The original validation involved 84 Edinburgh mothers at a 12/13 division (sensitivity 86%, specificity 78%); the pooled modern evidence supports threshold choice as a tradeoff and no universal pregnancy-versus-postpartum split; father and partner cutoffs in the literature run lower and vary with the target construct, so partners screen as separate respondents with their own thresholds, consent, and records. The scale is translated into more than 60 languages, but translation availability is not local validation, reviews have found uneven translation quality, and evidence in Indigenous populations is notably limited, so the chart names the exact version and language. The EPDS-US adaptation and the proposed EPDS-R revision are named versions, not silent substitutes. A nonzero item 10 is a strong relative signal with low absolute predictive value, which is why the rule is assess every nonzero response, neither ignore nor automatically hospitalize. On rights: the Edinburgh Postnatal Depression Scale is copyright of the Royal College of Psychiatrists; current terms permit individual own-use photocopying of complete, attributed copies and require written permission for distribution or republication in any medium. BastionGPT is not affiliated with the Royal College of Psychiatrists or the instrument authors. This page describes the instrument in original prose and reproduces no items, response options, or scoring materials. ↑ Back to the template and downloads ## Common EPDS documentation errors reviewers flag The numbers behind these errors are precise. In the pooled individual-participant data, a threshold of 13 or higher runs sensitivity 0.66 against specificity 0.95 , missing about one-third of major-depression cases, while 11 or higher gives the best combined balance; in a screening cohort of 10,000 postpartum women, 22.6% of assessed screen-positive women had bipolar disorders ; a 2026 Danish cohort of 170,218 births found a nonzero item 10 carried an adjusted odds ratio of 10.91 for later recorded self-harm while the absolute rate stayed at 0.27%; and measured US screening rates remain in the single digits to low teens across health plans. The BastionGPT Clinical Advisory Board sees the same errors most often in EPDS documentation reviews: A threshold charted as a diagnosis. "EPDS 14, postpartum depression" converts a screening cutoff into a clinical conclusion, and pages that misassign operating characteristics between thresholds make it worse. Chart the total, the threshold applied, its purpose, and "screen positive, assessment indicated." Item 10 buried in the arithmetic. A low total used to skip follow-up of a nonzero self-harm item, every nonzero response treated as an automatic emergency transfer, or "denies SI" charted against a nonzero item with no reconciliation. The item gets its own line, a same-day assessment whatever the total, and a disposition matched to assessed acuity. The scale treated as public domain. The College's terms permit own-use photocopying of complete, attributed copies; embedding the questionnaire in an EHR shared with others, a website, or a handout library needs written permission, and altered circulating copies (a documented wording anomaly among them) compromise validity as well as rights. The wrong chart at well-child visits. A detailed maternal mental-health narrative in the infant's record, a maternal chart opened solely to hold a screen, or the caregiver code and the patient code swapped. Infant chart, minimum necessary, caregiver pathway billed under the infant. Screens without a pathway. A positive result ending at a passive referral order, no follow-up within the 30-day window quality measures count, or a declined screen entered as a zero. Document the offer, the decline, the re-offer plan, and for positives, the handoff, receipt, and verified connection. One cutoff for everyone. The maternal threshold copied into partner screening (father cutoffs in the literature run lower and vary by target), an English threshold applied to a translation without local validation, or a pregnancy-versus-postpartum split invented that the pooled data do not support. Name the population, version, and threshold together. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version and language, mode, perinatal timing, total, threshold and purpose, item-10 status, actions and referrals) and it drafts the full entry: the screening line, modest interpretation, the separate item-10 documentation, the closed referral loop, and the well-child chart-placement conventions, ready for your review. Cross-check a finished note for the gaps reviewers flag: a threshold charted as a diagnosis, item 10 buried in the total, a maternal narrative in an infant chart, a declined screen entered as a score, or a positive screen with no follow-up plan. Draft the rescreen and follow-up documentation: prior results with timing anchors, connection verified or not, and the next screening date with its owner, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do EPDS scores actually mean? Ten items scored 0 to 3 sum to a total of 0 to 30 over the past seven days. Thresholds are sensitivity-specificity tradeoffs, not diagnostic lines: in the pooled individual-participant data, 10 or higher runs 0.85 sensitivity with 0.84 specificity, 11 or higher gives the best combined balance, and 13 or higher reaches 0.95 specificity while missing roughly one-third of major-depression cases. Accuracy is broadly similar in pregnancy and postpartum. A positive screen means elevated symptoms warranting clinical assessment; it is never itself a diagnosis, and the defensible chart line names the total, the threshold applied, and its purpose. What is the item 10 rule? Any nonzero response to the self-harm item requires a same-day, direct risk assessment regardless of the total score, a rule that traces to the instrument's developers and is operationalized by both US and Australian guidance. The disposition matches the assessed acuity: not every nonzero response is an emergency transfer, and a large 2026 cohort shows why both extremes fail (the relative association with later recorded self-harm was strong, an adjusted odds ratio near 11, while the absolute rate stayed near 0.27%). Record item 10 on its own line even when zero, document the assessment separately without method or means detail in the screening note, and reconcile any apparent contradiction rather than charting "denies SI" against a nonzero item. Whose chart does a well-child screen go in, and how is it billed? The infant's chart, because the screen is performed for the infant's benefit: pediatric guidance says to record the instrument, result, discussion, and follow-up plan there, and not to open a maternal chart solely for the screening encounter. Keep it to the minimum necessary, since proxies with access to the child's record may see it, and explain that before administering. The caregiver-instrument pathway is billed under the infant with 96161; 96127 is for a brief assessment of the billed patient themselves, and some payers direct one pathway or the other, so the code follows the payer. Detailed maternal assessment and treatment belong in her own record with her own clinicians; consent rules for sharing results (explicit in Illinois and Louisiana law) are documented where they apply. Can we copy the EPDS into our EHR, website, or handout library? Not without permission. The scale is copyright of the Royal College of Psychiatrists and is not public domain: the current terms allow an individual clinician or researcher to photocopy a complete, unmodified copy for their own use with the full citation, and require written permission for copying or distribution to others or republication in any medium, which covers websites, downloadable libraries, and EHR embedding shared across users. The older attribution-only footer still circulating on legacy PDFs was superseded. Two practical checks: verify the form in local use matches the original (a published letter documented circulating copies with altered item wording), and route republication requests through the College's permissions process rather than relying on what other sites appear to get away with. Which cutoff should we use, and does pregnancy change it? Choose by purpose and program, then document the choice. A threshold of 10 or higher favors sensitivity and is what US quality-measure logic counts as positive; 11 or higher maximized the combined balance in the pooled data; 13 or higher favors specificity and is Australia's further-assessment threshold, with scores of 10 to 12 generally repeated within two to four weeks there. The pooled evidence found broadly similar accuracy in pregnancy and postpartum, so there is no universal timing split, and population matters more: authorized translations may carry locally validated thresholds, and lower thresholds may fit some migrant and refugee populations. What fails review is not the choice of cutoff; it is a chart that never says which one was applied. Can we screen fathers and partners with the same form? Offering partner screening is supported (Australia's guideline recommends it), but the maternal threshold does not transfer. Father cutoffs in the literature run lower and vary with the target: one study found 5/6 useful for broader distress, another found above 10 best for major depression, a systematic review found candidate cutoffs from 7 through 10, and measurement-invariance work shows the scale does not function identically across parents. Screen each partner as a separate respondent with their own consent, their own documented threshold and rationale, and their own referral pathway, and do not place one partner's detailed mental-health information in the other's record or in the infant's chart beyond the minimum necessary. What does a positive or negative EPDS not tell you? A positive screen is not a diagnosis, and it is not even specific to unipolar depression: in the largest screening cohort, 22.6% of assessed screen-positive women had bipolar disorders, which changes treatment materially. A negative screen rules nothing out, and the scale does not detect postpartum psychosis at all; acute confusion, mania, delusions, hallucinations, or rapidly changing behavior is an emergency on clinical grounds regardless of any total. The items 3 to 5 anxiety cluster can flag anxiety content for conversation, but the evidence does not support it as a stand-alone anxiety screen; where anxiety is the question, use a validated anxiety instrument or a diagnostic assessment. How do I document a declined screen or the rescreen cadence? A declined screen is charted as offered and declined, with any reason the patient volunteered, resources provided, and the re-offer plan, and it is never entered as a zero or a negative result. Arkansas law makes this concrete, requiring the record to state that the patient was not screened because she refused; Australian Medicare items likewise require the result or the decision not to participate to be recorded. Cadence follows the named authority: ACOG's initial-prenatal, later-pregnancy, and postpartum pattern; AAP's 1-, 2-, 4-, and 6-month well-child visits; Australia's early-and-later pregnancy plus six-to-12-weeks and once more in the first year; and in Canada, name whose guidance applies, since the national task force and the obstetric society genuinely diverge. Can BastionGPT draft my EPDS documentation? Yes. Give it the facts (version and language, mode, perinatal timing, total, threshold and purpose, item-10 status, actions, referrals, and the chart placement) and it drafts the full entry: the screening line, modest interpretation, separate item-10 documentation, the closed referral loop with owners and dates, and the well-child conventions, ready for your review. It can also check a finished note for thresholds charted as diagnoses, item 10 buried in the total, maternal detail in an infant chart, declined screens entered as scores, and positives with no follow-up. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Cox JL, Holden JM, Sagovsky R, 1987, British Journal of Psychiatry, the EPDS publication (architecture, original 12/13 validation in 84 mothers); Cox J, 2019, the thirty-year review (60-plus translations; the item-10 assess-any-nonzero position). Royal College of Psychiatrists, rating-scales copyright and permissions ; the RCPsych reproduction terms as documented in the CamCOPS project record (own-use photocopying; written permission for distribution or republication). Levis B and colleagues, 2020, BMJ, individual-participant-data meta-analysis (58 studies, 15,557 participants; threshold operating characteristics; similar accuracy by perinatal timing). ACOG, 2023, Clinical Practice Guideline No. 4 and the perinatal mental health screening resources (cadence, systems requirement, immediate assessment on an affirmative self-harm response). AAP, 2019 (reaffirmed 2025), Pediatrics, perinatal depression in pediatric practice (well-child cadence, infant-chart placement); AAP coding guidance, caregiver health-risk-assessment code changes (96161 versus 96127, bare codes). USPSTF, 2023, screening for depression and suicide risk in adults (Grade B including pregnant and postpartum persons; no instrument or interval mandated). Canadian Task Force on Preventive Health Care, 2022, perinatal depression screening guideline (conditional recommendation against universal instrument-based screening); SOGC, 2024, Guideline No. 454 (risk- and concern-based assessment with validated tools). COPE, 2023, Australian perinatal mental health guideline (NHMRC-approved; universal EPDS cadence; the 13-or-greater and 10-to-12 handling; item-10 rule; partner screening); MBS Online, July 2026 schedule (items 16590, 16591, 16407 and telehealth equivalents; bare numbers only). Wisner KL and colleagues, 2013, JAMA Psychiatry, screening 10,000 postpartum women (14% positive; 19.3% of positives endorsed the self-harm item; 22.6% of assessed positives had bipolar disorders); Howard LM and colleagues, 2011, item-10 analysis ; Liu and colleagues, 2026, Danish cohort of 170,218 births (adjusted odds ratio 10.91; absolute rate 0.27%). CMS, May 11, 2016, informational bulletin on maternal depression screening in Medicaid (EPSDT coverage as a service to the child); KFF, postpartum coverage extension tracker (49 states plus DC as of July 2026); NCQA, depression-measure specifications memo (the EPDS measure threshold and 30-day follow-up logic). State law examples: Arkansas, Act 316 of 2023 (documented refusal with liability consequence); Illinois, Public Act 95-0469 ; California, Health and Safety Code 123640 ; Louisiana, Act 188 of 2022 (consent; no relationship created by screening alone). Shrestha SD and colleagues, 2016, translation validation review (uneven quality across 14 languages); NSW Health, authorized translated forms (population-specific threshold cautions). ## Related templates PHQ-9 Score Documentation BDI-II Score Documentation Suicide Risk Assessment General Risk Screening Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mini-cog All note templates # Mini-Cog Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Mini-Cog is a three-minute cognitive screen scored 0 to 5, combining three-item recall with a pass-or-fail clock drawing, published by Soo Borson and colleagues in 2000. Primary care, geriatrics, and hospital teams use it at wellness visits and as a first-pass check when memory concerns arise. It screens; it never diagnoses. This page covers how to document and interpret Mini-Cog results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any trained clinician or staff member (medical assistants, nurses, physicians) after brief training; interpreted by the treating clinician; no certification required Audience Primary care and geriatrics teams, memory clinics and neuropsychologists, hospital and post-acute teams, families receiving results, payers and quality reviewers Typical length 3 to 8 chart lines · administration about 3 minutes, plus time to disclose the result and plan Format family Clinician-administered cognitive screen (three-item recall 0 to 3 plus pass-or-fail clock 0 or 2, total 0 to 5) When it's used Medicare annual wellness visits, first-pass screening on patient or family concern, hospital and perioperative baselines, triage before a longer instrument Standards context Copyright Soo Borson (mini-cog.com): free for noncommercial clinical use unmodified; commercial and research use need permission; no authority mandates it by name On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Mini-Cog? The Mini-Cog is a brief cognitive screen that pairs uncued three-item delayed recall (0 to 3 points) with a clock drawing scored pass or fail (2 or 0, no intermediate point), for a total of 0 to 5. Soo Borson and colleagues published it in 2000 as a cognitive vital-signs measure for dementia screening in a multilingual older population, and the scoring-algorithm paper followed in 2001. Two versions share one algorithm: the standardized numerical form (dated 01.19.16, with six numbered word-list versions so that repeat testing can use a different list) and the graphical form, which keeps the scoring of the 2000 proof-of-concept study. The rights holder recommends the standardized form for repeated or longitudinal use and lists 23 language versions as of August 2026. Administration takes about three minutes; the rights holder's own instructions add that time is needed to explain the procedure, discuss the result, and set the next step, so three minutes is the test, not the encounter. The load-bearing fact is that the Mini-Cog is a screen, never a diagnosis. On the current scoring page, a total of 0 to 2 indicates a higher likelihood of clinically important cognitive impairment and 3 to 5 a lower likelihood of dementia that does not rule out some degree of impairment; the cut point was set to balance sensitivity and specificity for dementia detection, and it estimates no severity, no cause, no decisional capacity, and no driving fitness. A positive result means enough concern to justify a fuller cognitive and medical evaluation, which is why the chart carries the components, the version, the validity conditions, and the follow-up chain rather than a bare total. The MoCA and the MMSE are the usual longer comparators; the free SLUMS is the closer cousin in rights terms, and none of the four independently establishes a neurocognitive diagnosis, capacity, or fitness to drive. ## Who uses Mini-Cog documentation and when Primary care is the main user: the Medicare annual wellness visit requires detection of cognitive impairment (by observation, patient report, and concerns raised by family or caregivers) without naming any tool, and the Mini-Cog's brevity and no-certification administration make it a common choice for that slot, for a first-pass check when a patient or family member voices concern, and for the triage step before a longer instrument. Geriatrics, hospital medicine, and perioperative teams use it for baselines and delirium-adjacent screening, where the validity conditions matter more than the number; skilled-nursing assessment uses its own CMS instrument instead. In Canada, the national task force recommends against instrument-based screening of asymptomatic older adults, so the indication (concern, not routine) leads the note; in Australia the 75-and-over health assessment must cover cognition without naming a tool, and local guidance more often reaches for the GPCOG, RUDAS, or KICA. When a screen is positive, the fuller workup belongs in a neuropsychological report or a longer bedside instrument, the bedside cognitive findings sit in the mental status exam , and any question of decisional ability goes to a capacity evaluation , because the screen decides none of those things. ## How to document Mini-Cog results in the chart No US, Canadian, or Australian authority prescribes a Mini-Cog note format, and none makes a score diagnostic. What survives review is a record that names the indication, both component scores with the form and word-list version, the language and administrator, the conditions that could invalidate the result, a screen-only interpretation, and a closed follow-up loop. Each element below carries the pitfall that most often undermines it. Context and indication. Record why the screen was done: the cognitive component of a wellness visit, a patient or family concern, follow-up of a prior abnormal screen, a hospital or perioperative baseline, or part of another assessment. The Medicare requirement is detection of impairment; the tool is your choice. Pitfall: "Mini-Cog required by Medicare" (no regulation names it) or a score with no indication, which leaves a reviewer unable to tell routine screening from a symptomatic evaluation. The exact result, both components. Chart recall x/3, clock x/2 (2 = normal, 0 = abnormal; inability or refusal to draw scores 0), and total x/5. The rights holder's FAQ asks for at least the word-list version and each component score in a discrete EHR field. Pitfall: A bare total. A 2/5 built from two words and a failed clock is a different picture from no words and a normal clock, and the next examiner cannot tell which without the components. Form, version, and language. Name the standardized numerical or graphical form, the word-list version (1 to 6), the language version, and whether administration was language-concordant or through a qualified interpreter. On repeat testing use a different list and chart both. Pitfall: The recall words written into the narrative (they belong on the form, not the note), a translation or improvised interpretation charted as plain "Mini-Cog," or the same list reused at follow-up without saying so. Administrator and conditions affecting validity. Record who administered and scored it (any trained staff member may), then the conditions: hearing and vision aids, motor limitation, pain, fatigue, acute illness, sedation, delirium or fluctuating attention, education and literacy, analog-clock familiarity, interruptions, and any departure from standard instructions. Close with a validity conclusion: standard and interpretable, interpretable with caution, or not interpretable. Pitfall: A precise total charted beside a condition that made it uninterpretable, or a motor, visual, or literacy barrier scored as a cognitive clock failure. Document the barrier; do not invent an adjusted score. Interpretation kept to a screen. State the classification (0 to 2 positive; 3 to 5 lower likelihood, impairment not excluded) and the rule your program applies if it treats scores under 4 as a sensitivity trigger, then say what the result does not establish: mild cognitive impairment, dementia, cause, decisional capacity, or driving fitness. Note the reversible contributors considered. Pitfall: "Positive dementia screen" or a dementia diagnosis code supported only by the screen; a normal score used to close the question when the family reports decline. Disclosure, collateral, and function. Document what was explained in plain, non-diagnostic terms, who was present with the patient's permission, and the patient's response; then the collateral history (informant, onset, course) and instrumental function (medications, finances, appointments, transportation, cooking, safety). Pitfall: No record of disclosure, a diagnostic label delivered from a screen, or a positive result with no informant sought. Action, owner, and the closed loop. Record the reversible-contributor review, medication reconciliation, investigations, the fuller cognitive and functional assessment or referral, interim safety steps, and who owns follow-up by what date; a declined screen is charted as offered and declined with the alternative used and the re-offer plan. Pitfall: A positive screen that ends at a number, or a refusal entered as a zero or a positive result. ### Blank template (copy and adapt) MINI-COG DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Indication: [wellness visit / patient or family concern / follow-up of a prior screen / baseline before ...] Form: [standardized numerical / graphical] Word-list version: [1-6] Language: [ ] Interpreter: [none / qualified interpreter / language-concordant examiner] Result: recall [ ]/3 clock [ ]/2 (2 = normal, 0 = abnormal) total [ ]/5 Conditions: [hearing/vision aids, motor limits, pain, fatigue, acute illness, sedation, attention, education/literacy, analog-clock familiarity, deviations from standard administration] Validity: [standard and interpretable / interpretable with caution / not interpretable: reason] Interpretation: [screen positive (0-2) / screen negative (3-5); does not establish MCI, dementia, cause, capacity, or driving fitness; reversible contributors considered] Disclosure: [what was explained, who was present with permission, response] Collateral + function: [informant; onset/course; medications, finances, driving, appointments] Plan: [medication review / investigations / fuller cognitive and functional assessment or referral / interim safety steps] Follow-up owner + date: [ ] Prior result + list version, if repeat: [ ] Declined: [offered + declined + reason if volunteered + alternative assessment + re-offer plan; never entered as a score] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the word lists, clock instructions, or form. ## Sample Mini-Cog documentation (fictional) Scenario: a positive screen at a subsequent Medicare annual wellness visit in primary care, standard administration in the patient's language, with a family informant present and the follow-up chain set in motion. All details are fictional. Patient: M.R., 74 · Setting: Primary care, subsequent annual wellness visit · Clinician: L. Navarro, MD · Note date: 08/11/2026 Cognitive screening: Cognitive assessment component of today's subsequent annual wellness visit, with an added indication: the patient's daughter voiced concern about her memory at check-in. Mini-Cog, standardized numerical form, Spanish version, word-list version 3, administered and scored by J. Rivera, RN, in a quiet room before the visit; language-concordant, no interpreter. Patient wore her reading glasses and hearing aids, was alert with steady attention, no acute illness, no motor limitation, no sedating medication changes. Recall 1/3, clock 0/2 (abnormal), total 1/5: screen positive. Prior Mini-Cog 08/2025 at last year's visit: 5/5, word-list version 1. Interpretation: Standard administration; result interpretable. A positive cognitive screen with a documented one-year change; it does not establish mild cognitive impairment, dementia, a cause, decisional capacity, or driving fitness. Reversible contributors considered: no features of delirium, no new anticholinergic or sedating medications on reconciliation, PHQ-2 negative today, sleep and hearing reviewed. Disclosure and collateral: With the patient's permission the result was reviewed with her and her daughter in plain terms: the screen suggests memory and thinking should be looked at more closely, and it is not a diagnosis. Daughter describes about nine months of repeated questions, two missed utility payments, and a pill organizer found double-filled last month; patient acknowledges some forgetting, denies getting lost, still drives locally in daylight, manages cooking and self-care independently. Plan: Medication reconciliation completed today. Laboratory evaluation for reversible contributors ordered per clinic protocol. Dedicated 40-minute cognitive and functional assessment visit scheduled 08/25/2026 with the daughter present, to include a longer cognitive instrument, an instrumental-activities inventory, and a structured driving history; memory-clinic referral to be decided at that visit. Interim: daughter to oversee the pill organizer and bill payments with the patient's agreement; patient advised to defer any unfamiliar or night driving until the assessment. Follow-up owner: Dr. Navarro; nursing call within one week to confirm the appointment. Administrative: Result entered in the discrete cognitive-screen field with the form, word-list version, language, and both component scores; the clock drawing scanned to the chart. Wellness-visit cognitive detection documented for this encounter; no diagnosis coded from the screen. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, versions, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works Both component scores, the form, the word-list version, the language, and the administrator are named, so a reviewer can see what drove the classification and the next test can use a different list. The validity conditions are charted and a validity conclusion is drawn before the number is interpreted, which is what separates an abnormal screen from an uninterpretable attempt. The screen stays a screen: no diagnosis, no capacity or driving conclusion, reversible contributors reviewed, and the prior year's result gives the change its meaning. Disclosure and collateral are recorded, with the informant's concrete examples of instrumental decline reconciled against the score rather than left beside it. The loop closes: a dated assessment visit with a longer instrument, interim safety steps, a named owner, a confirmation call, and discrete-field entry that serial reviewers can find. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the requirement to look is law; the tool is a convention. Section 1861(hhh) of the Social Security Act and 42 CFR 410.15 make detection of cognitive impairment a required element of the initial and subsequent Medicare annual wellness visit (G0438, G0439), defined as assessment by direct observation with due consideration of patient report and concerns raised by family, friends, or caregivers, and no instrument is named (LAW). CMS deliberately declined to name one, reasoning, as Cordell and colleagues characterized it in 2013, that "there is no single, universally accepted screen," and its 2026 wellness-visit guidance tells clinicians to consider brief tools without naming any (PAYER POLICY). Bills in the CHANGE Act lineage before the 119th Congress would require an NIA-identified detection tool at the wellness visit and documentation of the tool used; as of August 2026 they are proposed, not law, so a chart that says Medicare required the Mini-Cog is wrong today and may need a different sentence later. The cognitive assessment and care plan service (99483, which replaced G0505 in 2018) is PAYER POLICY with documented elements far beyond any screen: a cognition-focused history and examination with an independent historian, functional assessment including decision-making capacity, standardized dementia staging, medication reconciliation, neuropsychiatric symptom evaluation, home and driving safety, caregiver identification, and a written care plan. A Mini-Cog result may be the initiating finding; it is not by itself 99483 documentation. Medicare Advantage risk-adjustment validation asks whether a submitted diagnosis is supported by the record (PAYER POLICY): an abnormal screen supports cognitive concern under evaluation, never a dementia code on its own. Hospitals must keep records that support diagnosis and response (42 CFR 482.24), and skilled-nursing facilities assess cognition through the CMS resident assessment instrument, whose cognitive section is its own interview (LAW for the record and the assessment; no Mini-Cog mandate in either). Driving fitness and decisional capacity are decision-specific evaluations governed by state law and clinical standards (CONVENTION on the screen's role): a positive Mini-Cog can justify a driving history, a functional or on-road assessment, or a capacity evaluation ; it decides neither. Canada and Australia frame the same instrument differently, which is itself the documentation point. The Canadian Task Force on Preventive Health Care recommends against instrument-based screening of asymptomatic community-dwelling adults 65 and older (a strong recommendation on low-certainty evidence, reaffirmed 2024) while expressly excluding anyone with patient, family, or clinician concern or functional change; the 2020 Canadian consensus conference supports targeted assessment with any of several tools, so a Canadian chart records the indication (concern, not routine) and the tool chosen (CONVENTION). Physician billing is provincial: Ontario's July 2026 bulletin lists dementia-assessment fee codes (A071, C071, A073, C073, A074, C074, A078) without an age restriction and without naming an instrument (PAYER POLICY, one province). Australia's Medicare health assessment for people 75 and older (items 701, 703, 705, 707; MBS note AN.0.39, March 2026) must include psychological function including cognition and mood, must be recorded, and must offer the patient a written report; it names no instrument, and Australian guidance more often reaches for the locally developed GPCOG, the RUDAS for culturally and linguistically diverse or lower-education patients, and the KICA for older Aboriginal and Torres Strait Islander people, so a Mini-Cog result there is one input to a broader recorded assessment (PAYER POLICY for the item, CONVENTION for the tool). Austroads and the Canadian driver-fitness standard both use condition-, function-, and risk-based assessment, and neither treats a brief screen as a licensing determination. On the evidence: the 2000 development cohort (249 people, about half with dementia, uncertain cases excluded) produced the 99% sensitivity figure that still headlines most pages; the population-based 2003 sample gave 76% and 89%; the 2024 meta-analysis pooled 73% and 84% across eight primary-care and community studies (5,620 participants) with very high heterogeneity and high or unclear selection risk in most studies; a Cochrane review summarized by AAFP in 2022 found too little evidence to support routine Mini-Cog screening in primary care; a Brazilian study of 211 adults with four or fewer years of schooling found 60% and 65%; and USPSTF's 2020 I statement (an update research plan posted June 2025) finds insufficient evidence for or against screening asymptomatic older adults, while the Australian government expects updated dementia guidelines in late 2026. Report ranges and name the population; do not chart the 99%. Rights and versions reward precision. The Mini-Cog is copyright of Soo Borson and distributed through mini-cog.com, whose 2026 permission terms tier use by who is using it: universities, foundations, health professionals, hospitals, clinics, and public-health institutes may use, reproduce, and distribute the unmodified instrument for clinical and educational purposes without permission or a licensing agreement, provided it stays free to patients; commercial entities and pharmaceutical companies need prior written permission and a license even when the end user pays nothing; and research use requires permission for everyone, with a licensing agreement when commercially funded. No express carve-out is published for a public self-scoring web tool, so a commercial operator that reproduces the stimuli and applies the algorithm should assume permission is required, and modification of any kind voids the no-permission position. Two forms share one algorithm (the standardized numerical form, dated 01.19.16, with six numbered word lists, and the graphical form that keeps the 2000 scoring), 23 language versions are listed, and availability is not validation for every population: the rights holder itself warns that people without basic literacy or analog-clock familiarity may be cognitively normal and still fail the clock. Mini-Cog is copyright Soo Borson; this page links to the official source and hosts nothing. BastionGPT is not affiliated with, or endorsed by, the rights holder or the instrument's authors. This page reproduces no word lists, clock instructions, forms, or scoring materials. ↑ Back to the template and downloads ## Common Mini-Cog documentation errors reviewers flag The numbers behind these errors are specific. The original 2000 development cohort gave 99% sensitivity in a sample where about half the participants had dementia; the 2024 meta-analysis pooled 73% sensitivity and 84% specificity in primary-care and community settings, so roughly one in four people with dementia screens negative; untrained raters agreed with experts on only about 60% of mildly impaired clocks in the 2001 rater study ; and a Brazilian cohort with four or fewer years of schooling produced 60% sensitivity and 65% specificity. The BastionGPT Clinical Advisory Board sees the same errors most often in Mini-Cog documentation reviews: A screen charted as a diagnosis. "Mini-Cog 1/5, positive dementia screen" or a dementia code supported only by the screen. Chart the classification, the statement that it establishes no diagnosis, cause, capacity, or driving fitness, and the assessment that follows; code cognitive concern under evaluation until an evaluation supports more. A total without its parts. "Mini-Cog 3/5" with no recall and clock components, no form, no word-list version, and no language. The rights holder asks for the version and each component in a discrete field; without them the next test cannot avoid the same list and the reviewer cannot see what drove the result. An invalid attempt scored as abnormal. A patient with hemiparesis, tremor, uncorrected vision, delirium, an improvised family translation, or no analog-clock experience scored 0 on the clock and charted as a positive screen. Document the barrier, call the result nonstandard or not interpretable, and use a screen that does not depend on the impaired function. "Medicare requires the Mini-Cog". The wellness-visit regulation requires detection by observation, patient report, and collateral concern and names no tool; 99483 documentation is a full cognitive assessment and care plan, not a screen. Say how cognition was assessed and what was found; do not cite a mandate that does not exist. A positive screen that ends at a number. No disclosure recorded, no informant sought, no reversible contributors reviewed, no fuller assessment or referral, no owner or date. Reviewers read a 1/5 followed by silence as an assessment performed and ignored. A normal screen used to close the question. "Mini-Cog 5/5, no concerns" when a spouse has described nine months of missed bills. The rights holder states that a normal score does not rule out impairment and that concern from patient, clinician, or family warrants further evaluation regardless; chart the collateral examples and the next step. The reverse error, a declined screen entered as zero, belongs here too. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (form and word-list version, language, administrator, both component scores, validity conditions, collateral, prior result) and it drafts the documentation block: components stated, validity conclusion in place, the screen-not-diagnosis line, disclosure, and the follow-up chain with owner and date, ready for your review. Cross-check a finished note for the gaps reviewers flag: a total without components, a "positive dementia screen," a motor or delirium-invalidated attempt scored as abnormal, a declined screen entered as zero, or a positive result with no plan. Draft the follow-up documentation: prior result and list version, the interval change, what the longer assessment found, and the reconciled plan, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Mini-Cog scores actually mean? Two components sum to a total of 0 to 5: three-item delayed recall scored 0 to 3, and a clock drawing scored 2 (normal) or 0 (abnormal) with no intermediate point; inability or refusal to draw scores 0. On the rights holder's current scoring page, 0 to 2 indicates a higher likelihood of clinically important cognitive impairment (a positive screen) and 3 to 5 a lower likelihood of dementia that does not rule out some degree of impairment. The cut point was set empirically to balance sensitivity and specificity for dementia detection; it estimates no severity, no cause, and no capacity or driving fitness. Chart the components, not just the total: a 2/5 built from two words and a failed clock is a different picture from no words and a normal clock, and the follow-up examiner needs to know which. Is a Mini-Cog score of 3 positive or negative? Under the conventional rule, 3 is negative (3 to 5 means lower likelihood). The 2016 standardized form adds that a cut point below 4 may be used when greater sensitivity is desired, and some care pathways route anyone under 4 to a longer instrument; the current mini-cog.com page, however, states that treating 0 to 3 as positive has been used in some studies but has not been adequately tested against formal cognitive-disorder diagnoses. Document which rule your program applies and why, and chart the number itself with its components. A 3 built from one word and a normal clock, in a patient whose family reports decline, is exactly the case where the screen should not end the conversation whatever rule you use, and a normal clock in a patient with three words recalled and no concern is a routine negative. Does Medicare require the Mini-Cog at the annual wellness visit? No. The statute (Social Security Act 1861(hhh)) and 42 CFR 410.15 require detection of any cognitive impairment at the initial and subsequent wellness visit (G0438, G0439), defined as assessment by direct observation with due consideration of patient report and concerns raised by family, friends, or caregivers, and CMS deliberately named no instrument, reasoning that no single universally accepted screen exists. Using a brief validated tool is a convention CMS encourages, and the Mini-Cog is one option among several. Two cautions: bills in the CHANGE Act lineage before the 119th Congress would require an NIA-identified tool and documentation of the tool used, and as of August 2026 they are proposed, not law; and 99483 is a separate cognitive assessment and care plan service whose documented elements (independent historian, functional and capacity assessment, staging, medication reconciliation, safety, caregiver, care plan) go far beyond any screen. Can we print the Mini-Cog, put it in our EHR, or build a web scorer? The answer turns on who you are and whether the instrument stays unmodified and free to patients. Universities, foundations, health professionals, hospitals, clinics, and public-health institutes may use, reproduce, and distribute the current form for clinical and educational purposes without permission or a licensing agreement, which covers a printed form and an internal EHR build; the rights holder's FAQ actively recommends a discrete EHR field capturing the word-list version and both component scores. Commercial entities and pharmaceutical companies need prior written permission and a licensing agreement even when the tool is free to end users, research use always requires permission, and modification voids the no-permission position for everyone. No express carve-out is published for a public self-scoring web tool, so a commercial operator that reproduces the stimuli and applies the algorithm should ask first. Terms are as stated at mini-cog.com in August 2026; confirm with the rights holder before commercial or research use. How do I score an ambiguous clock, or a patient who cannot draw? The clock is binary by design (2 or 0), which keeps scoring reliable for untrained staff on clearly normal and clearly abnormal drawings; the honest caveat is that agreement between untrained and expert raters fell to about 60% for mildly impaired clocks in the 2001 rater study, and the misses ran one way (subtly abnormal clocks called normal). When the drawing is borderline, get a second trained rating, keep the drawing (scan it to the chart), and state the uncertainty if it changes the classification. A patient who cannot hold a pen, cannot see the page, or has never used an analog clock has not demonstrated a cognitive clock failure even though the form would score 0: document the barrier, do not invent an adjusted score, call the result nonstandard or not interpretable, and use a screen that does not depend on the impaired function. The screen is normal but the family reports decline. Which do I believe? Believe the history, then test it. The rights holder states that a normal Mini-Cog does not rule out clinically relevant impairment, especially very mild impairment or a person with high prior ability, and that concern from a patient, clinician, or family member warrants additional evaluation whatever the score; pooled sensitivity in primary care runs about 73%, so roughly one in four people with dementia screens negative. Chart both the score and the collateral examples (onset, course, medication management, finances, driving), state that the normal screen does not close the question, and document the next step. The reverse case follows the same logic: a positive screen with no reported concern first gets a validity check (acute illness, delirium, sedating drugs, depression, sensory or language barrier, literacy, clock familiarity), then collateral and a fuller assessment; absence of concern does not erase an abnormal result. How do I document a declined, refused, or invalid screen? Record that cognitive assessment was offered and why, that the patient declined (or which component), the reason if volunteered, how cognition was assessed instead (observation, patient report, collateral), any observed concern, the alternative plan or re-offer, and any immediate safety issue addressed. Do not enter a total, a zero, or a positive screen. For a US wellness visit the underlying requirement is detection by observation and history, which can still be met and documented without a completed tool. An invalid attempt (delirium or fluctuating attention, acute illness, an improvised family translation, a motor or visual barrier) is charted as attempted, nonstandard, and not interpretable, with the reason and the plan to reassess after stabilization or with an authorized language version and qualified interpreter; it is not charted as a positive screen. Mini-Cog, MoCA, SLUMS, or MMSE: which one, when? Choose by the question and the setting. The Mini-Cog answers one question, whether there is enough concern to justify a fuller evaluation, in about three minutes, by any trained staff member, without certification, and free for noncommercial clinical use; its clock component is its weakness in low-literacy or non-analog-clock populations. The MoCA samples more domains and is generally regarded as more sensitive to mild cognitive impairment, at the cost of about 10 minutes and a training and certification regime for anything beyond total-score screening. The SLUMS is a free, longer screen with education-adjusted interpretation and its own training expectation. The MMSE is a licensed, per-form instrument chosen mainly for continuity with existing records or drug-coverage rules that still key to it. Many pathways run the Mini-Cog first and escalate positives (or scores under 4) to a longer instrument; no peer-reviewed same-cohort Mini-Cog-versus-MoCA accuracy study had been published as of August 2026, so claims of superiority either way outrun the evidence. Can BastionGPT draft my Mini-Cog documentation? Yes. Give it the facts (form and word-list version, language, administrator, both component scores, validity conditions, disclosure, collateral, prior result, and the plan) and it drafts the full entry: components stated, validity conclusion, the screen-not-diagnosis line, disclosure and collateral, and the follow-up chain with owners and dates, ready for your review. It can also check a finished note for a total without components, a screen charted as a diagnosis, an invalidated attempt scored as abnormal, a declined screen entered as zero, and a positive result with no plan. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Borson S, Scanlan J, Brush M, Vitaliano P, Dokmak A, 2000, International Journal of Geriatric Psychiatry, the Mini-Cog publication (249 multilingual older adults; development-cohort accuracy); Scanlan J, Borson S, 2001, the scoring-algorithm and rater study (about 60% naive-rater agreement on mildly impaired clocks); Borson S and colleagues, 2003, Journal of the American Geriatrics Society, population-based validation (n = 1,119; 76% and 89%). Mini-Cog rights holder (Soo Borson), mini-cog.com, accessed August 2026: permission tiers (noncommercial clinical and educational use without permission; commercial, pharmaceutical, and research use with permission), scoring page (0 to 2 and 3 to 5 interpretation; the 0 to 3 caveat), FAQ (discrete EHR field with version and component scores; disclosure guidance; normal scores do not rule out impairment), standardized and graphical forms , language versions (23 listed), instrument development (literacy and analog-clock caution). Abayomi SN and colleagues, 2024, PLoS One, systematic review and meta-analysis (14 studies; primary care and community 73% and 84% across 5,620 participants; risk-of-bias findings). Cochrane Dementia and Cognitive Improvement Group, Mini-Cog accuracy in general practice ; AAFP, 2022, American Family Physician, Cochrane summary (four primary-care studies; sensitivity 76% to 100%, specificity 27% to 85%; the one low-risk study at 76% and 73%). Ribeiro Filho ST, Lourenço RA, 2009, Dementia and Neuropsychologia, low-education validation (n = 211, four or fewer years of schooling; 60% and 65%); Carnero-Pardo C and colleagues, 2013, International Journal of Alzheimer's Disease, Spanish primary-care analysis (n = 307; AUC 0.78; education effect; post hoc construction acknowledged); Li X and colleagues, 2018, Medicine, Mini-Cog versus MMSE for MCI (referred cohort, n = 229). USPSTF, February 2020, screening for cognitive impairment in older adults (I statement) and June 2025, final research plan for the update . eCFR, 42 CFR 410.15 (annual wellness visit; detection of cognitive impairment; no instrument named), 42 CFR 482.24 (hospital medical records), 42 CFR 483.20 (resident assessment); CMS, annual wellness visit guidance and cognitive assessment and care plan services (99483 elements; page last modified May 2026); Cordell CB and colleagues, 2013, Alzheimer's and Dementia, Medicare AWV cognitive-detection algorithm (CMS rationale for naming no tool); US Congress, 119th, H.R. 3501 (proposed NIA-tool requirement, pending). Canadian Task Force on Preventive Health Care, cognitive impairment guideline (against screening asymptomatic adults 65 and older; reaffirmed 2024); Government of Ontario, July 2026 OHIP bulletin (dementia-assessment fee codes without age restriction); CCMTA, Determining Driver Fitness in Canada . MBS Online, note AN.0.39 (health assessment for people 75 and older; items 701, 703, 705, 707; cognition and mood; no instrument named); Australian Government Department of Health, Focus on dementia, July 2026 (updated guidelines anticipated late 2026); Austroads, Assessing Fitness to Drive . Comparator rights pages: MoCA permission terms ; Saint Louis University SLUMS page ; GPCOG ; Dementia Australia, RUDAS . ## Related templates MoCA Score Documentation MMSE Score Documentation SLUMS Score Documentation Neuropsychological Evaluation Report Capacity & Competency Evaluation Mental Status Exam (MSE) Note Clock Drawing Test Score Documentation Geriatric Depression Scale (GDS-15) Score Documentation CAM-ICU Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/dass-21 All note templates # DASS-21 Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The DASS-21 (Depression Anxiety Stress Scales, 21-item short form) is a public-domain self-report measure with three seven-item subscales, published by Lovibond and Lovibond in 1995 and scored over the past week. Psychologists, GPs, and counselors use it at intake and review, especially in Australia. It measures symptom severity; it never diagnoses. This page covers how to document and interpret DASS-21 results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report (adult form from age 14 with adequate language; DASS-Y for younger children); scored and interpreted by the clinician; no license or qualification level required Audience Psychologists and counselors, GPs and prescribed medical practitioners under Better Access, headspace and PHN-commissioned services, primary care and university counseling teams, payers and auditors Typical length 3 to 8 chart lines (three subscales, doubled, with labels and change) · patient completion about 5 minutes Format family Self-report symptom scales (21 items in three 7-item subscales scored 0 to 3, past-week window; each subscale doubled to a 0 to 42 metric) When it's used Intake baselines, treatment reviews and course-of-treatment reports, Better Access outcome measurement, program evaluation and research Standards context Public domain (Psychology Foundation of Australia): copy freely, do not modify, sell, or host on a public website or app; named with the K10 as an example in Better Access guidance, mandated nowhere On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the DASS-21? The Depression Anxiety Stress Scales (DASS) are a set of self-report scales published by S. H. Lovibond and P. F. Lovibond, whose 1995 manual (second edition, Psychology Foundation, Sydney) is the controlling citation. The original instrument has 42 items, 14 per scale; the DASS-21 keeps seven per scale, chosen so that each short-form subscale runs at about half the full-scale value, which is why every adult DASS-21 subscale is multiplied by two before it is read against the manual's norms and the conventional severity ranges. Each item is rated 0 to 3 for the past week, so the raw subscale range is 0 to 21 and the doubled range is 0 to 42. The scales are dimensional and public domain, distributed by the Psychology Foundation of Australia through the UNSW DASS website, and a separate youth form (DASS-Y, 2022) exists for children and adolescents and is scored on its own metric, not doubled. The load-bearing facts for documentation are three. First, there are three scores, not one: Depression, Anxiety, and Stress are scored and interpreted separately, and the rights holder's FAQ describes the Stress scale as a factorially distinct syndrome of nervous tension, difficulty relaxing, and irritability, not a count of life stressors, which is exactly what a single-number distress measure such as the K10 cannot show. Second, the severity labels (normal, mild, moderate, severe, extremely severe) are conventional population-relative ranges set at percentile cut points, and the rights holder calls the choice of any single clinical threshold necessarily arbitrary; a label describes where a score sits in the population, never a diagnosis. Third, the DASS-21 has no suicide or self-harm item, by design, so risk is assessed and documented separately. The disorder-anchored PHQ-9 and GAD-7 are the usual comparators in US and Canadian practice; the DASS-21 earns its place when the three-way profile matters. ## Who uses DASS-21 documentation and when Australia is the DASS-21's home market. The Better Access initiative expects a mental health treatment plan to include an outcome measurement tool unless clinically inappropriate, with the choice at the GP's or prescribed medical practitioner's discretion, and Commonwealth guidance (March 2026) names the K10 and the DASS-21 as examples, so GPs, psychologists, and counselors chart it at intake, at review points, and in the course-of-treatment report back to the referrer described on the Better Access treatment plan page. The K10 remains the incumbent for national reporting (the primary mental health care minimum data set, headspace, several state services), so many services run the K10 for comparability and add the DASS-21 when the depression, anxiety, and stress profile changes what they do. In Canada and the United States it lives in university counseling, integrated primary care, workers' compensation and insurer programs, and research, alongside or instead of the PHQ-9 and GAD-7 ; no payer in either country requires it. Serial results follow the conventions of the outcome measure note , and any risk finding is documented in a suicide risk assessment , because the DASS-21 cannot carry it. ## How to document DASS-21 results in the chart No law in Australia, Canada, or the United States prescribes a DASS-21 note format, and the rights holder publishes scoring and cautions, not a template. What survives review is a record that names the version, language, and timeframe; reports three doubled subscale scores with their conventional labels; keeps the labels descriptive; states change per subscale; and records the separate risk assessment. Each element below carries the pitfall that most often undermines it. Version, language, and timeframe. Chart adult DASS-21, DASS-42, or DASS-Y; the language and, where possible, the specific translation (the DASS site posts translations but does not verify their quality); the standard past-week timeframe; and the administration conditions (self-completed, interviewer-assisted, paper, or a restricted electronic workflow), with any literacy, comprehension, or missing-item concern and the prorating rule if one was used. Pitfall: "DASS-21" with no language or version, DASS-Y results trended against adult scores, or a changed timeframe that silently breaks comparability with the norms. Three subscales, doubled. Sum the seven items in each subscale (raw 0 to 21), multiply each subscale by two, and chart both metrics: Depression raw 8/21, doubled 16/42. A fully completed, simply doubled adult DASS-21 gives even numbers; an odd or decimal score means prorating or a different metric and needs a note. DASS-Y is never doubled. Pitfall: Raw sums read against the doubled severity ranges (two tables circulate online), or a doubled score charted without saying so, so the next reader cannot tell which metric they are looking at. Severity labels kept descriptive. Attach the conventional label to each doubled subscale: Depression 0 to 9 normal, 10 to 13 mild, 14 to 20 moderate, 21 to 27 severe, 28 and above extremely severe; Anxiety 0 to 7, 8 to 9, 10 to 14, 15 to 19, 20 and above; Stress 0 to 14, 15 to 18, 19 to 25, 26 to 33, 34 and above. The labels sit at population percentile cut points (about the 78th, 87th, 95th, and 98th) and describe range, not disorder. Pitfall: "DASS confirms severe anxiety disorder," "mild" read as a mild disorder, or extremely severe stress charted as an emergency in itself. No unexplained total. Report the three subscales separately. If a composite is used, say so, name the method (the FAQ's approach converts each scale to a norm-referenced z score and averages), and name the norm set; never apply the subscale severity ranges to a sum. Pitfall: "DASS score 48, severe," a number no reader can decode: one subscale, an undoubled 21-item sum, a sum of doubled subscales, or a standardized composite. The Stress scale and discordant profiles. Read Stress as tension, irritability, over-reactivity, and difficulty relaxing over the week, and preserve discordance: an extremely severe Stress score beside a normal Depression score is information about where the symptoms sit, to be examined against triggers, sleep, pain, substances, and function. Pitfall: The three scores averaged into one impression, or Stress reported as a measure of life stressors or burnout. Change per subscale, in context. State the prior date, prior scores, and the change for each subscale (Anxiety 18 to 8, change of 10) on the same version, language, timeframe, and metric. Use descriptive language unless the service has adopted and named a reliable-change method and norm set. Note recent acute events: the past-week window means a result two days after a job loss is valid for that week but may not be a stable baseline. Pitfall: "DASS improved from 66 to 52," "reliably improved" with no method, or a single-week spike read as treatment failure. Risk assessed separately. State that the DASS-21 contains no suicide or self-harm item and record the direct risk assessment, its findings, and the plan in the appropriate section; a low Depression score documents nothing about ideation, and an extremely severe score documents no intent. Pitfall: "Depression 6, normal, no risk concerns" with no direct inquiry, or the DASS-21 entered as the service's suicide screen. ### Blank template (copy and adapt) DASS-21 DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Time point: [intake / review / course-of-treatment report / discharge] Version: [adult DASS-21 / DASS-42 / DASS-Y] Language + translation: [ ] Mode: [self-completed / interviewer-assisted; paper / restricted electronic] Timeframe: [past week (standard) / other: state and note comparability] Completeness: [all items answered / missing items + prorating rule used] Depression: raw [ ]/21 -> x2 = [ ]/42 [normal / mild / moderate / severe / extremely severe] Anxiety: raw [ ]/21 -> x2 = [ ]/42 [label] Stress: raw [ ]/21 -> x2 = [ ]/42 [label] (DASS-Y: report on its own metric, not doubled) Prior administration: [date] D [ ] A [ ] S [ ] Change per subscale: [ ] Interpretation: [labels are conventional population-relative ranges, not diagnoses; no total reported; profile and context: acute events, sleep, pain, substances, function; congruence with interview] Risk: [DASS-21 has no suicide item; direct risk assessment done: findings, formulation, plan documented at ...] Plan: [treatment focus / next administration and purpose / report to referrer] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale items or forms. ## Sample DASS-21 documentation (fictional) Scenario: an Australian private-practice review at the end of an initial Better Access course of treatment, with a repeat DASS-21 showing improvement in two subscales and persistent stress, risk assessed separately, and the report back to the referring GP. All details are fictional. Patient: T.L., 34 · Setting: Private psychology practice, Better Access review (session 6) · Clinician: R. Whitfield, Clinical Psychologist · Note date: 08/12/2026 Measure: DASS-21, adult English version, self-completed on paper in the waiting room before session 6, standard past-week timeframe, all 21 items answered. Raw subscale sums: Depression 8/21, Anxiety 4/21, Stress 14/21. Doubled for comparison with DASS norms and the conventional ranges: Depression 16/42 (moderate), Anxiety 8/42 (mild), Stress 28/42 (severe). Baseline at intake, 07/01/2026, same version and conditions: Depression 24/42 (severe), Anxiety 18/42 (severe), Stress 30/42 (severe). Interpretation: Change by subscale: Depression 24 to 16 (down 8), Anxiety 18 to 8 (down 10), Stress 30 to 28 (down 2). Depressive and anxiety symptoms decreased over the six sessions; tension, irritability, and difficulty relaxing remain in the severe range, consistent with the ongoing workplace conflict T.L. described during both measurement weeks. Labels are conventional population-relative ranges, not diagnoses; no total is reported; the practice has not adopted a reliable-change rule for this measure, so change is described rather than labelled significant. Result congruent with today's interview: improved mood and fewer panic-like episodes, continued poor sleep before shifts and irritability at home. Risk: The DASS-21 contains no suicide or self-harm item. Risk assessed directly today: denies thoughts of self-harm or suicide since the last review, no plan or intent, no self-harm behavior, protective factors unchanged (partner, work role, treatment engagement); formulation and safety planning recorded in the risk section of the record. Plan and report: Initial course of treatment complete. Recommend a further course under the existing plan focused on arousal reduction, sleep, and workplace boundaries; re-administer the DASS-21 at the final session of the next course. Written report to the referring GP today as required at the end of the course: intake and current scores for all three doubled subscales with their ranges, change per subscale, current functioning, the separate risk conclusion, and the recommendation. Session claimed under item 80010 per the referral. Administrative: Scores entered in the practice's outcome-measure record with version, language, timeframe, and both metrics; the K10 is not used by this practice for this client, and the referral did not specify an instrument. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works Version, language, timeframe, mode, and completeness are named, so the scores are reproducible and comparable with the baseline. All three subscales appear in both metrics with their conventional labels, and no total is offered, so a reader knows exactly what each number is. Change is stated per subscale against a dated baseline on the same conditions, described rather than over-claimed, and reconciled with the interview and the week's context. The missing suicide item is named and the direct risk assessment is documented separately, so a normal-range Depression score is never mistaken for a risk finding. The Better Access loop closes: the end-of-course report to the referrer carries the three scores, change, function, risk, and the recommendation, and the item claimed matches the referral. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations Australia: outcome measurement is expected; the instrument is a choice. No Australian law names the DASS-21. Under Better Access (PAYER POLICY, via Medicare eligibility and claiming rules), Commonwealth guidance updated in March 2026 says a mental health treatment plan must include an assessment, an outcome measurement tool unless clinically inappropriate, and a provisional or formal diagnosis, with the tool at the GP's or prescribed medical practitioner's clinical discretion and the K10 and DASS-21 named as examples; there is no mandated DASS-21 template, format, or interval. Current plan-preparation items (Services Australia, June 2026) are 2700, 2701, 2715, and 2717 for GPs and 272, 276, 281, and 282 for prescribed medical practitioners, with video equivalents 92112, 92113, 92116, 92117, 92118, 92119, 92122, and 92123. The dedicated review and ongoing-management items (2712, 2713, and their equivalents) were removed on 1 November 2025 and reviews now use applicable general attendance items, so any page still directing a GP to 2712 is obsolete; a plan does not expire, a new plan is generally not prepared within 12 months, and reviews ordinarily occur no more than once every three months and not within four weeks of plan preparation (PAYER POLICY, and rules about the plan, not a DASS administration schedule). The K10 is the system-level incumbent (CONVENTION): it is the standard consumer measure in the primary mental health care minimum data set, headspace's principal outcome measure, and mandatory in several state and territory specialized services, which is why many services keep the K10 for comparability and add the DASS-21 for the profile. The November 2025 Better Access reforms carry a post-implementation review with no fixed date. United States and Canada: no requirement, and specific billing limits. No US federal law or payer names the DASS-21; PHQ-9 and GAD-7 hold the primary-care and quality-measure slots. Brief standardized emotional or behavioral assessment is reported under 96127 (PAYER POLICY): the code is structured per instrument, one completed, scored, and documented instrument is one unit, three DASS-21 subscales do not automatically justify three units absent explicit payer direction, commercial payers commonly cap the code at four units per date of service and traditional Medicare at three, and Medicare's annual depression screening runs through G0444 rather than 96127; the CY 2027 physician fee schedule was still a proposed rule in August 2026, so recheck payer policy after it finalizes. In Canada, no federal or provincial law or payer names the DASS-21: Ontario Health's anxiety quality standard calls for validated tools and regular measurement without prescribing one, WSIB's community mental health program requires its progress form after every sixth session or eighth week without naming an instrument, and family health teams publish DASS-21 scoring guides for local use (CONVENTION and program PAYER POLICY; verify each contract). On the evidence: the 1995 severity framework rests on a development sample of 717 first-year psychology students, the UK adult norms are from 2005 (1,794 adults; alphas .88 Depression, .82 Anxiety, .90 Stress, .93 for the 21-item total), and Australian adult percentile norms from 2011, so a percentile or z score should name its norm set. The scales intercorrelate around .5 to .7, and the modern factor-analytic literature (Henry and Crawford 2005; Osman and colleagues 2012; Zanon and colleagues across eight countries in 2021 and nine in 2025) finds a strong general-distress factor beneath the three specific factors, which is the reason to read small subscale differences modestly, not a reason to stop reporting three scores. A 2019 COSMIN review of 48 studies rated responsiveness evidence low for Depression and Stress and very low for Anxiety, so no universal minimal-change threshold is established; a Chichewa-language study in Malawi reported an Anxiety alpha of 0.29, a reminder that translation availability is not local validation. Rights are precise and easy to overstate. The DASS is described by its rights holder as public domain: the forms may be downloaded and copied without restriction, but they may not be modified or sold for profit, computerized administration is permitted with results going to the clinician or researcher, automated interpretation returned to respondents is discouraged as misleading and potentially dangerous, and a website or app open to members of the public (a clinical practice website is the FAQ's own example) is not permitted; the allowed digital model is a restricted, defined group such as enrolled patients with results routed to the clinician. A clinician may charge for an assessment that includes the DASS; the scales themselves may not be sold. Translations must remain public domain and cannot be copyrighted or sold, and no separate translator permission is needed for versions posted on the DASS site, which does not verify their quality. The correct citation is the 1995 manual, and professional publications are asked to cite the permanent DASS website. The DASS-Y (ages 8 to 17 on the site) is a standalone youth form on its own metric. The Depression Anxiety Stress Scales are distributed by the Psychology Foundation of Australia through the UNSW DASS website. BastionGPT is not affiliated with, or endorsed by, the Psychology Foundation of Australia, UNSW, or the instrument's authors. This page reproduces no items, response options, or forms and describes the scoring conventions in original prose. ↑ Back to the template and downloads ## Common DASS-21 documentation errors reviewers flag The numbers behind these errors are specific. The conventional severity labels sit at population percentile cut points of about the 78th, 87th, 95th, and 98th , derived from a 1995 development sample of 717 first-year psychology students; the three scales intercorrelate around .5 to .7 and share a strong general-distress factor in eight-country and nine-country analyses; a 2019 COSMIN review of 48 studies rated responsiveness evidence low or very low, so no universal minimal-change threshold exists; and no published audit quantifies how often the multiply-by-two step is skipped, only that two incompatible cutoff tables circulate online. The BastionGPT Clinical Advisory Board sees the same errors most often in DASS-21 documentation reviews: Raw sums read against the doubled ranges. "Anxiety 9, mild" when 9 was the undoubled sum (18 doubled, severe), or a doubled score charted with no indication of the metric. Chart raw and doubled together (raw 9/21, doubled 18/42) and label the range from the doubled value; a fully completed adult DASS-21 doubles to an even number, so odd or decimal scores need a prorating note. A single "DASS score" with a severity label. "DASS-21 total 52, moderate" has no official interpretive basis: there is no five-band table for a raw grand total, and the subscale ranges do not apply to a sum. Report three subscales; if a composite is used, name the method (standardized z scores averaged) and the norm set. Labels charted as diagnoses or urgency triggers. "DASS confirms severe anxiety disorder," "mild depression" read as a mild disorder, or extremely severe Stress treated as an emergency on its own. The rights holder calls the labels arbitrary population-relative conventions; chart the range and interpret with the interview, impairment, context, and differential. Discordance collapsed or misread. Three scores averaged into one impression, or Stress reported as a measure of life stressors or burnout. Stress is tension, irritability, and difficulty relaxing over the week; an extremely severe Stress score beside a normal Depression score is the profile, and it is why the three scores exist. Risk inferred from the Depression score. "Depression 6, normal, no risk concerns" with no direct inquiry, or the DASS-21 entered as the service's suicide screen. The instrument has no suicide item by design; a low score documents nothing about ideation and a high score nothing about intent. Assess directly and document separately. Version, language, timeframe, and metric drift. DASS-Y results doubled or trended with adult scores, a translation charted as plain "DASS-21," a changed timeframe, a repeat on different conditions, or "DASS improved from 66 to 52" with no subscale detail. Name the version, language, timeframe, and metric every time and state change per subscale. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version and language, timeframe, mode, raw subscale sums or the completed form, prior scores and dates, context, risk assessment findings) and it drafts the documentation block: doubled scores with their ranges, change per subscale, a descriptive interpretation, and the separate risk line, ready for your review. Cross-check a finished note for the gaps reviewers flag: raw sums read against the doubled ranges, an unexplained total, a label charted as a diagnosis, discordance flattened, or a risk conclusion drawn from the Depression score. Draft the Better Access report back to the referrer: intake and current scores for all three subscales, change, function, the risk conclusion, and the recommendation, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do DASS-21 scores mean, and do I really multiply by two? Yes, for the adult DASS-21. Each of the three subscales (Depression, Anxiety, Stress) sums seven items rated 0 to 3 for the past week, giving a raw 0 to 21; multiply each subscale by two to reach the 0 to 42 metric on which the manual's norms and the conventional severity ranges are defined. On that doubled metric, Depression runs 0 to 9 normal, 10 to 13 mild, 14 to 20 moderate, 21 to 27 severe, and 28 and above extremely severe; Anxiety 0 to 7, 8 to 9, 10 to 14, 15 to 19, and 20 and above; Stress 0 to 14, 15 to 18, 19 to 25, 26 to 33, and 34 and above. Chart both metrics (raw 8/21, doubled 16/42, moderate) so the next reader knows which they are looking at, and remember that a fully completed, simply doubled adult DASS-21 gives even numbers. Do not multiply a grand total, and never multiply the DASS-Y, which is scored on its own metric. Is there a total DASS-21 score? Not in ordinary clinical scoring. The three subscales are reported and interpreted separately, and there is no official severity table for a raw grand total, so "DASS score 48, severe" cannot be decoded (one subscale? an undoubled 21-item sum? a sum of doubled subscales?). The rights holder's FAQ does permit a composite measure of negative emotional symptoms, mainly for research, and recommends converting each scale to a norm-referenced z score and averaging the standardized values; if you use one, say so, name the method and the norm set, and never apply the subscale ranges to it. Collapsing the three scores also throws away the discordance (severe Stress with normal Depression, say) that is the instrument's advantage over a single-number distress measure. Are the severity labels diagnoses? No. The labels mark positions in a population distribution, at percentile cut points of about the 78th (normal to mild), 87th (mild to moderate), 95th (moderate to severe), and 98th (severe to extremely severe), and the rights holder's own FAQ says the selection of any single clinical threshold is necessarily arbitrary and that "mild" means above the population mean while still well below the level typical of people seeking treatment, not a mild disorder. Extremely severe Stress is not a stress disorder, a burnout diagnosis, or an emergency by itself; normal Depression does not exclude a depressive disorder or any risk. Chart the range and interpret it with the interview, duration, impairment, context, and differential, in language like "score falls in the conventional extremely severe range; interpret with the clinical assessment," never "DASS confirms severe anxiety disorder." Does Better Access, or any payer, require the DASS-21? No authority requires it by name. In Australia, Better Access guidance (March 2026) says a mental health treatment plan must include an outcome measurement tool unless clinically inappropriate, with the choice at the GP's or prescribed medical practitioner's discretion, and names the K10 and DASS-21 as examples; the K10 remains the incumbent for national reporting, so services often keep it for comparability and add the DASS-21 for the profile. Current plan-preparation items are 2700, 2701, 2715, 2717, 272, 276, 281, and 282 with video equivalents; the dedicated review items (2712 and its equivalents) were removed on 1 November 2025 and reviews now use general attendance items, and neither the plan rules nor the reviews set a DASS administration schedule. In the United States there is no mandate; brief standardized assessment is reported per instrument under 96127, with commercial caps commonly at four units a day and Medicare at three, and three subscales do not automatically make three units. In Canada, no federal or provincial payer names it; verify each program's own form and cadence (WSIB, for example, requires its progress form after every sixth session or eighth week without naming a tool). See the Better Access treatment plan page for the plan and referral rules. Can we put the DASS-21 in our EHR, on our website, or in an app? EHR yes, public website or app no. The rights holder describes the DASS as public domain: the forms may be downloaded, printed, and copied without restriction, and computerized administration is permitted, provided the scales are not modified or sold and the results go to the clinician or researcher rather than being interpreted automatically for the respondent, which the FAQ calls misleading and potentially dangerous. A website or app intended for or open to members of the public (the FAQ's own example is a clinical practice website) is not permitted, whether or not it is free; the allowed digital model is a restricted, defined group such as enrolled patients or research participants with results routed to the clinician. A clinician may charge for an assessment that includes the DASS; the scales themselves may not be sold, and translations must remain public domain and cannot be copyrighted or sold. Terms are as stated on the UNSW DASS site in August 2026; cite the 1995 manual. My client is extremely severe on Stress and normal on the other two. What does that mean? It means the client endorsed a high level of tension, irritability, over-reactivity, and difficulty relaxing during the measurement week relative to the conventional population ranges, and it is a legitimate, informative profile rather than an error. It does not by itself establish a stress disorder, generalized anxiety disorder, burnout, malingering, or a need for emergency care. Assess the precipitating events, duration, sleep, pain, substances, medical contributors, agitation, trauma-related symptoms, and functional impairment, and let the interview carry the diagnostic weight; the modern factor literature also finds a strong general-distress component under all three scales, so read small subscale gaps modestly. Preserve the three scores in the chart; a discordant profile is exactly the case for reporting them separately. What do I do about the missing suicide item? Treat the DASS-21 as unable to rule suicide or self-harm risk in or out. The scales contain no suicide item because candidate items did not load on any of the three scales during development, and the rights holder's guidance says experienced clinicians must assess risk separately in seriously disturbed patients. So: a low Depression score documents nothing about ideation, an extremely severe score documents no intent, and the DASS-21 cannot satisfy a suicide-screening workflow. Run the service's direct risk assessment whenever it is clinically indicated, and record in the DASS entry that risk was assessed separately, with the findings, formulation, and plan in a suicide risk assessment or the risk section of the record. If your workflow needs a self-harm item in the questionnaire itself, the PHQ-9 has one, with its own follow-up rules. DASS-21, K10, PHQ-9 and GAD-7, DASS-42, or DASS-Y: which one, when? Choose the DASS-21 when the three-way profile matters, especially to isolate the tension and over-arousal domain that a single-number measure cannot show, and when past-week monitoring fits the cadence. Choose the K10 when one global distress score over four weeks, lower burden, and compatibility with Australian national and service reporting are the priorities; running both needs a defined purpose. Choose the PHQ-9 and GAD-7 when disorder-anchored measures with validated cutoffs, a self-harm item, and US and Canadian payer familiarity matter. Choose the DASS-42 when more reliable subscales and more symptom detail justify twice the length; the rights holder describes the full form as often preferable for clinical work. Use the DASS-Y for children and adolescents (ages 8 to 17 on the DASS site; the adult form is acceptable from 14 with adequate language), scored on its own metric, never doubled, and never trended against adult scores. Can BastionGPT draft my DASS-21 documentation? Yes. Give it the facts (version and language, timeframe, mode, raw subscale sums or the completed form, prior scores and dates, context, and the risk assessment findings) and it drafts the full entry: doubled scores with their ranges, change per subscale, a descriptive interpretation, the separate risk line, and the report back to the referrer, ready for your review. It can also check a finished note for raw sums read against the doubled ranges, an unexplained total, a label charted as a diagnosis, discordance flattened, and a risk conclusion drawn from the Depression score. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Lovibond SH, Lovibond PF, 1995, Manual for the Depression Anxiety Stress Scales, 2nd ed., Psychology Foundation, Sydney, as cited on the DASS publications page ; Psychology Foundation of Australia, UNSW, DASS home (site updated January 2025), FAQ (public domain; computerized administration; no public website or app; not for sale; composite by standardized z scores; missing-item rule of thumb; translations; DASS-42 first-page correction factors), download terms (copy without restriction; do not modify or sell), overview (no suicide item; assess risk separately), DASS-Y (ages 8 to 17; not doubled). Australian Government Department of Health, Disability and Ageing, Comorbidity Guidelines, appendix K (the doubling instruction and conventional severity ranges). Crawford JR, Henry JD, 2003, British Journal of Clinical Psychology, the DASS in a non-clinical sample (percentile basis of the severity labels; student-weighted 1995 norms); Henry JD, Crawford JR, 2005, the short-form DASS-21 in a large non-clinical sample (1,794 UK adults; alphas; general-distress factor); Crawford J and colleagues, 2011, Australian Psychologist, Australian percentile norms . Osman A and colleagues, 2012, further examination of DASS-21 dimensions ; Lee J and colleagues, 2019, COSMIN systematic review of DASS-21 measurement properties (48 studies; responsiveness evidence low or very low); Zanon C and colleagues, 2021, Assessment, eight-country invariance study , and 2025, nine-country replication ; Szabo M, Lovibond PF, 2022, Frontiers in Psychology, DASS-Y development . Australian Government Department of Health, Disability and Ageing, Better Access fact sheet for health professionals (March 2026; outcome measurement tool unless clinically inappropriate; K10 and DASS-21 as examples); Services Australia, MBS billing rules for mental health services (June 2026; current plan items; review timing; bare item numbers only); MBS Online, Better Access arrangements for GPs and prescribed medical practitioners (review and ongoing-management items removed 1 November 2025). Kessler RC and colleagues, 2002, Psychological Medicine, short screening scales for non-specific psychological distress (the K10); headspace, program evaluation (K10 as the principal outcome measure); AIHW, consumer outcomes in mental health services . American Medical Association, behavioral health coding guide (March 2025; 96127 as a per-instrument code; bare numbers only); CMS, CY 2027 physician fee schedule proposed rule . Ontario Health, anxiety disorders quality standard (validated tools and regular measurement; no instrument prescribed); Hamilton Family Health Team, DASS-21 scoring guide (Canadian primary-care use). Kroenke K, Spitzer RL, Williams JB, 2001, the PHQ-9 (the disorder-anchored comparator with a self-harm item). ## Related templates PHQ-9 Score Documentation GAD-7 Score Documentation Better Access Mental Health Treatment Plan Referral Outcome Measure Review Note Suicide Risk Assessment Treatment Plan Review & Update Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/vanderbilt All note templates # Vanderbilt Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The NICHQ Vanderbilt Assessment Scales are parent- and teacher-completed ADHD rating scales published by NICHQ with the American Academy of Pediatrics in 2002, with initial and follow-up forms and a third edition in 2019. Pediatric and primary care clinicians use them to gather cross-setting symptom, impairment, and follow-up data. They support a diagnosis; they never make one. This page covers how to document Vanderbilt results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Parents or caregivers and teachers complete the forms; the clinician scores and interprets; no qualification level, but the third-edition toolkit is licensed by the AAP Audience Pediatric and family-medicine clinicians, child psychiatry and psychology, school teams and 504 or IEP coordinators, payers and quality reviewers Typical length 4 to 10 chart lines per informant set · each form takes an informant about 10 minutes Format family Informant-rated ADHD scale (18 DSM-aligned symptom items scored 0 to 3, performance items scored 1 to 5, associated-condition screens; parent and teacher, initial and follow-up) When it's used Pediatric ADHD evaluation (validated at ages 6 to 12), cross-setting collateral, associated-condition screening, medication titration and side-effect follow-up Standards context First edition (2002) free from NICHQ with attribution; third-edition toolkit (2019) licensed by the AAP and EHR integration needs an institutional license; no authority mandates it On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Vanderbilt? The NICHQ Vanderbilt Assessment Scales are informant-completed ADHD rating scales: a parent initial form, a teacher initial form, and parent and teacher follow-up forms that add a medication side-effect section. NICHQ published the first edition with the American Academy of Pediatrics in 2002, adapted from the Vanderbilt rating scales developed by Mark Wolraich and colleagues, so the accurate attribution is NICHQ and the AAP, not Vanderbilt University. A second edition followed in 2011 and the AAP's third-edition toolkit in 2019 (scoring instructions copyright 2020, DSM-5 aligned), with a fourth edition in development and no release date as of August 2026. Each form rates the 18 DSM-aligned ADHD symptom behaviors on a four-level frequency scale scored 0 to 3, rates performance in academic and social domains on a five-level scale, and screens for associated conditions (oppositional, conduct, and anxiety or depression symptoms on the parent form; a combined oppositional-conduct section, an anxiety or depression section, and an academic-performance flag on the teacher form). Two facts govern the documentation. First, the Vanderbilt's own scoring is a two-part rule, not a total: an ADHD domain is scoring-positive when at least six of the nine behaviors in that domain are rated 2 or 3 and the performance-impairment criterion is met, which on the current edition means at least two performance ratings of 4 or one of 5 (the 2002 edition accepted any single 4 or 5). A symptom count without impairment is not a positive Vanderbilt result, and there is no official grand-total cutoff that establishes ADHD, even though raw sums and follow-up means exist. Second, the scale supports the diagnosis and never makes it: the AAP guideline requires DSM-5 criteria with symptoms and impairment in more than one major setting, established through parent, teacher or other school-personnel, and clinician information, and the community validation studies put the parent scale's positive predictive value at .19 and the teacher scale's at .32 against negative predictive values of .98 and .96, so a negative form excludes well and a positive form is a signal to evaluate. The full workup belongs in an ADHD evaluation report ; the norm-referenced Conners 4 is the licensed comparator, and the ASRS and CAARS 2 take over at the adult boundary. ## Who uses Vanderbilt documentation and when Pediatric and family-medicine practices are the core users, because the AAP's 2019 guideline expects a clinician to gather symptom and impairment information from more than one major setting before diagnosing ADHD and the Vanderbilt was built for exactly that collateral workflow: parent and teacher forms at evaluation, associated-condition screens alongside, and the follow-up forms for titration and side-effect monitoring once treatment starts. School teams see the results in school collateral notes and in the input that feeds a Section 504 statement or IEP input , where the form is one source among several and never the eligibility decision. In Canada, CADDRA's child pathway defaults to SNAP-IV parent and teacher forms with the Vanderbilt named as an acceptable substitute; in Australia the AADPA guideline (updated January 2026) lists it among example child instruments while insisting no diagnosis rests on scales alone. Direct validation is at ages 6 to 12; the AAP permits DSM-anchored use in preschoolers and adolescents without norms, and no Vanderbilt-specific preschool or adolescent diagnostic-accuracy study had been published as of August 2026, so those charts say so. Adult screening moves to the ASRS , and any diagnostic conclusion moves to the ADHD evaluation report . ## How to document Vanderbilt results in the chart No US, Canadian, or Australian authority prescribes a Vanderbilt note format, and none makes a score diagnostic. What survives review is a record that names the edition and form, the informant and setting for each form, both symptom-domain counts, whether the edition-specific impairment criterion was met, each associated-condition screen as a screen, an honest cross-setting synthesis, and, on follow-up, means against baseline with function and side effects. Each element below carries the pitfall that most often undermines it. Edition and form. Name the edition or exact source (2002 first edition, 2011 second, 2019 third-edition toolkit) and whether each form is initial or follow-up. The editions differ in the impairment threshold, the length of the parent conduct section (14 versus 15 behaviors), and the follow-up arithmetic, so a reviewer cannot reproduce "Vanderbilt positive" without the edition. Pitfall: The 2002 impairment rule (any single performance rating of 4 or 5) applied to a current form, which manufactures false positives, or an indexed second-edition copy mistaken for the third because both use the two-4s-or-one-5 rule. Informant, setting, date, and medication status. For each form, record who completed it (relationship; teacher role, subject, and how long they have observed the child), the completion date, and whether the child was unmedicated or on a named dose during the observation period, and when the rater usually sees the child relative to dosing. Pitfall: "Parent and teacher forms positive" with no dates, no observation window, and no medication context, which makes concordance meaningless and follow-up uninterpretable. Domain counts, not a total. Chart inattentive x/9 and hyperactive-impulsive x/9 for each informant, counting behaviors rated 2 or 3 as symptom-positive; six or more of nine is the domain threshold, and both domains together is the combined pattern. Raw sums and means exist for research and follow-up, but no official grand-total cutoff establishes ADHD. Pitfall: "Total 42, diagnostic of ADHD," or one combined count that hides which domain carried it. The performance-impairment rule. Record the number of performance ratings of 4 and 5 and whether the edition-specific criterion was met (current edition: at least two 4s or one 5). A domain is scoring-positive only when the symptom count and the impairment criterion are both met; you may separately document impairment established by interview, but do not alter the instrument classification. Pitfall: Performance items treated as optional, a symptom count called "positive" without impairment, or a rating of 3 read as impairment (on the current scale, 4 and 5 mark a problem). Associated-condition screens as screens. Chart each screen with its numerator and denominator against the edition's threshold (for example, oppositional 4 of 8 on the parent form) plus the impairment finding, and write "screen positive" or "screen negative." The teacher form has one combined oppositional-conduct section and an academic-performance flag that points to records and achievement testing, not a diagnosis. Pitfall: "Vanderbilt diagnosed depression" or "conduct disorder per Vanderbilt": the 2012 validation found the parent comorbidity cutoffs had limited rule-in utility, and the oppositional cutoff was useful mainly for ruling out. Concordance and discordance, without averaging. State which domains and impairments each informant reports and whether they agree, then interpret disagreement as information about settings, observation opportunity, structure, medication timing, learning demands, or family stress. When a teacher form is missing, document the attempts, the alternate school source used (another teacher, counselor, records, behavior plans), and whether the more-than-one-setting criterion is established, uncertain, or not met. Pitfall: Parent and teacher counts averaged into an artificial child, one rater dismissed as unreliable, or a diagnosis finalized on a parent form alone with no second-setting evidence. Follow-up: means, function, side effects. On the current follow-up forms, calculate the mean rating for the nine inattentive and nine hyperactive-impulsive behaviors separately and compare each with its own baseline mean; record performance change and the side-effect section, with the medication, dose, dosing time, and observation interval, and the resulting decision. Pitfall: Follow-up scored by counting positives, baseline and follow-up on different informants or conditions, or side effects charted nowhere while the dose goes up. ### Blank template (copy and adapt) VANDERBILT DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Child age: [ ] Edition / source: [2002 first / 2011 second / 2019 third-edition toolkit] Form A: [parent initial / follow-up] Informant: [ ] Completed: [date] Medication status during observation: [none / drug + dose + timing] Inattentive [ ]/9 Hyperactive-impulsive [ ]/9 (rated 2 or 3 = positive; 6+ of 9 = domain threshold) Performance: ratings of 4 [ ] of 5 [ ] -> impairment criterion [met / not met] (current edition: two 4s or one 5) Screens: oppositional [ ]/8 conduct [ ]/15 anxiety-depression [ ]/7 -> [screen positive / negative], not diagnoses Form B: [teacher initial / follow-up] Informant + role + observation period: [ ] Completed: [date] Medication status: [ ] Inattentive [ ]/9 Hyperactive-impulsive [ ]/9 Performance: [ ] Combined oppositional-conduct [ ]/10 anxiety-depression [ ]/7 Academic-performance flag: [ ] Cross-setting synthesis: [concordant / discordant: pattern by setting; missing form: attempts + alternate school source; more-than-one- setting criterion established / uncertain / not met] Interpretation: [Vanderbilt supports, does not establish, the diagnosis; integrated with history, onset, duration, differential, records] Follow-up (if applicable): baseline means I [ ] HI [ ] -> current I [ ] HI [ ]; performance change; side effects; drug/dose/timing; decision Plan: [evaluation steps / school supports / treatment / next forms + date] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale items, forms, or scoring sheets. ## Sample Vanderbilt documentation (fictional) Scenario: concordant parent and teacher third-edition initial forms reviewed at the second visit of a pediatric ADHD evaluation, with impairment met in both settings, negative associated-condition screens, and the diagnosis made on the full evaluation rather than the forms. All details are fictional. Patient: D.K., 8 · Setting: Pediatric primary care, ADHD evaluation visit 2 · Clinician: A. Mehta, MD · Note date: 08/13/2026 Forms reviewed: NICHQ Vanderbilt initial scales, AAP toolkit third edition, no ADHD medication during either observation period. Parent initial form completed by mother 08/05/2026: inattentive 7/9, hyperactive-impulsive 6/9; performance ratings include three 4s and no 5s, impairment criterion met. Associated-condition screens: oppositional 2/8, conduct 0/15, anxiety-depression 1/7, all screen negative. Teacher initial form completed 06/2026 by the third-grade homeroom teacher over the final term of last school year (about nine months of daily observation), returned through the school office 08/07/2026: inattentive 8/9, hyperactive-impulsive 7/9; performance ratings include one 5 and one 4, impairment criterion met; combined oppositional-conduct 1/10, anxiety-depression 0/7, academic-performance flag not raised. Synthesis: Parent and teacher reports are concordant for clinically significant inattentive and hyperactive-impulsive symptoms with impairment at home and school, meeting the current symptom-count and impairment rules for a combined-domain Vanderbilt result in both settings. Vanderbilt results support, but do not independently establish, the diagnosis. Integrated with the developmental and medical history from visit 1 (symptoms noted by kindergarten, persistent, in excess of developmental level), report cards and the second-grade behavior plan, today's interview, normal hearing and vision screens, no sleep-disorder symptoms, no learning-disorder signal on records or teacher flag, and negative anxiety and mood screens, findings meet criteria for ADHD, combined presentation. No better explanation identified. Discussion and plan: Diagnosis, options, and expected course reviewed with the mother and D.K. in age-appropriate terms. Plan by shared decision: behavioral parent-training referral placed; letter to school requesting a Section 504 evaluation with the parent's consent, enclosing a summary (school eligibility is the school's determination); stimulant trial discussed with cardiovascular history reviewed and no contraindication, family to decide at a phone check-in within one week; if medication starts, parent and teacher follow-up forms in four weeks under the same informants and observation conditions, first prescriber follow-up within 30 days, side effects reviewed at each contact. Administrative: Both forms scanned to the chart with completion dates and informant details; domain counts, impairment findings, and screen results entered in the structured ADHD flowsheet with the edition named. Two standardized informant instruments scored and interpreted this encounter, reported under 96127 per payer rules; no diagnosis coded from the forms alone. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The edition, each form, each informant, the observation period, and the medication status are named, so the two-part scoring rule can be reproduced and the follow-up forms will have a true baseline. Both domain counts and the impairment finding are charted per informant, and the associated-condition screens are reported as screens with their denominators. Concordance is stated by domain and setting, and the diagnosis rests on the integrated evaluation (history, onset, records, interview, differential), with the Vanderbilt explicitly supporting rather than establishing it. The school step is framed correctly: a request for the school's own evaluation, not a claim that the forms decide eligibility. The loop closes: a dated follow-up-form plan on the same informants and conditions, the 30-day prescriber follow-up if medication starts, side effects owned at every contact, and structured-field entry a reviewer can find. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: guideline convention, education law, and payer measures pull in different directions, and none names the Vanderbilt. The AAP's 2019 clinical practice guideline (CONVENTION) requires that DSM-5 criteria be met with documentation of symptoms and impairment in more than one major setting, drawing on parents or guardians, teachers, other school personnel, and involved mental health clinicians, and that alternative causes be ruled out; it requires a process to screen for associated conditions; and it does not name an instrument or require two positive forms, so "two informants" is shorthand for cross-setting evidence, which another credible school source can supply when a teacher form never returns. Under IDEA, 34 CFR 300.304 (LAW) requires a variety of assessment tools and forbids a single measure as the sole criterion for eligibility, and Section 504 evaluation duties (LAW, with 2016 federal guidance) turn on suspected disability and functional limitation, not on crossing a Vanderbilt cutoff, so a medical diagnosis and a positive form together do not produce an IEP or 504 plan; the school team decides on multiple sources. Stimulant prescribing carries state controlled-substance and PDMP obligations (LAW, varying by state) that attach to the prescription, not to a score, and no PDMP rule reviewed makes a Vanderbilt a prerequisite; defensible prescribing notes still carry the diagnostic basis, targets, cardiovascular and medical history, response, function, and adverse effects, which the follow-up forms document efficiently. AAP coding guidance maps an informant-completed Vanderbilt to 96127 rather than 96110 (CONVENTION), while units, bundling, and modifiers are PAYER POLICY. The HEDIS follow-up measure for children aged 6 to 12 newly prescribed ADHD medication (PAYER POLICY) counts a prescriber follow-up within 30 days of the first dispensing and, for the continuation phase, at least 210 treatment days with two additional follow-ups within 270 days after the initiation phase; it does not require a Vanderbilt result, so the record must support the dispensing chronology and the dated encounters, and the forms give those encounters their clinical content. Canada and Australia name other defaults and stricter framing. CADDRA's current child flowchart (2025) places parent and teacher SNAP-IV forms, the CADDRA teacher assessment form, and a functional-impairment measure in its default workflow and expressly allows the Vanderbilt, Conners, or the Strengths and Difficulties Questionnaire instead (CONVENTION); its practice guidelines (4.1, 2021) call rating scales efficient but insufficient for diagnosis, because anxiety, depression, sleep apnea, anemia, and other conditions produce positive responses. Australia's AADPA guideline (2022, updated January 2026, NHMRC approval stated through 29 July 2027) says ADHD should not be diagnosed solely from rating scales or observation, requires information from more than one setting and reporter for children, and lists the Vanderbilt among example instruments (CONVENTION); the stimulant-authority layer is state law that names no scale and has been changing quickly through 2025 and 2026 (endorsed or authorized GP pathways in New South Wales, South Australia, Tasmania, the ACT, and Victoria; Queensland's post-December 2025 framework; monitored-medicines checks such as SafeScript, QScript, ScriptCheckWA, and Canberra Script), so an Australian note separates the clinical diagnostic basis from the authority, approval, and monitored-medicines requirements. On the evidence: the parent scale in a community sample (2013) showed sensitivity .80, specificity .75, positive predictive value .19, and negative predictive value .98, and the teacher scale .69, .84, .32, and .96, which is why the authors themselves read the numbers as an argument for multiple observers; a 2015 study of 6,659 children found parents rate higher than teachers, with weak agreement on inattention and stronger agreement on hyperactivity-impulsivity, so discordance is expected, not error; the 2022 nationally representative parent norms (1,570 caregivers, ages 5 to 12) were built on the older 55-entry form and did not create demographic cutoffs; a 2012 study found the parent comorbidity cutoffs had limited rule-in utility; and a 2024 AAP-AHRQ review of 231 studies found tools perform worse when separating ADHD from other referred conditions than from community controls. Direct validation is at ages 6 to 12: no Vanderbilt-specific preschool diagnostic-accuracy study was found, and the registered adolescent-modification study had no posted or published results as of August 2026. Rights are more layered than "the Vanderbilt is free." NICHQ offers the 2002 first edition for download without charge and asks publications that use it to reference NICHQ; the PDF carries the 2002 AAP and NICHQ copyright and no license authorizing re-typesetting, modification, removal of notices, or unrestricted redistribution. The 2019 third-edition toolkit is sold and licensed by the AAP: its product page states that the single-user license does not support EHR integration and that an institutional license is required for multi-user access or EMR and EHR integration (live August 2026), AAP permissions require material to be reproduced exactly with changes needing written consent, Spanish forms ship in the toolkit, and no NICHQ or AAP source publishes a carve-out for a public website that reproduces the items and scores responses, whether or not it is free; a calculator that accepts already-derived counts without displaying item wording is lower-risk, not AAP-approved. The NICHQ Vanderbilt Assessment Scales are published by NICHQ with the American Academy of Pediatrics, and the third-edition toolkit is copyright American Academy of Pediatrics. BastionGPT is not affiliated with, or endorsed by, NICHQ, the American Academy of Pediatrics, or the scales' authors. This page reproduces no items, forms, or scoring sheets and describes the scoring architecture in original prose. ↑ Back to the template and downloads ## Common Vanderbilt documentation errors reviewers flag The numbers behind these errors are specific. In the community validation studies the parent scale's positive predictive value was .19 and the teacher scale's .32, against negative predictive values of .98 and .96; a study of 6,659 children found parent-teacher agreement weak for inattention; the 2002 first edition accepted any single performance rating of 4 or 5 as impairment while the current edition requires two 4s or one 5; and the AAP's third-edition license excludes EHR integration without an institutional license. The BastionGPT Clinical Advisory Board sees the same errors most often in Vanderbilt documentation reviews: A positive form charted as the diagnosis. "Vanderbilt positive, ADHD combined" from a form alone. Chart the domain counts, the impairment finding, and the statement that the scale supports the diagnosis, then the history, onset, cross-setting evidence, differential, and records that make it. Community positive predictive values of .19 and .32 are the reason. Symptom counts without the impairment rule. A 7/9 inattentive count called positive with performance items skipped, a rating of 3 read as impairment, or the 2002 any-single-4-or-5 rule applied to a current form. Record the number of 4s and 5s and whether the edition's criterion was met; a count without impairment is not a positive Vanderbilt result. The edition left unnamed. "Vanderbilt" with no edition, form, or source, when the 2002, 2011, and 2019 versions differ in impairment threshold, parent conduct-section length, and follow-up arithmetic. Name it every time; a two-4s-or-one-5 rule alone does not prove a third-edition form. Informants averaged, dismissed, or missing without a trail. Parent and teacher counts averaged, a discordant teacher called unreliable, or a diagnosis finalized on a parent form with no attempts, no alternate school source, and no statement of whether the more-than-one-setting criterion is met. Preserve each informant's data and document the trail. Screens charted as diagnoses. "Vanderbilt diagnosed anxiety" or "conduct disorder per teacher form." The associated-condition sections are screens with limited rule-in utility; chart numerator, denominator, impairment, and "screen positive," then the targeted assessment, risk inquiry where indicated, and plan. Follow-up scored like an intake. Follow-up forms counted for positives, compared across different informants or medication conditions, or filed with no side-effect entry while the dose changes. Follow-up uses the mean per domain against its own baseline, the same raters under comparable conditions, and a side-effect line at every step, with the drug, dose, and dosing time recorded. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (edition, each form's informant, dates, and medication status, the domain counts and performance ratings, the screen results, and the history) and it drafts the documentation block: counts and impairment per informant, screens as screens, the cross-setting synthesis, the supports-not-establishes line, and the plan, ready for your review. Cross-check a finished note for the gaps reviewers flag: a form charted as the diagnosis, a symptom count without the impairment rule, an unnamed edition, informants averaged or a missing form with no trail, or a screen written up as a diagnosis. Draft the follow-up documentation: baseline and current domain means, performance change, side effects with drug, dose, and timing, and the titration decision, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Vanderbilt scored? As a two-part rule per informant, not a total. On the initial forms, each of the 18 DSM-aligned ADHD behaviors is rated 0 to 3, a rating of 2 or 3 counts as symptom-positive, and six or more of the nine behaviors in a domain meets the inattentive or hyperactive-impulsive threshold, with both domains together giving the combined pattern. Separately, the performance items rate functioning on a five-level scale, and the impairment criterion on the current edition is at least two ratings of 4 or one rating of 5 (the 2002 first edition accepted any single 4 or 5). A domain is scoring-positive only when both parts are met; a symptom count without impairment is not a positive Vanderbilt result. The associated-condition sections have their own thresholds and are screens. Raw sums and means exist for research and follow-up, but no official grand-total cutoff establishes ADHD, so "total 42, diagnostic" is a made-up rule. Which edition am I using, and does it matter? It matters. The 2002 first edition (free from NICHQ) used any single performance rating of 4 or 5 as impairment, a 14-behavior parent conduct section, and a single combined symptom sum on its follow-up forms. The 2011 second edition adopted the two-4s-or-one-5 impairment rule but kept the 14-behavior conduct section. The 2019 third-edition AAP toolkit (scoring instructions copyright 2020) is DSM-5 aligned, has a 15-behavior parent conduct screen, uses two 4s or one 5, and tracks follow-up as separate inattentive and hyperactive-impulsive means; a fourth edition is in development. Freely circulating PDFs on hospital and university sites are mostly first- and second-edition copies. Name the edition or exact source in the chart, and never apply the 2002 impairment rule to a current form; the impairment rule alone cannot tell you which edition you hold, because the second and third share it. Do I need both a parent and a teacher form to diagnose ADHD? You need evidence of symptoms and impairment in more than one major setting; you do not need two positive Vanderbilt forms. The AAP guideline draws that evidence from parents or guardians, teachers, other school personnel, and involved mental health clinicians, and it names no instrument, so a missing teacher form does not automatically block a diagnosis when reliable school-setting information arrives another way (a second teacher, counselor, or school psychologist; report cards, behavior plans, disciplinary or psychoeducational records; direct school communication). Document the attempts and dates, the alternate source, and whether the more-than-one-setting criterion is established, uncertain, or not met. When only a parent form exists and no second-setting evidence does, chart a positive parent screen and a pending evaluation, not a completed diagnosis. Return rates are a structural problem: a 195-child study of parent-return agreements found teacher forms came back 71% of the time before and 76% after, no significant change. The parent form is strongly positive and the teacher form is negative. What now? Do not choose one by default and do not average them. Preserve each informant's counts and impairment findings, describe the setting-specific pattern, and treat the disagreement as information: parents rate higher than teachers on average, agreement is weak for inattention and stronger for hyperactivity-impulsivity across 6,659 children, and different observers validly see different contexts. Then investigate: classroom structure and size, academic level and homework conflict, sleep, anxiety, learning, family stress, medication timing if applicable, observation opportunity, another teacher or activity setting, and school records. If credible information shows impairment only at home, the more-than-one-setting criterion is not yet established; the home concern may still warrant intervention and differential assessment, and the chart says exactly that. Is the Vanderbilt free? Can we put it in our EHR or on our website? The 2002 first edition is free to download from NICHQ, which asks for a reference to NICHQ in publications and grants no license to re-typeset, modify, remove notices, or redistribute without limit; the PDF carries the 2002 AAP and NICHQ copyright. The 2019 third-edition toolkit is sold and licensed by the AAP, and its product page states that the single-user license does not support EHR integration and that an institutional license is required for multi-user access or EMR and EHR integration; AAP permissions require exact reproduction with written consent for changes, and Spanish forms are part of the licensed toolkit. No NICHQ or AAP source publishes a carve-out for a free public website that reproduces the items and scores responses, so a tool that displays item wording or rebuilds the form should obtain written permission; a calculator that only accepts already-derived counts is lower-risk, not approved. Terms as stated on the NICHQ and AAP pages in August 2026; confirm before building anything. How do I chart a positive anxiety, depression, oppositional, or conduct screen? As a screen. Record the exact numerator and denominator against the edition's threshold (the parent form separates oppositional and conduct sections and has an anxiety-depression section; the teacher form has one combined oppositional-conduct section, an anxiety-depression section, and an academic-performance flag), the impairment finding, and the words "screen positive." Then document the targeted history, severity, duration, settings, a risk inquiry when clinically indicated, collateral, the differential, and the plan for assessment, treatment, or referral. The 2012 validation study found the parent comorbidity cutoffs generally lacked adequate utility for confirming or excluding structured-interview diagnoses, with the oppositional cutoff useful mainly for ruling out oppositional defiant disorder, which is why "Vanderbilt diagnosed depression" is wrong twice over. The teacher's academic flag points to records and achievement testing, not to a learning-disorder diagnosis. How do I document follow-up on medication with the Vanderbilt? With the follow-up forms, scored differently from intake. On the current edition, calculate the mean rating for the nine inattentive and the nine hyperactive-impulsive behaviors separately and compare each with its own baseline mean; record performance change (how many 4s and 5s remain), the side-effect section, the medication, dose, dosing time, and the rater's observation interval relative to dosing, and the decision (continue, adjust, change). Use the same informants under comparable conditions, and repeat when results can inform titration, after meaningful dose changes, and periodically in maintenance; no official source fixes an interval. HEDIS timing (a prescriber follow-up within 30 days of the first dispensing for children 6 to 12, then two more visits in the continuation phase) is an encounter requirement, not a form requirement, and PDMP checks attach to the prescription under state law. Charting side effects nowhere while the dose rises is the follow-up error reviewers see most. Vanderbilt, Conners 4, SNAP-IV, or ASRS: which one, when? Choose by age, question, and setting. The Vanderbilt suits pediatric primary care and school-collateral workflows at ages 6 to 12 when DSM-aligned domain counts, an impairment rule, associated-condition screens, and follow-up with side-effect tracking are what you need; it has no response-style index and its current edition is licensed, not simply free. The Conners 4 is a norm-referenced, licensed instrument for ages 6 to 18 with parent, teacher, and self-report forms, broader content, and response-style indicators, interpreted on its own norms and never on Vanderbilt thresholds. The SNAP-IV is the free scale CADDRA names as its default parent and teacher form in Canada. The ASRS and CAARS 2 are adult instruments, not pediatric substitutes. None of them is an evaluation: the diagnostic formulation belongs in the ADHD evaluation report , and a positive scale of any kind is a reason to do one. Can BastionGPT draft my Vanderbilt documentation? Yes. Give it the facts (edition, each form's informant, completion date, and medication status, the domain counts and performance ratings, the screen results, the history and records, and the plan) and it drafts the full entry: counts and impairment per informant, screens as screens, the cross-setting synthesis, the supports-not-establishes statement, and the plan with dates and owners, ready for your review. It can also check a finished note for a form charted as the diagnosis, a count without the impairment rule, an unnamed edition, averaged informants or a missing form with no trail, and follow-up scored like an intake. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: NICHQ, NICHQ Vanderbilt Assessment Scales (first-edition download and attribution request; edition history) and the 2002 scoring instructions ; American Academy of Pediatrics, Caring for Children With ADHD toolkit, third edition (2019; licensing terms including the EHR-integration statements), toolkit contents , and third-edition scoring instructions (copyright 2020; two-part rule; follow-up means; validation ages). Wolraich ML and colleagues, 2019, Pediatrics, AAP clinical practice guideline for ADHD (key action statements 1 to 3; more than one major setting; sources of information). Wolraich ML and colleagues, 2003, Journal of Pediatric Psychology, parent scale in a referred population ; Bard DE and colleagues, 2013, parent scale in a community population (sensitivity .80, specificity .75, PPV .19, NPV .98); Wolraich ML and colleagues, 2013, teacher scale in a community population (.69, .84, .32, .96); Anderson NP and colleagues, 2022, national parent norms (1,570 caregivers, ages 5 to 12; older 55-entry form). Narad ME and colleagues, 2015, parent-teacher agreement across development (6,659 children); Becker SP and colleagues, 2012, comorbidity screen utility ; Langberg JM and colleagues, 2010, learning-problem screening ; DuPaul GJ and colleagues, 2020, differential item functioning of ADHD symptom ratings (ADHD Rating Scale-5, not the Vanderbilt); AAP-AHRQ, 2024, systematic review of ADHD diagnosis and treatment (231 studies); Sawaya and colleagues, 2024, incremental validity review . eCFR, 34 CFR 300.304 (IDEA evaluation procedures; no single measure); US Department of Education, OCR, 2016, Section 504 and students with ADHD ; PDMP TTAC, state PDMP legislation ; AAP News, coding for informant-completed instruments (96127 versus 96110; bare codes); NCQA, follow-up care for children prescribed ADHD medication ; eCQI Resource Center, 2026 eCQM specification . CADDRA, child flowchart (2025; SNAP-IV default, Vanderbilt as an alternative) and Canadian ADHD practice guidelines 4.1 (2021). AADPA, Australian clinical practice guideline for ADHD, updated January 2026 ; state stimulant-authority pages: NSW ACI , Victoria , Queensland , South Australia , Tasmania , ACT . Comparator instruments: MHS, Conners 4 manual overview ; Harvard Medical School, ASRS . ## Related templates ADHD Evaluation Report Conners 4 Report Write-Up ASRS v1.1 Score Documentation CAARS-2 Report Write-Up School & Teacher Collateral Note Medication Management Follow-Up Note PSC-17 Score Documentation T.O.V.A. Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/y-bocs All note templates # Y-BOCS Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Y-BOCS (Yale-Brown Obsessive Compulsive Scale) is a clinician-rated measure of OCD severity published by Goodman and colleagues in 1989, scored 0 to 40 across obsession and compulsion subtotals, with a 0 to 50 second edition and child versions. OCD specialists, therapists, and prescribers use it for baselines and treatment tracking. It rates severity; it never diagnoses. This page covers how to document Y-BOCS results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Clinician-administered semi-structured interview (psychologists, psychiatrists, therapists trained on the scale); self-report and parent-report adaptations exist and are charted as such; use requires the rights holder's permission Audience OCD specialty and ERP programs, prescribers, utilization reviewers and payers, neurosurgical and DBS programs, clinical trials Typical length 3 to 8 chart lines per administration · severity interview about 15 to 30 minutes, longer with the checklist at intake (clinical team estimate) Format family Clinician-rated OCD severity scale (10 items scored 0 to 4; obsession and compulsion subtotals 0 to 20; total 0 to 40; second edition 0 to 5 per item, total 0 to 50) When it's used OCD baseline severity, ERP and medication response tracking, level-of-care and medical-necessity documentation, neurosurgical referral, trial endpoints Standards context Licensed intellectual property of OCD Scales, LLC: paper use needs permission, EHR, app, and web use need a paid license; response and remission definitions are expert consensus, not law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Y-BOCS? The Yale-Brown Obsessive Compulsive Scale is a clinician-administered, semi-structured measure of current OCD symptom severity, published in two 1989 Archives of General Psychiatry papers by Wayne Goodman, Lawrence Price, Steven Rasmussen, and colleagues. It has three working parts: a symptom checklist that establishes which obsession and compulsion content is present, a target list that names the symptoms the interview will rate, and a severity scale of ten items scored 0 to 4, five for obsessions and five for compulsions, giving two 0 to 20 subtotals and a 0 to 40 total for the past week. The checklist is never added into the total; the scale was built so that people with different symptom content can be compared on aggregate burden. The Y-BOCS-II (Storch, Rasmussen, Price, and colleagues, 2010) widened each item to 0 to 5 for a 0 to 50 total to relieve the ceiling, replaced the resistance-to-obsessions item (confusing for patients taught in ERP not to resist) with an obsession-free interval item, and folded avoidance into severity; the CY-BOCS (1997) and CY-BOCS-II (2019) adapt the interview for children and adolescents with caregiver input. Two facts govern the documentation. First, the number is a severity rating, not a diagnosis, a level-of-care rule, or a remission declaration: no band establishes OCD, a low score after treatment does not undo an established diagnosis, and a high score on a self-report adaptation does not create one. Second, the scale is licensed intellectual property. As stated at ocdscales.com in August 2026, any use requires explicit permission, noncommercial paper use for a clinician's own patients requires prior permission from Dr. Goodman, and a paid licensing agreement is required for commercial use, EHR incorporation, apps, and other digitalization, which the terms extend to ambient-AI capture of a paper administration; the many forms and calculators circulating online are not evidence of authorization. Charts therefore record scores, subtotals, targets, and interpretation without pasting item wording. Serial results follow the outcome measure note conventions, the score sits inside a psychotherapy progress note or medication management note , and the goals it anchors live in the treatment plan . ## Who uses Y-BOCS documentation and when OCD specialty clinics and ERP therapists use it as the severity anchor at intake and at treatment reviews, prescribers use it to document medication trials against a baseline, and intensive outpatient, partial hospital, and residential programs use it to communicate severity and trajectory to utilization reviewers, who apply generic level-of-care criteria that name no instrument. Neurosurgical and deep brain stimulation programs use it in eligibility workups, where three different authorities are routinely conflated (the FDA indication has no number, FDA labeling says benefit below 30 was not established, and the International OCD Foundation's clinical criteria use 28 or higher). Clinical trials use it as the standard primary outcome with protocol-defined responder rules. In Canada, Ontario Health's OCD quality standard names it as the recommended severity and monitoring instrument without setting any admission threshold, and Ontario's out-of-country funding checklist asks for standardized severity results with the CY-BOCS as an example; in Australia, Better Access requires an outcome measure of the clinician's choosing and OCD programs describe suitability for ERP, readiness, and risk rather than a cutoff. When a score is charted, it belongs in a psychotherapy progress note with the ERP work it informs, and the level-of-care rationale it supports belongs in the prior authorization request and the treatment plan review . ## How to document Y-BOCS results in the chart No US, Canadian, or Australian law prescribes a Y-BOCS note format, and no scale authority makes a band a diagnosis or a care level. What survives review is a record that names the version and administration mode, keeps the checklist and targets as the referent for the severity ratings, reports the total with both subtotals and denominators, names the band framework if one is used, states change with the definition it invokes, and pairs the number with function, avoidance, and insight. Each element below carries the pitfall that most often undermines it. Version and mode. Chart original Y-BOCS (0 to 40), Y-BOCS-II (0 to 50), CY-BOCS, or CY-BOCS-II, and whether it was a clinician interview, a self-report adaptation, or a parent report, with the date, rater, informants, and the past-week rating window. Pitfall: A Y-BOCS-II charted as 0 to 40 (or original bands pasted onto a 0 to 50 result), a self-report total spliced into a clinician-rated series, or a raw score trended across editions as if the metric had not changed. Checklist, targets, then severity. At intake, establish content with the symptom checklist and name the target symptoms before rating severity; at follow-up, review and update the targets so the ratings have a documented referent. Record the current content areas (for example contamination, responsibility and checking, covert reviewing, reassurance seeking) in your own words. Pitfall: A severity score with no identified symptoms behind it, or the checklist treated as optional at initial administration. Total with both subtotals. Report x/40 with the obsession and compulsion subtotals and their denominators (obsessions 11/20, compulsions 9/20, total 20/40); for the second edition, x/50. Keep the arithmetic to the ten severity items; the checklist never enters the total. Pitfall: "Y-BOCS 20" with no denominator, no subtotals, and no edition, which cannot be compared with anything later. Bands named by framework. If you use a severity band, name it: the legacy original-scale bands printed by the rights holder (0 to 7 subclinical, 8 to 15 mild, 16 to 23 moderate, 24 to 31 severe, 32 to 40 extreme); the 2022 empirical original-scale bands from 5,140 people (0 to 13, 14 to 21, 22 to 29, 30 to 40); or the 2025 empirical second-edition bands (0 to 14, 15 to 21, 22 to 34, 35 to 50). A 28/40 is "severe" under one framework and "moderate" under another, so report the raw score and the framework. Pitfall: "Y-BOCS 28, severe OCD" with no framework, "0 to 7 means no OCD," or 16 treated as a diagnostic cutoff (the 2022 study called that trial-entry threshold arbitrary). Change with its definition. State the baseline date and score, the current score, absolute change, and percentage reduction (baseline minus current, divided by baseline), on the same edition, mode, and interval. Then say which convention you invoke: the 2016 expert consensus defines full response as at least 35% reduction plus CGI-I of 1 or 2 sustained at least a week, partial response as 25% to under 35% with CGI-I no worse than 3, and remission as no longer meeting criteria or a total of 12 or less with CGI-S 1 or 2; a 2024 individual-participant meta-analysis found 30% and 15 statistically optimal and still endorsed the consensus. Pitfall: "36% reduction, responder" with no CGI anchor, remission inferred from a percentage alone, or 25% described as more conservative than 35% (it is less stringent). Function, avoidance, insight, accommodation. Document work, school, relationships, self-care, and community function; avoided settings and delegated tasks; covert rituals and reassurance seeking; family accommodation; and insight, separately from the total, because the original scale under-weights avoidance and the core ten items do not directly capture insight. Pitfall: A falling total read as restored function while the patient still cannot use the kitchen or attend work, or insight inferred from the number. Decision link, without item text. Tie the result to the ERP target or hierarchy change, the response-prevention focus, medication continuation or optimization, the level-of-care rationale, or a diagnostic reassessment when score and function diverge; and keep item wording, anchors, and forms out of the note, which the license does not permit you to reproduce. Pitfall: A number filed with no consequence, or a chart template that pastes the interview prompts into the record. ### Blank template (copy and adapt) Y-BOCS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Rater: [ ] Informants: [ ] Version: [original Y-BOCS (0-40) / Y-BOCS-II (0-50) / CY-BOCS / CY-BOCS-II] Mode: [clinician interview / self-report adaptation / parent report] Rating window: [past week] Assessment interval: [ ] Checklist / targets: [reviewed at intake / updated today; current content areas in own words: ...] Score: obsessions [ ]/20 compulsions [ ]/20 total [ ]/40 (or /50) Band (framework named): [legacy original bands / 2022 empirical original bands / 2025 second-edition bands]: [ ]; not a diagnosis Baseline: [date] [ ]/[ ] Change: [ ] points, [ ]% reduction Response classification (definition named): [2016 consensus: full (35%+ with CGI-I 1-2, sustained 1 week) / partial (25% to under 35%, CGI-I 3 or better) / remission (no longer meets criteria, or 12 or less with CGI-S 1-2) / protocol-specific / numerical change only] CGI-I / CGI-S: [ ] Duration sustained: [ ] Function, avoidance, insight, accommodation: [work, school, relationships, self-care; avoided settings; covert rituals; reassurance; family accommodation; insight] Decision link: [ERP targets / response prevention / medication / level of care / diagnostic reassessment] Next administration: [date + same version and mode] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale items, checklist, anchors, or forms. ## Sample Y-BOCS documentation (fictional) Scenario: a scheduled review during outpatient ERP, with a clinician-administered original Y-BOCS compared to the intake baseline, the change classified under the 2016 consensus with its CGI anchor, persistent avoidance documented separately from the falling score, and the treatment plan adjusted. All details are fictional. Patient: K.P., 31 · Setting: Outpatient OCD specialty clinic, ERP session 7 review · Clinician: N. Adeyemi, PhD · Note date: 08/12/2026 Measure: Y-BOCS, original 1989 version, clinician-administered by the treating psychologist, past-week window; symptom targets reviewed and updated before rating. Today: obsessions 11/20, compulsions 9/20, total 20/40. Baseline at intake 07/01/2026 (same version and mode, checklist completed): obsessions 15/20, compulsions 13/20, total 28/40. Change: 8 points, a 28.6% reduction over six weekly ERP sessions. Under the legacy original-scale bands today's total sits in the moderate range (baseline severe); under the 2022 empirical bands both fall in the mild-to-moderate boundary region. Raw scores govern; the band is descriptive only. Symptom targets and function: Current content areas: contamination-related threat with hand-washing, responsibility and checking of appliances, covert mental reviewing, and reassurance seeking from her partner. Ritual frequency and duration reduced by her report and partner report; covert reviewing persists in the evenings. Shared-kitchen avoidance remains marked (still preparing no meals at home, eating out or relying on her partner), and she has not yet returned to the two client sites she stopped visiting in the spring. Insight fair: recognizes the feared outcomes as unlikely most days. Family accommodation: partner still confirms appliance status on request several times a day. Response classification: CGI-I rated 3 (minimally improved), sustained for two weeks. Consistent with the 2016 expert-consensus definition of partial response (25% to under 35% reduction with CGI-I no worse than 3); does not meet full response (35% or more with CGI-I 1 or 2) or remission. Score change reflects reduced overt ritualizing more than restored function; avoidance and accommodation are carrying part of the improvement. Plan: Continue weekly ERP with the hierarchy re-weighted toward avoided settings (home kitchen meal preparation, the two client sites) and covert reviewing, and a response-prevention agreement with the partner to taper appliance reassurance starting this week. Between-session practice logged daily. Medication unchanged pending the prescriber review scheduled 08/26/2026; this score and the functional summary sent to the prescriber today. Next Y-BOCS at session 13, same version and mode. Outpatient level of care remains appropriate: engaged, completing practice, no safety concerns. Administrative: Score, subtotals, version, mode, and CGI entered in the outcome-measure record; no item wording or form content stored in the note. Instrument administered under the practice's permission from the rights holder. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, mode, rating window, and rater are named, and the baseline was taken on the same version and mode, so the comparison is valid and the second-edition metric cannot be confused with it. The total carries both subtotals and denominators, the band is reported with its framework and labelled descriptive, and the raw score is what governs. The change is stated as points and percentage, then classified under a named convention with its CGI anchor and duration, so "partial response" means exactly one thing. Function, avoidance, insight, and accommodation are documented separately from the score, and the note says plainly that avoidance is carrying part of the improvement. The number changes something: the hierarchy is re-weighted, a reassurance-taper agreement is made, the prescriber gets the data, the next administration is dated, and the level-of-care conclusion is reasoned. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no law requires the Y-BOCS, and the pressure comes from payers, programs, and one device authorization. Medicare's outpatient psychotherapy documentation expectations (PAYER POLICY) call for an individualized plan, diagnosis and goals, type and frequency of services, periodic evaluation, progress, and a reasonable expectation of improvement, and neither CMS nor a current contractor article names the Y-BOCS; a 2026 UnitedHealthcare residential policy requires a need for 24-hour structure, an individualized evidence-based plan with weekly updates, active treatment, a reasonable expectation of further improvement, and evidence of likely deterioration at a lower level, again with no score threshold, so specialty programs that use a Y-BOCS cutoff should attribute it to their own criteria, not to the instrument or the payer (CONVENTION). Do not assume 96127 applies to a clinician-administered severity interview or that its payment rules are uniform; record the service actually performed, the instrument and mode, the result and interpretation, and verify the payer (PAYER POLICY). Deep brain stimulation is where three authorities get collapsed into one number: the FDA humanitarian device exemption for bilateral stimulation of the anterior limb of the internal capsule (H050003, February 2009, adjunct to medication and alternative to anterior capsulotomy for chronic, severe, treatment-resistant OCD in adults who have failed at least three SSRIs) contains no Y-BOCS number in its indication (LAW); FDA labeling states that safety and probable benefit were not established below a Y-BOCS of 30, reflecting the 26-patient experience that selected above 30 (LAW, an evidence boundary); and the International OCD Foundation's clinical criteria use a Y-BOCS of 28 or higher with at least five years of primary OCD, three serotonin-reuptake inhibitor trials including clomipramine, an antipsychotic augmentation attempt, and at least 20 ERP sessions (CONVENTION). Two 2026 HDE supplements concerned manufacturing and process matters, not the indication. A neurosurgical referral note keeps those three apart and adds the insurer's own requirements. Trials use change from baseline as the continuous endpoint with protocol-defined responder rules; the historic entry threshold of 16 on the original scale was described as arbitrary by the 2022 multinational analysis, and the protocol, not a scoring website, governs the endpoint (CONVENTION). Canada and Australia name no threshold either, and the evidence rewards modesty about bands and change. Ontario Health's 2020 OCD quality standard names the Y-BOCS as the recommended severity and monitoring instrument (CONVENTION), Ontario's February 2024 out-of-country funding checklist requires specialist confirmation of medical necessity with diagnosis, comorbidity, standardized severity results (CY-BOCS given as an example), function, prior medication and psychotherapy, and outcomes but sets no cutoff (PAYER POLICY), Canada's leading intensive program screens admission severity with the Florida Obsessive Compulsive Inventory rather than a Y-BOCS number, and CADTH characterized DBS for OCD as emerging in 2023, with a Canadian research protocol using a minimum of 24. Australia's Better Access guidance (March 2026) requires an outcome tool of the clinician's choosing and does not name the Y-BOCS (PAYER POLICY), and public program descriptions turn on diagnosis, suitability for ERP, readiness, and risk (CONVENTION). On the evidence: the 1989 rater study (four raters, 40 patients) reported excellent inter-rater reliability, a 2015 synthesis of 144 studies found mean alpha .866, test-retest .848, and intraclass correlation .922 with real variation across samples, and the CY-BOCS validation reported alpha .87 with a compulsion-subtotal intraclass correlation of .68 in youth; the second edition's initial study (130 adults) reported severity alpha .89 and reliability coefficients above .85, and a 2025 Spanish Y-BOCS-II study of 1,805 adults found alpha .92 with only marginal fit for the usual two-factor model. The 2022 empirical bands (5,140 people aged 5 to 82) achieved only about 55% to 57% exact category agreement in validation samples and low negative predictive value at the diagnostic boundary, so bands are descriptors, not partitions, and a low score cannot rule OCD out. Self-report adaptations agree only moderately with clinician totals (correlations around .45 to .65 in one carefully diagnosed sample, with clinicians rating compulsions higher), the rights holder frames its self-rated versions as decision support, and no scale-author source prescribes an administration interval; charts state the local convention and keep version and mode constant. Rights are the fact most pages omit. As stated at ocdscales.com in August 2026, any use of the scales requires explicit permission: noncommercial paper use for a clinician's own patients requires prior permission from Dr. Goodman, permission granted to one clinician does not automatically extend to colleagues, and a paid licensing agreement is required for commercial use, EHR incorporation, apps, software, and other digitalization, which the terms extend to ambient-AI capture used to create an EHR record. No free-use exception covers a public web tool that scores responses; publication in any medium, website posting, and distribution are not permitted; adaptations and translations require prior authorization; permitted copies keep the copyright and proprietary markings and published work cites the primary papers. The one instrument in the suite the rights holder allows to be reproduced on paper without permission is the DSM-5 Level 2 repetitive thoughts and behaviors measure adapted from the FOCI. The site names the licensing entity inconsistently (OCD Scales, LLC and OCD Scales, Inc.), and OCD Scales, LLC filed US trademark applications for Y-BOCS and CY-BOCS in November 2024, so a license request should confirm the exact counterparty. Freely posted forms and calculators are not evidence of authorization; this page links to none and hosts nothing. Y-BOCS, CY-BOCS, and their second editions are the intellectual property of OCD Scales, LLC and the instrument's authors. BastionGPT is not affiliated with, or endorsed by, OCD Scales, LLC or the authors. This page reproduces no items, checklist content, anchors, or scoring forms and describes the scoring architecture in original prose. ↑ Back to the template and downloads ## Common Y-BOCS documentation errors reviewers flag The numbers behind these errors are specific. The 2022 empirical bands from 5,140 people reached only about 55% to 57% exact category agreement and place a 28/40 in "moderate" where the legacy bands say "severe"; the 2016 consensus makes 35% one component of response, not the definition, and a 2024 individual-participant meta-analysis of 1,235 trial participants found 30% and a post-treatment total of 15 statistically optimal; the FDA DBS indication carries no Y-BOCS number while its labeling says below 30 was not established and IOCDF criteria say 28 or higher; and the rights holder's terms require permission for paper use and a paid license for any digital use. The BastionGPT Clinical Advisory Board sees the same errors most often in Y-BOCS documentation reviews: A band charted as a diagnosis or a care level. "Y-BOCS 26, severe OCD, meets residential criteria." No scale authority makes a band a diagnosis, an admission threshold, or a remission declaration; payers apply generic level-of-care criteria. Chart the raw score, name the framework, and let function, risk, treatment history, and the plan carry the level-of-care case. The edition and mode left unnamed. A Y-BOCS-II read against 0 to 40 bands, a self-report total spliced into a clinician-rated series, or a score trended across editions. Name original versus second edition (0 to 40 versus 0 to 50), interview versus self-report, and keep them constant across the series. "35% = responder" with nothing else. The 2016 consensus requires the reduction plus a CGI-I of 1 or 2 sustained at least a week; remission requires no longer meeting criteria or a total of 12 or less with CGI-S 1 or 2. Write "meets the 2016 consensus definition," not "the Y-BOCS defines response as 35%," and never call 25% the more conservative threshold. A total with no referent. A severity score with no checklist review or named targets behind it, no subtotals, and no denominator. The ten ratings need identified symptoms; report obsessions x/20, compulsions x/20, total x/40, and the content areas in your own words. The score stands in for function. A 40% drop charted as recovery while the patient still cannot use the kitchen or return to work, insight inferred from the number, or accommodation never mentioned. Avoidance, covert rituals, family accommodation, and insight are documented separately every time; the original scale under-weights avoidance by design. Item text in the note, or a free scorer in the workflow. Interview prompts pasted into a chart template, a public calculator adopted because it was free, or an ambient-AI capture of a paper administration treated as license-free. The current terms require permission for paper use and a paid license for EHR, app, web, and ambient-AI digitalization; record scores and interpretation, not the instrument. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version and mode, targets, subtotals and total, baseline and current with dates, CGI, function and avoidance, medication status) and it drafts the documentation block: denominators in place, the band with its framework, change with the definition invoked, function and insight separate, and the decision link, ready for your review. Cross-check a finished note for the gaps reviewers flag: a band charted as a diagnosis or care level, an unnamed edition or mode, a percentage called response with no CGI anchor, a score with no targets, or a falling total standing in for restored function. Draft the referral or utilization summary: baseline and current scores on the same version and mode, the trajectory, treatment history, function, and the level-of-care or neurosurgical rationale with the right authority named, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Y-BOCS scored, and what do the severity bands mean? Ten clinician-rated items scored 0 to 4, five for obsessions and five for compulsions, give two 0 to 20 subtotals and a 0 to 40 total for the past week; the symptom checklist identifies content first and never enters the arithmetic. The Y-BOCS-II scores each item 0 to 5 for a 0 to 50 total. Bands are conventions, and there are three non-equivalent frameworks: the legacy original-scale bands printed by the rights holder (0 to 7 subclinical, 8 to 15 mild, 16 to 23 moderate, 24 to 31 severe, 32 to 40 extreme), the 2022 empirical original-scale bands from 5,140 people (0 to 13 subclinical, 14 to 21 mild, 22 to 29 moderate, 30 to 40 severe), and the 2025 empirical second-edition bands (0 to 14, 15 to 21, 22 to 34, 35 to 50). None is a diagnostic threshold or a level-of-care rule, exact agreement with clinician global severity in the 2022 validation samples was only about 55% to 57%, and a low score cannot rule OCD out. Chart the raw score with subtotals and name the framework if you use a band. What is the difference between the Y-BOCS, the Y-BOCS-II, and the CY-BOCS? The original 1989 scale rates ten items 0 to 4 for a 0 to 40 total and has more than three decades of trial, response-threshold, and longitudinal literature behind it. The 2010 second edition rates each item 0 to 5 for a 0 to 50 total to relieve the ceiling among the most severe presentations, replaces the resistance-to-obsessions item (which had the weakest psychometrics and confuses patients taught in ERP not to resist) with an obsession-free interval item, integrates avoidance into severity, and revises the checklist; a 25/40 and a 25/50 do not sit in the same part of their ranges, original bands must not be pasted onto second-edition results, and conversion back to the original requires additionally rating the replaced item. The CY-BOCS (1997) and CY-BOCS-II (2019) adapt the interview for children and adolescents and require developmentally informed reconciliation of child and caregiver report. Charts name the edition and never trend a raw score across editions. What counts as response or remission on the Y-BOCS? Whatever definition you name, and the best-known one has several parts. Percentage change is baseline minus current, divided by baseline. The 2016 international expert consensus defines full response as at least a 35% reduction plus a CGI-I of 1 or 2 sustained at least one week; partial response as 25% to under 35% with CGI-I no worse than 3; remission as no longer meeting diagnostic criteria or, without a structured interview, a total of 12 or less with CGI-S of 1 or 2, sustained a week; and recovery as remission held for at least a year. Earlier signal-detection work found around 30% (2005) and at least 35% (2013) best matched global improvement, and a 2024 individual-participant meta-analysis of 1,235 adult trial participants found 30% and a post-treatment total of 15 statistically optimal while endorsing continued use of the consensus. So distinguish numerical change ("decreased 36%"), study-specific responder status, consensus response with its CGI and duration components, and remission, which is never inferred from a percentage alone. Do I have to administer the symptom checklist every time? At initial administration, yes: the designed sequence is checklist and target list first, severity ratings second, so the ten ratings have identified symptoms as their referent. At follow-up you do not have to re-run the full checklist, but you should review and update the targets and record the current content areas in your own words before rating, and note any new content. What reviewers should not see is a severity number with no symptoms behind it, or the checklist treated as an optional extra. Two related administration questions: purely mental compulsions (reviewing, neutralizing, internal checking, repeating) are compulsions and are rated in the compulsion half, described behaviorally without assuming that no visible ritual means no compulsion; and the interview stays neutral about feared content, clarifying what the patient does with uncertainty rather than adjudicating whether the feared event is likely, which the note can state in a sentence. Can we put the Y-BOCS in our EHR, on our website, or capture it with ambient AI? Not without a license. As stated at ocdscales.com in August 2026, any use of the scales requires explicit permission; noncommercial paper use for your own patients requires prior permission from Dr. Goodman, and permission to one clinician does not automatically extend to colleagues; a paid licensing agreement is required for commercial use, EHR incorporation, apps, software, and other digitalization, and the terms expressly treat ambient-AI capture used to create an EHR record as digitalization. No free-use exception covers a public web tool that scores responses; publication in any medium, website posting, distribution, adaptation, and translation all require authorization. Recording a total, subtotals, and interpretation is different from reproducing prompts or a form, but item-level or automated workflows should be confirmed in the license. The site names its entity both as OCD Scales, LLC and OCD Scales, Inc., and trademark applications for Y-BOCS and CY-BOCS were filed in November 2024, so confirm the exact counterparty. The DSM-5 Level 2 repetitive thoughts and behaviors measure adapted from the FOCI is the one suite instrument the rights holder allows on paper without permission. Is a Y-BOCS score required for intensive treatment or deep brain stimulation? No authority reviewed requires a specific score for intensive outpatient, partial hospital, or residential care: payers apply generic level-of-care criteria (24-hour structure, an individualized evidence-based plan, active treatment, expected improvement, likely deterioration at a lower level) that name no instrument, and programs that use a cutoff should attribute it to their own criteria. The score is most useful beside concrete impairment, ERP participation and targets, trajectory, accommodation, and why a lower level cannot yet work. For deep brain stimulation, keep three authorities apart: the FDA humanitarian device exemption (H050003, 2009) for adults with chronic, severe, treatment-resistant OCD who have failed at least three SSRIs contains no Y-BOCS number in its indication; FDA labeling says safety and probable benefit were not established below a Y-BOCS of 30; and the International OCD Foundation's clinical criteria use 28 or higher with at least five years of primary OCD, three serotonin-reuptake inhibitor trials including clomipramine, an antipsychotic augmentation attempt, and at least 20 ERP sessions. "FDA requires 28" and "FDA requires 30" are both wrong. Canadian and Australian programs and funders ask for standardized severity results, function, risk, and prior treatment, with no national cutoff. The score dropped but the patient still avoids everything. How do I document that? Explicitly, and in two parts. State the numerical change ("total decreased from 30 to 18, a 40% reduction") and then the functional reality ("occupational and community functioning remain substantially restricted; the patient continues to avoid the kitchen, has not returned to work, and the family still performs the checking"). Consider whether the drop reflects real improvement, reduced contact with triggers, or a shift from overt rituals to avoidance and accommodation, and adjust the ERP targets toward avoided settings and covert rituals. Do not classify this as consensus full response or remission when the CGI-I sits outside the response range and impairment persists; write "40% numerical reduction with persistent major functional impairment." The original scale under-weights avoidance by design (the second edition integrated it), and insight is not directly captured by the core items, so both are charted separately every time. Are self-report Y-BOCS versions acceptable in progress notes, and when would I use the OCI-R or DOCS instead? Self-report adaptations are acceptable as clearly named self-report measures used through an authorized workflow and interpreted as adjunctive: record who completed it, whether it was reviewed with the clinician, and that the mode differs from an interview. They are not interchangeable with the clinician version (correlations around .45 to .65 in one carefully diagnosed clinical sample, with the worst agreement on resistance items and clinicians rating compulsions higher), the rights holder frames its self-rated versions as decision support that cannot generate a diagnosis, and a self-report total should never be spliced into a clinician-rated series without noting the mode change. For brief patient-completed screening or dimension-level tracking, the 18-item OCI-R and the 20-item DOCS are validated self-report options that do not replace a clinician severity interview for level-of-care or trial-comparable decisions; verify each instrument's own permission terms, since a license for one authorizes nothing for another. Can BastionGPT draft my Y-BOCS documentation? Yes. Give it the facts (version and mode, targets, subtotals and total, baseline and current with dates, CGI, function, avoidance, insight, accommodation, medication status, and the plan) and it drafts the full entry: denominators in place, the band with its framework, change with the definition invoked, function and insight documented separately, and the decision link with dates and owners, ready for your review. It can also check a finished note for a band charted as a diagnosis or care level, an unnamed edition or mode, a percentage called response with no CGI anchor, a score with no targets, and a falling total standing in for restored function, and it never needs the instrument's item text to do so. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Goodman WK, Price LH, Rasmussen SA and colleagues, 1989, Archives of General Psychiatry, the Y-BOCS development and reliability paper (ten-item architecture; four-rater reliability study) and the companion validity paper ; Storch EA and colleagues, 2010, Psychological Assessment, the Y-BOCS-II (0 to 5 items, 0 to 50 total, avoidance integrated, replaced resistance item); Scahill L and colleagues, 1997, the CY-BOCS ; Storch EA and colleagues, 2019, the CY-BOCS-II . OCD Scales, LLC, ocdscales.com and terms of use (accessed August 2026: permission required for any use; paid license for commercial, EHR, app, and digital use including ambient-AI capture; no public posting or distribution; legacy severity bands; self-rated versions as decision support); Justia Trademarks, OCD Scales, LLC filings (November 2024 applications). Cervin M and colleagues, 2022, empirical severity benchmarks for the Y-BOCS and CY-BOCS (5,140 people aged 5 to 82; exact agreement about 55% to 57%); Pinciotti CM and colleagues, 2025, empirical severity benchmarks for the Y-BOCS-II and CY-BOCS-II (2,982 participants, 13 countries). Mataix-Cols D and colleagues, 2016, World Psychiatry, international expert consensus on response, remission, and recovery in OCD ; Tolin DF and colleagues, 2005, defining response in clinical trials ; Farris SG and colleagues, 2013, treatment response, symptom remission, and wellness ; Ramakrishnan D and colleagues, 2024, Journal of Psychiatric Research, individual-participant meta-analysis of response and remission definitions (1,235 participants; 30% and 15 optimal; consensus retained). López-Pina JA and colleagues, 2015, reliability generalization of the Y-BOCS (144 studies); Deacon BJ, Abramowitz JS, 2005, factor structure of the Y-BOCS ; Federici A and colleagues, 2010, self-report versus clinician-administered Y-BOCS ; 2010 measurement-invariance study across ethnic groups ; 2025 Spanish Y-BOCS-II study (1,805 Hispanic and Latino adults). FDA, humanitarian device exemption H050003 and labeling (indication without a Y-BOCS number; benefit below 30 not established); International OCD Foundation, deep brain stimulation criteria (28 or higher; treatment history) and exposure and response prevention . CMS, outpatient psychiatric care compliance tips (documentation elements; no instrument named); UnitedHealthcare, 2026 Tennessee adult residential treatment policy (level-of-care criteria without a score threshold); Indiana Medicaid, bulletin BT2025114 (state-specific handling of brief assessment codes; bare numbers only). Ontario Health, OCD quality standard (Y-BOCS as the recommended severity and monitoring instrument; no threshold); Ontario Ministry of Health, out-of-country OCD treatment checklist (February 2024); Sunnybrook, intensive OCD treatment program referrals ; ClinicalTrials.gov, NCT04217408 (Canadian DBS protocol, minimum 24). Australian Government Department of Health, Disability and Ageing, Better Access treatment plan template information (March 2026; outcome tool of the clinician's choosing); St Vincent's Private Hospital, OCD program description . Comparators: Foa EB and colleagues, 2002, the OCI-R ; Abramowitz JS and colleagues, 2010, the DOCS ; CamCOPS, Y-BOCS task page (item text not embedded; reproduction not permitted). ## Related templates Outcome Measure Review Note Psychotherapy Progress Note Treatment Plan Treatment Plan Review & Update Medication Management Follow-Up Note Prior Authorization Request Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/stop-bang All note templates # STOP-Bang Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The STOP-Bang is an eight-item obstructive sleep apnea risk screen published by Frances Chung and colleagues at Toronto's University Health Network in 2008, scored 0 to 8 with one point per positive domain. Anesthesia, primary care, sleep, and occupational teams use it to stratify OSA risk before surgery and referral. It screens; it never diagnoses. This page covers how to document and interpret STOP-Bang results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report with clinician-measured items (BMI, neck circumference, blood pressure history); scored by nursing, anesthesia, or clinic staff; use is licensed through University Health Network Audience Anesthesia and preadmission teams, primary care and sleep medicine, occupational and driver medical examiners, bariatric programs, payers and auditors Typical length 2 to 6 chart lines plus the decision it drives · completion under 2 minutes Format family OSA risk-screening questionnaire (eight yes-or-no domains, one point each, total 0 to 8) When it's used Preoperative assessment, primary care and sleep referral, bariatric workup, occupational and driver fitness evaluations, Australia's direct-access sleep-study pathway Standards context Property of University Health Network: clinical, EHR, and web use go through UHN's express-licence process; named in Australia's MBS sleep-study pathway; mandated by no US or Canadian law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the STOP-Bang? The STOP-Bang questionnaire is an eight-domain screen for obstructive sleep apnea risk, published in Anesthesiology in 2008 by Frances Chung and colleagues at Toronto Western Hospital, University Health Network. The four STOP domains are symptom and history items (a clinically significant snoring burden, daytime tiredness, another person's observation of breathing pauses during sleep, and hypertension); the four Bang domains are anthropometric and demographic (a body-mass-index threshold, an age threshold, an enlarged neck circumference, and male sex). Each positive domain scores one point for an unweighted total of 0 to 8. Official versions are not identical: the current UHN interactive tool uses a single neck threshold of about 40 cm while some official 2015 country forms use sex-specific thresholds of about 43 cm for men and 41 cm for women, and the literature holds weighted, continuous, bicarbonate-augmented, and population-adjusted variants, so the chart names the form, locale, or institutional build, not just "STOP-Bang." Two facts govern the documentation. First, it is a screen and a risk stratifier, never a diagnosis: the American Academy of Sleep Medicine's guideline states that questionnaires and prediction algorithms must not be used to diagnose OSA without polysomnography or technically adequate home testing, and no score establishes OSA severity or treatment eligibility. Second, "positive" is ambiguous unless the algorithm is named. The original validation literature used 3 or more as the positive threshold; the current UHN interpretation runs three bands (0 to 2 low, 3 to 4 intermediate, 5 to 8 high) with a combination rule that moves an intermediate score to high risk when at least two STOP domains are positive together with a BMI above 35, an enlarged neck, or male sex. A chart that imports "3 or more equals high" from an older source will disagree with a chart built on the current algorithm, and both will be wrong about the sleepiness question, which belongs to the Epworth Sleepiness Scale : the two instruments measure different constructs and Australia's Medicare pathway requires both for exactly that reason. Insomnia complaints route to the Insomnia Severity Index , not to this screen. ## Who uses STOP-Bang documentation and when Anesthesia and preadmission teams are the largest users, because the Society of Anesthesia and Sleep Medicine's 2016 preoperative guideline recognizes validated questionnaires for OSA risk identification, the Society for Ambulatory Anesthesia recommended the STOP-Bang for ambulatory screening in 2012, and the American Society of Anesthesiologists' 2014 guideline asks anesthesiologists and surgeons to decide jointly between presumptive OSA management and further evaluation; none of the three mandates the instrument or a cancellation threshold. Primary care and sleep medicine use it to support referral for home sleep apnea testing or polysomnography, where payer rules key off clinical signs and symptoms rather than a score in the United States and where Australia's Medicare direct-access pathway names it explicitly alongside the Epworth score. Bariatric programs screen because OSA prevalence is high, under a locally chosen protocol rather than a national mandate. Occupational and commercial-driver examiners meet the largest myth in the field: FMCSA's regulations prescribe no OSA screening tool, referral rule, or score, and Canada's and Australia's driver standards treat the questionnaire as an aid to referral, not a licensing criterion. The perioperative reasoning is often documented in a consultation-liaison note , the sleep referral in a referral letter , and the safety-critical-work question in a fitness-for-duty evaluation , each carrying the score with the decision it informed. ## How to document STOP-Bang results in the chart No UHN, ASA, SASM, or AASM source prescribes a STOP-Bang note format, and no US or Canadian law names the instrument. What survives review is a record that names the version or build, reports the total with the positive domains and the source of each response, states the band under a named algorithm, keeps the result a screen, documents the decision and its reasoning, and assigns ownership of what happens next. Each element below carries the pitfall that most often undermines it. Version, build, and licence. Chart "STOP-Bang" plus the form, locale, or institutional build and the scoring algorithm in use (the current UHN three-band interpretation with its combination rule, or an older binary 3-or-more threshold), the completion date, and whether the responses were patient-reported, partner-reported, clinician-observed, imported, or historical. The instrument is UHN property; use it under the licence your organization holds. Pitfall: "STOP-Bang positive" with no version or algorithm, when official versions differ on the neck threshold and older and current interpretations disagree about a 3 or 4. The total and the positive domains. Report x/8 and which domains were positive (by letter or concept), rather than a bare total; note missing or disputed responses, who supplied them, which response was used for the formal score, and whether the alternate score would change the band. Pitfall: "STOP-Bang 4" with no domains, or a partner's report silently converted into a definitive yes when the patient disputes it. The band under a named algorithm. State low (0 to 2), intermediate (3 to 4), or high (5 to 8) risk under the UHN three-band algorithm, and whether an intermediate score was reclassified to high by the combination rule (at least two STOP domains plus BMI above 35, an enlarged neck, or male sex; age is not one of the three override variables). Pitfall: Every score of 3 or more charted as high risk from an older source, or a 4/8 that met the override recorded as merely intermediate. Screen, not diagnosis, with diagnostic status stated. Write "screen-positive, suspected OSA" or "previously confirmed OSA" with any prior AHI or REI, PAP prescription, and adherence, and state that the score establishes neither OSA nor its severity; only polysomnography or adequate home testing diagnoses. Pitfall: "OSA per STOP-Bang" or a severity grade inferred from the number, or a known PAP-treated patient re-screened as if undiagnosed. Context that changes the meaning. Record the procedure and its invasiveness, anticipated opioids or sedation, airway concerns, comorbid disease, and any evidence of hypoventilation, severe pulmonary hypertension, or unexplained resting hypoxemia; in primary care, the sleepiness picture, driving history, and safety-critical work; in women, thin patients, and East Asian populations, that the conventional threshold performs differently. Pitfall: A high score with no gas-exchange assessment, or a low score in a symptomatic woman charted as "no OSA." The decision and its reasoning. Document what the score changed: presumptive OSA precautions and proceeding, delay for evaluation, referral for testing, or a documented informed decline, with the reasoning. SASM advises against routine delay or cancellation solely for workup except with additional evidence of disturbed ventilation or gas exchange; ASA permits presumptive management or delay, including when suspicion first arises on the day of surgery. Pitfall: "Score high, surgery proceeded" with no reasoning, or a high score treated as an automatic cancellation or an automatic sleep-study order. Plan, ownership, and safety advice. Name who will order testing, communicate with anesthesia and the receiving unit, arrange follow-up, or reassess, and record sleepiness-related driving and occupational advice where relevant; a declined workup gets the offered evaluation, material risks, alternatives, capacity, the patient's reason, and the follow-up plan. Pitfall: "Patient refused" alone, or a referral with no owner in a chart where nothing happens next. ### Blank template (copy and adapt) STOP-BANG DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Version / build: [UHN three-band build / country form + year / other] Algorithm: [three-band with combination rule / binary 3-or-more] Source of responses: [patient / partner / clinician-observed / imported] Score: [ ]/8 Positive domains: [S T O P B A N G] Missing or disputed: [which; who reported; response used; alternate score [ ]/8; band changed? yes / no] Band: [low 0-2 / intermediate 3-4 / high 5-8]; override applied? [yes: two STOP domains + BMI over 35 / neck / male; no] Diagnostic status: [screen-positive, suspected OSA / confirmed OSA: prior AHI-REI, PAP prescribed, adherence]; screen does not diagnose Context: [procedure, opioids/sedation, airway; hypoventilation, severe pulmonary hypertension, resting hypoxemia; sleepiness, driving, safety-critical work; sex/body habitus/population caveats] Decision + reasoning: [proceed with presumptive OSA precautions / delay for evaluation / refer for testing / informed decline documented] Precautions or referral details: [ ] Owner + follow-up: [who orders / communicates / reassesses; date] Safety advice: [driving, occupational] Record: [full responses and scores retained where required (AU MBS)] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the questionnaire items or forms. ## Sample STOP-Bang documentation (fictional) Scenario: a high-risk screen at a preadmission visit before major abdominal surgery, with no prior sleep study, no additional gas-exchange concern, and a documented decision to proceed with presumptive OSA precautions and an owned outpatient referral. All details are fictional. Patient: J.M., 58 · Setting: Preadmission clinic, elective major abdominal surgery · Clinician: R. Iyer, MD (anesthesiology) · Note date: 08/13/2026 Screening: STOP-Bang completed today in the hospital's licensed EHR module (UHN three-band build, single 40 cm neck threshold), responses patient-reported with BMI, neck circumference, and blood pressure history entered by nursing. Score 6/8. Positive domains: S, O, P, B, A, G; T and N negative. Partner present and confirms the observed breathing pauses; patient agrees. High risk under the three-band algorithm (no override needed at 6). This is a screening result: no prior sleep study, no established OSA diagnosis, no PAP. Assessment: Denies resting dyspnea and morning headaches; room-air saturation 96% seated; no clinical or laboratory evidence of hypoventilation, no known severe pulmonary hypertension, no unexplained resting hypoxemia on the available record. Procedure is time-sensitive, laparoscopic with possible open conversion, moderate expected opioid requirement. Airway examination documented separately; predictors of difficult mask ventilation present. Screen indicates high probability of undiagnosed OSA; it does not establish severity. Decision and plan: Reviewed with the surgeon and the patient. Per SASM and ASA guidance, no delay for a preoperative sleep study in the absence of additional ventilation or gas-exchange concerns; proceed with presumptive OSA precautions: difficult-airway equipment immediately available, opioid-sparing multimodal analgesia with regional technique if feasible, no routine continuous basal opioid infusion, minimal sedative premedication, head-up positioning, continuous pulse oximetry in recovery with extended observation, and postoperative disposition decided on oxygenation, ventilation, sedation, and analgesic requirement. Recovery-area and receiving-unit handoff to include suspected OSA. Patient informed that outpatient diagnostic sleep evaluation remains recommended regardless of surgical course; referral responsibility assigned to primary care with a copy of this note, and the patient given sleepiness-related driving advice. Administrative: Full responses and score retained in the licensed module; score, domains, band, algorithm, and decision entered in the preoperative assessment. Instrument used under the hospital's UHN licence; no questionnaire text reproduced here. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, build, algorithm, and source of each response are named, so the 6/8 is reproducible and the reader knows which neck threshold and which interpretation produced it. The total is broken into its positive domains and the band is stated under a named algorithm, with the collateral report confirmed rather than silently adopted. The result stays a screen: diagnostic status is stated, severity is not inferred, and the gas-exchange concerns that would change the SASM reasoning are looked for and documented. The decision carries its reasoning: why surgery proceeds, what precautions substitute for a preoperative study, and how the information travels to recovery and the receiving unit. The loop has an owner: the outpatient sleep evaluation is assigned to primary care with a copy of the note, and the driving advice is recorded, so nothing depends on the score being noticed later. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: guideline conventions, payer policies, and one widely misquoted federal standard, none of which mandates the instrument. Ambulatory surgical centers participating in Medicare must keep a pre-surgical assessment policy grounded in age, diagnosis, procedure, comorbidities, anesthesia level, and nationally recognized standards (42 CFR 416.52, LAW), and the regulation prescribes no questionnaire. The ASA's 2014 perioperative OSA guideline (CONVENTION) asks anesthesiologists and surgeons to decide jointly between presumptive management and further testing and permits either when suspicion first arises on the day of surgery; the SASM 2016 preoperative guideline (CONVENTION) recognizes validated questionnaires and recommends against delaying or cancelling surgery solely for OSA workup except in patients with additional evidence of disturbed ventilation or gas exchange (hypoventilation, severe pulmonary hypertension, unexplained resting hypoxemia); the Society for Ambulatory Anesthesia recommended the STOP-Bang for ambulatory screening in 2012; MBSAQIP's 2026 bariatric standards require a committee-approved risk-assessment process without naming the tool (CONVENTION). On coverage (PAYER POLICY): Medicare's sleep-testing national coverage determination (240.4.1, 2009) requires clinical signs and symptoms indicative of OSA and an eligible test and names no questionnaire; among commercial payers, Cigna's delegated eviCore sleep-disordered-breathing guideline (V1 effective 1 May 2026, V2 scheduled 5 November 2026) names a validated questionnaire such as the STOP-Bang or Berlin as one element of the clinical evaluation, while the Aetna, UnitedHealthcare, and Carelon policies reviewed key off signs, symptoms, examination, and test suitability, with Carelon revisions scheduled for September and November 2026; a high score therefore strengthens a referral and guarantees no authorization. For commercial drivers, 49 CFR 391.41(b)(5) (LAW) bars a respiratory dysfunction likely to interfere with safe driving and names no instrument, FMCSA's Medical Examiner's Handbook (2024, CONVENTION without force of law) states that its regulations do not prescribe OSA screening, referral, preferred tests, treatment, compliance thresholds, waiting periods, maximum certification periods, or routine retesting, and the 2016 rulemaking was withdrawn in August 2017; examiners refer on multiple risk factors under the general standard, and "DOT requires the STOP-Bang" is false. Canada, Australia, and the evidence. Canada's June 2026 CCMTA driver-fitness model standard (CONVENTION, applied through provincial licensing LAW) names no questionnaire and turns on diagnosed severity, sleepiness, crash history, and treatment (untreated disease with an AHI below 30 and no admitted sleepiness is generally compatible with all classes; severe disease generally requires effective treatment or specialist assessment; commercial drivers with OSA are reviewed annually), the Canadian Anesthesiologists' Society 2026 guidelines set no STOP-Bang threshold, and British Columbia's adult OSA guideline names it as a tool that may help assess risk while routing testing through the provincial pathway (PROVINCIAL POLICY). Australia is the exception that names the instrument: for the direct GP-initiated adult sleep-study pathway (items 12203 and 12250, PAYER POLICY, current MBS), the patient must score 3 or more on the STOP-Bang, or 5 or more on the OSA50, or high risk on the Berlin, and 8 or more on the Epworth Sleepiness Scale, and the sleep-study provider must retain the fully administered questionnaires with responses and scores for possible audit; failing the thresholds blocks the direct pathway but not the specialist route, and an older, still-indexed fact sheet with a superseded threshold should not be relied on. Austroads' Assessing Fitness to Drive lists the STOP-Bang, OSA50, and Berlin as aids to deciding who to refer, not as licensing criteria (CONVENTION applied through state LAW), the Heavy Vehicle National Law's unfit-to-drive duty (1 August 2026) is a general fitness obligation, and ANZCA's pre-anaesthesia guidance requires risk assessment and documentation without prescribing a tool. On the evidence: the 2015 meta-analysis pooled 90%, 94%, and 96% sensitivity for any, moderate-to-severe, and severe OSA in sleep clinics at a threshold of 3 (specificity 49%, 34%, 25%) and 84%, 91%, and 96% in surgical populations (specificity 43%, 32%, 29%), with the probability of severe OSA at scores 3, 4, 5, 6, and 7 to 8 running about 25%, 35%, 45%, 55%, and 75% in sleep clinics and 15%, 25%, 35%, 45%, and 65% in surgical patients, and its detailed cut-point table came from only six pooled studies; the 2017 bivariate meta-analysis of 108 studies and 47,989 participants found the STOP-Bang more sensitive than the Berlin, STOP, and Epworth (88%, 90%, 93% by severity) and the Epworth more specific but far less sensitive; a 2022 surgical meta-analysis of 3,247 patients found 85%, 88%, and 90% sensitivity with 47%, 29%, and 27% specificity; the combination rule raised specificity for moderate-to-severe OSA from about 31% for any three positive domains to about 85% (two STOP domains plus BMI over 35), 79% (plus enlarged neck), and 77% (plus male sex) in the 2014 and 2016 analyses, while a 2020 external validation found the refinements added little in that surgical cohort; a midlife-women study found 77% sensitivity and 45% specificity at 3 or more (AUC 0.67), a 2021 meta-analysis of 47 studies and 26,547 sleep-clinic patients found substantially lower accuracy in East Asia, a multiethnic surgical study used a BMI threshold of 27.5 for Chinese and Indian patients, and USPSTF's 2022 I statement found insufficient evidence for screening asymptomatic adults. The chart names the population caveat where it applies. Rights are stricter than most pages say. The STOP-Bang is the property of University Health Network, Toronto, developed by Frances Chung and colleagues; the official screening page carries the UHN property notice and an all-rights-reserved footer, and the FAQ routes every permission request through UHN Commercialization's express-licence forms rather than an email or an attribution line, with an intake that expressly covers routine patient screening and clinical care, sponsored research, and clinical trials. UHN identifies software integration and redistribution as uses that may require permission or a licence, publishes no blanket clinical-use permission, no public fee schedule, no Creative Commons licence, and no exception for free web calculators, and describes a tiered model (free for non-profit use under agreement, an annual fee for corporate health partners, pay-per-administration for industry); its foundation reported nearly 600 licences and about $3 million in licensing revenue by 2023. The AASM's own patient site states that clinical use requires licensing through the official website. Storing a clinician's calculated score and interpretation is different from reproducing the instrument, but the public pages define no safe harbor for item text or scoring logic in an EHR, so an EHR build, a portal form, or a decision-support rule goes through the licence. Official translations exist in more than twenty languages, and neither translation availability nor free web calculators imply public-domain status. STOP-Bang is the property of University Health Network and its authors. BastionGPT is not affiliated with, or endorsed by, University Health Network or the instrument's authors. This page reproduces no questionnaire items or forms and describes the domains and scoring in original prose. ↑ Back to the template and downloads ## Common STOP-Bang documentation errors reviewers flag The numbers behind these errors are specific. At a threshold of 3, the 2015 meta-analysis pooled about 90% sensitivity for any OSA in sleep clinics against 49% specificity, so most screen-positive patients do not have severe disease; the combination rule for intermediate scores raises specificity for moderate-to-severe OSA from about 31% to as high as 85% in the 2016 two-step analysis ; in a midlife-women cohort a threshold of 3 caught only 77% of moderate-to-severe cases; the SASM guideline recommends against delay for workup except with disturbed ventilation or gas exchange; and FMCSA's handbook states its regulations prescribe no OSA screening tool. The BastionGPT Clinical Advisory Board sees the same errors most often in STOP-Bang documentation reviews: "Positive" with no algorithm or version. "STOP-Bang positive" or "high risk" for every 3 or 4 from an older binary source, when the current UHN interpretation makes 3 to 4 intermediate unless the combination rule applies, and official forms differ on the neck threshold. Chart the score, the domains, the band, the algorithm, and the build. A screen written as a diagnosis. "OSA per STOP-Bang," a severity grade inferred from the number, or "6/8, sleep study required" as if the score alone established medical necessity. State screen-positive versus confirmed, that only polysomnography or adequate home testing diagnoses, and the clinical signs and symptoms a payer actually keys off. A high score treated as a cancellation order. Surgery delayed on the number alone, or the opposite, "score high, proceeded" with no reasoning. Document the gas-exchange assessment, the procedure and opioid context, the joint decision, and the presumptive precautions that substitute for a preoperative study. The FMCSA myth. "DOT requires STOP-Bang" or a driver certification decision charted as mandated by federal OSA rules. The federal standard bars respiratory dysfunction likely to interfere with safe driving and names no tool; the handbook disclaims prescribed screening, referral, tests, treatment, compliance, and retesting rules. Chart the examiner's individualized reasoning. Disputed or missing responses flattened. A partner's report entered as a definitive yes over the patient's objection, or the observed-apnea domain scored zero because the patient lives alone, with no note of why. Record who reported, which response was used, and the alternate score and band; in women the observed-apnea domain is the most sensitive single item, so its absence is flagged, not silently zeroed. One threshold for every body. A low score in a symptomatic woman, a thin patient, or an East Asian patient charted as "no OSA," or a locally lowered BMI threshold used without saying so. Name the population caveat and any modified algorithm, and let symptoms and clinical suspicion drive referral when the score is reassuring. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version or build, score and positive domains, source of each response, diagnostic status, procedure and gas-exchange context, decision) and it drafts the documentation block: band under the named algorithm, screen-not-diagnosis line, the reasoning for proceeding or referring, precautions, owner, and safety advice, ready for your review. Cross-check a finished note for the gaps reviewers flag: a "positive" with no algorithm, a screen written as a diagnosis, a high score with no reasoning either way, a disputed response flattened, or a referral with no owner. Draft the referral or handoff: score and domains, band, symptoms and comorbidities the payer or the receiving team needs, prior testing and PAP status, and the safety advice given, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the STOP-Bang scored, and what do the bands mean? Eight yes-or-no domains score one point each for a total of 0 to 8: four STOP domains (a clinically significant snoring burden, daytime tiredness, another person's observation of breathing pauses in sleep, hypertension) and four Bang domains (a body-mass-index threshold, an age threshold, an enlarged neck circumference, male sex). Under the current UHN interpretation, 0 to 2 is low risk, 3 to 4 intermediate, and 5 to 8 high, and an intermediate score is reclassified to high when at least two STOP domains are positive together with a BMI above 35, an enlarged neck, or male sex (age is not one of the override variables). The older validation literature simply called 3 or more positive. The bands stratify probability, not disease: in the 2015 meta-analysis the estimated probability of severe OSA at scores of 3, 4, 5, 6, and 7 to 8 was about 25%, 35%, 45%, 55%, and 75% in sleep clinics and 15%, 25%, 35%, 45%, and 65% in surgical patients, which is why a calculator that shows one universal probability is misleading. Is a score of 3 or 4 high risk or intermediate? It depends on which algorithm your chart names, which is exactly why the algorithm must be named. The original 2008 validation and much of the older literature treated 3 or more as the positive threshold, and many pages still call every 3 "high risk." The current UHN three-band interpretation calls 3 to 4 intermediate unless the combination rule applies: at least two STOP domains plus a BMI above 35, an enlarged neck, or male sex moves the patient to high risk, a refinement that raised specificity for moderate-to-severe OSA from about 31% for any three positive domains to about 85%, 79%, and 77% for those three combinations in the 2014 and 2016 analyses, though a 2020 external surgical validation found the refinements added little there. So write "4/8, intermediate by the UHN three-band algorithm, not reclassified" or "4/8, high risk under the combination rule because two STOP domains and the BMI modifier were present," never just "positive." Does a high STOP-Bang mean the patient has OSA, and do we cancel surgery? No and usually no. The score establishes neither OSA nor its severity: the American Academy of Sleep Medicine's guideline says questionnaires must not be used to diagnose OSA without polysomnography or technically adequate home testing, and even at 7 to 8 roughly a quarter to a third of surgical patients do not have severe disease. On the day of surgery, the SASM 2016 guideline recommends against delaying or cancelling for OSA workup except in patients with additional evidence of disturbed ventilation or gas exchange (hypoventilation, severe pulmonary hypertension, unexplained resting hypoxemia), and the ASA 2014 guideline permits presumptive OSA management or an individualized delay decided jointly by anesthesiologist and surgeon. Document that the score was recognized as a screen, the gas-exchange assessment, the procedure and opioid context, existing PAP or prior testing, the precautions used (airway readiness, opioid-sparing multimodal analgesia, no routine basal opioid infusion, enhanced recovery monitoring, disposition by oxygenation and sedation), and the handoff. "Score high, proceeded" with no reasoning is the version reviewers reject. Does a high score justify or authorize a sleep study? It can justify a clinical referral when the whole evaluation raises suspicion; it authorizes nothing by itself in the United States or Canada, and it is required in Australia. Medicare's national coverage determination for sleep testing requires clinical signs and symptoms indicative of OSA and an eligible test type and names no questionnaire; Cigna's delegated eviCore guideline (2026) names a validated questionnaire such as the STOP-Bang or Berlin as one element of the clinical evaluation, while the Aetna, UnitedHealthcare, and Carelon policies reviewed key off symptoms, comorbidities, examination, and test suitability, and all apply plan-specific terms. In Canada, provincial pathways govern testing and none names a score. In Australia, the direct GP-initiated pathway (items 12203 and 12250) requires a STOP-Bang of 3 or more (or an OSA50 of 5 or more, or Berlin high risk) together with an Epworth of 8 or more, and the provider must retain the fully administered questionnaires with responses and scores for audit; failing the thresholds closes the direct pathway, not the specialist route. Document the symptoms, sleepiness, driving and safety context, prior testing, and suitability for home testing alongside the score. Can we print the STOP-Bang, put it in our EHR, or build a free calculator? Not on the strength of attribution or the tool being free. The STOP-Bang is the property of University Health Network; the official site's FAQ routes every permission request through UHN Commercialization's express-licence forms, whose intake expressly covers routine patient screening and clinical care, and UHN identifies software integration and redistribution as uses that may require permission or a licence. It publishes no blanket clinical-use permission, no public fee schedule, and no free-calculator exception; its tiered model runs free for non-profit use under agreement, an annual fee for corporate health partners, and pay-per-administration for industry, and by 2023 it reported nearly 600 licences. An EHR questionnaire, portal form, decision-support rule, or auto-scored flowsheet is a software integration even when only your clinicians see it; a commercial host of a free public scorer should use the commercial route. Storing a clinician's calculated score and interpretation is different from reproducing the items, but no published safe harbor defines how much text or logic may be stored without permission. Terms as stated at stopbang.ca and UHN Commercialization in August 2026; the signed agreement controls. Does FMCSA, or a Canadian or Australian driver authority, require the STOP-Bang? No. The US federal standard (49 CFR 391.41(b)(5)) bars a respiratory dysfunction likely to interfere with safe commercial driving and names no instrument, BMI rule, neck measurement, or screening score; FMCSA's 2024 Medical Examiner's Handbook states that its regulations do not prescribe OSA screening, referral, preferred testing, treatment, compliance thresholds, waiting periods, maximum certification periods, or routine retesting, and the 2016 rulemaking was withdrawn in August 2017. Examiners may refer a driver on multiple risk factors and may withhold or shorten certification while a condition that may interfere with safe driving is unresolved; treated OSA does not automatically preclude certification. Canada's CCMTA model standard turns on diagnosed severity, sleepiness, crash history within five years, and treatment, with provincial licensing law and reporting duties layered on; Australia's Austroads standard lists the STOP-Bang, OSA50, and Berlin as aids to deciding who to refer, with licensing decided on diagnosed OSA, significant sleepiness, crash risk, and effective treatment through state law. Chart the examiner's individualized reasoning, never a mandate that does not exist. The patient disputes the observed-apnea response, or lives alone. How do I score and document it? The instrument is binary and no UHN rule or validation study tells you to override the patient, adopt a partner's report automatically, or assign half credit, so document the source and run the sensitivity analysis: who reported the event, that the patient disputes or cannot confirm it, which response was used for the formal score, the resulting score and band, and the alternate score and whether it changes the band or the decision ("formal score 4/8 using the partner's report; 3/8 if that domain is excluded; both intermediate under the stated algorithm, so the referral decision is unchanged"). For a patient with no bed partner, record the domain as not endorsed for want of a witness and say why, rather than silently scoring zero; in women the observed-apnea domain is the most sensitive single item, so its absence is worth flagging. Silently converting disputed information into a definitive yes, or a missing witness into a no, is the less transparent choice either way. Does the STOP-Bang work the same in women, thin patients, and Asian populations, and how does it differ from the Epworth or Berlin? No, and the chart should say so where it matters. Male sex scores a point, so a woman with otherwise identical features sits one point lower and can cross a band boundary; in a midlife-women cohort a threshold of 3 had 77% sensitivity and 45% specificity for moderate-to-severe OSA (AUC 0.67), and observed apneas was the most sensitive single item. In a younger, thinner sleep-clinic sample a threshold of 3 had about 84% sensitivity but only 18% specificity, while in obese surgical cohorts 3 kept high sensitivity and higher cutoffs gained specificity for severe disease. A 2021 meta-analysis of 47 studies found accuracy for moderate-to-severe OSA substantially lower in East Asia, and a multiethnic surgical study used a BMI threshold of 27.5 for Chinese and Indian patients; a modified threshold is a modified instrument, to be named, validated, and licensed as such. The Epworth Sleepiness Scale measures subjective sleep propensity, not OSA risk, and is more specific but far less sensitive for OSA (about 54% sensitivity for any OSA in the 2017 meta-analysis versus 88% for the STOP-Bang); the Berlin is an alternative categorical risk tool with lower pooled sensitivity. They are not interchangeable, and Australia's pathway requires an OSA-risk instrument and the Epworth together for that reason. Can BastionGPT draft my STOP-Bang documentation? Yes. Give it the facts (version or build, score and positive domains, source of each response, diagnostic status, procedure and gas-exchange context, symptoms and comorbidities, decision, and owner) and it drafts the full entry: the band under the named algorithm, the screen-not-diagnosis line, the reasoning for proceeding, delaying, or referring, the precautions or referral details, the owner and follow-up, and the safety advice, ready for your review. It can also check a finished note for a "positive" with no algorithm, a screen written as a diagnosis, a high score with no reasoning either way, a disputed response flattened, and a referral with no owner. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Chung F, Yegneswaran B, Liao P and colleagues, 2008, Anesthesiology, the STOP questionnaire and STOP-Bang model (2,467 surgical patients screened; original validation); Chung F, Yang Y, Brown R, Liao P, 2014, alternative scoring models (the two-STOP-plus-BANG combinations); Chung F, Abdullah HR, Liao P, 2016, Chest, a practical approach to screening (three bands, two-step strategy, probabilities by score); 2016 weighted and continuous models ; 2020 external validation of the refinements . University Health Network, stopbang.ca , screening page (property notice; three-band interpretation and combination rules; 40 cm neck threshold), contact and FAQ (permission via UHN forms), translations index (country forms including the 2015 UK English form with sex-specific neck thresholds); UHN Commercialization, express licences (intake covering routine screening and clinical care); AASM Sleep Education, what is the STOP-Bang (owned by Dr. Chung and UHN; licence required for clinical use). Nagappa M and colleagues, 2015, PLoS One, systematic review and meta-analysis (17 studies; pooled sensitivity and specificity by setting; probabilities by score; the six-study cut-point table); Chiu HY and colleagues, 2017, Sleep Medicine Reviews, bivariate meta-analysis of Berlin, STOP-Bang, STOP, and Epworth (108 studies, 47,989 participants); Hwang M and colleagues, 2022, BMC Anesthesiology, surgical-population meta-analysis (3,247 patients). Mayo Clinic midlife-women validation, 2019, Maturitas, predictive ability in women ; 2021 geographic meta-analysis (47 studies, 26,547 patients; lower accuracy in East Asia); multiethnic surgical validation (population BMI thresholds); obese and morbidly obese surgical cohort ; younger, thinner sleep-clinic sample . American Academy of Sleep Medicine, 2017, clinical practice guideline for diagnostic testing for adult OSA (questionnaires not to be used to diagnose); American Society of Anesthesiologists, 2014, perioperative OSA practice guidelines ; Society of Anesthesia and Sleep Medicine, 2016, preoperative screening and assessment guideline (against routine delay except with disturbed ventilation or gas exchange); MBSAQIP, 2026 standards ; eCFR, 42 CFR 416.52 . CMS, NCD 240.4.1, sleep testing for OSA ; eviCore for Cigna, sleep-disordered breathing guideline V1.0.2026 and V2.0.2026 ; Carelon, sleep disorder management, November 2026 version ; UnitedHealthcare, sleep studies policy . eCFR, 49 CFR 391.41 ; FMCSA, Medical Examiner's Handbook, 2024 edition ; Federal Register, August 2017, withdrawal of the OSA rulemaking . CCMTA, National Safety Code Standard 6, Determining Driver Fitness in Canada (June 2026); Ontario, reporting for driver medical review ; Canadian Anesthesiologists' Society, 2026 guidelines to the practice of anesthesia ; Government of British Columbia, adult obstructive sleep apnea guideline . MBS Online, item 12203 (direct-access adult sleep-study pathway; STOP-Bang 3 or more, OSA50 5 or more, or Berlin high risk, with Epworth 8 or more; full questionnaires retained for audit; bare item numbers only); Austroads, Assessing Fitness to Drive ; National Heavy Vehicle Regulator, unfit-to-drive duty (1 August 2026); ANZCA, PG07 pre-anaesthesia consultation guideline . American Thoracic Society, Epworth Sleepiness Scale summary (sleepiness, not OSA risk); Berlin questionnaire primary-care validation, reference summary . ## Related templates Epworth Sleepiness Scale Documentation Insomnia Severity Index Documentation Hospital Consult & Liaison Note Fitness-for-Duty Evaluation Referral Letter Medication Management Follow-Up Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/epworth All note templates # Epworth Sleepiness Scale Documentation: Scoring & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Epworth Sleepiness Scale (ESS) is an eight-item self-report measure of usual daytime dozing propensity, published by Murray Johns in 1991 and scored 0 to 24. Sleep, respiratory, primary care, and occupational clinicians use it at evaluation and to track treatment. It measures sleepiness; it never diagnoses sleep apnea. This page covers how to document and interpret ESS results in the chart, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report (adult ESS); parent or carer assistance for younger children on the ESS-CHAD; scored by any clinician or staff; every use is licensed through Mapi Research Trust Audience Sleep medicine and respiratory clinics, primary care, PAP suppliers and payers, occupational and driver medical examiners, prescribers of wake-promoting agents Typical length 2 to 5 chart lines per administration · patient completion 2 to 3 minutes Format family Self-report sleepiness measure (eight everyday situations rated 0 to 3, total 0 to 24) When it's used Initial sleep evaluation, pretreatment baseline, PAP and medication follow-up, annual OSA review, driver and occupational fitness assessment, Australia's direct-access sleep-study pathway Standards context Copyright the estate of Murray W. Johns, licensed through Mapi for every use with or without a fee; named in Australia's MBS pathway (8 or more with an OSA-risk screen); mandated by no US or Canadian law On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Epworth Sleepiness Scale? The Epworth Sleepiness Scale is a self-report measure of a person's usual propensity to doze in eight everyday situations, developed by Murray W. Johns at Epworth Hospital in Melbourne and first published in Sleep in 1991, with the current standard adult wording dating from the 1997 revision. Each situation is rated on a four-level scale scored 0 to 3 for the respondent's average tendency to doze in recent times, and the eight ratings sum to a 0 to 24 total; a total with any situation unanswered is invalid and is not imputed. The recall frame is deliberately not a fixed number of days: the official explanation is that "recent times" spans roughly weeks to months so respondents have met most of the situations, and a version that fixes the interval (for treatment comparisons, say) is a specific version requiring permission. The ESS-CHAD, authorized in 2015 for children and adolescents, keeps the eight-situation architecture with developmentally appropriate wording and an explicit recent-month frame. Three facts govern the documentation. First, the construct is average sleep propensity, not fatigue, not tiredness at the moment of administration, not sleep quality, and not the risk of obstructive sleep apnea; the ESS correlates only weakly with apnea severity and with objective sleepiness tests, so a low score cannot exclude OSA and a high one identifies no cause. Second, the score is a symptom measure with a conventional threshold: on the official interpretation, 0 to 10 is the normal range (0 to 5 lower, 6 to 10 upper), 11 to 12 mild, 13 to 15 moderate, and 16 to 24 severe excessive daytime sleepiness, so 11 is the first elevated score, and policy thresholds such as Australia's 8 or more or a US utilization manager's above 10 are access criteria, not clinical redefinitions. Third, it is licensed intellectual property: the copyright belongs to the estate of Dr Johns, Mapi Research Trust administers licences and authorized translations worldwide, and a licence is required for every use, whether or not a fee is payable. The OSA-risk question belongs to the STOP-Bang , the insomnia question to the Insomnia Severity Index , and Australia's Medicare pathway requires the ESS together with an OSA-risk instrument for exactly that reason. ## Who uses Epworth Sleepiness Scale documentation and when Sleep medicine and respiratory clinics chart it at the initial evaluation, as the pretreatment baseline, at PAP and medication follow-ups, and at least annually in ongoing OSA care, which is the cadence the American Academy of Sleep Medicine's quality framework describes alongside drowsy-driving assessment and adherence review. Primary care uses it to quantify a sleepiness complaint before referral, PAP suppliers and payers see it in the clinical evaluation that supports testing and in the reevaluation that supports continued coverage, and prescribers of wake-promoting agents meet it on prior-authorization forms that ask for a validated sleepiness scale without, in current policies, a uniform numeric rule. Occupational and driver medical examiners use it as one input in a fitness assessment that turns on actual sleepiness, crash history, and treatment, never on the score alone; the official ESS site itself cautions against using the scale on its own where a legal outcome such as a driving privilege depends on it. The score belongs beside the OSA-risk result on the STOP-Bang page, in the medication management note when a wake-promoting agent is titrated, and in the fitness-for-duty evaluation or return-to-work letter when safety-critical work is the question. ## How to document Epworth Sleepiness Scale results in the chart No US, Canadian, or Australian law prescribes an ESS note format, and no authority makes a score diagnostic. What survives review is a dated, patient-reported result with the version, mode, and validity stated, the total written out of 24 with its official band, the sleep-opportunity and medication context that explains it, the safety history, a modest interpretation, and a comparison with the prior score under comparable conditions. Each element below carries the pitfall that most often undermines it. Version, mode, respondent, and assistance. Chart adult ESS (1997 standard) or ESS-CHAD, the language and whether the translation is authorized, the mode (paper, licensed portal or EHR form, interview, telephone), who answered (patient; parent or carer under ESS-CHAD guidance), and any reading help, interpreter, or partner prompting. Pitfall: "Epworth 14" with no version, mode, or respondent, or a partner-completed adult form charted as if the patient answered it. The total, out of 24, and its validity. Write x/24 and confirm all eight situations were answered; a total with any missing response is invalid and is not imputed, and a half-point total is rounded up per the official guidance. Note the standard recall frame (recent times) and do not insert a fixed interval into a locally reproduced form. Pitfall: A total scored from seven answers, or a locally edited form with a new recall window that breaks comparability with the norms. The band, with 10 kept normal. State the official band: 0 to 10 normal (0 to 5 lower, 6 to 10 upper), 11 to 12 mild, 13 to 15 moderate, 16 to 24 severe excessive sleepiness. Above 10 and 11 or more describe the same threshold; a policy using 8 or more (Australia's MBS pathway) or above 10 (a US utilization manager) is an access criterion and the CMS PAP policy appendix's 0 to 9 "normal" is a legacy label, not the rights holder's interpretation. Pitfall: A 10 charted as abnormal, an 8 charted as excessive sleepiness because a payer pathway uses it, or the bands treated as diagnostic categories. What it does not measure. State that the score reflects reported dozing propensity and identifies no cause: it is not an OSA-risk score, an apnea severity measure, a sleep-quality or insomnia measure, a substitute for polysomnography, home testing, MSLT, or MWT, or a real-time alertness or crash-risk test; a low score does not exclude OSA. Pitfall: "ESS 15, consistent with sleep apnea," or "ESS 6, sleep apnea unlikely," when the ESS-apnea correlation runs about 0.1 to 0.2. Context that explains the number. Record usual sleep opportunity and recent restriction, shift work or circadian factors, sedating and stimulating medications, alcohol and other substances, caffeine, mood, acute illness, and treatment adherence; then the safety history: dozing while driving or operating equipment, near-misses, crashes, unintentional sleep episodes, and occupational exposure. Collateral goes in separately with its source. Pitfall: An elevated score attributed to OSA in a rotating-shift worker sleeping five hours on a sedating antihistamine, or a low score accepted at face value beside admitted sleep-at-the-wheel episodes. Change over time, described not over-claimed. Give the prior score, date, and conditions, the absolute change, and whether administration was comparable; group-level minimally important difference estimates in OSA run about 2 to 3 points, but in a short-interval repeatability study 63% of retested patients differed by at least 2 points and 48% by at least 3, so pair the change with objective PAP use and efficacy, function, and safety history before calling it response. Pitfall: "Responded to CPAP, ESS down 2" with no usage data, or worsening blamed on treatment failure before sleep opportunity, schedule, and medication are reviewed. Plan, follow-up, and the licence. Link the result to the decision: objective testing, medication review, sleep-extension trial, PAP optimization, MWT where wakefulness in safety-sensitive work is the question, driving advice, and the next administration on the same version and mode; and note that the instrument was administered under the practice's licence, with no items reproduced in the note. Pitfall: A number filed with no consequence, or the eight situations pasted into a chart template or an unlicensed EHR form. ### Blank template (copy and adapt) EPWORTH SLEEPINESS SCALE DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Time point: [initial / pretreatment baseline / PAP day __ / medication follow-up / annual] Version: [adult ESS 1997 / ESS-CHAD] Language: [ ] Translation: [ ] Mode: [paper / licensed portal or EHR / interview / telephone] Respondent: [patient / parent or carer (ESS-CHAD)] Assistance: [none / reading help / interpreter / prompting] Total: [ ]/24 Validity: [all 8 answered / missing: invalid] Band (official): [0-10 normal (0-5 lower, 6-10 upper) / 11-12 mild / 13-15 moderate / 16-24 severe]; policy thresholds (MBS 8+, payer over 10) are access criteria, not this band Context: [usual sleep opportunity, recent restriction, shift work, sedating or stimulating medications, alcohol, caffeine, mood, illness, treatment adherence] Safety: [dozing while driving or operating equipment, near-misses, crashes, unintentional sleep episodes, occupational exposure] Collateral (separate, with source): [ ] Interpretation: [reported dozing propensity in context; not an OSA-risk score, not a diagnosis, not a driving-fitness determination] Prior: [date] [ ]/24 under [comparable / different] conditions Change: [ ] points; paired with [objective PAP use, function] Plan: [testing / medication review / sleep-extension trial / PAP optimization / MWT / driving advice / next administration + date] Licence: [administered under practice licence; no items in note] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's situations, response options, or forms. ## Sample Epworth Sleepiness Scale documentation (fictional) Scenario: a PAP follow-up inside the Medicare reevaluation window, with a repeat ESS compared to the pretreatment baseline, objective usage and efficacy data alongside the score, the safety history reviewed, and the change described rather than over-claimed. All details are fictional. Patient: C.D., 47 · Setting: Sleep medicine clinic, PAP follow-up (day 39) · Clinician: H. Brennan, NP · Note date: 08/14/2026 Measure: Epworth Sleepiness Scale, adult 1997 version, English, patient self-completed on paper in the waiting room before the visit; all eight situations answered; no proxy, prompting, or reading assistance. Total 7/24, within the official normal range (upper portion). Baseline at the initial evaluation 07/01/2026, same version, mode, and conditions: 17/24 (severe range). Change: 10 points. Context: Auto-titrating PAP started 07/06/2026 for moderate OSA on home testing (result and settings documented separately). Device download for the last 30 days: use on 28 of 30 nights, average 6 hours 20 minutes per night, residual event burden low, no significant leak. Usual sleep opportunity now about 7.5 hours; no shift work; no sedating medication; alcohol occasional. Reports waking refreshed most days and no unintentional daytime dozing. Direct inquiry: no sleep-at-the-wheel events, near-misses, or crashes since starting treatment; drove to the visit. Partner's observation of much quieter sleep noted separately as collateral. Interpretation: Substantial improvement in reported dozing propensity, from the severe range to the normal range, corroborated by objective use above the 4-hour, 70%-of-nights pattern, low residual events, and improved daytime function; the ESS itself does not measure apnea control, and the score is read with the device data rather than as proof of treatment success on its own. Group-level meaningful-change estimates run about 2 to 3 points, and short-interval retest noise can reach several points, which a 10-point change well exceeds. Plan: Continue PAP at current settings; supplier notified that the clinical reevaluation within the day-31-to-91 window is complete with documented symptom improvement and objective adherence. Mask fit and humidification reviewed. Counseled to avoid driving when sleepy and to report any return of dozing. Repeat ESS on the same version and mode at the annual review, or sooner if symptoms change; annual drowsy-driving and adherence review scheduled. Administrative: Score, version, mode, band, baseline, and change entered in the sleep flowsheet; usage report filed. Instrument administered under the clinic's licence through Mapi Research Trust; no scale items reproduced in the record. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, mode, respondent, assistance, and validity are named, so the 7/24 and the 17/24 are comparable and reproducible. The band uses the official normal range and the note keeps the score a symptom measure, read with objective device data rather than in place of it. The context that could explain the number (sleep opportunity, shift work, sedating medication, alcohol) is charted, and the safety history is asked directly and recorded, with collateral kept separate. The change is described as points and range with the published meaningful-change and retest-noise figures in view, and the response claim rests on usage, efficacy, and function together. The payer loop closes inside the required window with the elements the policy actually asks for, and the next administration and driving advice are dated and owned. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: payer policies use the score in different ways, and no law names it. Carelon's sleep-disorder management guideline (effective November 2025, codes updated July 2026, a new edition scheduled for 19 September 2026) accepts an ESS above 10 as one way to document excessive daytime sleepiness in a sleep-study pathway that ordinarily needs a combination of clinical criteria (PAYER POLICY); Medicare's sleep-testing and PAP policies recognize documented excessive sleepiness as a symptom that can support coverage without requiring a score or a numeric threshold, and its PAP local coverage determination reproduces the scale in an appendix (with a legacy permission line and a 0 to 9 "normal" label that conflicts with the rights holder's 0 to 10) while conditioning continued coverage on a clinical reevaluation between day 31 and day 91, documented symptom improvement, and objective use of at least 4 hours a night on 70% of nights in a consecutive 30-day period within the first three months, not on the ESS falling below 11 (PAYER POLICY). Current modafinil and armodafinil prior-authorization policies reviewed (CVS Caremark, Cigna, Aetna Better Health, Highmark, Anthem) ask for diagnosis, objective criteria, exclusion of insufficient sleep, adequate PAP treatment, and clinical response, or for a validated sleepiness scale, without a uniform numeric ESS rule; the archived policies that used 10 explain the memory of a required Epworth and should not be generalized (PAYER POLICY). For commercial drivers, 49 CFR 391.41(b)(5) bars a respiratory dysfunction likely to interfere with safe driving and names no OSA test or ESS (LAW); FMCSA's 2024 handbook states its regulations prescribe no screening method, referral rule, testing, treatment, or compliance threshold (CONVENTION), and the Medicare 4-hour, 70% pattern is not a federal driver-certification standard. FAA aviation materials (May 2026) treat untreated or poorly controlled OSA as potentially disqualifying, set no ESS number, and use a stricter PAP expectation of treatment during at least 75% of sleep periods averaging at least six hours (CONVENTION), so an occupational note names the applicable protocol rather than importing the Medicare minimum. Narcolepsy and idiopathic hypersomnia workups (CONVENTION) require adequate sleep documented by diary and often actigraphy, other sleep disorders stabilized, and PSG with MSLT; the ESS quantifies reported sleepiness and supplies no diagnostic threshold, and the AASM's quality framework calls for sleepiness assessment at initial evaluation and at least annually in treatment. Canada, Australia, and the evidence. Canada's June 2026 CCMTA driver-fitness model standard (CONVENTION, applied through provincial licensing LAW) treats the ESS as subjective, sets no numeric cutoff, decides on disease severity, admitted sleepiness, sleepiness-related crashes, treatment effectiveness, and ongoing commercial review, and contains a primary-source error worth knowing: it describes the response scale as 1 to 4 where the authorized architecture is 0 to 3, so its description should not be copied into local forms; British Columbia, Ontario, and Quebec run their own reporting duties and medical forms with no ESS trigger. Australia's Austroads standard (2022, CONVENTION applied through state LAW) states that the ESS is neither sensitive nor specific for diagnosing OSA, depends on honest reporting, and is one component of assessment; state and territory notification duties turn on impairment, not a total. Australia's Medicare direct-access adult sleep-study pathway (items 12203 and 12250, PAYER POLICY) requires an ESS of 8 or more together with a STOP-Bang of 3 or more, an OSA50 of 5 or more, or a high-risk Berlin, with the complete questionnaires, responses, and scores retained for audit, and the PBS narcolepsy authority criteria for wake-promoting agents reference objective testing rather than an ESS score. On the evidence: the 1992 reliability study found five-month test-retest correlation 0.82 and alpha 0.88 in medical students; a 2014 systematic review of 35 studies found internal consistency about 0.73 to 0.86 and pooled correlations of about minus 0.43 with the MWT, minus 0.27 with the MSLT, and 0.11 to 0.23 with sleep-apnea measures, judging the evidence better for group comparison than precise individual inference; OSA trial analyses estimate a minimally important difference near 2 points (639 participants) and 2 to 3 points, while a short-interval repeatability study of 40 sleep-clinic patients found retest differences of at least 2 points in 63%, 3 in 48%, 5 in 20%, and 7 in 8%; and when 122 patients were told about licensing implications, the group mean did not change but about 9% crossed the sleepy-versus-not classification, which supports caution about threshold-based legal decisions without assuming every low score is dishonest. Normative work varies by country: 507 Australian workers without a sleep disorder averaged 4.6 with 10.9% of the whole sample above 10, a US community sample of 703 adults showed little effect of age or sex, and a German population sample of 9,711 found about 23% above 10, so scores are recorded as standard totals and read in context, not age-adjusted informally. Rights are broader than most pages say. The copyright in the Epworth Sleepiness Scale and the ESS-CHAD belongs to the estate of Dr Murray W. Johns (who died in 2021), and Mapi Research Trust administers usage, licences, and authorized translations worldwide through its ePROVIDE platform. The official position, as stated in August 2026, is that a licence is required for every use, whether or not a licence fee is payable: students, physicians in individual clinical practice, and non-funded academic users may register for direct access, funded academic projects, healthcare organizations, commercial users, and IT companies submit a request and fees may apply, electronic implementations require an agreement for all user classes with implementation and screenshot review, and no universal price list is published because fees are quoted by project, language, and scope. Printing the form into an intake packet, scanning it into a forms library, building it into an EHR flowsheet, portal, app, or decision-support rule, or hosting a free public calculator are all reproduction or electronic use, and neither wide online circulation nor an older permission footnote (Austroads' note that individual doctors may use it without permission describes past practice) changes the current terms; documenting a result such as "ESS 14/24" is different from reproducing the instrument. The Epworth Sleepiness Scale is copyright of the estate of Dr Murray W. Johns and licensed worldwide through Mapi Research Trust. BastionGPT is not affiliated with, or endorsed by, the Johns estate or Mapi Research Trust. This page reproduces none of the scale's situations, response options, or forms and describes the instrument in original prose. ↑ Back to the template and downloads ## Common Epworth Sleepiness Scale documentation errors reviewers flag The numbers behind these errors are specific. The ESS correlates about 0.11 to 0.23 with sleep-apnea measures and minus 0.27 with the MSLT across 35 studies; the official normal range is 0 to 10 while a Medicare policy appendix labels 0 to 9 normal and Australia's pathway uses 8 or more; group-level meaningful change is about 2 to 3 points while 63% of retested patients differed by at least 2 points over a short interval; and every use requires a licence whether or not a fee is payable. The BastionGPT Clinical Advisory Board sees the same errors most often in ESS documentation reviews: The score charted as an apnea finding. "ESS 15, consistent with OSA" or "ESS 6, OSA unlikely." The scale measures reported dozing propensity, correlates weakly with apnea severity, and identifies no cause; chart it as a symptom result read with the sleep history, the OSA-risk screen, and objective testing. 10 charted as abnormal, or 8 as excessive. The official normal range is 0 to 10, and 11 is the first elevated score. A payer's above-10 route and Australia's 8-or-more pathway criterion are access rules; a Medicare appendix's 0 to 9 label is legacy. Name the official band and the policy separately. A total with no version, mode, respondent, or validity. "Epworth 12" with no 1997-versus-CHAD, language, paper-versus-portal, who answered, or confirmation that all eight situations were completed; a total scored from seven answers is invalid and is not imputed. Chart x/24 with the metadata that makes it comparable. Response declared from a 2-point change. "Responded to CPAP, ESS down 2" with no usage download, efficacy data, or function. Group-level meaningful change runs about 2 to 3 points and short-interval retest noise reaches several points; describe the change, then rest the response claim on objective use, residual events, function, and safety history. The context left out. An elevated score attributed to OSA in a shift worker sleeping five hours on a sedating antihistamine, or a low score accepted at face value beside admitted sleep-at-the-wheel episodes and a licence at stake. Record sleep opportunity, schedule, medications, substances, mood, the direct safety inquiry, and collateral with its source; chart inconsistencies factually rather than editing the patient's answers. The items in the note, or an unlicensed form in the workflow. The eight situations pasted into a chart template, a locally edited version with a new recall window, an EHR flowsheet or public calculator built without an agreement, or a partner-completed adult form charted as the patient's. Record scores and interpretation, use the instrument under a licence, keep proxy scores labelled and separate, and never modify or translate it yourself. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version and mode, respondent, total and validity, sleep opportunity and medications, safety history, prior score and date, device data, plan) and it drafts the documentation block: the score out of 24 with its official band, the context and safety lines, a modest interpretation, the change described against the prior result, and the plan, ready for your review. Cross-check a finished note for the gaps reviewers flag: a score charted as an apnea finding, a 10 called abnormal, a total with no version or respondent, response declared from a 2-point change with no usage data, or a safety history never asked. Draft the payer or occupational summary: baseline and current scores on the same version, the reevaluation elements the policy actually asks for, objective use and efficacy, function, and the safety and driving advice given, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Epworth Sleepiness Scale scored, and what do the bands mean? Eight everyday situations are each rated 0 to 3 for the respondent's usual chance of dozing in recent times, and the ratings sum to a total of 0 to 24; a total with any situation unanswered is invalid and is not imputed, and a half-point total is rounded up. On the official interpretation, 0 to 10 is the normal range (0 to 5 lower, 6 to 10 upper), 11 to 12 mild, 13 to 15 moderate, and 16 to 24 severe excessive daytime sleepiness, so above 10 and 11 or more describe the same conventional threshold. The bands describe reported dozing propensity, not a diagnosis and not a cause: an elevated score is compatible with insufficient sleep, shift work, sedating medication, depression, sleep-disordered breathing, central disorders of hypersomnolence, or several of these together, and the differential lives in the note, not in the number. Is an Epworth score of 10 normal or abnormal? Normal, on the current official interpretation: 0 to 10 is the normal range and 11 is the first elevated score. Three sources make clinicians unsure. Australia's Medicare direct-access sleep-study pathway uses 8 or more, but only combined with a separate OSA-risk instrument, as an access criterion, not a redefinition of abnormal sleepiness; a US utilization manager's pathway uses above 10, which is the same boundary as 11 or more; and the Medicare PAP policy appendix labels 0 to 9 as normal, a legacy label that is neither the rights holder's interpretation nor a coverage threshold. Historically the author suggested 16 in 1991 and above 9 in 1992 before the modern above-10 convention settled. Chart the official band and, where a policy threshold matters, name the policy. Does a high or low Epworth diagnose or rule out sleep apnea? Neither. The ESS measures average sleep propensity, and across 35 studies its correlation with sleep-apnea measures ran about 0.11 to 0.23; the American Academy of Sleep Medicine's guideline says questionnaires must not be used to diagnose OSA without polysomnography or technically adequate home testing, and Australia's Austroads standard says the ESS is neither sensitive nor specific for OSA. A high score raises the question of cause (insufficient sleep, circadian misalignment, sedating drugs, mood, sleep-disordered breathing, hypersomnolence disorders); a low score can reflect limited awareness of dozing, fatigue without recognized dozing, adaptation, strategic reporting, or an OSA phenotype without much subjective sleepiness. "My doctor says a low Epworth proves I don't have sleep apnea" is a common patient report and it is wrong; testing decisions use the full history, examination, the STOP-Bang or another risk screen, and objective testing criteria. Can we print the Epworth, put it in our EHR, or use a free online calculator? Only under a licence. The copyright belongs to the estate of Dr Murray W. Johns and Mapi Research Trust administers it worldwide; the official position is that a licence is required for every use, whether or not a licence fee is payable. Students, physicians in individual clinical practice, and non-funded academic users may register for direct access; funded academic projects, healthcare organizations, commercial users, and IT companies submit a request and fees may apply; electronic implementations (an EHR flowsheet, portal questionnaire, app, REDCap build, or vendor scoring component) require an agreement for all user classes with implementation and screenshot review; and there is no public price list because fees are quoted per project, language, and scope. A free public calculator on a commercial site is reproduction and public display, not an exemption. Documenting "ESS 14/24" is different from reproducing the eight situations, and the widespread online copies, government-hosted appendices, and older permission footnotes prove circulation, not permission. Terms as stated on the official ESS and Mapi pages in August 2026. Does Medicare require the Epworth to normalize for continued PAP coverage? No. The current Medicare PAP local coverage determination conditions continued coverage on a clinical reevaluation no earlier than day 31 and no later than day 91 after initiation, documentation that OSA symptoms have improved, and objective use of at least 4 hours a night on 70% of nights in a consecutive 30-day period within the first three months; it does not require the ESS to fall below 11 or improve by any number of points, and its appendix's 0 to 9 "normal" label is not a coverage threshold. Medicare's sleep-testing policy likewise recognizes documented excessive sleepiness as a supporting symptom without a score. Among commercial payers, one utilization manager accepts an ESS above 10 as one route to document sleepiness in a pathway that generally needs a combination of criteria; wake-promoting-agent prior authorizations ask for diagnosis, objective findings, adequate PAP, and clinical response, or a validated sleepiness scale, without a uniform numeric ESS rule in current policies. The reevaluation visit is a sensible time to repeat the ESS, paired with the usage download, but the download and the symptom statement are what the policy reads. Do FMCSA, the FAA, or Canadian and Australian driver authorities require an Epworth score? No authority reviewed sets an ESS number as a licensing or certification rule. The US federal standard for commercial drivers (49 CFR 391.41(b)(5)) bars a respiratory dysfunction likely to interfere with safe driving and names no OSA test or score, and FMCSA's 2024 handbook states its regulations prescribe no screening method, referral rule, testing, treatment, or compliance threshold; the FAA's May 2026 OSA materials treat untreated or poorly controlled OSA as potentially disqualifying, set no ESS cutoff, and use a PAP expectation of at least 75% of sleep periods averaging at least six hours, stricter than the Medicare coverage minimum. Canada's June 2026 CCMTA model standard treats the ESS as subjective and decides on severity, admitted sleepiness, crashes, and treatment (and, oddly, misdescribes the response scale as 1 to 4); provincial reporting duties turn on impairment, not a total. Australia's Austroads standard says the ESS depends on honest reporting and is one component of assessment, with state notification duties keyed to impairment. Chart the score, the direct safety inquiry, and the individualized reasoning under the applicable rule, and where wakefulness in safety-sensitive work is the question consider objective testing such as the MWT. How much change on the Epworth is meaningful, and how often should I repeat it? Describe the change; do not over-claim it. OSA trial analyses estimate a group-level minimally important difference of about 2 points (639 participants across three trials) and 2 to 3 points, but in a short-interval repeatability study of 40 sleep-clinic patients, sequential scores differed by at least 2 points in 63%, 3 in 48%, 5 in 20%, and 7 in 8%, with a standard error near 2 points, so a 2-point change is meaningful in a group and unproven in an individual. Write "ESS decreased by 3 points" (factual) and "within published group-level estimates of meaningful change" (defensible), and rest a response claim on objective PAP use and efficacy, function, and safety history together; review sleep opportunity, schedule, and medication before calling a rise treatment failure. There is no ESS-only interval: a pretreatment baseline, an early follow-up when adherence and efficacy are reviewed (the Medicare day-31-to-91 window is a natural point), a repeat at meaningful changes in symptoms, schedule, medication, or treatment, and at least annually in ongoing OSA care per the AASM quality framework, always on the same version and mode. The patient is minimizing because of a licence or job, or the partner wants to fill it in. How do I document that? Record the patient's own score exactly and label it self-report; never substitute the clinician's estimate or edit answers to match the impression. Chart inconsistencies factually (a low ESS beside acknowledged sleep-at-the-wheel episodes), obtain collateral separately with its source, ask directly about involuntary sleep, near-misses, crashes, work incidents, and countermeasures, review sleep duration, schedule, medications, substances, and treatment data, and consider objective testing such as the MWT when the question is ability to stay awake in a safety-sensitive setting; then apply the jurisdiction's actual reporting law, which turns on impairment or danger, not a number. When 122 patients were told about licensing implications the group mean did not change but about 9% crossed the classification boundary, so the concern is real without every low score being dishonest. The adult ESS is patient-reported: a partner may help with reading or comprehension if that is documented, partner observations go in as collateral, and a proxy score is labelled as such and kept separate; the ESS-CHAD expressly allows more parent or carer involvement for younger children. Can BastionGPT draft my Epworth Sleepiness Scale documentation? Yes. Give it the facts (version and mode, respondent and assistance, total and validity, sleep opportunity, schedule, medications and substances, safety history, prior score and date, device data, and the plan) and it drafts the full entry: the score out of 24 with its official band, the context and safety lines, a modest interpretation, the change described against the prior result with the objective data beside it, and the plan with the next administration dated, ready for your review. It can also check a finished note for a score charted as an apnea finding, a 10 called abnormal, a total with no version or respondent, response declared from a 2-point change with no usage data, and a safety history never asked, and it never needs the scale's items to do so. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Johns MW, 1991, Sleep, a new method for measuring daytime sleepiness ; Johns MW, 1992, Sleep, reliability and factor analysis of the ESS (test-retest 0.82; alpha 0.88); Johns MW, Hocking B, 1997, Sleep, daytime sleepiness and sleep habits of Australian workers (507 workers). Official ESS site, about the ESS (1997 standard version; scoring integrity; official bands; recall frame; ESS-CHAD) and licences (a licence for every use, with or without a fee); Dr Murray Johns estate, contact page (estate owns the copyright; enquiries to Mapi); Mapi Research Trust ePROVIDE, ESS and ESS-CHAD (user classes; electronic agreements; July 2026 catalogue). Kendzerska TB and colleagues, 2014, systematic review of ESS measurement properties (35 studies; correlations with MWT, MSLT, and apnea measures); Crook S and colleagues, 2019, Thorax, minimum important difference of the ESS in OSA (639 participants); Patel S and colleagues, 2018, ESS minimally important difference in CPAP-treated OSA ; Grewe FA and colleagues, 2020, low repeatability of the ESS (40 patients); Ayeni A and colleagues, 2019, fitness-to-drive framing study (122 patients); Sanford SD and colleagues, 2006, US community norms ; Sander C and colleagues, 2016, German population norms (9,711). American Academy of Sleep Medicine, quality measures for the care of adult patients with OSA (sleepiness assessment at evaluation and annually) and MSLT and MWT protocol guidance ; central disorders of hypersomnolence assessment . CMS, LCD L33718, positive airway pressure devices for OSA (day-31-to-91 reevaluation; 4-hour, 70% usage; appendix reproducing the scale with a 0 to 9 label) and LCD L35050, polysomnography and sleep testing ; Carelon, sleep disorder management (November 2025, updated July 2026) and September 2026 edition ; CVS Caremark, modafinil criteria ; Highmark, modafinil prior authorization form ; DailyMed, modafinil labeling . eCFR, 49 CFR 391.41 ; FMCSA, Medical Examiner's Handbook, 2024 edition ; FAA, OSA reference materials for aviation medical examiners (May 2026). CCMTA, National Safety Code Standard 6, June 2026 ; RoadSafetyBC, sleep disorders medical standard ; Ontario, reporting for driver medical review ; SAAQ, medical examination form . Austroads, Assessing Fitness to Drive, 2022 ; MBS Online, item 12250 (ESS 8 or more with an OSA-risk instrument; questionnaires retained for audit; bare item numbers only); Transport Victoria, sleep apnoea and driving ; PBS, armodafinil listing . Comparators: Chung F and colleagues, 2008, the STOP-Bang questionnaire (OSA risk, not sleepiness). ## Related templates STOP-Bang Score Documentation Insomnia Severity Index Documentation Medication Management Follow-Up Note Fitness-for-Duty Evaluation Return-to-Work & Accommodation Letter Hospital Consult & Liaison Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/insomnia-severity-index All note templates # Insomnia Severity Index Documentation: Scoring & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Insomnia Severity Index (ISI) is a seven-item self-report measure of perceived insomnia severity and impact, developed by Charles Morin and validated by Bastien and colleagues in 2001, scored 0 to 28. CBT-I therapists, primary care, and sleep clinics use it at baseline, mid-course, and completion. It measures perceived severity; it never diagnoses insomnia disorder. This page covers how to document and interpret ISI results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report (clinician and significant-other versions exist and are charted as such); scored by the treating clinician; obtained through Mapi Research Trust, with direct access for individual clinicians Audience CBT-I therapists and behavioral sleep medicine, primary care, sleep clinics, prescribers, digital-therapeutic programs, payers and auditors Typical length 3 to 8 chart lines per administration plus the diary summary · patient completion under 5 minutes Format family Self-report insomnia severity measure (7 items scored 0 to 4, total 0 to 28; one-month standard recall, two-week version for treatment monitoring) When it's used CBT-I baseline, mid-course, completion, and follow-up; insomnia screening in primary care and VA settings; digital CBT-I outcome tracking; comorbid insomnia and OSA care Standards context Copyright Charles M. Morin, distributed by Mapi: direct access for individual clinicians, licences for organizations, EHR, and web use; named as an example in AASM quality measures and VA/DoD guidance; mandated by no payer On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Insomnia Severity Index? The Insomnia Severity Index is a seven-item self-report measure of perceived insomnia severity and impact, developed by Charles M. Morin (the official catalog dates development to 1985, with copyright dates of 1993, 1996, 2000, and 2006) and validated clinically by Bastien, Vallieres, and Morin in 2001 and psychometrically by Morin and colleagues in 2011. Each item is scored 0 to 4 and the seven values sum to a total of 0 to 28, higher meaning greater perceived severity; no reverse-scoring step is needed. The total integrates the respondent's appraisal of three nocturnal problem domains, dissatisfaction with sleep, daytime interference, how noticeable the impairment seems to others, and distress about it, which is why it measures perceived severity and impact and not sleep physiology or nightly continuity. Patient, clinician, and significant-other versions exist and give complementary rather than identical information; the current catalog calls the last month the standard recall interval and accepts a two-week version, common in CBT-I monitoring, provided the same interval is used at every assessment; and the ISI-3 is a separate three-item derivative (2021), not another version of the seven-item scale. Three facts govern the documentation. First, the descriptive bands (0 to 7 no clinically significant insomnia on the scale, 8 to 14 subthreshold, 15 to 21 moderate, 22 to 28 severe) are severity labels, not diagnostic rules; the 2011 case-finding cutoffs (10 in the community sample, 11 in the clinical sample) are a different, purpose-specific thing, and insomnia disorder is established by interview. Second, the change conventions have different provenances and are never interchangeable: a six-point minimally important difference from an eszopiclone trial, an eight-point-or-more responder rule and a current-score-below-eight operational remission from Morin's CBT-I trial convention, and a mean change of about minus 8.4 for moderate improvement in the 2011 study, so the chart names the rule it invokes. Third, the ISI is copyright of Dr Morin and distributed by Mapi Research Trust: individual clinicians get direct access, while organizations, EHR builds, and public web tools need a licence. Serial results follow the outcome measure note conventions, the sleepiness question belongs to the Epworth Sleepiness Scale , and the apnea-risk question to the STOP-Bang . ## Who uses Insomnia Severity Index documentation and when CBT-I therapists and behavioral sleep medicine programs are the core users: the ISI is the perceived-severity anchor at baseline, at a clinically meaningful midpoint, at completion, and at follow-up, always beside the prospective sleep diary that drives the behavioral prescription, and its response and remission conventions are how CBT-I trials and the FDA-cleared digital CBT-I products (Somryst in 2020, Sleepio in 2024) report outcomes. Primary care and VA settings use it as a validated screen that identifies who needs further evaluation, which is how the 2025 VA/DoD guideline and Australian primary-care guidance frame it, and sleep clinics use it to track insomnia burden in comorbid insomnia and OSA, where a trial of 145 patients found CBT-I before CPAP improved insomnia symptoms and CPAP uptake. Payers and quality programs see it as an example: the AASM's 2025 insomnia quality measures name the ISI and PSQI as acceptable questionnaires and expect a clinician's global summary rather than a filed form, and Australia's Better Access program requires an outcome measure of the clinician's choosing. The score sits inside a psychotherapy progress note with the CBT-I work it informs, the goals it anchors live in the treatment plan , and medication changes that alter it belong in the medication management note . ## How to document Insomnia Severity Index results in the chart No US, Canadian, or Australian authority prescribes an ISI note format, and none makes a band diagnostic. What survives review is a versioned score with its denominator and band, a change against a named criterion, the diary summary kept as a separate data stream, a clinician's interpretation that reconciles the two, and the treatment decision that follows. Each element below carries the pitfall that most often undermines it. Version: respondent, recall, language, mode. Chart the patient, clinician, or significant-other version, the recall interval (one month standard, or the two-week version kept consistent across the episode), the authorized language and country version, the mode (paper from the distributor, licensed portal, interview), and the administration date. Pitfall: "ISI administered" with no version, a two-week and a one-month administration compared as if identical, or an ISI-3 result trended against seven-item scores. Total out of 28 with its band, kept a label. Write x/28 and the descriptive band (0 to 7, 8 to 14 subthreshold, 15 to 21 moderate, 22 to 28 severe) as a scale classification; the 2011 case-finding cutoffs (10 community, 11 clinical) are screening findings, not new bands, and insomnia disorder rests on the interview (duration, frequency, opportunity, impairment, differential). Pitfall: "ISI 19, moderate clinical insomnia" entered as the diagnosis, or a band mapped straight to a treatment ladder (15 means CBT-I, 22 means medication). Change against a named criterion. State the baseline score and date, the current score, and the arithmetic change with its sign, then name the rule: at least an eight-point reduction as the CBT-I trial responder convention, a current score below eight as operational remission, or the six-point minimally important difference estimate, citing which; the 2011 study's mean change of about minus 8.4 for moderate improvement is supporting evidence, and the five-point figure from 2024 applies to a modified six-item score, not the standard total. Pitfall: "Clinically significant improvement" with no criterion, six and eight points cited as one rule, or remission declared from a percentage. The diary, as a separate stream. Summarize the prospective diary window (nights completed; mean time in bed, total sleep time, sleep-onset latency, wake after sleep onset, terminal wakefulness, sleep efficiency, schedule regularity), and say whether it agrees with the ISI. The ISI converges moderately with diaries and weakly with polysomnography by design, so discordance is information, not error. Pitfall: "ISI improved, sleep normalized" with no diary, or a genuine drop in distress dismissed because sleep efficiency has not moved. Clinical interpretation and comorbid sleep disorders. Give the global summary the AASM quality measure expects: what drove the score (distress, interference, satisfaction, nocturnal change), daytime function, adherence to the sleep window and stimulus control, medication changes, mood, and the status of comorbid conditions; document OSA testing, PAP status, sleepiness, and safety separately, because the ISI cannot screen or grade apnea. Pitfall: A bare total filed with a form, or an ISI drop read as apnea control in a patient who also has OSA. Treatment decision and relapse plan. Record what the result changed: the sleep-window prescription, component adjustments, coordination with the prescriber, the next administration on the same version, and at completion the relapse-prevention instructions and recontact thresholds, with the classification named (operational remission, response, neither). Pitfall: A number with no consequence, or "cured" written where the note should say "operational ISI remission, diagnostic and functional status documented separately." The licence, and no items in the note. Note that the instrument was obtained through the distributor and used under its terms (direct access for an individual clinician; a project licence for an organization, EHR build, digital product, or public tool), and keep item wording, anchors, and forms out of the record; a result line is documentation, a reproduced questionnaire is not. Pitfall: The seven items retyped into a chart template or an unlicensed portal form, or a public calculator adopted because it was free. ### Blank template (copy and adapt) INSOMNIA SEVERITY INDEX DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Time point: [baseline / mid-course session __ / completion / follow-up] Version: [patient / clinician / significant-other] Recall: [one month (standard) / two weeks, kept consistent] Language + country: [ ] Mode: [paper from distributor / licensed portal / interview] Total: [ ]/28 Band: [0-7 / 8-14 subthreshold / 15-21 moderate / 22-28 severe]; a scale label, not a diagnosis Baseline: [date] [ ]/28 Change: [+/-] [ ] points Criterion named: [8-point-or-more reduction = CBT-I trial response / current score below 8 = operational remission / 6-point MID estimate] -> [met / not met] Diary (window, nights completed): TIB [ ] TST [ ] SOL [ ] WASO [ ] terminal wake [ ] SE [ ]%; schedule regularity [ ] Concordance: [ISI and diary agree / discordant: what drove the ISI change; diary completeness, schedule, meds, adherence reviewed] Interpretation: [distress, interference, satisfaction, daytime function, adherence, medication changes, mood; global summary] Diagnosis + comorbid sleep disorders: [insomnia disorder by interview; OSA testing / PAP status / sleepiness / safety documented separately] Plan: [sleep window / components / prescriber coordination / next administration (same version) / relapse plan + recontact thresholds] Licence: [obtained through distributor; used under terms; no items here] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the ISI's items, anchors, or forms. ## Sample Insomnia Severity Index documentation (fictional) Scenario: a mid-course CBT-I review where the ISI has improved by six points while the diary has not yet moved, documented as meaningful perceived improvement under a named criterion, short of the episode's response rule, with the discordance examined rather than resolved by fiat. All details are fictional. Patient: R.T., 52 · Setting: Outpatient behavioral sleep medicine, CBT-I session 4 of 6 · Clinician: E. Marsh, PsyD · Note date: 08/13/2026 Measure: Insomnia Severity Index, authorized patient version, two-week recall (the version used throughout this episode), English, paper form from the distributor, completed before the session. Today: 12/28, subthreshold band. Baseline at session 1, 07/16/2026, same version and recall: 18/28, moderate band. Change: minus 6 points. Criterion and diary: The six-point reduction exceeds the six-point minimally important difference estimate (Yang and colleagues, 2009) and does not yet meet this episode's prospectively chosen response rule of an eight-point-or-more reduction; the current score remains above the operational remission threshold of below 8. Sleep diary, 7 of 7 mornings for the past week: mean time in bed 465 minutes, total sleep time 335 minutes, sleep-onset latency 38 minutes, wake after sleep onset 72 minutes, terminal wakefulness 20 minutes, sleep efficiency about 72% (baseline week about 70%). Rise time held within 15 minutes on 6 of 7 days; time in bed still about 45 minutes above the prescribed window on 3 nights. Interpretation: Perceived improvement is real and driven by lower distress and less daytime interference (concentration at work improved, less worry about sleep) rather than by nocturnal change; the diary shows persistent initiation and maintenance difficulty and incomplete adherence to the sleep window, so ISI and diary are discordant in the expected direction. Diary completeness and clock-checking reviewed; no medication changes; no new stressors; naps denied. Insomnia disorder was established at intake by interview; OSA screening at intake was low risk and is documented separately with the sleepiness result. Plan: Continue CBT-I: sleep window held (no extension yet, given sleep efficiency under 85%), stimulus-control review with problem-solving for the late-evening drift, cognitive work on residual sleep-related worry. Repeat ISI on the same version at session 6 (completion) with a full seven-day diary; classification at completion to be stated against the same eight-point response and below-8 remission rules. Prescriber informed that no medication change is requested. Administrative: Score, version, recall, band, baseline, change, and criterion entered in the outcome-measure record with the diary summary; the questionnaire itself was obtained through the distributor and used under its clinical-practice terms, with no items reproduced in this note. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The respondent version, recall interval, language, and mode are named and held constant, so the six-point change is a real comparison and not an artifact of mixed versions. The band is charted as a scale label, the diagnosis rests on the intake interview, and the OSA and sleepiness findings are documented separately where they belong. The change is measured against named criteria with their sources, and the note says plainly which it meets and which it does not, without inventing an official cutoff. The diary is summarized as its own data stream, the discordance is examined (adherence, completeness, schedule, medications, naps) rather than smoothed over, and the interpretation says what drove the ISI. The number changes something: the sleep window is held on the diary evidence, the next administration is dated on the same version, and the completion classification is pre-committed to the same rules. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: guidelines and quality measures name it as an example, payers do not require it, and the digital-therapeutic rules sit alongside. The AASM's 2021 clinical practice guideline strongly recommends multicomponent CBT-I for chronic insomnia and treats response and remission as critical outcomes without naming an instrument or a schedule (CONVENTION); its 2025 insomnia quality measures name the ISI and PSQI as examples of validated questionnaires, expect sleep quality or satisfaction and daytime function to be assessed when insomnia is addressed, and expect the clinician's global summary or interpretation when a questionnaire or diary is filed (CONVENTION); the 2025 VA/DoD insomnia and OSA guideline names the ISI and the Athens Insomnia Scale as example screening instruments that identify who needs further evaluation, distinguishes screening from diagnosis, strongly recommends CBT-I, and states that it is not itself VA or DoD policy (CONVENTION). No national Medicare source reviewed names the ISI as required for 96127, requires that code during CBT-I, or adopts the six-, eight-, or below-eight thresholds as coverage rules; payment turns on payer, practitioner type, setting, medical necessity, documentation, and bundling (PAYER POLICY), and a score in the note does not by itself establish separate payment. Prescription digital CBT-I is a distinct layer: Somryst (cleared March 2020) and Sleepio (cleared August 2024) are FDA-cleared computerized behavioral therapy devices under 21 CFR 882.5801 whose efficacy dossiers used the ISI and diaries (LAW as to device status; the clearance confers no ISI licence and no payer mandate), Sleepio's summary used a six-point response threshold in its prespecified analysis and eight points post hoc with below-8 remission for both, and Medicare covers qualifying digital mental health treatment devices (codes G0552, G0553, G0554, bare numbers) when ordered, used incident to professional behavioral health services, and incorporated into an ongoing plan of care, with 2026 expansion to another device classification and no ISI threshold (PAYER POLICY); the CY 2027 physician fee schedule proposed rule (July 2026) contained no ISI-specific provision and should be rechecked when final. Canada, Australia, and the evidence. Canada has no current national insomnia practice guideline (a 2024 Delphi panel of 16 experts produced 37 consensus recommendations centered on CBT-I access, expert consensus rather than law), the Canada Health Act creates no psychologist-delivered CBT-I benefit or ISI requirement, and provincial and private coverage of psychology, sleep-clinic care, and digital programs varies, so a Canadian chart satisfies the actual program or contract (LAW for the Act; PAYER POLICY provincially). Australia's 2026 joint position statement holds that insomnia is diagnosed on self-reported symptoms, that comorbid insomnia needs concurrent targeted management, that sleep hygiene alone is inadequate, and that multicomponent CBT-I is first-line (CONVENTION); RACGP guidance names the ISI as a global severity measure and recommends a prospective diary alongside, warning that the questionnaire reflects symptoms rather than diagnosis and that people below a given score may still warrant CBT-I, and Australian primary-care sleep resources pair an ISI of 15 or more with the Sleep Condition Indicator and a seven-day diary for follow-up (CONVENTION); Better Access requires an evidence-based outcome measure at the clinician's discretion, names the K10 and DASS-21 as examples and not the ISI, and requires the treating professional to report assessments, treatment, and recommendations to the referrer after a course (PAYER POLICY). On the evidence: internal consistency was about .74 in the 2001 clinical sample and .90 to .91 in the 2011 community and clinical samples, a reminder that reliability belongs to scores in a sample; the 2011 study found the total correlated about .59 with diary total wake time, minus .54 with total sleep time, and minus .59 with sleep efficiency, while most polysomnography relationships were small or nonsignificant, which is why diary and objective discordance are expected; the six-point minimally important difference came from an eszopiclone trial of 828 adults whose authors cautioned about generalizability, the eight-point responder and below-eight remission rules from Morin's CBT and medication trial, and a 2024 review of 81 insomnia trials found within-person thresholds ranging from 3 to 8 points and inconsistent terminology; factor solutions vary by population (two components among 1,670 cancer patients, three factors in US college samples, a six-item two-factor model in lemborexant trials), so the clinical output remains the total; and in comorbid insomnia and OSA, a randomized trial of 145 patients found CBT-I before CPAP improved insomnia, CPAP acceptance, and average use, while the ISI still cannot screen or grade apnea. Rights are more restrictive than the free PDFs imply. The Insomnia Severity Index is copyright of Charles M. Morin (the distributor lists 1993, 1996, 2000, and 2006) and is distributed on his behalf by Mapi Research Trust through its ePROVIDE platform. Individual clinicians, physicians, students, clinical-practice users, and non-funded academics have a direct-access pathway after registration and acceptance of terms; funded academic projects, healthcare organizations, commercial users, and IT companies submit a project request and a licence agreement and fees may apply; electronic versions are treated as derivative works, so embedding the wording, response controls, or scoring in an EHR, portal, app, or digital product requires its own permission process with screenshot review and possibly fees; translation goes through the distributor's process, and self-translation by a bilingual clinician is not authorized; and no public fee schedule exists. Direct clinical access is not public domain: it confers no right to retype, redistribute, modify, translate, or put the questionnaire online, the web-based ISI was validated in 2011 but validation is not permission, and a free public calculator hosted by a commercial entity is commercial and electronic use whatever it charges. Charting "ISI 12/28, two-week patient version, administered today" is documentation of a result, not reproduction, and the authorized copyright block on the distributed form is retained on any permitted copy. The Insomnia Severity Index is copyright Charles M. Morin and distributed by Mapi Research Trust. BastionGPT is not affiliated with, or endorsed by, Dr Morin or Mapi Research Trust. This page reproduces none of the ISI's items, response anchors, or forms and describes the instrument in original prose. ↑ Back to the template and downloads ## Common Insomnia Severity Index documentation errors reviewers flag The numbers behind these errors are specific. The six-point change figure comes from an eszopiclone trial of 828 adults and the eight-point response and below-8 remission rules from a CBT and medication trial , and a 2024 review of 81 insomnia trials found within-person thresholds ranging from 3 to 8 points; the total correlates about .59 with diary wake time and mostly not at all with polysomnography; and the current distributor catalog calls one month the standard recall while accepting a consistent two-week version, so the two-week window most pages present as the only ISI is one authorized version among several. The BastionGPT Clinical Advisory Board sees the same errors most often in ISI documentation reviews: A band charted as the diagnosis. "ISI 19, moderate clinical insomnia" in the diagnosis line, or a band mapped straight to a treatment ladder. The bands are scale labels; the 2011 case-finding cutoffs (10 community, 11 clinical) are screening findings; insomnia disorder rests on the interview, and treatment follows diagnosis, duration, impairment, preference, comorbidity, and safety. The version and recall left unnamed. "ISI administered" with no respondent version, recall interval, language, or mode; a two-week and a one-month result compared as if identical; an ISI-3 trended against seven-item scores. Name the version every time and keep it constant across the episode. One "official" change threshold. Six points and eight points cited as the same rule, "clinically significant improvement" with no criterion, or the 2024 five-point figure applied to the standard total when it belongs to a modified six-item score. Name the criterion and its source with every change statement. Below 8 written as cured. A current score under 8 recorded as diagnostic remission or discharge on the number alone. Write "operational ISI remission" and document interview criteria, diary sleep, daytime function, residual symptoms, comorbid sleep disorders, and ongoing treatment separately. The diary missing or overruled. "ISI improved, sleep normalized" with no diary summary, or a genuine drop in distress dismissed because sleep efficiency has not moved. The ISI converges moderately with diaries and weakly with polysomnography by design; chart both streams and explain discordance instead of picking one as correct. The instrument in the note, or an unlicensed build in the workflow. The seven items retyped into a template or an unlicensed portal form, a self-made translation, or a free public calculator adopted because it charged nothing. Direct clinical access is not public domain; organizations, EHR builds, digital products, and public tools need the distributor's licence, and the note carries the result, not the questionnaire. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (version and recall, total, baseline and date, diary summary, adherence, medications, comorbid sleep disorders, plan) and it drafts the documentation block: the score with its band as a label, the change against the criterion you name, the diary as a separate stream, the reconciling interpretation, and the plan, ready for your review. Cross-check a finished note for the gaps reviewers flag: a band charted as the diagnosis, an unnamed version or recall, six and eight points cited as one rule, below 8 written as cured, or a diary that is missing or overruled. Draft the completion or follow-up summary: baseline and current scores on the same version, the classification against the pre-committed rules, diary and functional recovery, comorbid status, and the relapse plan with recontact thresholds, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Insomnia Severity Index scored, and what do the bands mean? Seven items are each scored 0 to 4 and summed to a total of 0 to 28; higher totals mean greater perceived insomnia severity and impact, and no reverse-scoring step is needed with the authorized form. The descriptive bands are 0 to 7 (no clinically significant insomnia on the scale), 8 to 14 (subthreshold), 15 to 21 (moderate clinical insomnia), and 22 to 28 (severe clinical insomnia). They are severity labels, not diagnostic rules: they establish nothing about duration, frequency, sleep opportunity, impairment on independent interview, substance or medication effects, or another sleep disorder. The 2011 psychometric study also derived case-finding cutoffs (10 was optimal in the community sample, with 86.1% sensitivity and 87.7% specificity; 11 in the clinical sample), which are screening findings for those samples and purposes, not new bands. Insomnia disorder is diagnosed by interview; the ISI measures how severe and impairing the patient perceives it to be. Does the ISI use a two-week or a one-month recall period? Both versions are authorized, and the chart must say which. The current distributor catalog calls the last month the standard interval and expressly accepts a two-week interval, especially for shorter treatment studies and CBT-I monitoring, provided the same interval is used at baseline, post-treatment, and follow-up; the 2011 psychometric paper used the one-month window while most CBT-I clinical forms use two weeks, which is why so many pages present two weeks as the only ISI. Patient, clinician, and significant-other versions exist and give complementary information, and the ISI-3 is a separate three-item derivative with its own two-week and one-month forms. Name the respondent version, the recall interval, the language and country version, and the mode at every administration, and never compare a two-week result with a one-month result as if they were the same measurement. What counts as response or remission on the ISI, and is it 6 or 8 points? Whichever rule your episode names, because the numbers have different provenances. The six-point figure is a minimally important difference estimated from an eszopiclone trial of 828 adults using quality-of-life, fatigue, and productivity anchors, whose authors called for study of generalizability. The eight-point-or-more reduction is the responder convention from Morin's CBT and medication trial, and a current score below 8 is that trial's operational remission rule; both reappear in the FDA summaries for digital CBT-I products, and one dossier used a six-point threshold in its prespecified analysis and eight points post hoc. The 2011 study found a mean change of about minus 8.4 for moderate improvement rated by an independent evaluator, supporting evidence rather than a rule; a five-point figure published in 2024 applies to a modified six-item score, not the standard total; and a 2024 review of 81 trials found within-person thresholds ranging from 3 to 8 points. So write "minus 6, exceeds the six-point MID estimate; does not meet this episode's eight-point response rule; above the below-8 remission threshold," and never "clinically significant" without a source. The ISI improved but the sleep diary has not. What do I document? Both results and the discordance, then the examination of it. Check diary completeness and timing, sleep opportunity and schedule, arithmetic, clock-checking, medications, adherence to the sleep window and stimulus control, acute stress, napping, and whether the ISI recall window and the diary week actually overlap. Then ask what drove the ISI change: less distress, greater satisfaction, better daytime function, less perceived interference, or true nocturnal improvement. Moderate convergence with diaries and weak convergence with polysomnography are expected because the ISI measures perceived severity and impact (in the 2011 clinical sample the total correlated about .59 with diary wake time and mostly not at all with polysomnography), so do not erase a real reduction in distress because sleep efficiency has not normalized, and do not claim sleep-continuity recovery when the diary is still impaired. A defensible sentence: "ISI improved by six points, driven clinically by lower distress and daytime interference; diary sleep efficiency and nocturnal wakefulness remain impaired, so behavioral treatment continues." Can I call it remission when the ISI is under 8? Call it operational ISI remission under the below-8 convention, and say so in those words. Then separately state whether the patient currently meets interview-based insomnia criteria, whether diary sleep and daytime function have recovered, what residual symptoms remain, whether medication or another active treatment is maintaining the result, and whether OSA, a circadian disorder, mood symptoms, or another condition is still active. The below-8 rule comes from treatment-trial conventions and the digital-CBT-I regulatory summaries; it is not a stand-alone diagnostic determination, and "cured" is not a chart word. At completion, record the diary, functional recovery, residual symptoms, comorbidity status, and the relapse plan with recontact thresholds beside the classification. Can we print the ISI, put it in our EHR, or use a free online calculator? It depends on who you are and what you are building. The ISI is copyright of Charles M. Morin and distributed by Mapi Research Trust: an individual clinician, physician, student, or non-funded academic can register for direct access and use the authorized paper form under its terms at no fee; funded academic projects, healthcare organizations, commercial users, and IT companies submit a project request, and a licence and fees may apply. Electronic versions are treated as derivative works, so embedding the wording, response controls, or scoring in an EHR flowsheet, patient portal, app, or digital therapeutic requires its own permission process with screenshot review, and a free public calculator hosted by a commercial entity is commercial and electronic use whatever it charges; the web-based ISI was validated in 2011, but validation is not permission. Direct access is not public domain: no retyping, redistribution, modification, translation, or online posting without authorization, and self-translation by a bilingual clinician is not authorized. Charting "ISI 12/28, two-week patient version" is documentation of a result, not reproduction. Terms as stated on the distributor's catalog in June 2026; no public fee schedule is published. Does any guideline or payer require the ISI, and how does it fit 96127 and digital CBT-I? Guidelines name it as an example; no payer reviewed requires it. The AASM's 2021 guideline strongly recommends multicomponent CBT-I and treats response and remission as critical outcomes without naming an instrument; its 2025 quality measures name the ISI and PSQI as acceptable questionnaires and expect a clinician's global summary rather than a filed form; the 2025 VA/DoD guideline names the ISI as an example screening instrument that identifies who needs further evaluation. No national Medicare source names the ISI as required for 96127 or adopts its thresholds as coverage rules, so billing turns on payer, practitioner, setting, medical necessity, and documentation, and a score in the note does not itself create separate payment. Prescription digital CBT-I products (Somryst, cleared 2020; Sleepio, cleared 2024) used the ISI in their FDA dossiers, and Medicare covers qualifying digital mental health treatment devices when ordered, used incident to professional services, and incorporated into an ongoing plan of care under codes G0552 through G0554, with no ISI threshold; the clearance confers no licence and no mandate. In Australia, Better Access requires an outcome measure of the clinician's choosing (the ISI is not named), and RACGP guidance recommends the ISI with a diary; in Canada, no federal rule names it. Can I use the ISI when the patient also has sleep apnea, and how does it differ from the PSQI, the Epworth, and the diary? Yes for insomnia severity and treatment monitoring, no for diagnosing or grading OSA: document apnea testing, respiratory severity, PAP or other treatment, adherence, residual sleepiness, and driving or occupational safety separately, and note that a randomized trial of 145 patients with comorbid insomnia and OSA found CBT-I delivered before CPAP improved insomnia symptoms, CPAP acceptance, and average use. The instruments answer different questions: the ISI measures retrospective perceived insomnia severity, dissatisfaction, interference, visibility, and distress; the PSQI measures broader past-month sleep quality and disturbance across duration, efficiency, disturbances, medication use, and daytime dysfunction; the Epworth Sleepiness Scale measures propensity to doze and belongs in the sleepiness and safety picture; the STOP-Bang estimates apnea risk; and the prospective diary supplies the nightly variables that prescribe and adjust CBT-I. A sophisticated insomnia record uses the ISI plus the diary, adds the Epworth when sleepiness matters and the PSQI only when broad sleep quality is the construct, and verifies each instrument's own permissions. Can BastionGPT draft my Insomnia Severity Index documentation? Yes. Give it the facts (version and recall, total, baseline and date, the criterion your episode uses, the diary summary, adherence, medications, comorbid sleep disorders, and the plan) and it drafts the full entry: the score with its band as a label, the change against the named criterion, the diary as a separate stream, the reconciling interpretation, and the plan with the next administration dated, ready for your review. It can also check a finished note for a band charted as the diagnosis, an unnamed version or recall, six and eight points cited as one rule, below 8 written as cured, and a diary that is missing or overruled, and it never needs the instrument's items to do so. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Bastien CH, Vallieres A, Morin CM, 2001, Sleep Medicine, validation of the ISI as an outcome measure for insomnia research ; Morin CM, Belleville G, Belanger L, Ivers H, 2011, Sleep, psychometric indicators to detect insomnia cases and evaluate treatment response (bands, case-finding cutoffs, diary and polysomnography correlations, minus 8.4 anchor); Morin CM and colleagues, 2009, JAMA, CBT and medication trial (eight-point response and below-8 remission convention); Yang M and colleagues, 2009, minimally important difference of the ISI (six points; 828 adults). Mapi Research Trust ePROVIDE, Insomnia Severity Index catalog page (copyright holder and distributor; respondent and recall versions; direct-access and project pathways; electronic versions; translations; June 2026) and ISI-3 ; Thorndike FP and colleagues, 2011, validation of the web-based ISI . Qin and colleagues, 2024, review of meaningful-change thresholds in 81 insomnia trials ; Lenderking WR and colleagues, 2024, meaningful within-individual change on a modified six-item score ; Savard MH and colleagues, 2005, ISI in cancer patients ; Emert SE and colleagues, 2024, ISI factor structure in college samples ; Sweetman A and colleagues, 2019, Sleep, CBT-I before CPAP in comorbid insomnia and OSA (145 patients). Edinger JD and colleagues, 2021, JCSM, AASM clinical practice guideline for behavioral and psychological treatments of chronic insomnia ; Lloyd RM and colleagues, 2025, AASM quality measures for adult insomnia ; VA/DoD, 2025, clinical practice guideline for insomnia and OSA ; Carney CE and colleagues, 2012, the consensus sleep diary . FDA, Somryst 510(k) summary K191716 (2020) and Sleepio 510(k) summary K233577 (2024); CMS, Medicare mental health coverage (March 2026; digital mental health treatment devices; bare codes only) and CY 2027 physician fee schedule proposed rule . Morin CM and colleagues, 2024, Canadian Delphi consensus on insomnia care ; Health Canada, the Canada Health Act . Sullivan and colleagues, 2026, Sleep Advances, Australian joint position statement on insomnia ; Sweetman A and colleagues, 2023, AJGP, GP assessment and management of insomnia ; Australian Government Department of Health, Disability and Ageing, Better Access fact sheet for health professionals (March 2026). Comparators: Buysse DJ and colleagues, 1989, the Pittsburgh Sleep Quality Index ; Johns MW, 1991, the Epworth Sleepiness Scale . ## Related templates Epworth Sleepiness Scale Documentation STOP-Bang Score Documentation Psychotherapy Progress Note Outcome Measure Review Note Treatment Plan Medication Management Follow-Up Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/slums All note templates # SLUMS Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The SLUMS (Saint Louis University Mental Status examination) is a 30-point cognitive screen developed by Tariq and colleagues at Saint Louis University with the St. Louis VA and published in 2006, interpreted through education-adjusted bands. Primary care, geriatrics, and VA teams use it as a no-fee alternative to the licensed MMSE. It screens; it never diagnoses. This page covers how to document SLUMS results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Health-care and social-service professionals after reviewing SLU's training video (annual review advised); interpreted by the treating clinician; official paper form used at no fee, copyright retained by SLU Audience Primary care and geriatrics teams, VA and post-acute clinicians, memory clinics and neuropsychologists, payers and case-management programs, families receiving results Typical length 4 to 8 chart lines · administration about 7 minutes (clinical team estimate) Format family Clinician-administered cognitive screen (11 items, total 0 to 30, education-adjusted three-band interpretation) When it's used Cognitive screening on patient or family concern in adults 60 and older, geriatric and post-acute baselines, Medicare wellness-visit cognitive detection, medication-authorization pathways in Alberta Standards context Free for professional clinical use on SLU's official paper form, copyright retained (not public domain); no US, Canadian, or Australian law mandates it; named by Alberta Blue Cross and one Missouri program On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the SLUMS? The Saint Louis University Mental Status examination is a clinician-administered, 11-item cognitive screen scored out of 30, published in 2006 by Syed Tariq, Nina Tumosa, John Chibnall, Mitchell Perry, and John Morley of Saint Louis University's Division of Geriatric Medicine, developed and first studied with the Geriatric Research, Education and Clinical Center at the St. Louis VA Medical Center, which is why the official form is headed VAMC SLUMS Examination. In aggregate the items sample orientation, registration and attention, delayed verbal recall, calculation, timed animal naming, working memory, clock construction, figure recognition and size judgment, and recall from a brief narrative; the developer describes the functions each item taps, but item groups are not validated stand-alone domain scores. SLU intends it for adults 60 and older, advises against use for cognitive dysfunction attributable to traumatic brain injury, developmental disability, or schizophrenia, recommends administration no more than annually in most cases, and lists 23 professional language and regional localizations (English-Canada and English-Australia among them). The current US-English form carries no version number or revision date, so a reproducible note records the form source and access date. Two facts govern the documentation. First, the official interpretation is not a single cutoff but a three-band, education-adjusted system printed on the form: for a person with a high-school education, 27 to 30 is the normal band, 21 to 26 the mild neurocognitive disorder band, and 1 to 20 the dementia band; for less than a high-school education, 25 to 30, 20 to 24, and 1 to 19. A raw 26 is therefore normal under one table and in the mild band under the other, a chart that records only "SLUMS 26/30" cannot be reproduced, and the printed labels are screening bands, not diagnoses; SLU states plainly that a diagnosis requires fuller assessment by a qualified clinician. Second, the SLUMS is not public domain, whatever many pages say (the label traces to a 2011 New England Journal of Medicine essay contrasting it with the newly enforced MMSE copyright): SLU retains the copyright and charges no fee for health professionals to use the official paper form in clinical work and research, and the LOINC entry for the instrument carries a Saint Louis University and Department of Veterans Affairs copyright notice. The licensed MMSE is the instrument it is most often chosen to replace; the MoCA and the Mini-Cog are the other common comparators, and none of the four independently establishes a neurocognitive diagnosis, capacity, or fitness to drive. ## Who uses SLUMS documentation and when Primary care and geriatrics teams use it when a patient, family member, or clinician raises a cognitive concern in an adult 60 or older, and as the cognitive-detection element of a Medicare annual wellness visit, where federal regulation requires detection of impairment by observation and report and names no instrument. VA clinicians use it heavily by convention and workflow (it was developed with the St. Louis VA GRECC and disseminated through VA training), though national VHA materials do not require it by name; post-acute and home-based primary care teams use it for baselines, while skilled-nursing assessment uses the CMS resident assessment instrument's own cognitive interview instead. Two payers and programs name it: Alberta Blue Cross accepts the SLUMS and RUDAS in place of the MMSE for dementia-medication authorization, and Missouri's home and community based services manual requires it when there is concern about a participant's ability to self-direct care, while forbidding its use as the sole basis for denying services. In Australia the 75-and-over health assessment must cover cognition without a named tool and dementia-drug subsidy remains keyed to the MMSE, and Canada's national task force recommends against instrument-based screening of asymptomatic older adults, so the indication leads the note. A positive screen routes to a neuropsychological report or a fuller memory workup, the bedside findings sit in the mental status exam , and any decision-specific question goes to a capacity evaluation , because the screen decides none of those things. ## How to document SLUMS results in the chart SLU publishes administration instructions and printed bands, not a note template, and no US, Canadian, or Australian authority prescribes one. What survives review is a record that names the form and access date, records education and the interpretation table applied, reports the total out of 30 within the printed band as a screening band, documents the conditions that could invalidate the result, keeps the interpretation to a screen with function and collateral beside it, and closes the referral loop. Each element below carries the pitfall that most often undermines it. Form, localization, and access date. Chart the official SLU form and its localization (US-English, English-Canada, another language) with the access date, since the current form carries no version number, and that it was administered in person on paper by a professional who has reviewed SLU's training video. SLU says the examination was not designed for telephone or virtual administration, that fully virtual accommodations remain unvalidated, and that item wording is not to be changed. Pitfall: "SLUMS 24" with no form, language, or mode, a video or telephone administration charted as standard, or a translation improvised at the bedside. Education recorded, table named. Record highest completed grade or years of schooling, highest credential, and country and language of schooling, then name the table applied (high-school education or less than high-school education). SLU publishes no rule for a GED, interrupted schooling, or foreign credentials; when the choice would change the band, chart the raw total with both band implications and say the interpretation rests on history and function. Pitfall: "26/30, normal" with no education line, when the same 26 sits in the mild band under the other table; a GED silently recoded as high school. Total out of 30, in the printed band, as a screen. Write x/30 and the printed screening band for the named table ("23/30; high-school table; falls in the printed 21 to 26 mild neurocognitive disorder screening band"). Do not report a percentage or a pass or fail; the form's lowest bands begin at 1, so a total of zero has no printed band and is described as profoundly abnormal performance requiring interpretation, not assigned to one. Pitfall: "Patient has mild neurocognitive disorder per SLUMS," "failed the SLUMS," or a zero forced into the dementia band the form never printed. Administration conditions and validity. Document the setting, alertness and engagement, primary language and any interpreter, corrective lenses and hearing aids in place and adequate (SLU says not to administer when the patient cannot see or hear adequately despite correction), pain, fatigue, mood, acute illness, sedation, interruptions, deviations, whether every item was completed (no partial or prorated total), and whether the result is valid and interpretable. Pitfall: A precise total charted beside inadequate hearing or an unavailable language localization; a partial score generated because the patient stopped. Pattern in prose, no domain subscores. Describe relative performance broadly (greater difficulty with delayed retrieval and clock construction, stronger orientation and attention) without reproducing item content, and do not chart item groups as validated stand-alone domain measures; factor studies disagree about the instrument's structure. Pitfall: "Memory 9/13, visuospatial 3/5" reported as domain scores, or the answers and stimuli written into the note. Screen, not diagnosis, with function and collateral beside it. State that the SLUMS is a screening measure and does not independently establish a neurocognitive diagnosis, decisional incapacity, or driving unfitness; then record patient and informant concerns, change from baseline, medication and medical review, delirium and mood assessment, ADL and IADL status, and safety (medications, finances, driving, wandering, firearms). Pitfall: A normal band used to close the question when the spouse reports new medication errors, or a low band written up as dementia. Referral chain, owner, and the record. Document collateral history, functional assessment, reversible contributors, indicated examination and laboratory workup, referral for comprehensive cognitive, geriatric, neurologic, or neuropsychological assessment where concern persists, a decision-specific capacity or driving evaluation where those questions arise, the owner and date, and the completed form scanned to the record; a declined screen is charted as offered and declined with the alternative assessment used, never as a zero or a band. Pitfall: A positive screen that ends at a number, or a refusal entered as 0/30. ### Blank template (copy and adapt) SLUMS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Form: [official SLU form; localization: US-English / English-Canada / other language] Accessed: [date] Mode: [in person, paper] Trained administrator: [training video reviewed; annual review date] Education: [highest grade / years; credential; country + language of schooling; GED or foreign credential noted] Table applied: [high-school education / less than high-school education] Conditions: [quiet setting; alert and engaged; primary language; interpreter; lenses and hearing aids adequate; pain, fatigue, mood, illness, sedation; interruptions or deviations; all items completed] Validity: [valid and interpretable / not interpretable: reason; no partial total] Total: [ ]/30 Printed band for the named table: [normal 27-30 or 25-30 / mild neurocognitive disorder 21-26 or 20-24 / dementia 1-20 or 1-19]; zero: no printed band, described not assigned Band under the other table (if education uncertain): [ ] Pattern (prose, no subscores): [ ] Interpretation: [screening band only; does not establish a neurocognitive diagnosis, incapacity, or driving unfitness] Concerns + function: [patient / informant concerns; change from baseline; ADLs, IADLs; safety: medications, finances, driving] Reversible contributors: [delirium, depression, medications, sensory, metabolic] Plan: [collateral / functional assessment / workup / referral / capacity or driving evaluation if indicated] Owner + date: [ ] Record: [completed form scanned] Declined: [offered + declined + alternative assessment; never a score] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the examination's items, stimuli, or scoring instructions. ## Sample SLUMS documentation (fictional) Scenario: a cognitive-concern visit in geriatric primary care where the raw score lands in the printed normal band under the correct education table but would sit in the mild band under the other, and a spouse's report of medication errors keeps the evaluation open. All details are fictional. Patient: W.H., 79 · Setting: Geriatric primary care clinic, cognitive concern visit · Clinician: T. Nakamura, DO · Note date: 08/12/2026 Screening: Saint Louis University Mental Status examination, official SLU US-English paper form (accessed 08/2026), administered in person in a quiet private room by the clinic's nurse (training video reviewed 02/2026), scored and interpreted by me. Indication: spouse's concern about medication management raised at the last visit. Patient alert and engaged; corrective lenses and hearing aids in place and adequate; primary language English, no interpreter; no acute illness, no sedating medication changes; all items completed. Education: 10 completed years, no high-school credential, US schooling in English; less-than-high-school table applied. Result: Total 26/30, which falls within the printed normal screening band of 25 to 30 for the less-than-high-school table. For transparency: the same total would fall in the printed 21 to 26 mild neurocognitive disorder band under the high-school table. Relative pattern: greater difficulty with delayed verbal retrieval, stronger performance across the other broadly sampled functions; no domain subscores reported. Prior formal screen: none on record. Result is a screening finding and does not establish or exclude a neurocognitive diagnosis, decisional incapacity, or driving unfitness. Concerns and function: With the patient's permission the spouse was interviewed: about six months of a pill organizer found double-filled or skipped, two missed appointments, and repeating questions; patient acknowledges occasional forgetting, denies getting lost, manages meals and self-care, drives locally, and handles cash but the spouse now pays the bills. PHQ-2 negative; no features of delirium; sleep and hearing reviewed. Because the collateral history describes functional change despite a normal-band screen, the evaluation stays open. Plan: Medication reconciliation completed and pharmacy synchronization arranged; laboratory evaluation for reversible contributors ordered per clinic protocol; structured instrumental-activities inventory and a fuller cognitive assessment scheduled 08/26/2026 with the spouse present, memory-clinic referral to follow if concern persists; interim: spouse to supervise the pill organizer with the patient's agreement, driving discussed with a plan to assess formally at the follow-up visit; return precautions for acute change. Follow-up owner: Dr. Nakamura; nursing call within one week to confirm the appointment. Administrative: Completed official form scanned to the chart; total, table applied, band, and interpretation entered in the cognitive-screen field with the form access date. Cognitive detection for the wellness-visit component documented from observation, patient report, and spouse report; no diagnosis coded from the screen; SLU's guidance of no more than annual administration noted for the next routine screen. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The form, localization, access date, mode, and trained administrator are named, so the result is reproducible from a form that carries no version number. Education is recorded in enough detail to justify the table, the table is named, and the note shows the other table's band rather than hiding the boundary case. The band is charted as a printed screening band and the screen-not-diagnosis line is explicit; the pattern is described in prose without item content or invented subscores. Function and collateral sit beside the score and override the false reassurance of a normal band, which is exactly the case SLU's own guidance anticipates. The loop closes: reversible contributors, a dated fuller assessment with the informant present, interim safety steps, an owner, a confirmation call, and the scanned form in the record. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the requirement to look is federal, the tool is a choice, and payer acceptance is not universal. The Medicare annual wellness visit regulation (42 CFR 410.15, LAW) requires detection of cognitive impairment by direct observation with due consideration of patient report and concerns raised by family, friends, or caregivers, and names no instrument; the SLUMS may serve that component when clinically appropriate, but a chart should never say Medicare required it. The cognitive assessment and care plan service (99483, PAYER POLICY, CMS page modified May 2026) requires a far broader evaluation with an independent historian, function, safety, medications, behavioral symptoms, supports, and a written care plan, and a SLUMS result alone does not meet it; a Palmetto GBA article effective January 2026 names the Mini-Cog, GPCOG, or short MoCA for the documented cognitive assessment under that service, does not name the SLUMS, and requires the full raw scoring to be available for review, in its jurisdictions only (PAYER POLICY), so payer acceptance must be verified per policy and region. Missouri's home and community based services manual is a rare authority that names the SLUMS (PAYER POLICY, one state program): it shall be used when there is concern about a participant's ability to self-direct care, cannot be the sole basis for denying covered services, requires the last completed grade or degree to be recorded and the completed form scanned into the electronic case record, and requires further capacity evaluation. National VHA materials do not require the SLUMS by name despite its VA origins (CONVENTION), skilled-nursing assessment uses the resident assessment instrument's own cognitive interview (PAYER POLICY, with FY2027 MDS expansion effective October 2026 carrying no SLUMS element), and capacity and driving remain decision-specific evaluations that a screen can prompt but never decide (CONVENTION). Canada, Australia, and the evidence. Canada's national task force reaffirmed in 2024 its recommendation against instrument-based screening of asymptomatic community-dwelling adults 65 and older, excluding anyone with symptoms or concern (CONVENTION); Alberta Health Services recommends the SLUMS among several tools, published an April 2026 administration guide that treats the clock's two criteria as all-or-none, obtained provincial copyright permissions before implementing cognitive tools in Connect Care, and reports that Alberta Blue Cross accepts the SLUMS and RUDAS in place of the MMSE for medication authorization (CONVENTION and PAYER POLICY, one province); the June 2026 CCMTA driver-fitness model standard says no cognitive test can be the sole determinant of driving ability (CONVENTION applied through provincial LAW). Australia's MBS health assessments (items 224, 225, 226, 227, 701, 703, 705, 707) cover cognition and mood without a named tool, a March 2026 change removed outdated screening-test requirements from affected assessments, dementia-drug subsidy remains keyed to the MMSE, and Austroads uses individualized functional assessment (PAYER POLICY and CONVENTION); an English-Australia localization exists and establishes no Australian norms or endorsement. On the evidence: the 2006 study of 702 VA patients (mean age 75) reported sensitivity and specificity of 95% and 76% for mild neurocognitive disorder and 98% and 100% for dementia in the high-school stratum, and 92% and 81% and 100% and 98% in the less-than-high-school stratum, comparable to the MMSE for dementia and better for mild impairment; a 2014 comparison in 136 educated veterans found statistically equivalent discrimination for the SLUMS, MoCA, and Short Test of Mental Status (AUC 0.74, 0.77, and 0.77 for MCI; 0.98, 0.96, and 0.97 for dementia); the validation program was concentrated in male veterans (all 357 surviving participants in a 7.5-year follow-up were men); a 2023 study of 602 Black and White veterans found Black veterans 1.99 times more likely to be classified in the dementia band; a Chinese-version study of 367 participants found normal-versus-MCI AUCs of only 0.32 to 0.54; a 2024 study of 263 neuropsychology referrals found the official cutoffs classified 55.1% correctly and proposed research thresholds (24 or below for MCI, 17 or below for dementia) that have not replaced the form; reliability data are thin (alpha .709 and one-year stability .723 in 108 veterans; an earlier report near .57; no item-level inter-rater coefficient for the clock or figure items), and a 2022 review of 20 psychometric studies advised cautious use. Report the original figures with their strata and population, and never age-adjust or transport the bands to a translation without local validation. Rights are precise: free is not public domain. Saint Louis University makes the official SLUMS form available at no fee for health-care and social-service professionals to administer in clinical work and research after reviewing its training video, and states that it retains the copyright; the LOINC entry for the instrument carries a Saint Louis University and Department of Veterans Affairs copyright and used-with-permission notice; SLU's copyright policy notes that absence of a notice does not place a work in the public domain and that reproduction, distribution, public display, and derivatives are owner-controlled; and the widely repeated "public domain" label traces to a 2011 New England Journal of Medicine essay written to contrast the SLUMS with the MMSE's enforced copyright. Three uses differ: entering the total, education, band, and interpretation in a record and scanning the completed official form are ordinary clinical documentation; reproducing the items, stimuli, wording, or scoring logic inside an EHR, or building a derivative electronic form, has no public authorization on SLU's current page (Alberta obtained permissions before its EHR build); and a commercial public web test that reproduces the examination has no published licence, while a score-only calculator that accepts a total and an education category without displaying items is a lower-risk question that SLU has not endorsed. Wording is not to be changed, translations come from SLU's 23 localizations rather than the bedside, and telephone or virtual administration is outside the design. The Saint Louis University Mental Status examination is copyright of Saint Louis University, developed with the St. Louis VA Geriatric Research, Education and Clinical Center, and the official form is available from SLU. BastionGPT is not affiliated with, or endorsed by, Saint Louis University or the Department of Veterans Affairs. This page reproduces no items, stimuli, or scoring instructions and describes the instrument in original prose. ↑ Back to the template and downloads ## Common SLUMS documentation errors reviewers flag The numbers behind these errors are specific. A raw score of 26 is in the printed normal band under the less-than-high-school table and in the mild neurocognitive disorder band under the high-school table; the form's lowest bands begin at 1, leaving zero unassigned; the original 2006 study of 702 VA patients reported 95% sensitivity for mild impairment in the high-school stratum while a 2024 study of 263 neuropsychology referrals found the official cutoffs classified 55.1% correctly and Black veterans were 1.99 times more likely to land in the dementia band; and SLU retains the copyright it does not charge for. The BastionGPT Clinical Advisory Board sees the same errors most often in SLUMS documentation reviews: The education table left unnamed. "SLUMS 26/30, normal" with no education line, when the official interpretation depends on which of two tables applies. Record grade, credential, country and language of schooling, name the table, and when a GED, interrupted schooling, or foreign credential makes the choice uncertain, chart both band implications and rest the interpretation on history and function. A printed band charted as a diagnosis. "Mild neurocognitive disorder per SLUMS," "dementia score," or "failed the SLUMS." Write "falls within the printed 21 to 26 screening band" and the statement that the screen does not establish a diagnosis, incapacity, or driving unfitness; diagnosis needs decline, function, exclusions, collateral, and clinical synthesis. The lower bound rewritten. A total of zero forced into a "0 to 20" dementia band the form does not print, or secondary tables copied with the bounds silently changed. The printed bands begin at 1; describe a zero as profoundly abnormal performance requiring interpretation, and quote the official table as printed. An invalid attempt scored. A total generated with inadequate hearing or vision despite correction, an unavailable language localization, an improvised translation, a telephone or video administration, or missing items prorated. SLU says stop rather than change wording, and Alberta's guide says no partial score; chart attempted, not interpretable, and the reason. A normal band used to close the question. "25/30, normal, no concerns" when a spouse describes six months of medication errors, or a low band written up with no informant, no reversible-contributor review, no referral, and no owner. Function and collateral sit beside the score; the loop closes with a dated next step. "Public domain" and everything that follows from it. The examination retyped into an EHR template or a public web tool because it was free, wording edited, domain subscores invented, or a translation done at the bedside. SLU retains the copyright and permits no-fee professional paper use; scan the completed form, chart the result in prose, and obtain permission before item-level digital reproduction. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (form and localization, education and table, conditions, total, pattern, informant report, function, prior results) and it drafts the documentation block: the band for the named table with the other table's band where relevant, the validity conclusion, the screen-not-diagnosis line, function and collateral, and the referral chain with owner and date, ready for your review. Cross-check a finished note for the gaps reviewers flag: an unnamed education table, a band charted as a diagnosis, a zero forced into a band, an invalidated attempt scored, a normal band used to dismiss reported decline, or a positive screen with no plan. Draft the follow-up documentation: prior total, table, and access date, the interval change, what the fuller assessment found, and the reconciled plan, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the SLUMS scored, and what do the bands mean? Eleven items are scored to a total out of 30, sampling orientation, registration and attention, delayed verbal recall, calculation, timed animal naming, working memory, clock construction, figure recognition and size judgment, and recall from a brief narrative. Interpretation is education-adjusted, with two tables printed on the official form: with a high-school education, 27 to 30 is the normal band, 21 to 26 the mild neurocognitive disorder band, and 1 to 20 the dementia band; with less than a high-school education, 25 to 30, 20 to 24, and 1 to 19. The labels are screening bands derived from the original 2006 study's clinical classifications, not diagnoses, and SLU states that a diagnosis requires fuller assessment by a qualified clinician. Two details most pages miss: the lowest bands begin at 1, so a total of zero has no printed band, and a raw 26 changes bands solely with the education table, which is why the table is always named. Do not report a percentage, a pass or fail, or item-group subscores as validated domain measures. Which education table do I use, and what about a GED or foreign schooling? The official form offers two categories, high-school education and less than high-school education, and SLU asks administrators to record years of schooling or highest degree; it publishes no rule for a GED obtained later, interrupted schooling, foreign credentials, learning disability, or language of instruction. So document what you know ("completed grade 10; GED at 32; English-language US schooling; reads independently"), name the table you applied, and when the choice would change the band, chart the raw total with both band implications and say the interpretation rests on history and function ("26/30: normal band under the less-than-high-school table, mild band under the high-school table; interpretation qualified"). Do not convert a foreign credential to "high school" from years alone unless a recognized equivalency or local policy supports it, and do not silently recode a GED. That dual statement is a documentation solution, not an SLU rule, and it is far better than hiding the uncertainty. Is the SLUMS public domain? Can we put it in our EHR or a free online calculator? No, and not without asking. Saint Louis University makes the official paper form available at no fee for health-care and social-service professionals to administer in clinical work and research after reviewing its training video, and states that it retains the copyright; the LOINC entry for the instrument carries a Saint Louis University and Department of Veterans Affairs copyright and used-with-permission notice, SLU's copyright policy says the absence of a notice does not make a work public domain, and the popular "public domain" label traces to a 2011 New England Journal of Medicine essay contrasting the SLUMS with the MMSE's enforced copyright. Three uses differ: charting the total, education, band, and interpretation and scanning the completed form are ordinary documentation; reproducing the items, stimuli, wording, or scoring logic inside an EHR or a derivative electronic form has no public authorization on SLU's current page (Alberta Health Services obtained provincial permissions before its Connect Care build); and a commercial public web test that reproduces the examination has no published licence, while a score-only calculator that takes a total and an education category is lower-risk but not SLU-endorsed. Wording may not be changed, and translations come from SLU's 23 localizations. Contact SLU at the address on its page before any item-level digital build. Is the SLUMS accepted for the Medicare wellness visit, 99483, or by other payers? It depends on the exact policy, so verify rather than assume. The Medicare annual wellness visit regulation requires detection of cognitive impairment by observation and report and names no instrument, so the SLUMS may serve that component when clinically appropriate and no test is federally mandated. The cognitive assessment and care plan service (99483) requires a much broader evaluation (independent historian, function, safety, medications, behavioral symptoms, supports, a written care plan), which a SLUMS alone cannot satisfy, and a Palmetto GBA article effective January 2026 names the Mini-Cog, GPCOG, or short MoCA for the documented cognitive assessment under that service, does not name the SLUMS, and requires the full raw scoring to be available, in its jurisdictions only. Missouri's home and community based services manual requires the SLUMS for a specific self-direction concern and forbids using it alone to deny services; Alberta Blue Cross accepts the SLUMS and RUDAS in place of the MMSE for medication authorization; Australia's health-assessment items require cognition to be assessed without a named tool, and dementia-drug subsidy there is keyed to the MMSE. "A validated cognitive screen" does not mean every payer accepts every instrument for every service. SLUMS, MMSE, MoCA, or Mini-Cog: which one, when? Choose the SLUMS when a no-fee official paper screen with an education-adjusted three-band output is wanted, mild impairment detection is an explicit goal, and the patient fits the design (60 or older, adequate sensory function, in-person administration in an available localization). Keep or choose the MMSE when continuity with prior serial scores or a payer or pathway that keys to it (Australian dementia-drug subsidy, some Canadian plans) matters and the organization has licensed access; the 2006 study found the two comparable for dementia and the SLUMS better for mild neurocognitive disorder. Choose the MoCA when a local referral, research, or payer pathway expects it or longitudinal MoCA data exist, mindful of its permission and certification rules; a 2014 comparison in educated veterans found statistically equivalent discrimination for the SLUMS, MoCA, and Short Test of Mental Status. Choose the Mini-Cog when a three-minute first-pass screen by any trained staff member is the need, accepting less multidomain narrative and no education adjustment; and consider the RUDAS when cultural, language, or educational fairness is central. None establishes a diagnosis, capacity, or driving fitness. How do I document a refusal, a partial administration, or an invalid attempt? Never as a score. For a refusal: record that the SLUMS was offered and its purpose explained, that the patient declined, the reason only if volunteered, that no items were administered and no total or band was assigned, how cognition was assessed instead (observation, history, collateral, functional review), any communication, sensory, distress, or language factors, and the follow-up plan; do not enter 0/30 or imply that refusal is evidence of impairment. For a partial administration: no prorated total, per Alberta's guide, and per SLU no wording changes to get through an item. For an invalid attempt (hearing or vision inadequate despite correction, no suitable language localization, delirium or acute illness, a telephone or video attempt): chart "attempted but not completed; no total or interpretation assigned; administration not valid because..." and the accommodations or alternative evaluation arranged. Approved enlarged visual material with lenses in place is a documented accommodation, not an invalidation. Can I administer the SLUMS by phone or video, or adjust the visual items for low vision? Not by phone or video as a standard administration: SLU states the examination was not designed for telephone or virtual administration and that a valid study of fully virtual accommodations has not been done, so a remote attempt is documented as nonstandard and not compared with the printed bands. For low vision, SLU allows corrective devices in place and enlarged visual material in specified circumstances, and Alberta's guide likewise permits enlarged versions while prohibiting partial scoring; if the patient still cannot adequately see or hear despite correction, do not administer, and chart the attempt as not completed with no score. Document the accommodation exactly ("approved enlarged visual material used; lenses in place; material adequately perceived"). Item wording is not to be changed, and there is no published SLUMS-specific inter-rater reliability for the clock or figure items; Alberta's all-or-none rule for each clock criterion standardizes scoring without being a reliability study. How accurate is the SLUMS, and in whom? Good in its original setting, less certain elsewhere. The 2006 study of 702 VA patients reported sensitivity and specificity of 95% and 76% for mild neurocognitive disorder and 98% and 100% for dementia in the high-school stratum, and 92% and 81% and 100% and 98% in the less-than-high-school stratum, comparable to the MMSE for dementia and better for mild impairment; a 2014 comparison in 136 educated veterans found statistically equivalent discrimination for the SLUMS, MoCA, and Short Test of Mental Status. The limits: the validation program was concentrated in male veterans (all 357 survivors in a 7.5-year follow-up were men); a 2023 study of 602 Black and White veterans found Black veterans 1.99 times more likely to be classified in the dementia band; a Chinese-version study of 367 participants found normal-versus-MCI AUCs of only 0.32 to 0.54 despite dementia AUCs near 0.8; a 2024 study of 263 neuropsychology referrals found the official cutoffs classified 55.1% correctly and proposed research thresholds that have not replaced the form; reliability data are thin (alpha about .71 in one veteran sample, an earlier estimate near .57); and a 2022 review of 20 psychometric studies advised cautious use pending better norms and reliability evidence. Report the numbers with their strata and population, and never transport the bands to a translation without local validation. Can BastionGPT draft my SLUMS documentation? Yes. Give it the facts (form and localization, education and the table applied, conditions, total, pattern, informant report, function, prior results, and the plan) and it drafts the full entry: the printed band for the named table with the other table's band where the education choice is uncertain, the validity conclusion, the screen-not-diagnosis line, function and collateral, and the referral chain with owner and date, ready for your review. It can also check a finished note for an unnamed education table, a band charted as a diagnosis, a zero forced into a band, an invalidated attempt scored, a normal band used to dismiss reported decline, and a positive screen with no plan, and it never needs the examination's items to do so. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Tariq SH, Tumosa N, Chibnall JT, Perry MH, Morley JE, 2006, American Journal of Geriatric Psychiatry, the SLUMS validation study (702 VA patients; education-stratified sensitivity and specificity; comparison with the MMSE); Saint Louis University, SLUMS page (free exam for professional use after training; population, cautions, no phone or virtual administration, no wording changes; 23 localizations) and official form (printed education-adjusted bands; no version number); SLU legacy assessment-tools FAQ as indexed, copyright retained, no charge for health professionals ; SLU, copyright compliance and use policy ; LOINC, SLUMS total score entry (SLU and VA copyright notice). Cummings-Vaughn LA and colleagues, 2014, SLUMS, MoCA, and Short Test of Mental Status compared (136 veterans); Roberg BL and colleagues, 2023, race and SLUMS classification in veterans (602 veterans); Yang and colleagues, 2021, Chinese-version validation (367 participants); Merz ZC, Lace JW, 2024, SLUMS classification against neuropsychological conclusions (263 referrals); Noyes and colleagues, 2022, reliability in home-based primary care veterans ; Spencer RJ and colleagues, 2022, Clinical Gerontologist, systematic review of SLUMS psychometrics (20 studies); 2026 factor-structure study in neuropsychology referrals ; 7.5-year follow-up of the validation cohort . eCFR, 42 CFR 410.15 (annual wellness visit; no instrument named); CMS, cognitive assessment and care plan services (99483 elements) and Palmetto GBA article A59036 (named cognitive tools; raw scoring available for review); CMS, resident assessment instrument manual and FY 2027 SNF final rule ; Missouri DHSS, home and community based services manual (SLUMS for self-direction concerns; not the sole basis for denial; scanned form); NHTSA, clinician's guide to assessing and counseling older drivers . Canadian Task Force on Preventive Health Care, cognitive impairment guideline (reaffirmed 2024); Alberta Health Services, cognitive screening FAQ (copyright permissions; Connect Care; Alberta Blue Cross acceptance) and SLUMS administration guide (April 2026; education recorded; no partial score; all-or-none clock criteria); CCMTA, National Safety Code Standard 6 (June 2026). MBS Online, note AN.0.39 (health assessments; cognition and mood; no instrument named) and March 2026 changes to health-assessment items ; RACGP, dementia guidance ; Austroads, Assessing Fitness to Drive . Comparator rights and evidence: PAR, MMSE permission requirements ; MoCA Cognition, training and certification ; Mini-Cog, permission terms ; Storey JE and colleagues, 2004, the RUDAS . ## Related templates Mini-Cog Score Documentation MoCA Score Documentation MMSE Score Documentation Neuropsychological Evaluation Report Capacity & Competency Evaluation Mental Status Exam (MSE) Note Clock Drawing Test Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/stroop-test All note templates # Stroop Test Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Stroop test is a family of timed color-word interference measures, not one instrument: Golden, Victoria, Comalli and Kaplan, Trenerry SNST, and the D-KEFS Color-Word Interference Test each use different scores and norms. Neuropsychologists and psychologists use them to assess performance under response conflict. A score is uninterpretable without its version and norm source. This page covers how to document Stroop results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists and neuropsychologists select and interpret it; trained psychometrists may administer and score under supervision; purchaser qualification level B or C depending on version and seller Audience Neuropsychologists and psychologists, referring physicians, ADHD and dementia evaluators, schools and disability reviewers, courts and attorneys, payers reviewing testing claims Typical length 4 to 10 chart lines or a short report paragraph · administration about 4 to 10 minutes per version, plus scoring and integration Format family Timed color-word interference test family (fixed-time counts or fixed-card completion times, depending on version) When it's used Neuropsychological and psychoeducational batteries, ADHD and executive function evaluations, dementia workups, medicolegal, driving fitness, and fitness-for-duty assessments Standards context Version-specific products from Stoelting, PAR, and Pearson plus academic versions; no authority mandates any version by name; described here for documentation purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Stroop test? The Stroop test is any of several standardized clinical instruments built on the interference effect J. R. Stroop published in 1935: reading a color word is more automatic than naming the ink color it is printed in, so naming the ink color of an incongruent color word costs time and errors. The 1935 article appears to be in the US public domain, but every modern clinical packaging is proprietary. The versions in current clinical use differ in structure, raw metric, and rights holder: the Golden Stroop Color and Word Test, whose current edition is the Normative Update (Golden, Freshwater, Syzdek, and Ailes; Stoelting lists a 2022 publish date, adult and children's versions, with PAR and WPS as distributors); the Victoria modification (Regard, 1981), a brief three-card version sold through the University of Victoria Psychology Clinic to verified professionals; the Comalli and Kaplan lineage (Comalli, Wapner, and Werner, 1962), which has no current commercial kit; the Stroop Neuropsychological Screening Test (Trenerry and colleagues, 1989; PAR); and the D-KEFS Color-Word Interference Test (Pearson, 2001), joined in July 2025 by the separately normed, digital-only D-KEFS Advanced. As of August 2026 no successor edition has been announced for the Golden, Victoria, or SNST products. The load-bearing fact for documentation is that there is no such thing as a generic Stroop score. Golden and the SNST count items completed in a fixed time; Victoria, Comalli and Kaplan, and the D-KEFS time completion of a fixed display; derived interference is a predicted-score residual in Golden, a ratio or difference in Victoria and legacy versions, and an age-scaled contrast score in the D-KEFS. A chart entry of "Stroop = 42, impaired" could be a count, a time in seconds, a T score, a scaled score, or a derived index, and nothing in the number says which. That is why a defensible note names the version, edition, publisher, norm source, and unit for every score it reports. The D-KEFS report page covers the full nine-test battery this family's four-condition member belongs to, and the Trail Making Test page covers the set-shifting task it is most often paired with. ## Who uses Stroop test documentation and when Neuropsychologists and clinical psychologists administer a Stroop version inside most adult executive function batteries: ADHD and executive complaints, dementia and mild cognitive impairment workups, traumatic brain injury, epilepsy and movement disorder evaluations, and medicolegal or fitness-for-duty questions where response control matters. School and pediatric psychologists reach for the child-normed options (the Golden children's version from age 5, the D-KEFS from age 8, or the related NEPSY-II inhibition subtest covered on the NEPSY-II page ). Screening-oriented settings sometimes use the SNST for a brief adult check. The documentation reader is rarely just the author: the next evaluator needs the version and norms to test for change, payers reviewing 96130 to 96139 claims need the medical-necessity chain, and in forensic work the version and norm choice is routinely cross-examined. Results land in a neuropsychological report or feed an ADHD evaluation report ; the RBANS page covers the screening battery it sometimes accompanies. ## How to document Stroop test results in the chart No US, Canadian, or Australian authority prescribes a Stroop documentation format, and none requires a specific version. What survives review is a record that pins the result to one named instrument, one named norm system, and explicit units, then keeps the interpretation inside what a timed interference task can support. Each element below carries the pitfall that most often undermines it. Version, edition, and publisher. Name the exact product: for example "Stroop Color and Word Test Normative Update, adult version (Stoelting)" or "D-KEFS Color-Word Interference Test (Pearson, 2001 norms)", never "the Stroop." For D-KEFS work, state whether scores come from the 2001 battery or the separately normed D-KEFS Advanced (2025). For a Comalli or Kaplan administration, cite the actual protocol and publication followed, because no current commercial product defines one. Pitfall: "Stroop" with no version. Fixed-time counts and fixed-card completion times are different scales, so an unnamed score cannot be interpreted, compared, or defended later. Norm source and corrections applied. Name the normative set actually used (publisher norms; Norman and colleagues' 2011 HNRC norms; the Revised Comprehensive Norms; Mayo's Older Americans studies; Troyer's 2006 Victoria data; a published regression source for the SNST) and which demographic corrections it applies: age only, age and education, or fuller demographic models. Say "Norman et al. 2011" rather than the ambiguous "Heaton norms." Pitfall: A T score or scaled score reported as if demographically neutral. What was corrected depends entirely on the norm system, and mismatched norms are a standard cross-examination target. Raw scores with their units, by condition. Report each administered condition with its unit: items completed in the interval (Golden, SNST) or seconds to completion (Victoria, Comalli and Kaplan, D-KEFS), plus the standardized metric (T score, scaled score, percentile) for each. Keep baseline word-reading and color-naming results in the record; they determine what the interference condition can mean. Pitfall: A bare number. "Stroop 42" could be a count, a time, or a standardized score, and a low color-word result means something different over intact versus slowed baselines. Derived interference score, named. State which derived index was computed: Golden's predicted-score residual, a Victoria-style ratio or difference, or a D-KEFS contrast score, and report it beside the condition-level scores rather than instead of them. Keep the computation in the authorized manual or scoring software; the note needs the name and the standardized result, not the formula. Pitfall: Letting a derived score do decisional work alone. Difference and contrast scores compound the error of both components; in the D-KEFS reliability analysis none of the 51 contrast coefficients reached 0.70. Errors, self-corrections, and observations. Record uncorrected errors and self-corrections separately where the version scores them, and describe the performance: impulsive starts, loss of set, slowed articulation, repeated checking, fatigue, frustration, strategy statements. A slow-and-accurate record and a fast-and-error-prone record can produce similar times with different meanings. Pitfall: Time or count charted as the whole result. Victoria and D-KEFS interpretation can turn on the error pattern, and the observations are what make a low score explainable. Validity conditions, screened and stated. Document the color-vision check before timing (congenital deficiency affects up to about 8 percent of males), corrected acuity, reading fluency and history (the paradigm presumes automatized reading), language of administration, dominance, and interpreter status, speech-motor factors, effort indicators, and any deviation from standard administration. Close with a validity conclusion: interpretable, interpretable with caution, or not interpretable. Pitfall: A color-vision, reading, language, or motor-speech barrier scored as executive impairment. Document the barrier and use a measure that does not depend on the impaired function; do not invent an adjusted score. Interpretation inside the battery. Describe the finding as performance under color-word conflict relative to the person's own baselines, state what it does not establish (no version diagnoses ADHD or dementia by itself), and integrate it with history, other executive and attention measures, and validity evidence. For serial testing, compare only within the same version and norm system, with prior exposure and interval documented. Pitfall: "Stroop confirms ADHD," or decline calculated between a Golden count and a D-KEFS time. Group-level effects do not support individual diagnosis, and there is no valid cross-version conversion. ### Blank template (copy and adapt) STROOP TEST DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Referral question / battery context: [ ] Version + edition + publisher: [Golden SCWT Normative Update (Stoelting) / Victoria / Comalli or Kaplan protocol (cite source) / SNST (PAR) / D-KEFS Color-Word Interference (Pearson; 2001 battery or Advanced 2025)] Norm source + corrections: [publisher / Norman 2011 HNRC / RCN / MOANS / Troyer 2006 / other; age / age+education / demographic model] Language + administration: [language, dominance, interpreter, paper or digital, deviations from standard procedure] Results (each condition, with units): Word reading: [raw + unit] [std score + metric] Color naming: [raw + unit] [std score + metric] Color-word / inhibition: [raw + unit] [std score + metric] Switching condition (D-KEFS): [raw + unit] [std score + metric] Derived interference index: [predicted residual / ratio / difference / contrast] Result: [ ] Errors: [uncorrected] Self-corrections: [ ] Observations: [ ] Validity: [color-vision check + method, acuity, reading fluency, language, speech-motor, effort] Conclusion: [interpretable / with caution / not interpretable: reason] Interpretation: [performance under conflict vs own baselines; what it does not establish; convergence with other measures] Prior testing: [version, date, norms; same-version change only] Plan / integration: [ ] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces stimuli, card layouts, scoring formulas, or norm values. ## Sample Stroop test documentation (fictional) Scenario: a Golden Stroop administration inside an adult ADHD evaluation, with intact reading and a normal color-vision check, documented at the level a records reviewer or later evaluator needs. All details are fictional. Patient: J.T., 28 · Setting: Outpatient neuropsychology clinic, ADHD evaluation · Clinician: R. Okafor, PhD · Note date: 08/14/2026 Measure and administration: Stroop Color and Word Test Normative Update, adult version (Stoelting), administered in English, paper format, by this examiner as part of a four-hour attention and executive function battery. English is the patient's first and dominant language; he reads for work daily and reported no reading history. Corrected vision worn; he named all test colors accurately on the pre-timing check, with no reported color-vision history. Standard administration, no deviations. Scoring used the publisher's adult age-and-education norms; T scores, mean 50, SD 10. Results: Word reading raw 96 items in the interval, T 49. Color naming raw 71, T 46. Color-word raw 34, T 37. Interference score (predicted-score residual method) T 41: observed color-word performance modestly below the level predicted from his own reading and naming rates. Two self-corrections, no uncorrected errors. He started quickly on each card, slowed visibly midway through the color-word condition, and twice paused with a whispered restart; effort was engaged throughout and embedded validity indicators across the battery were unremarkable. Validity: Standard administration in the dominant language with intact baselines, an accurate pre-timing color check, and adequate corrected acuity: result interpretable. Baseline reading and naming within expected limits, so the interference result reflects performance under response conflict rather than a reading or naming bottleneck. Interpretation: Mildly reduced efficiency when a dominant reading response must be suppressed, relative to his own intact baselines. This is a nonspecific finding: it is consistent with the referral concern but does not establish ADHD or any diagnosis by itself, and no driving, capacity, or fitness conclusion follows from it. It converges with reduced performance on the other conflict-loaded task in the battery and with collateral ratings of distractibility, while sustained attention and processing speed measures were within expected limits. Integration and comparison note: Findings integrated in the evaluation summary with developmental history, symptom ratings, collateral report, and the remaining battery. A 2021 outside report listed "Stroop within normal limits" without version, norms, or scores; that result is treated as version and norms unspecified, not interpretable for change, and no decline or improvement is calculated against it. Any retest should use this same version and norm system with the interval and prior exposure documented. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No test stimuli, formulas, or norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The version, edition, publisher, norm source, and correction variables are all named, so the next evaluator can reproduce the frame and test for change instead of guessing what "Stroop" meant. Every score carries its unit and metric, condition-level results stay beside the derived interference index, and the index is named as the predicted-residual method rather than a generic "interference score." The validity screens are documented in the order that matters: color vision checked before timing, reading and language status stated, and a validity conclusion drawn before any interpretation. The interpretation stays inside the task: reduced efficiency under response conflict, integrated with the battery and history, with the non-diagnostic boundary stated instead of implied. The unusable outside result is handled the defensible way: labeled version and norms unspecified rather than silently compared, and the retest rule (same version, same norms) is set in advance. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the requirement is a defensible rationale, not a brand. Psychological and neuropsychological testing is billed under the 96130 to 96139 code family (with 96136 and 96137 for administration and scoring, and 96146 for a single automated instrument), and Medicare contractor policy (for example LCD L34646 and its billing articles) requires documented medical necessity: the diagnostic question, why testing was needed, why these tests were selected given education, culture, sensory and physical status, fatigue, and premorbid function, the tests given, the time, and integrated findings (PAYER POLICY). No CMS policy names a Stroop version, so the chart must show why this version and norm system fit this examinee, which is exactly what the version-and-norms line provides. For ADHD referrals, the American Academy of Pediatrics' 2019 guideline states that neuropsychological testing generally does not improve diagnostic accuracy for ADHD, useful as that testing is for profiling strengths and weaknesses, so a Stroop result is adjunctive evidence, never the diagnosis (CONVENTION). In forensic and disability work, APA forensic guidelines and test-security standards make the version, edition currency, norm fit, confound handling, and derived-score restraint foreseeable cross-examination targets (CONVENTION). Driving and return-to-duty evaluations use interference tasks as supporting evidence only; no US authority sets a Stroop cutoff for either decision. Canada and Australia change the funding frame, not the documentation standard. Canadian provincial plans generally do not pay for outpatient psychological testing: the Ontario Psychological Association notes OHIP does not cover private psychological services, and access runs through hospital programs, auto insurers, workers' compensation, or private benefits, none of which names a Stroop version (PAYER POLICY, provincial). British Columbia's driver-fitness guidance treats office cognitive screening as informative while recognizing that functional driving assessment may better determine capability (REGULATORY GUIDANCE). In Australia, psychometric and neuropsychological testing sits outside the Better Access rebate structure, so Stroop-inclusive assessments are typically private, NDIS, or privately insured, and no MBS item names an instrument (PAYER POLICY); Austroads' Assessing Fitness to Drive standard is condition- and function-based and mandates no cognitive test by name (REGULATORY STANDARD). The evidence base argues for the same restraint everywhere: the apparent ADHD interference deficit moves from about 0.24 to about 1.11 standardized units depending on how interference is computed, meta-analytic work on the paradigm rejects any single-mechanism reading of the effect, and reduced interference performance should be written as consistent-with, never diagnostic-of (CONVENTION). Rights and test security reward the same precision as psychometrics. The 1935 Stroop article is a public-domain scientific finding, but the clinical packagings are not: Stoelting publishes the Golden Normative Update (PAR and WPS distribute it, and seller qualification levels differ, Level B at Stoelting versus Level C listings elsewhere, so state the seller when qualification matters); the University of Victoria Psychology Clinic sells the Victoria materials, by arrangement with the copyright holders, only to professionals with verified qualifications; PAR publishes the SNST; Pearson publishes the D-KEFS and D-KEFS Advanced. Under APA Ethics Standards 9.04 and 9.11, the patient's responses, scores, and your interpretation are test data and belong in the record; stimulus cards, record-form layouts, administration scripts, scoring constants, and conversion tables are test materials and stay out of the chart, the patient portal, and any public page, with no blanket web or EHR carve-out from any of these publishers. A practice that wants a web-based or in-EHR administration needs a written, product-specific arrangement, not an assumption. The Stroop Color and Word Test, the Stroop Neuropsychological Screening Test, and the D-KEFS are products and trademarks of their respective publishers (Stoelting Co., PAR, Inc., and NCS Pearson, Inc.). BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Stroop test documentation errors reviewers flag The numbers behind these errors are specific. In the 2008 reliability analysis of 51 D-KEFS contrast coefficients, none reached 0.70 (mean 0.27), and in clinical patients the inhibition/switching condition was not reliably harder than inhibition (57.1 percent showed an atypical pattern); the 2007 ADHD meta-analysis found the interference effect ranged from about 0.24 with difference scores to about 1.11 with time-per-item methods; and congenital color-vision deficiency affects up to about 8 percent of males . The BastionGPT Clinical Advisory Board sees the same errors most often in Stroop documentation reviews: A score with no version. "Stroop = 42, impaired." Golden and the SNST count items in a fixed time; Victoria, Comalli and Kaplan, and the D-KEFS time a fixed display; the number could also be a T score, scaled score, or derived index. Name the product, edition, and publisher, or the result is uninterpretable and, in forensic settings, impeachable. Cross-version comparison charted as change. Decline computed between a Golden fixed-time count and a D-KEFS completion-time scaled score, or between publisher norms and a research norm set. There is no valid conversion between versions or norm systems; describe qualitative convergence if warranted and label the rest not interpretable for change. "Heaton norms" and other unnamed norm sources. A demographically corrected T score with no named source, or "Heaton norms" where the verifiable source is Norman and colleagues' 2011 HNRC norms or the separately titled Revised Comprehensive Norms. Name the set and the corrections it actually applies, and never mix D-KEFS Advanced (2025) scores with 2001 battery norms. An interference score charted as a diagnosis. "Interference T 38, consistent with ADHD" as the decisive line. The measured deficit varies several-fold with the computation method, professional guidance treats testing as non-decisive for ADHD, and no version diagnoses dementia, ADHD, or fitness to drive alone. Write consistent-with language tied to the battery and history. An invalid administration scored as impairment. A color-vision deficiency, uncorrected acuity, slow or non-automatized reading, a non-dominant language, an improvised translation, or dysarthria scored as executive dysfunction. Screen and document these before timing, chart the barrier, call the attempt nonstandard or not interpretable, and use a measure that does not depend on the impaired function. A derived score doing all the work. A ratio, difference, or contrast score reported without its condition-level components, errors, and observations. Derived scores compound measurement error and none is process-pure; report baselines and the interference condition with units, keep self-corrections and uncorrected errors distinct, and let the pattern, not one index, carry the interpretation. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Give it the facts (version, edition, publisher, norm source, condition scores with units, derived index, errors, validity screens, observations) and it drafts the documentation block or report paragraph: version and norms named, units stated, validity conclusion in place, interpretation kept to consistent-with language, ready for your review. Cross-check a finished note or results section for the gaps reviewers flag: a score with no version or norm source, a derived index without its components, a cross-version comparison, missing color-vision or reading documentation, or diagnostic language a timed interference task cannot support. Draft the records-review paragraph for mixed prior testing: what each report used, which results are comparable, which are version or norms unspecified, and what the defensible retest plan is, ready to confirm against the records. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do Stroop test scores actually mean? It depends entirely on the version. The Golden test and the SNST count how many items were completed in a fixed time; Victoria, Comalli and Kaplan, and the D-KEFS Color-Word Interference Test record seconds to complete a fixed display, so a raw "42" is a count on one instrument and a time on another. Raw scores convert to the version's standardized metric (T scores for Golden, scaled scores for the D-KEFS), and a separate derived interference index estimates the cost of naming ink colors against an incongruent word: Golden predicts expected color-word performance from the person's own reading and naming speeds and scores the difference, Victoria-style approaches use ratios or differences, and the D-KEFS uses contrast scores. A defensible note reports the condition scores with units, names the derived index, and names the norm system, because none of these numbers is interpretable without that frame. Is there a cutoff score where the Stroop becomes abnormal? No universal cutoff exists, and any page or report that states one without naming a version and norm set is wrong on its face. Impairment ranges depend on the version, the raw metric, the normative sample, and the corrections that sample applies: publisher norms for the Golden Normative Update incorporate age and education, Norman and colleagues' 2011 HNRC norms apply fuller demographic models, Mayo's older-adult studies adjust for age and estimated ability, Victoria and SNST research norms have their own samples and variables, and standard D-KEFS scores correct for age only. The same raw performance can sit at different standardized levels under different systems, which is why the note names the system rather than asserting a free-floating threshold, and why consumer websites' millisecond bands for browser Stroop tasks have no clinical standing. Which Stroop version should our practice buy and document? Choose by referral mix, age range, and norm fit, then document whichever you chose by name. The Golden Normative Update (Stoelting; adult and children's versions from age 5) is a brief, inexpensive standalone with current age-and-education norms; note that seller qualification levels differ, Level B at Stoelting versus Level C on some distributor listings. The Victoria modification is the shortest fixed-card option, sold through the University of Victoria to verified professionals, with norms in the research literature rather than a publisher manual. The SNST (PAR) is a brief adult screen with two broad publisher age bands and newer independent regression norms. The D-KEFS Color-Word Interference Test (Pearson) adds an inhibition/switching condition and formal error scoring inside a full executive battery, with the digital-only, separately normed D-KEFS Advanced available since 2025. A Comalli or Kaplan administration has no current commercial kit, so cite the protocol and norms you actually followed. Can I compare a Golden result with a D-KEFS Color-Word result across evaluations? Not quantitatively. Golden scores are counts in a fixed interval with a predicted-residual interference index on T-score norms; D-KEFS scores are completion times on age-scaled norms with contrast-score architecture, and the D-KEFS Advanced adds a third, separate norm system. A records reviewer can describe a shared qualitative pattern, for example that both evaluations showed reduced efficiency under color-word conflict, but calculating decline, converting scores, or comparing percentile movement across versions has no psychometric basis. The defensible language for a prior result reported without version or norms is "version and norms unspecified, not interpretable for change." For what the full nine-test battery's scores mean, see the D-KEFS report write-up guide . Does any payer or guideline require the Stroop, and does it diagnose ADHD? No and no. US neuropsychological testing coverage (the 96130 to 96139 family) turns on documented medical necessity and clinician-justified test selection; no CMS policy names a Stroop version (PAYER POLICY). Canadian provincial plans generally do not fund outpatient psychological testing, and Australian Medicare's Better Access structure does not rebate psychometric testing, so access runs through hospitals, insurers, NDIS, or private pay, again with no named instrument (PAYER POLICY). On diagnosis, the American Academy of Pediatrics' ADHD guideline states that neuropsychological testing generally does not improve diagnostic accuracy, and the meta-analytic ADHD interference effect swings from about 0.24 to about 1.11 depending on the computation method, so a low interference score is written as consistent with executive inefficiency inside a fuller formulation, never as the diagnosis (CONVENTION). How do I document a color-blind, low-literacy, or non-native-English examinee? Screen first, then let the screen decide what the scores can mean. Check color discrimination on the actual test colors before timing (formal plates where history or behavior raises doubt) and record the method and result; congenital red-green deficiency affects up to about 8 percent of males, and an examinee who cannot reliably discriminate the hues has not shown executive impairment by failing a color condition. Reading must be automatic for the paradigm to work, so document reading history and baseline word-reading performance, and treat reduced interference over effortful reading as a reading finding, not intact inhibition. Record the language of administration, dominance, proficiency, and interpreter status; an improvised translation is a nonstandard test without norms. In each case, chart the barrier, mark the attempt interpretable with caution or not interpretable, and assess the construct with a measure that does not depend on the impaired function. Can I put Stroop stimuli in our EHR, print our own cards, or build a scoring web tool? Not with any published clinical version's materials. Under APA Ethics Standards 9.04 and 9.11, the patient's responses, scores, and your interpretation are test data and belong in the record, while stimulus cards, record-form layouts, administration scripts, scoring constants, and conversion tables are test materials: publishers treat them as protected, security-sensitive content, and none of Stoelting, PAR, or Pearson publishes a free-web or EHR reproduction carve-out. PAR states that individual Golden items are not available for licensing, and digital administration runs through written, product-specific arrangements such as Pearson's platforms. What a practice may build freely is an original Stroop-like task with independently created stimuli and its own validation, described as such, never presented or scored as the Golden, Victoria, SNST, or D-KEFS. The same boundary governs this page and its templates: structure in words, no stimuli, no norms. Can BastionGPT draft my Stroop test documentation? Yes. Give it the facts (version, edition, publisher, norm source, condition scores with units, derived index, errors and self-corrections, validity screens, observations, and prior testing) and it drafts the documentation block or report paragraph: version and norms named, units stated, validity conclusion drawn, and interpretation kept to what an interference task supports, ready for your review. It can also cross-check a finished note for the errors reviewers flag: a score with no version, an unnamed norm source, a cross-version comparison, or missing confound documentation. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher and distributor pages, accessed August 2026: Stoelting Co., Stroop Color and Word Test Normative Update (adult and children's versions, qualification, and norms description) and test disclosure policy ; PAR, Inc., Golden adult listing (distributor status; items not available for licensing) and Stroop Neuropsychological Screening Test ; University of Victoria Psychology Clinic, Victoria Stroop test sales information ; Pearson, D-KEFS and D-KEFS Advanced . Stroop JR, 1935, Journal of Experimental Psychology, Studies of interference in serial verbal reactions (the public-domain paradigm); MacLeod CM, 1991, Psychological Bulletin, half a century of research on the Stroop effect (against single-mechanism readings); Scarpina F, Tagini S, 2017, Frontiers in Psychology, the Stroop Color and Word Test (scoring-method proliferation as a validity concern). Crawford JR, Sutherland D, Garthwaite PH, 2008, Journal of the International Neuropsychological Society, reliability of D-KEFS contrast scores (none of 51 coefficients above 0.70; mean 0.27); Lippa SM, Davis RN, 2010, Archives of Clinical Neuropsychology, inhibition/switching is not always harder (57.1 percent atypical pattern in clinical patients). ADHD evidence: Lansbergen MM, Kenemans JL, van Engeland H, 2007, Neuropsychology, Stroop interference and ADHD meta-analysis (about 0.24 with difference scores versus 1.11 with time-per-item methods); van Mourik R, Oosterlaan J, Sergeant JA, 2005, Journal of Child Psychology and Psychiatry, the Stroop revisited (17 studies, 1,395 participants; method drives the finding); American Academy of Pediatrics, 2019, ADHD clinical practice guideline (testing does not improve diagnostic accuracy). Norm systems named on this page: Norman MA and colleagues, 2011, HNRC demographically corrected norms ; Steinberg BA and colleagues, 2005, Mayo's Older Americans Normative Studies ; Troyer AK, Leach L, Strauss E, 2006, Victoria Stroop adult lifespan norms ; Hankee LD and colleagues, 2016, Framingham Heart Study norms ; Kang C and colleagues, 2013, Kaplan-version older-adult norms ; 2026 SNST lifespan regression study, PMC ; Kirsebom BE and colleagues, 2024, Norwegian D-KEFS regression norms . Validity confounds: Simunovic MP, 2010, Eye, colour vision deficiency (prevalence up to about 8 percent of males); van Boxtel MP and colleagues, 2001, visual function and Stroop components ; Protopapas A and colleagues on reading automaticity, reading and interference ; Rosselli M and colleagues, 2002, and Suarez PA and colleagues, 2014, Journal of the International Neuropsychological Society, bilingual Stroop performance . United States coverage and standards: CMS Medicare Coverage Database, LCD L34646, psychological and neuropsychological testing (medical necessity and documentation elements; no instrument named); APA, specialty guidelines for forensic psychology ; APA Ethical Principles of Psychologists, Standards 9.04 and 9.11 (test data versus test materials). Canada and Australia: Ontario Psychological Association, OHIP and psychological services ; RoadSafetyBC, cognitive impairment driver-fitness standard ; MBS Online, Medicare Benefits Schedule (Better Access structure; psychometric testing not rebated as such); Austroads, Assessing Fitness to Drive . ## Related templates D-KEFS Report Write-Up Trail Making Test Score Documentation Neuropsychological Evaluation Report NEPSY-II Report Write-Up RBANS Report Write-Up ADHD Evaluation Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/trail-making-test All note templates # Trail Making Test Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Trail Making Test is a timed paper-and-pencil measure of visual search, sequencing, and set shifting: Part A connects numbered circles, Part B alternates numbers and letters. Neuropsychologists, psychologists, and occupational therapists use it in cognitive batteries and driving screens. Raw seconds mean nothing without a named norm system. This page covers how to document and interpret Trail Making results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Neuropsychologists and psychologists select and interpret it; trained psychometrists may administer and score under supervision; occupational therapists use it in driving and functional screens Audience Neuropsychologists and psychologists, referring physicians, occupational therapists and driving evaluators, licensing authorities, courts and attorneys, payers reviewing testing claims Typical length 4 to 10 chart lines or a short report paragraph · administration about 5 to 10 minutes, plus scoring and integration Format family Timed visual search and set-shifting test (two paper conditions; completion time in seconds is the raw metric) When it's used Neuropsychological batteries for dementia, TBI, ADHD, and executive complaints; older-driver and fitness-to-drive screens; serial monitoring with alternate forms Standards context Classic form from the 1944 Army battery with contested rights messaging; commercial relatives from Pearson, PRO-ED, and PAR; no authority mandates it or sets a universal cutoff; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Trail Making Test? The Trail Making Test is the most widely administered paper-and-pencil measure of visual search, sequencing speed, and set shifting. Part A asks the examinee to connect 25 numbered circles in order; Part B alternates between numbers and letters (1, A, 2, B), adding the demand of holding and switching between two sequences. The task began as John Partington's late-1930s distributed-attention test, entered the record as a component of the US Army Individual Test Battery published by the War Department in 1944, and was adopted and clinically validated by Ralph Reitan in the 1950s within the Halstead-Reitan tradition, whose laboratory later published the standard administration and scoring manual. The primary raw score for each part is elapsed completion time in seconds; there is no standardized score until a named normative system converts it. The load-bearing fact is that the classic two-part test is only one member of a family whose scores do not interchange. The D-KEFS Trail Making Test (Pearson, 2001) splits the task into five conditions with its own scaled and contrast scores, and its number-letter switching condition is not classic Part B; the digital D-KEFS Advanced version (2025) is normed separately again. The Comprehensive Trail-Making Test, Second Edition (PRO-ED, 2020) uses five trails and proprietary T scores, and the Color Trails Test and Oral Trail Making Test were built to reduce alphabet and motor demands, at the price of separate norms and no score conversion. A chart entry that says only "Trails B = 130, impaired" is therefore close to meaningless: the reviewer needs the exact form, the raw seconds, the error count, and the norm system, because the same 130 seconds is unremarkable for an 80-year-old with eight years of education and clearly abnormal for a 30-year-old graduate. The Stroop test page covers the interference measure this test is most often paired with in executive batteries. ## Who uses Trail Making Test documentation and when Neuropsychologists and clinical psychologists administer it inside nearly every adult battery: dementia and mild cognitive impairment workups, traumatic brain injury, ADHD and executive complaints, epilepsy, and medicolegal evaluations, with results integrated into a neuropsychological report . Occupational therapists and physicians use Part B in older-driver screens: the AGS and NHTSA clinician's guide keeps it in its office battery (renamed CADReS in the 2019 edition), and the Canadian Medical Association's driver's guide discusses a Part B screening trigger, in both cases as one input to a functional decision, never a pass or fail. Pediatric evaluators use age-appropriate systems instead (the D-KEFS from age 8, the CTMT-2, or the Children's Color Trails Test; the NEPSY-II page covers the pediatric battery context). The documentation reader is usually downstream: the next evaluator re-deriving a standardized score, a payer reviewing a 96130-series claim, a licensing authority weighing a driving referral, or an attorney testing whether the norms fit the examinee, which is why the raw data, error counts, and norm identification carry the note. The RBANS page covers the repeatable screening battery it often runs alongside. ## How to document Trail Making Test results in the chart No US, Canadian, or Australian authority prescribes a Trail Making documentation format, mandates the test, or sets a universal cutoff. What survives review is a record from which an experienced reader could re-derive the standardized result: the exact form, raw seconds, errors under the standard correction convention, completion status, the named norm system with its demographic inputs, and the confounds that bound interpretation. Each element below carries the pitfall that most often undermines it. Exact test identity and form. Name the instrument precisely: classic paper TMT (with vendor or source of the form), D-KEFS Trail Making Test (2001 battery or the digital Advanced version), CTMT-2, Color Trails, Oral TMT, or a named digital task, plus the language, alternate form if any, modality (paper, tablet, oral), and hand used. Digital and paper performances are not automatically equivalent, especially for Part B. Pitfall: "Trails" without a version, or a D-KEFS switching score reported as "TMT-B." The relatives use different conditions, scores, and norms, and none converts to a classic result. Raw times in seconds, both parts. Report Part A and Part B as elapsed seconds, each with completion status. Seconds are the raw datum a later reviewer needs to re-derive any standardized score, so they stay in the record even after conversion, together with the standardized metric and its direction. Pitfall: A standardized score with no raw time, or a raw time with no norm system. Either half alone cannot be checked, compared, or defended. Errors and the correction convention. Under the standard administration the examiner points out each error immediately, the examinee corrects it, and the clock keeps running, so the cost is absorbed into the time with no separate deduction. State that convention and still record the number and type of errors per part: sequencing slips versus losses of the alternating set carry different information. Pitfall: Errors left uncounted because "only time scores." Part B errors carry independent clinical information; in one lesion study the errors, not the completion time, tracked the frontal finding. Discontinuation and the ceiling, sourced. If a part is stopped, record the elapsed time at discontinuation, the prespecified ceiling and where it comes from (protocols differ: 150 or 100 seconds for Part A and 300 for Part B in common research protocols, a five-minute clinic convention, or a norm system's own rule), the errors to that point, and the reason: time, set loss, fatigue, distress, or a motor or visual barrier. Pitfall: Silently charting 300 or 301 seconds as an observed completion time. Score at ceiling only where the named norm system directs it, and say the ceiling creates a floor effect that limits severity grading. Norm system and demographic inputs. Name the normative source and its corrections, and record the inputs it used: exact age at testing and completed education at minimum. Options differ materially: Tombaugh's 2004 age-and-education strata (ages 18 to 89), the Heaton Revised Comprehensive Norms (PAR, demographic T scores), Mayo's older-adult systems and their 2024 update, population-specific sets, and national norms abroad. Pitfall: "Age norms" or "Heaton norms" without the edition and inputs, or strata from one system combined with corrections from another. The same raw seconds can shift by roughly half a standard deviation between accepted systems. Derived indexes, named and supported. If you report B minus A or B divided by A, name the formula and the normative or regression source that supports interpreting it. The difference score still tracks baseline speed and intelligence, the ratio destabilizes when Part A is unusually fast or slow, and large clinical studies found neither adds classification value over the two direct times. Pitfall: "Trails ratio 3.2, impaired." No universal B/A boundary exists, and a derived score reported without its components and source is uninterpretable. Validity qualifiers and interpretation. Document tremor, weakness, arthritis, or other motor limits with the hand used; visual acuity, field loss, or neglect and any asymmetric search behavior; literacy and alphabet familiarity for Part B (years of schooling is an imperfect proxy, and non-Latin-alphabet backgrounds change the task); medications, sleep, and engagement. Then interpret inside the battery: Part B adds set alternation to a task already loaded on search and speed, so write pattern-level conclusions, and frame any driving concern as a referral trigger, not a determination. Pitfall: All slowing read as executive dysfunction, or a driving conclusion from one office score. Guidelines in all three countries treat the test as one input that can trigger functional or on-road assessment. ### Blank template (copy and adapt) TRAIL MAKING TEST DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Referral question / battery context: [ ] Test identity: [classic paper TMT (form source/vendor) / D-KEFS TMT (2001 or Advanced 2025) / CTMT-2 / Color Trails / Oral TMT / named digital task] Form + modality: [standard or alternate form, language, paper or digital, hand used, deviations from standard administration] Part A: [ ] seconds Errors: [ ] Completed: [Y/N] Part B: [ ] seconds Errors: [ ] ([sequencing / set-loss]) Completed: [Y/N] Error convention: [errors corrected in real time; timing continued; no deduction] Discontinuation (if any): [elapsed time, prespecified ceiling + its source, errors to that point, reason; no invented completion time] Norm system: [Tombaugh 2004 / Heaton RCN 2004 (PAR) / MOANS / MOAANS / updated Mayo 2024 / other] Inputs used: [age, education, others] Standardized scores: [metric + direction; Part A, Part B] Derived index (optional): [B-A / B/A, formula named, supporting source, result] Observations: [search pattern, pencil lifts, set losses, self-corrections, pace, fatigue, frustration] Validity: [motor, visual, literacy and alphabet familiarity, language, medications, engagement] Conclusion: [interpretable / with caution / not interpretable: reason] Interpretation: [pattern within the battery; what it does not establish; driving framed as referral trigger only, if relevant] Prior testing: [date, form, modality, norms; practice effects and reliable-change framing for serial results] Plan / integration: [ ] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the test form, circle layouts, norms, or scoring tables. ## Sample Trail Making Test documentation (fictional) Scenario: a dementia workup in which Part B is discontinued at the clinic's prespecified ceiling, documented without inventing a completion time, with the driving question framed as a referral. All details are fictional. Patient: H.L., 79 · Setting: Outpatient neuropsychology clinic, cognitive decline evaluation · Clinician: M. Beaudry, PhD · Note date: 08/18/2026 Measure and administration: Trail Making Test, classic paper form, standard administration in English with the dominant right hand, administered by this examiner within a dementia evaluation battery. Corrected vision worn; no field cut on confrontation screening earlier in the battery. Sample items completed correctly on both parts after one repetition of the Part B instructions. Errors were pointed out and corrected in real time with timing continuing throughout, per the standard convention; no deduction applied. Results: Part A: 74 seconds, completed, one examiner-corrected sequencing error. Part B: discontinued at the clinic's prespecified 300-second ceiling, not completed, four examiner-corrected errors (one sequencing, three losses of the alternating set, each a return to the number sequence). Part A was scored against Tombaugh (2004) age-and-education norms (age 79, 10 years of education, documented in the scoring record), falling in the low-average range for her stratum. Because Part B was not completed, no completion-time standardized score was assigned; the discontinuation, elapsed ceiling, and error pattern are reported in place of a number, and the ceiling's floor effect is noted as limiting severity grading. No derived index was computed on an incomplete part. Observations and validity: Visual search on Part A was slow but systematic, with no lateralized omissions. On Part B she articulated the rule correctly, managed the first three transitions, then repeatedly reverted to consecutive numbers despite prompts, appearing effortful and increasingly frustrated; no tremor or motor slowing interfered with drawing. Attention was adequate and effort was engaged across the battery, with validity indicators unremarkable: the pattern is read as cognitive, not motor, visual, or effort-related. Interpretation: Marked difficulty maintaining and alternating between two sequences, beyond the generalized slowing evident on Part A, convergent with the memory and category fluency findings elsewhere in the battery and with the informant's report of declining instrumental function. The result is one component of the evaluation, not a stand-alone diagnostic finding, and no severity grade is assigned from a ceiling score. Driving and plan: Because she drives locally, the performance exceeds the screening triggers discussed in the older-driver guidance (Part B beyond three minutes with three or more errors) and supports referral, not a determination: an occupational therapy driving evaluation with on-road assessment was recommended and discussed with the patient and her son, consistent with guidance that no single office test decides fitness to drive. Full integration, diagnosis, and recommendations appear in the evaluation summary; any retest will use an alternate form with the interval and practice effects addressed by a reliable-change method. This sample is fictional and for educational purposes. It does not describe a real patient or record; the times, errors, dates, and details are invented to show documentation structure and are not clinical guidance. No test form, layout, or norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The raw seconds, error counts, completion status, norm system, and demographic inputs are all present, so a later reviewer can re-derive the Part A score and see exactly why Part B has no number. The discontinuation is handled the defensible way: elapsed time at a named, prespecified ceiling, the error pattern in place of an invented completion time, and the floor effect stated. Errors are counted and typed (sequencing versus set loss) even though the correction convention absorbs them into the time, which is what makes the set-shifting interpretation checkable. Motor, visual, and effort explanations are explicitly examined and set aside before the slowing is read as cognitive, and the interpretation stays at the pattern level inside the battery. The driving concern becomes a referral with its guidance frame named, not a pass-fail call, and the serial-testing rule (alternate form, reliable change) is set before any retest. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the external requirements are documentation requirements, not test requirements. Psychological and neuropsychological testing is billed under the 96130 to 96139 family (96136 and 96137 for administration and scoring), and the governing CMS billing articles (A57780 and A57481, current revisions effective October 2025) expect the record to show the referral question, medical necessity, the tests administered with time, scoring, interpretation, and recommendations, with test selection accounting for age, education, ethnicity, and sensory or physical limitations (PAYER POLICY); the Trail Making Test has no standalone code and no CMS policy names it. Driving is state law: licensing and medical review boards decide fitness, clinician reporting duties vary by state (LAW), and the AGS and NHTSA clinician's guide keeps Part B in its CADReS office battery as screening guidance only (CONVENTION); the American Academy of Neurology's dementia-driving guideline found the evidence insufficient for any neuropsychological test to determine driving risk by itself, and FMCSA sets no Trail Making requirement for commercial drivers. In aviation work, the FAA's current guide requires complete score reporting with the normative comparison group identified, using pilot norms when available, but its core-test list has been moved behind a secure site, so a report should not assert that the FAA requires the TMT (CONVENTION). Concussion frameworks (the Amsterdam consensus, CDC guidance, the military's progressive-return protocols) individualize return decisions and mandate no Trail Making test (CONVENTION). Canada and Australia put the same test inside different frames. Canadian driver fitness runs on provincial law with physician reporting mandatory in most provinces (LAW); the Canadian Medical Association's driver's guide states that no single office test determines fitness and discusses a Part B screening trigger of three minutes or three or more errors, drawn from a systematic review of 47 studies that itself found the underlying cutoff literature methodologically weak, with candidate cutoffs scattered from 90 to 180 seconds (CONVENTION); outpatient neuropsychological testing is generally privately funded, through auto insurers, workers' compensation, or extended benefits rather than provincial plans (PAYER POLICY). In Australia, Austroads' Assessing Fitness to Drive (sixth edition, 2022, applied by licensing authorities and under a national review announced for 2026 to 2028) does not name the Trail Making Test at all: it directs medical, functional, occupational therapy, and on-road evidence (REGULATORY STANDARD), and Australian discrepancy data show that the expected difference between Parts B and A depends on the level of Part A performance, one more reason a raw difference is not a fixed rule. Across all three countries the widely circulated absolute thresholds (a 273-second Part B "impairment" line, a universal B/A ratio of 3, a single NHTSA seconds figure) trace to secondary handouts, single protocols, or conflicting program documents rather than to any demographically corrected norm system, and should be described that way when a record must address them (CONVENTION). Rights and versions demand unusual care here, because even federal sources disagree. The classic form descends from the 1944 War Department publication, which has a strong basis for US public-domain treatment as a government work, yet one NIH institute's copyright statement describes the test as copyrighted and directs licensing through the current Halstead-Reitan vendor while another NIH page calls it public domain (and misdates the Army battery), and the vendor sells current forms, manuals, and Canadian and Australian English translations under stated use restrictions. The defensible position for a clinic is to separate three questions: the historical task structure can be described freely; a specific modern form, manual text, or translation should be treated as the vendor's product unless counsel has verified the exact edition's provenance; and the norm systems that give raw seconds meaning are unambiguously proprietary (the Heaton Revised Comprehensive Norms are a commercial PAR product, Tombaugh's tables sit in a copyrighted journal article, and D-KEFS, CTMT-2, and Color Trails norms belong to their publishers), so scores and interpretations belong in the record while reproduced forms, tables, and scoring software output stay out of public pages and portals, and a free web task may not present itself as producing classic TMT or any publisher's scores without validation and permission. One scheduled change is worth noting in serial-testing plans: Pearson has announced that support for the original D-KEFS desktop scoring software ends from 2027, while the 2001 battery and the separate D-KEFS Advanced continue. The D-KEFS is a product and trademark of NCS Pearson, Inc.; the CTMT-2 of PRO-ED, Inc.; the Color Trails Test of PAR, Inc.; Reitan-branded materials are sold through the Reitan Neuropsychology Laboratory and the Neuropsychology Center. BastionGPT is not affiliated with, or endorsed by, any of these publishers or vendors. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Trail Making Test documentation errors reviewers flag The numbers behind these errors are specific. A 2021 NACC analysis found accepted norm systems moved the same raw times by up to about half a standard deviation; the driving-cutoff systematic review of 47 studies found none justified its sample size and candidate cutoffs scattered from 90 to 180 seconds; in a 2015 lesion study Part B errors, not completion time, tracked the right frontal finding; and in 571 people with TBI derived indexes added nothing over the two direct times. The BastionGPT Clinical Advisory Board sees the same errors most often in Trail Making Test documentation reviews: A time with no norms, or norms with no inputs. "Trails B = 130 seconds, impaired." The same time is unremarkable at 80 with eight years of education and abnormal at 30 with a degree. Chart raw seconds plus the named norm system, its edition, and the exact age and education inputs, so the standardized score can be re-derived and checked. Internet cutoffs charted as authority. The circulating 78-second and 273-second "deficient" lines, a universal ratio of 3, and single NHTSA seconds figures come from secondary handouts and conflicting program documents, not from any demographically corrected norm system. If a referral source cites one, address it as an unsourced heuristic and report the normed result instead. An invented completion time at the ceiling. A discontinued Part B charted as "300 seconds" as if observed, or a percentile assigned to an unfinished part. Record discontinued status, the elapsed time, the ceiling and its source (protocols disagree: 150 or 100 seconds for Part A, 300 for Part B, five-minute conventions), the errors, and the floor effect; apply an above-ceiling code only where the named norm system directs it. Derived indexes doing decisional work. B minus A called a pure executive score, or a B/A ratio deciding impairment. The difference still tracks baseline speed and ability, the ratio destabilizes at fast or slow Part A times, large TBI studies found no added classification value, and the expected discrepancy itself varies with Part A. Report the two direct times first and any index with its formula and source. The wrong instrument's score in the classic slot. A D-KEFS switching scaled score reported as "TMT-B," a CTMT-2 composite abbreviated to "Trails," or an Oral TMT, Color Trails, or unvalidated digital task scored with paper norms. Each relative has its own conditions, norms, and scores, and none converts to a classic result; name what was actually given. Confounded slowing read as executive impairment. Tremor, hemiparesis, field loss or neglect, low literacy, or unfamiliarity with the Latin alphabet scored as set-shifting failure, or a driving determination made from one office score. Chart the confound, qualify or withhold interpretation, consider the alternatives built for these cases, and frame driving as a referral for functional assessment. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Give it the facts (form, modality, raw seconds, errors and their types, completion status, ceiling, norm system and inputs, observations, confounds) and it drafts the documentation block or report paragraph: units stated, discontinuation handled without invented times, norms named, interpretation kept to the pattern, ready for your review. Cross-check a finished note or results section for the gaps reviewers flag: a time without norms, an absolute cutoff doing decisional work, a percentile on an unfinished part, a derived index without its components, or a D-KEFS score sitting in a classic TMT slot. Draft the serial-testing or driving-referral paragraph: prior form and interval, practice-effect and reliable-change framing, or the screening-trigger language that supports an on-road referral without overstepping into a fitness determination. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What is a normal Trail Making Test score? There is no normal range without a norm system. The raw score is seconds to complete each part, and its meaning depends on age and education at minimum: published systems include Tombaugh's 2004 age-and-education strata from 911 adults aged 18 to 89, the Heaton Revised Comprehensive Norms with demographic T scores (a commercial PAR product), Mayo's older-adult systems and their 2024 update, and population-specific sets. A 2021 analysis found that accepted systems standardize the same raw times differently, with average differences reaching about half a standard deviation, so the note names the system and the inputs rather than citing a free-floating range. The widely copied absolute lines (about 29 seconds average and 78 seconds "deficient" for Part A, 75 and 273 for Part B) are unsourced heuristics that over-pathologize older and less-educated examinees. Do errors count against the Trail Making Test score? Not as a deduction, and yes as information. Under the standard administration the examiner points out each error the moment it happens, the examinee corrects it, and the clock keeps running, so the error's cost is already inside the completion time; no points are subtracted afterward. Record the count and type anyway, per part: sequencing slips differ from losses of the alternating set, and the research argues errors carry independent signal, from geriatric samples where time and errors each discriminated impairment groups to a lesion study where Part B errors, not time, tracked the right frontal finding. Two patients with identical times, one clean and one with repeated set losses, are different clinical pictures, and only the error record shows it. What do I document when a patient cannot finish Part B? Chart it as discontinued, then give the data that replaces the number: elapsed time at discontinuation, the prespecified ceiling and its source (protocols genuinely differ: a common clinic convention stops Part B at five minutes, major research protocols use 300 seconds with 150 or 100 for Part A, and some norm systems have their own above-ceiling coding), the error count and pattern to that point, whether the rule was understood, and the reason: time, set loss, fatigue, distress, or a motor or visual barrier. Do not chart the ceiling as an observed completion time or assign a completion-time percentile to an unfinished part unless the named norm system explicitly directs that coding, and say that scoring at ceiling creates a floor effect that limits severity grading among the most impaired examinees. Which norms should I use for an older adult with limited education? Use a system whose sample, age range, and corrections actually fit, and say why. For an 85-year-old with eight years of education: Tombaugh (2004) stratifies by age and education through 89; the Heaton system applies simultaneous demographic corrections but tops out around 85, putting this patient at its boundary; Mayo's older-adult systems reach into the 90s with an ability-adjusted model, with a 2024 update modeling age, sex, and education; and population-specific sets exist where they match the examinee. Never combine one system's strata with another's corrections, and remember that years of schooling is an imperfect proxy: reading ability explains performance beyond education in some groups, and Part B assumes automatic Latin-alphabet knowledge, which is why alphabet-reduced alternatives like the Color Trails Test exist for the cases where that assumption fails. Can the Trail Making Test decide whether someone is fit to drive? No, in all three countries, and the note should say what role it did play. In the US, licensing and medical review are state law, the AGS and NHTSA clinician's guide keeps Part B in its CADReS office screen as guidance, and the neurology dementia-driving guideline found no neuropsychological test sufficient to determine risk alone. In Canada, the CMA driver's guide states no single office test determines fitness and discusses a screening trigger of three minutes or three or more errors on Part B, but the systematic review behind such cutoffs found the literature methodologically weak, with candidates scattered from 90 to 180 seconds. Austroads in Australia does not name the test at all, directing medical, functional, and on-road evidence. Defensible language: performance exceeded the screening trigger and supports referral for occupational therapy and on-road assessment; the licensing authority makes the determination. Is the Trail Making Test public domain? Can we print it or build it into our EHR? The honest answer is that the rights messaging conflicts, so separate the three questions. The task descends from a 1944 US War Department publication, which gives the exact federal edition a strong basis for US public-domain treatment, and yet one NIH institute's copyright statement calls the test copyrighted and directs licensing through the current Halstead-Reitan vendor while another NIH page labels it public domain; the vendor meanwhile sells current forms, manuals, and restricted Canadian and Australian translations. Practically: the task structure can be described freely; treat any specific modern form, manual, or translation as the vendor's product unless the exact edition's provenance has been verified; and the norm tables that make seconds meaningful are unambiguously proprietary (Heaton through PAR, Tombaugh in a copyrighted journal, D-KEFS and CTMT-2 and Color Trails through their publishers). Scores, times, errors, and interpretations belong in the EHR; reproduced stimulus sheets and norm tables do not belong in portals or public pages, and a web task may not present itself as producing classic or publisher scores without validation and permission. Is the D-KEFS Trail Making Test the same as classic Trails B? No. The D-KEFS version restructures the task into five conditions (visual scanning, number sequencing, letter sequencing, number-letter switching, motor speed) on its own page layout, converts times to age-scaled scores, records error types formally, and evaluates switching through contrast scores; its number-letter switching condition is analogous to Part B but normed and scored differently, and the digital D-KEFS Advanced (2025) is normed separately again, so no D-KEFS result converts to a classic TMT score in either direction. The same boundary holds for the CTMT-2's five trails and composite T scores, the Color Trails Test, and the Oral TMT. In serial testing, say plainly that the instrument changed and describe qualitative convergence instead of calculating change. The D-KEFS report write-up guide covers the full battery's structure and scores. Can BastionGPT draft my Trail Making Test documentation? Yes. Give it the facts (form and modality, raw seconds, errors and types, completion status, any ceiling and its source, norm system and demographic inputs, observations, confounds, and prior testing) and it drafts the documentation block or report paragraph: units stated, the correction convention named, discontinuation handled without invented times, norms identified, and interpretation kept to the pattern, ready for your review. It can also cross-check a finished note for a time without norms, an absolute cutoff doing decisional work, a percentile on an unfinished part, or a D-KEFS score in a classic slot. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Lineage and administration: Smithsonian Institution, Army Individual Test Battery catalog record (War Department, 1944; National Research Council committee development); Reitan RM, 1955 and 1958 validity studies; Bowie CR, Harvey PD, 2006, Nature Protocols, administration and interpretation of the Trail Making Test (standard error-correction convention). Rights record: NIH/NINDS, TMT copyright statement (describes the test as copyrighted; licensing via the vendor); NIDA Data Share, TMT entry (labels it public domain); US Copyright Office, 1909 Copyright Act ; Neuropsychology Center, current Halstead-Reitan materials including restricted translated forms . Norm systems: Tombaugh TN, 2004, Archives of Clinical Neuropsychology, age-and-education norms, 911 adults ; PAR, Inc., Heaton Revised Comprehensive Norms ; Steinberg BA and colleagues, 2005, Mayo's Older Americans Normative Studies ; Lucas JA and colleagues, 2005, MOAANS ; Karstens AJ and colleagues, 2024, updated Mayo Normative Studies ; Patel BM and colleagues, 2021, norm-system comparison in the NACC cohort (about half a standard deviation between systems). Errors, derived indexes, and construct: Ashendorf L and colleagues, 2008, TMT errors in normal aging, MCI, and dementia ; Kopp B and colleagues, 2015, errors, not time, and right frontal lesions ; Sanchez-Cubillo I and colleagues, 2009, construct validity ; Martin TA and colleagues, 2003, ratio-score utility in TBI ; Lange RT and colleagues, 2005, derived indexes in 571 TBI cases ; Senior G and colleagues, 2018, Australian B-minus-A discrepancy data . Ceilings and protocols: NHTSA and AGS older-driver materials, TMT protocol with 150- and 300-second ceilings ; CENTER-TBI, 100- and 300-second protocol ; Ahern L and colleagues, 2019, efficiency scoring for noncompleters (research approach). Serial testing and validity: Dikmen SS and colleagues, 1999, test-retest reliability and practice effects ; Calamia M and colleagues, 2012, practice-effect meta-analysis ; LoSasso GL and colleagues, 1998, alternate-form difficulty differences ; AACN, 2021, performance-validity consensus ; Iverson GL and colleagues, 2002, extreme TMT scores as low-sensitivity red flags . Confounds and adaptations: Schneider BC, Lichtenberg PA, 2011, reading ability beyond years of education ; Waggestad TH and colleagues, 2023, Frontiers in Psychology, alphabet knowledge and misclassification ; Abeare CA and colleagues, 2019, demographic bias in raw cutoffs ; PAR, Inc., Color Trails Test ; digital-paper equivalence, computerized trail-making comparison . Driving and coverage: Roy M, Molnar F, 2013, Canadian Geriatrics Journal, systematic review of TMT driving cutoffs (47 studies); Canadian Medical Association, CMA Driver's Guide, 10th edition ; American Academy of Neurology, 2010, dementia and driving guideline ; Austroads, Assessing Fitness to Drive ; FAA, neurocognitive evaluation protocol ; CMS billing articles A57780 and A57481 ; Pearson product pages for the D-KEFS and PRO-ED for the CTMT-2 . ## Related templates D-KEFS Report Write-Up Stroop Test Score Documentation Neuropsychological Evaluation Report RBANS Report Write-Up NEPSY-II Report Write-Up Capacity & Competency Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/clock-drawing-test All note templates # Clock Drawing Test Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The clock drawing test is a brief constructional screen in which the patient draws a clock face and sets a requested time. It has no single version: at least a dozen named scoring systems use different ranges, conditions, and anchors. Primary care, geriatrics, and neuropsychology teams use it in cognitive screening. A score means nothing without its system. This page covers how to document clock drawing results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any clinician after brief familiarization (physicians, nurses, occupational therapists, psychologists); embedded versions follow their parent instrument's rules; detailed systems reward trained raters Audience Primary care and geriatrics teams, memory clinics and neuropsychologists, hospital teams, driving and capacity evaluators, families receiving results, payers and quality reviewers Typical length 3 to 8 chart lines · administration about 2 to 5 minutes, plus disclosure and planning Format family Brief constructional cognitive screen (command or copy drawing, scored by a named system; ranges run from 0 to 3 up to 0 to 20) When it's used Medicare wellness visits and memory concerns, screens alongside recall tasks, bedside cognitive checks, older-driver screens, serial monitoring of documented decline Standards context A family of scoring systems, not one test; embedded MoCA and Mini-Cog clocks governed by their rights holders; no authority mandates any system; no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the clock drawing test? The clock drawing test asks a patient to draw a clock face, place the numbers, and set the hands to a requested time, and it is best understood as a family of tasks rather than one instrument. Recorded clinical use goes back to at least 1915 with head-injured soldiers, and the modern screening role grew out of the Boston tradition in the 1980s, but standardization never converged: the major named scoring systems (Shulman, 1986; Sunderland, 1989; Mendez, 1992; Rouleau, 1992; Freedman, 1994; the CLOX executive task, 1998; Freund's driving-oriented system, 2005) use different ranges, score directions, conditions, and requested times, and the clocks embedded in the MoCA (0 to 3) and Mini-Cog (0 or 2, deliberately binary) are components of those instruments, not standalone scores. Even "Shulman score" is not self-defining: current implementations number the same six categories 0 to 5 or 1 to 6, so a charted "clock 4" cannot be read without its anchors. The load-bearing rule for documentation follows directly: a clock score is uninterpretable without five facts, the scoring system with its range and direction, the condition (command from instruction, or copy of a model), the drawing substrate (blank page or pre-drawn circle), the exact time requested, and the qualitative errors in words. Those variables change the task itself: command drawing taxes generation and planning while copy supplies the model, a pre-drawn circle removes contour planning, and the conventional "ten past eleven" is deliberate, forcing both halves of the clock face into use and requiring the spoken "ten" to be recoded to the numeral 2, which is exactly where stimulus-bound errors show themselves. This page covers the standalone clock; the MMSE page covers the screen that, contrary to common belief, contains no clock at all, and the SLUMS page covers the free instrument whose embedded clock is one of its items. ## Who uses clock drawing documentation and when Primary care and geriatrics teams use a clock, alone or inside the Mini-Cog , for the Medicare annual wellness visit's required cognitive detection and whenever a patient or family raises a memory concern; the regulation requires detection by observation and collateral report and names no tool. Memory clinics and neuropsychologists use standalone administrations, often command plus copy, because the command-copy contrast is the clock's most clinically specific signal, and they retain the drawing for later re-scoring. Hospital teams reach for it at the bedside, where an abnormal clock during acute illness documents impairment at that moment but cannot separate delirium from baseline without collateral and a delirium instrument. Driving evaluators use the clock inside the AGS and NHTSA office battery (scored with Freund's system) as a referral signal, and Canada's 2026 national driver-fitness standard names an abnormal clock as a trigger for deeper assessment while expressly rejecting any cognitive test as the sole determinant. Capacity questions stay decision-specific in all three countries, with the clock as collateral evidence only, feeding a capacity evaluation rather than deciding it; fuller workups land in a neuropsychological report . ## How to document clock drawing test results in the chart No US, Canadian, or Australian authority prescribes a clock documentation format or mandates a scoring system. What survives review is a record that fixes the five facts that give a clock score meaning, describes the errors in words, keeps the interpretation to a screen, and preserves the drawing. Each element below carries the pitfall that most often undermines it. Context and indication. Record why the clock was drawn: the wellness-visit cognitive component, a patient or family concern, a bedside check during admission, a driving screen, or part of a larger instrument or battery. If the clock sits inside a MoCA, Mini-Cog, or SLUMS, say so and chart it as that instrument's component. Pitfall: An embedded clock detached from its parent: "MoCA clock 2/3" charted as a standalone CDT, or a Mini-Cog clock re-scored with a standalone rubric its conditions were never validated for. Scoring system, range, and direction. Name the system and its anchors: "Shulman six-category system, implemented 0 to 5, higher better," "CLOX1 and CLOX2, each 0 to 15, lower worse," "Mini-Cog clock, 0 or 2." Implementations genuinely differ (Shulman circulates as both 0 to 5 and 1 to 6), so the name alone is not enough. Pitfall: "Clock 4." Without the system, range, and direction, the number cannot be interpreted, compared, or re-derived, and systems' totals (4/5, 4/7, 4/10, 4/15, 4/20) describe entirely different performances. Condition, substrate, and requested time. Chart command versus copy (and both, in order, if both were given), blank page versus pre-drawn circle, and the exact time requested. These are test specifications, not trivia: a pre-drawn circle removes contour planning, copy removes generative demands, and 11:10 exists to test both hemifields and the ten-to-2 recoding that exposes stimulus-bound errors. Pitfall: Scores compared across different conditions or times as if serial data. A copy clock is a different observation from a command clock, and norms travel with the administration that produced them. Administration integrity. Record repetitions, clarifications, cueing, restarts and which attempt was scored, spontaneous versus prompted correction, refusal, and any time limit. Keep the first attempt: abandoned contours and marginal work carry process information the final drawing hides. Pitfall: A silently discarded first clock. Protocols differ on restarts and self-correction, and a re-drawn clock scored as the only attempt misstates what happened. Score plus the errors in words. Report the numeric score, then describe the drawing: number omissions, duplications, crowding or one-sided spacing, sequence, hand placement and length, stimulus-bound responses (hands drawn to the 10 and 11), perseveration, spatial disorganization, graphic difficulty. The qualitative account is what a later clinician can actually use. Pitfall: A score with no description. The number discards the pattern, and reviewers cannot tell a planning failure from a motor one, or re-score the performance, from "3/5" alone. Confounds and validity. Document vision and hearing correction, dominant-hand motor status and tremor, language and literacy, education, analog-clock familiarity (now a demonstrated confound in younger, digital-native adults), medications, acute illness, and delirium status for inpatients. Conclude: interpretable, interpretable with caution, or not interpretable. Pitfall: An abnormal inpatient clock read as dementia during delirium, or a younger patient's unfamiliarity with analog clocks scored as cognitive impairment. Interpretation, retention, and the next step. Keep the conclusion to a screen: what the result raises, what it does not establish (no clock diagnoses dementia, decides capacity, or clears driving), and the follow-up chain with an owner: collateral history, functional review, medication reconciliation, fuller assessment or referral. Scan the drawing into the record; the image outlives the score and permits later re-scoring. Pitfall: A screen charted as a diagnosis, a normal clock used to close a question the family has kept open, or the drawing binned so only an unreviewable number remains. ### Blank template (copy and adapt) CLOCK DRAWING TEST DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Indication: [wellness visit / memory concern / bedside check / driving screen / component of instrument or battery] Standalone or embedded: [standalone / MoCA clock x/3 / Mini-Cog clock 0 or 2 / SLUMS item; embedded clocks stay with their parent instrument] Scoring system + range + direction: [e.g. Shulman six-category, 0-5, higher better / CLOX1 + CLOX2, each 0-15 / other named system] Condition: [command / copy / both, in order] Substrate: [blank page / pre-drawn circle] Time requested: [e.g. ten past eleven, 11:10] Administration integrity: [repetitions, cueing, restarts, which attempt scored, spontaneous vs prompted correction, refusal] Score: [ ] Qualitative description: [omissions, duplications, spacing, sequence, hand placement and length, stimulus-bound response, perseveration, spatial planning, graphic difficulty] Copy condition result, if given: [score + description; command-copy contrast noted] Confounds: [vision, hearing, motor and tremor, language, literacy, education, analog-clock familiarity, medications, acute illness, delirium screen result if inpatient] Validity: [interpretable / with caution / not interpretable: reason] Interpretation: [screen result only; does not establish dementia, MCI, capacity, or driving fitness; considered with history and collateral] Drawing retained: [scanned to record: Y/N; first attempt kept] Plan: [collateral, functional review, medication reconciliation, fuller assessment or referral, repeat after stabilization] Owner + date: [ ] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces any instrument's form, rubric text, or scoring tables. ## Sample clock drawing test documentation (fictional) Scenario: an abnormal command clock with a markedly better copy at a memory-concern visit, documented with the system, condition, and time named, the errors in words, and the follow-up chain set. All details are fictional. Patient: R.D., 81 · Setting: Geriatric primary care, memory concern visit · Clinician: T. Nguyen, MD · Note date: 08/12/2026 Cognitive screening: Standalone clock drawing test administered after R.D.'s daughter reported six months of missed bills and repeated questions. Command condition first: blank page, instructed to draw a clock, put in all the numbers, and set the time to ten past eleven (11:10). Scored with the Shulman six-category system, implemented as 0 to 5 with higher scores better: 2/5. Reading glasses worn, hearing aids in, dominant right hand steady, no tremor; instructions repeated once at her request, single attempt, no restart. She has used analog clocks all her life. Qualitative description: Contour drawn adequately. Numbers 1 through 8 present with 9 through 12 omitted, remaining numbers crowded into the right half of the face. Hands drawn pointing to the 10 and to the 11, a stimulus-bound response to the spoken instruction rather than the requested time. One number (4) written twice. No perseverative looping; pencil control intact. Copy condition: A copy condition followed (examiner's model clock visible, same 11:10 setting): 5/5 by the same implementation, organized and complete, with accurate hand placement. The command-copy contrast, impaired generation with intact copying, suggests the burden falls on planning and executive control more than on visuoconstruction, and it is a screening observation, not a localization or a diagnosis. Interpretation and disclosure: Abnormal cognitive screen at today's visit, consistent with the family's report; it does not by itself establish dementia, mild cognitive impairment, a cause, decisional capacity, or driving fitness. Result reviewed in plain terms with the patient and her daughter with the patient's permission: the drawing suggests thinking and planning should be looked at more closely, and it is not a diagnosis. Both clocks scanned to the diagnostic-media section of the record; the first attempt retained. Plan: Collateral history and instrumental-function review completed today (finances, medications, cooking, driving: drives short daytime trips only). Medication reconciliation done; no new sedating or anticholinergic agents. Laboratory workup for reversible contributors ordered. Dedicated cognitive assessment visit scheduled 08/28/2026 with a longer instrument and functional inventory, neuropsychology referral to be decided there; driving discussed, with an occupational therapy driving evaluation to be arranged if the fuller assessment sustains concern. No dementia code entered from a screen. Follow-up owner: Dr. Nguyen; nurse call in one week to confirm the appointment. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, drawings, dates, and details are invented to show documentation structure and are not clinical guidance. No instrument's rubric text or scoring tables are reproduced. ↑ Back to the template and downloads ### Why this sample works The five facts that give a clock score meaning are all fixed: the named system with its range and direction, the condition, the substrate, the exact time requested, and the score, so the result can be interpreted and re-derived later. The errors are described in words (omissions, crowding, the stimulus-bound hands, the duplication), which is what lets a later clinician re-score the performance or track its pattern over time. The command-copy contrast is documented as a screening observation with its interpretive weight stated carefully, not converted into a localization or a diagnosis. Confounds are handled before interpretation: vision, hearing, motor status, analog-clock familiarity, and administration integrity are all on the record, so the abnormal score is defensible. The loop closes the way reviewers expect: disclosure in plain terms, collateral and function documented, reversible contributors addressed, a dated next assessment with an owner, both drawings retained, and no diagnosis coded from a screen. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the duty is to detect, never to use one tool. The Medicare annual wellness visit regulation (42 CFR 410.15) requires detection of cognitive impairment by direct observation with due consideration of patient, family, and caregiver report, and CMS deliberately names no instrument (LAW); a clock, alone or inside the Mini-Cog, is one conventional way to meet it. The cognitive assessment and care plan service (99483) is PAYER POLICY with documented elements far beyond any screen (independent historian, function and safety, neuropsychiatric symptoms, medication reconciliation, caregiver assessment, care plan), and a regional Medicare LCD illustrates the documentation principle for testing generally: name the instrument, summarize findings, and keep raw results available on request. An abnormal clock can initiate that pathway; it is not by itself 99483 documentation, and the CMS GUIDE dementia model standardizes care without mandating an instrument. Hospital pathways screen delirium with dedicated instruments (CAM, 4AT), and a clock drawn during acute illness documents impairment at that moment only (CONVENTION). The USPSTF's standing position on screening asymptomatic older adults is an insufficient-evidence statement, which makes the indication line (concern versus routine) worth charting. In driving evaluations, the AGS and NHTSA clinician's guide includes a clock, scored with Freund's system, in its CADReS office battery as a referral signal, and the underlying validation treated low scores as a reason for formal driving evaluation, not a licensing decision (CONVENTION). Canada and Australia frame the clock explicitly as a trigger, not a verdict. Canada's National Safety Code Standard 6 for driver fitness (June 2026) names abnormalities on clock drawing, among other screens, as findings that should prompt deeper assessment, and in the same document states that cognitive tests lack established safe-or-unsafe cut points and must not be the sole determinant of driving competence; provincial licensing authorities make the decision, and physician reporting duties are provincial law (LAW for reporting, CONVENTION for the tool). Capacity is decision-specific in all three countries: Ontario's consent framework ties capacity to the particular decision, US federal and professional guidance analyzes understanding, appreciation, reasoning, and choice, and Australian frameworks presume capacity and require decision-specific, supported assessment, so a clock score is collateral cognitive evidence in a capacity evaluation , never the determination (LAW, varying by jurisdiction). Australia's Austroads driving standard does not name the clock at all, directing individualized medical, functional, and on-road evidence (REGULATORY STANDARD), and the national delirium care standard requires validated, culturally appropriate screening without naming a clock (CONVENTION). The evidence base supports the same restraint everywhere: pooled dementia accuracy is moderate and system-dependent, sensitivity for mild cognitive impairment is poor (about 0.63 for paper scoring in the largest meta-analysis, whether scoring was brief or detailed), and cultural, educational, and literacy effects are pervasive, so cutoffs do not travel across systems or populations (CONVENTION). Rights and versions differ by system, and the clock now has a cohort-validity problem worth charting. Under the general US principle that copyright protects published expression rather than a method, the standalone scoring approaches can be applied clinically and described in original words, but the published rubric tables and figures belong to their journals and publishers: the rights holder of the Shulman system has stated the test is not copyrighted, the Freedman book is an all-rights-reserved Oxford University Press publication, CLOX reuse runs through the journal's permissions channel, and no blanket license was located for the Sunderland, Mendez, Rouleau, or Freund rubrics, so paraphrase, do not transcribe. The embedded clocks carry their instruments' terms: the MoCA permits qualifying clinical reproduction but prohibits web posting and unauthorized electronic versions, and its current policy still points to certification when an individual task like the clock is interpreted beyond the total score; the Mini-Cog is copyrighted with free unmodified clinical use and discrete-field EHR documentation recommended by its rights holder; the SLUMS is freely available. Digital clock products are proprietary: the leading device is FDA-listed under a premarket-exempt product code, which is not the same as FDA clearance or approval, and its several-hundred-feature process scores are not interchangeable with any paper system's score. Finally, analog-clock familiarity is now a documented confound: cognitively normal younger adults underperform on the clock (a 2024 study found Generation Z adults averaging 8.1 of 10 despite otherwise normal testing, and 2026 studies found clock errors in about 30 percent of younger adults against 7 percent of older adults), with no corrected norms yet, so familiarity belongs in the validity line. The MoCA, Mini-Cog, and DCTclock are trademarks or copyrighted instruments of their respective rights holders (MoCA Test Inc., Soo Borson, and Linus Health). BastionGPT is not affiliated with, or endorsed by, any of these rights holders or publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common clock drawing test documentation errors reviewers flag The numbers behind these errors are specific. In the largest meta-analysis (90 studies, 22,567 participants), paper clock scoring pooled a sensitivity of about 0.63 for mild cognitive impairment whether scoring was brief or detailed; naive and trained raters agreed on only about 60 percent of mildly impaired clocks (against more than 98 percent for clearly normal or impaired ones); pooled dementia accuracy differs by system (about 82 percent sensitivity for Shulman versus 73 for Sunderland ); and cognitively normal younger adults now underperform on the clock for familiarity reasons, with no corrected norms yet. The BastionGPT Clinical Advisory Board sees the same errors most often in clock drawing documentation reviews: A score with no system, range, or direction. "Clock 4." Systems score out of 5, 6, 7, 10, 15, or 20, directions differ, and even Shulman implementations number the same categories 0 to 5 or 1 to 6. Name the system and anchors, the condition, the substrate, and the requested time, or the number cannot be read at all. An embedded clock detached from its instrument. A MoCA clock charted as "standalone CDT 2/3," or a Mini-Cog or MoCA clock re-scored with Rouleau or Shulman. Embedded clocks were validated only inside their parent instruments, under their own conditions; chart them as components (MoCA clock x/3; Mini-Cog clock 0 or 2) and administer a standalone clock if a standalone score is wanted. A cutoff quoted without its system. "Score of 3 or below flags impairment" with no scale named. Cutoffs are system- and population-specific, accuracy differs by system and setting, and complexity does not rescue them: detailed and brief paper scoring showed the same 0.63 sensitivity for mild impairment. Report the named system's result and let the battery, history, and collateral carry the decision. A screen charted as a verdict. "Abnormal clock, consistent with dementia," a clock deciding capacity, or a driving conclusion from one office drawing. Every framework that names the clock treats it as a trigger: Canada's 2026 driver-fitness standard says cognitive tests must not be the sole determinant, capacity is decision-specific in all three countries, and no clock diagnoses anything. Chart the referral chain instead. Confounds unexamined, delirium above all. An inpatient clock during acute illness read as baseline dementia, a tremor or field cut scored as cognitive failure, or a 20-year-old's analog-clock unfamiliarity charted as impairment. Screen delirium with a dedicated instrument, document vision, motor, literacy, and clock familiarity, and mark the attempt not interpretable when a barrier explains it. The drawing discarded, the errors undescribed. A bare number with no qualitative account and no retained image. The pattern (omissions, crowding, stimulus-bound hands, perseveration) is the clinically usable content, and the scanned drawing is what lets a later clinician re-score or compare; keep the first attempt, note which attempt was scored, and describe the errors in words. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (system with range and direction, condition, substrate, time requested, score, the errors you observed, confounds, prior results) and it drafts the documentation block: five facts fixed, errors described in words, screen-not-diagnosis line in place, and the follow-up chain with owner and date, ready for your review. Cross-check a finished note for the gaps reviewers flag: a score without its system or condition, an embedded clock charted as standalone, a cutoff without anchors, a missing validity line, or a screen carrying diagnostic, capacity, or driving weight it cannot bear. Draft the follow-up or serial-comparison documentation: what changed against the prior clock under the same system and condition, the command-copy contrast in plain language, and the referral chain, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the clock drawing test scored? By whichever named system the clinician chose, and the systems genuinely differ. Simple ordinal systems (Shulman's six categories; Sunderland's 1 to 10) rate the whole clock; detailed point systems (Mendez's 20 points, Freedman's 15, Rouleau's 10 plus a qualitative error taxonomy, the two 15-point CLOX conditions, Freund's 7-point driving-oriented scale) award points for discrete features; and the embedded clocks in the MoCA (0 to 3) and Mini-Cog (0 or 2, deliberately binary so untrained raters can apply a clean pass or fail) belong to their parent instruments. Ranges, directions, conditions, and requested times all vary, and even Shulman implementations number the same categories 0 to 5 or 1 to 6. That is why the documentation rule is to name the system with its range and direction, the condition, the substrate, and the exact time requested alongside the score. Is there a cutoff score where a clock becomes abnormal? Only within a named system, in a population it was studied in, and never as a diagnosis. Pooled accuracy differs by system (one meta-analysis reported about 82 percent sensitivity and 76 percent specificity for Shulman scoring against about 73 and 88 for Sunderland), comparative studies find no system consistently superior, and the field's own review concluded that for screening, simpler is about as good as complex. The weak point is mild impairment: across 90 studies, paper scoring pooled a sensitivity near 0.63 for mild cognitive impairment whether brief or detailed, so a normal clock cannot rule out early decline, and naive raters agree with experts on only about 60 percent of mildly impaired clocks. Report the named system's result, describe the errors, and let history, collateral, and a fuller assessment carry any decision. Why do clinicians ask for ten past eleven? Because 11:10 works the whole task. Setting it requires placing hands in both halves of the clock face, which can expose a neglected hemifield, and it requires recoding the spoken "ten" into the minute hand at the numeral 2 rather than pointing at the 10, which is exactly where a stimulus-bound response shows itself: hands drawn to the 10 and the 11 because those are the numbers that were said. Systems vary (CLOX uses 1:45, some protocols use 2:45 or 8:20, and Freedman's group noted that the popular "3 o'clock" tests neither hemifield balance nor recoding), which is one more reason the requested time is a test specification to chart, not a stylistic choice. A literal hand placement is an observation about how the drawing failed; it does not by itself identify a disease or a lesion. What does it mean when the command clock is bad but the copy is good? It is the clock's most informative pattern, and it should be charted as a contrast, not a conclusion. Command drawing requires generating the concept, plan, and layout from instruction alone; copying supplies the model and drops those generative and executive demands while keeping the visuospatial and constructional ones. A markedly better copy therefore shifts suspicion toward planning and executive control rather than visuoconstruction, which is exactly what the CLOX task was designed to separate with its two 0-to-15 conditions. The defensible note reports both conditions with the same named system, describes each drawing, states the discrepancy as suggesting disproportionate executive burden, and routes it into history, collateral, medication review, and fuller assessment rather than into a diagnosis or an anatomical claim. Can I score the clock inside a MoCA or Mini-Cog as a standalone clock test? No. The embedded clocks were validated only inside their instruments, under their own conditions: the MoCA clock is drawn with no pre-drawn circle and scored 0 to 3 on contour, numbers, and hands, and current MoCA policy still points to certification when an individual task is interpreted beyond the total; the Mini-Cog clock uses a permitted pre-printed circle and a deliberate 0-or-2 binary. Re-scoring either with Shulman, Rouleau, or another standalone rubric applies cutoffs that were never validated on those administrations. Chart embedded clocks as components ("MoCA clock 2/3," "Mini-Cog clock 0/2") with a descriptive note if useful, and administer a standalone clock, under a named system and condition, when a standalone score is actually wanted. The SLUMS clock item follows the same rule, and the MMSE contains no clock at all. The clock is normal but the family reports decline. What now? Keep evaluating; the screen cannot close that question. Paper clock scoring has poor sensitivity for mild impairment (about 0.63 pooled, brief or detailed), reviews have concluded it should not be used to screen for mild cognitive impairment, and high-reserve patients can produce a normal drawing while declining. The Medicare wellness-visit regulation itself requires due consideration of concerns raised by family, friends, and caregivers, not just performance, so the collateral report is a required input, not an anecdote. Chart the specific examples (missed bills, repeated questions, medication slips), state that the normal screen does not exclude impairment, and document the next step: a longer instrument, an informant tool, functional review, and follow-up with an owner and date. Do younger patients fail the clock because they never use analog clocks? Sometimes, and it is now a documented validity issue rather than a hunch. A 2024 study found cognitively normal Generation Z adults averaging 8.1 of 10 on a standard clock system despite otherwise normal testing, generational work has shown the effect independent of measured cognition (some younger examinees write a digital "11:10" inside the circle), and 2026 studies reported clock errors in about 30 percent of younger adults against about 7 percent of older adults, with analog-clock literacy predicting clock performance but not a non-clock visuospatial task. No corrected norms or accepted replacement task exist yet, so the practical rule is documentary: ask about and record analog-clock familiarity, describe atypical responses rather than just scoring them, interpret cautiously in younger or digital-native patients, and never chart unfamiliarity as cognitive impairment. Can we print clock sheets, put a scoring rubric in the EHR, or use a digital clock app? It depends on whose material you touch. A clinic-designed generic sheet (its own instruction wording, blank page or its own circle) is the low-risk route, and the standalone scoring methods can be operationalized as EHR fields (system name, range, condition, time, score, errors) because a method is not copyrightable, though published rubric tables and figures belong to their journals and publishers, so paraphrase rather than transcribe, and the Freedman book is expressly all-rights-reserved. Branded instruments carry their own terms: the MoCA prohibits web posting and unauthorized electronic versions, and the Mini-Cog permits free unmodified clinical use, with its rights holder recommending discrete EHR fields for the component scores. Digital products are proprietary: the leading digital clock device is FDA-listed under a premarket-exempt product code, which is not FDA clearance or approval, and its machine-scored process metrics do not interchange with any paper system's score, so name the product and report its output as its own scale. Can BastionGPT draft my clock drawing test documentation? Yes. Give it the facts (system with range and direction, condition, substrate, time requested, score, the errors you observed, confounds, disclosure, and plan) and it drafts the full entry: the five specification facts fixed, the qualitative description in words, the screen-not-diagnosis line, and the follow-up chain with owner and date, ready for your review. It can also cross-check a finished note for a score without its system, an embedded clock charted as standalone, a missing validity line, or diagnostic weight a screen cannot carry, and it can draft the serial-comparison paragraph against a prior clock under the same system and condition. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: System publications: Shulman KI and colleagues, 1986, International Journal of Geriatric Psychiatry (six-category system; see also Shulman's 2000 review); Sunderland T and colleagues, 1989, Journal of the American Geriatrics Society (1 to 10); Mendez MF and colleagues, 1992, Clock Drawing Interpretation Scale ; Rouleau I and colleagues, 1992, quantitative and qualitative analyses ; Freedman M and colleagues, 1994, Clock Drawing: A Neuropsychological Analysis (Oxford University Press); Royall DR and colleagues, 1998, CLOX ; Freund B and colleagues, 2005, Journal of General Internal Medicine, drawing clocks and driving cars ; history in Spenciere B and colleagues, 2017, scoring systems, a historical review . Accuracy and comparison: Park J and colleagues, 2018, meta-analysis by scoring method ; Storey JE and colleagues, 2001, five systems compared in Australian outpatients ; Mainland BJ, Amodeo S, Shulman KI, 2014, is it time for simpler systems ; Scanlan JM and colleagues, 2002, naive versus trained raters ; Chan JYC and colleagues, 2022, 90-study meta-analysis, paper and digital ; Ehreke L and colleagues, 2010, CDT and MCI screening review ; Powlishta KK and colleagues, 2002, Neurology (poor screen for very mild dementia). Digital-native cohort evidence: Vishnevsky G, Fisher T, Specktor P, 2024, Journal of the Neurological Sciences, the clock drawing test in the digital era ; DeMessie B and colleagues, 2026, clock errors in younger versus older adults ; Deterding R, Smith AM, 2026, analog-clock literacy and CDT performance . Digital products and machine scoring: FDA AccessGUDID, DCTclock device listing (premarket-exempt product code; listing is not clearance or approval); Linus Health, process-feature approach ; Chang and colleagues, 2026, npj Digital Medicine, digital clock meta-analysis . United States: eCFR, 42 CFR 410.15 (annual wellness visit; detection with collateral consideration; no instrument named); CMS Medicare Coverage Database, LCD L39266 (name the instrument; raw results available on request); USPSTF, cognitive-impairment screening statement ; AGS and NHTSA, Clinician's Guide to Assessing and Counseling Older Drivers (CADReS battery). Canada and Australia: CCMTA, National Safety Code Standard 6, Determining Driver Fitness in Canada (abnormal clock as a trigger; no test the sole determinant); Austroads, Assessing Fitness to Drive (no clock named); ACSQHC, Delirium Clinical Care Standard ; Ontario, Health Care Consent Act (decision-specific capacity). Rights and instrument terms: US Copyright Office, Circular 33 (methods and systems versus expression); ePROVIDE (Mapi Research Trust), CDT rights entry (Shulman: not copyrighted, per the author); MoCA Test Inc., permission terms and training and certification policy ; Mini-Cog rights holder, FAQ and EHR documentation guidance ; Saint Louis University, SLUMS availability . ## Related templates MoCA Score Documentation MMSE Score Documentation SLUMS Score Documentation Mini-Cog Score Documentation Neuropsychological Evaluation Report Capacity & Competency Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/boston-naming-test All note templates # Boston Naming Test Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Boston Naming Test (BNT-2) is a 60-item confrontation naming test: the examinee names line drawings that run from common to rare words. Neuropsychologists and speech-language pathologists use it in aphasia, dementia, and epilepsy surgery evaluations. A total is uninterpretable without the form, edition, cue accounting, and norm source. This page covers how to document BNT results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Neuropsychologists, psychologists, and speech-language pathologists; US publisher qualification level C for the kit (listed levels differ by seller and territory) Audience Neuropsychologists and SLPs, neurologists and epilepsy surgery teams, dementia and memory clinics, payers reviewing testing claims, attorneys in medicolegal work Typical length 3 to 8 chart lines or a short report paragraph · administration about 10 to 20 minutes for the full form, plus scoring and integration Format family Confrontation naming test (60 pictured items in a difficulty gradient; short forms are separate score systems) When it's used Aphasia characterization, dementia and MCI evaluations, epilepsy presurgical and postoperative workups, broader neuropsychological batteries, serial monitoring of language change Standards context BNT-2 published by PRO-ED (US) with regional distributors; no authority mandates it and no universal cutoff exists; described for documentation purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Boston Naming Test? The Boston Naming Test is the most widely used confrontation naming measure in North American practice: 60 black-and-white line drawings, ordered roughly from high-frequency to rare words, that the examinee names one at a time. It began as a 1978 Boston VA research edition by Edith Kaplan, Harold Goodglass, and Sandra Weintraub, was first published commercially by Lea & Febiger in 1983, and exists today as the Boston Naming Test, Second Edition (BNT-2, 2001), published in the United States by PRO-ED, which also handles permissions; PAR sells the BDAE-3 package that contains it, Pearson Clinical distributes it in Australia and New Zealand, and the Lippincott name is the historical imprint. Two currency facts matter for records: no third edition exists ("BDAE-3" modifies the aphasia battery, not the naming test), and PRO-ED now offers a no-cost stimulus upgrade replacing Item 48, a picture the publisher itself describes as racially and culturally inappropriate, while keeping the 60-point maximum, so a report should say which stimulus set was used. The load-bearing facts are the cue asymmetry and the form question. The countable total is spontaneous correct responses plus responses correct after a stimulus (semantic) cue, the clarification used when the drawing was visually misperceived; responses retrieved only after a phonemic cue, which supplies the word's sound, are recorded separately and never added to the total. And "the BNT" is really a family of forms: the full 60, the publisher's own 15-item short form, the separately permissioned CERAD 15-item set, odd and even 30-item halves, and several research-built short forms, whose scores are not interchangeable; the CERAD 15 in particular cannot be validly prorated to a 60-item score. A record that says "BNT = 42, impaired" is therefore uninterpretable. Naming results usually sit inside a neuropsychological report ; the RBANS page covers the screening battery whose brief naming component is not a substitute for this test. ## Who uses Boston Naming Test documentation and when Neuropsychologists administer the BNT inside dementia and mild cognitive impairment workups (where the error and cue pattern helps separate retrieval difficulty from semantic loss), in aphasia and stroke characterization alongside a full language battery, and in epilepsy surgery programs, where naming is a core presurgical and postoperative measure: left temporal resections carry a documented naming-decline risk, so the baseline form, norms, and a reliable-change plan get set before surgery. Speech-language pathologists document confrontation naming within evaluations billed under the 92521 to 92524 family and use the cue profile to shape therapy targets. Geriatric and memory clinics meet it inside CERAD-derived protocols, which use the separately permissioned CERAD 15-item form; US Alzheimer's research centers replaced the BNT with the Multilingual Naming Test in their uniform dataset back in 2015, a fact worth knowing when research records cross into clinical charts. Readers are rarely just the author: the next evaluator needs the form and norms to test change, payers reviewing 96130-series claims need the medical-necessity chain, and the NEPSY-II page covers the pediatric language instruments used instead of adult naming norms. Executive-function measures like those on the D-KEFS page assess fluency, which is not confrontation naming. ## How to document Boston Naming Test results in the chart No US, Canadian, or Australian authority prescribes a BNT note format, mandates the test, or sets a universal cutoff. What survives review is a record from which a reader can reconstruct exactly what was administered and how the total was assembled: the form and edition, the administration route and stop rule, the cue accounting, the named norms, and the language context. Each element below carries the pitfall that most often undermines it. Form, edition, and stimulus set. Name the exact form: 60-item BNT-2, the publisher's 15-item short form, the CERAD 15-item set, a named 30-item half, or a named research short form, plus the edition and whether the original or upgraded stimulus set (the Item 48 replacement) was used. The maximum stays 60 across stimulus sets, so only the note preserves the difference. Pitfall: "BNT-15" or "BNT-30" with no source. The publisher 15 and the CERAD 15 are different forms, the halves differ, and none of their scores interchange with the full form or each other. Administration route and stop rule. State whether the classic basal-and-discontinuation route was used (starting mid-list, establishing a basal, stopping after a run of consecutive failures) or the all-items revised procedure that administers all 60 with defined response intervals. Name the discontinuation rule as applied: the failure-run length differs by edition (six in 1983, eight in the 2001 edition), and say whether phonemic-cued successes interrupted the failure run. Pitfall: Silence on the stop rule. The two published interpretations of the failure run changed final scores in 31 percent of an Alzheimer's sample, by as much as 16 points in older adults, so an unstated convention makes serial comparison unsafe. Cue accounting, made explicit. Report the arithmetic: spontaneous correct plus stimulus-cued correct equals the countable total, with phonemic-cued successes counted separately and excluded. The stimulus cue exists to repair a visual misperception and restore a valid naming opportunity; the phonemic cue hands over the word's sound, so counting it would inflate the construct. Pitfall: "Answers after cues count." Collapsing the two cue types, or crediting phonemic-cued responses, changes the score and destroys comparability with every properly scored administration. Error pattern, in words. Describe the errors: semantic paraphasias, phonemic paraphasias, circumlocutions, visual misperceptions, omissions, perseverations. The pattern carries the clinical meaning: strong phonemic-cue benefit suggests retrieval difficulty with the lexicon intact, while semantic errors and no cue benefit point toward semantic degradation. Pitfall: A total with no error description. Two identical totals can represent different syndromes, and the treating SLP or the next evaluator cannot use a bare number. Norm source and corrections. Name the normative system and the corrections actually applied: the publisher manual, Tombaugh and Hubley (1997), Mayo's older-adult systems (MOANS, MOAANS) and the 2024 Mayo regression norms (4,428 adults aged 30 to 91, with age, sex, and education terms), Zec (2007), Heaton's demographically corrected system, or CERAD norms for the CERAD form only. Record the exact age and education inputs. Pitfall: "Below cutoff" with no system. No universal BNT cutoff exists, systems correct for different variables, and race-stratified historical norms are sample calibrations, not biological claims; mismatched norms are a standing cross-examination target. Language, culture, and familiarity. Document the testing language, the examinee's language history (dominance, age of acquisition, education language, daily use), and cultural and generational familiarity considerations. Record other-language correct responses separately rather than adding them to the English total, and consider a validated adaptation or the Multilingual Naming Test for bilingual examinees; monolingual English norms overestimate impairment in bilinguals. Pitfall: Anomia diagnosed from an English administration of a culturally loaded stimulus set. Twelve items showed differential functioning between demographically matched groups, and matched young adults hit "impaired" cutoffs at 18 versus 4 percent, so the confound analysis belongs in the note. Interpretation and the change plan. Interpret the naming result inside the evaluation: what it converges with, what it does not establish (no naming score is independently diagnostic of dementia, aphasia, or seizure laterality), and the follow-up or comparison plan. For serial and surgical work, keep the same form, stimulus set, procedure, and norms, and use a reliable-change frame; post-left-temporal-resection decline is common and expected to exceed chance variation. Pitfall: Change computed across forms or conventions: a CERAD 15 subtracted from a prior 60-item total, a short form doubled into a full-form estimate, or a decline claimed without the stop rule and norms held constant. ### Blank template (copy and adapt) BOSTON NAMING TEST DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Administered by: [ ] Interpreted by: [ ] Referral question / battery context: [ ] Form + edition: [60-item BNT-2 / publisher 15-item / CERAD 15-item / named 30-item half / named research short form] Stimulus set: [original / upgraded (Item 48 replacement)] Administration route: [classic basal + discontinuation (rule length as applied; phonemic-cued successes interrupt run: Y/N) / all-items revised procedure] Language of testing: [ ] Cue accounting: Spontaneous correct: [ ] Stimulus-cued correct (counted): [ ] Countable total: [ ]/[60 or form maximum] Phonemic-cued correct (recorded, NOT counted): [ ] Error pattern: [semantic paraphasias / phonemic paraphasias / circumlocutions / visual misperceptions / omissions / perseverations] Norm source + corrections: [system + year; age, education, sex, other inputs actually used] Standardized result: [ ] Language + culture: [dominance, acquisition, education language, daily use; other-language responses recorded separately; adaptation or MINT considered: Y/N] Validity: [vision, hearing for cues, motor speech, familiarity factors] Conclusion: [interpretable / with caution / not interpretable] Interpretation: [pattern within the evaluation; retrieval vs semantic shape; what the score does not establish] Prior testing / change plan: [same form, set, procedure, norms; reliable change frame; no cross-form arithmetic] Plan / integration: [ ] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces pictures, target words, record forms, or norm tables. ## Sample Boston Naming Test documentation (fictional) Scenario: a full-form administration inside a memory-clinic dementia evaluation, documented with the cue arithmetic explicit, the norms named, and the retrieval-versus-semantic reasoning shown. All details are fictional. Patient: E.S., 74 · Setting: Outpatient neuropsychology, memory clinic evaluation · Clinician: K. Whitfield, PhD · Note date: 08/19/2026 Measure and administration: Boston Naming Test, Second Edition, 60-item form, upgraded stimulus set, administered in English (the patient's first and dominant language) using the all-items procedure with standard response intervals and the authorized cue sequence. Corrected vision worn and adequate for the drawings; hearing adequate for cues; no motor-speech limitation. Administered and scored by this examiner within a three-hour dementia evaluation battery. Results: Spontaneous correct 40; correct after stimulus clarification 3 (each following an initial visual misperception); countable total 43/60. Seven additional targets were retrieved only after phonemic cues and are recorded here but excluded from the total. Errors were predominantly circumlocutions conveying knowledge of the object without its name, with four semantic paraphasias on lower-frequency items and two visual misperceptions; no perseverations. Scored against the 2024 Mayo regression norms with age, sex, and education entered as that model specifies (age 74, 14 years of education, documented in the scoring record): performance fell below the expected range. Interpretation: Clinically significant naming weakness for age and education. The shape of the performance matters as much as the level: the strong phonemic-cue benefit and knowledge-revealing circumlocutions indicate that lexical retrieval, more than semantic storage, is failing, a pattern consistent with the amnestic-plus-language presentation elsewhere in the battery and distinct from the semantic degradation profile in which cues stop helping. The naming score is one language marker within the evaluation and is not independently diagnostic of a neurodegenerative disorder. Validity and context: Standard administration, dominant language, adequate sensory function, engaged effort with unremarkable validity indicators across the battery: result interpretable. Lifelong US resident with schooling in English; no cultural or generational familiarity concern was evident for this stimulus set, and the upgraded set was used and is named here so any future comparison holds it constant. Comparison and plan: No prior naming testing exists. For future monitoring, the same 60-item form, stimulus set, all-items procedure, and norm system should be repeated with a reliable-change frame; a short form, and any CERAD 15-item administration inside a research protocol, would be reported as a separate measure rather than compared numerically with today's total. Findings are integrated with the memory, fluency, and functional results in the evaluation summary, with feedback scheduled 08/31/2026. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No test pictures, target words, or norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The cue arithmetic is explicit (spontaneous plus stimulus-cued equals the total; phonemic-cued recorded and excluded), so the score means the same thing it means in every properly scored administration. The form, edition, stimulus set, procedure, and language are all named, which is what makes the result reproducible and the future comparison plan possible. The error pattern is described and then used: the phonemic-cue benefit and circumlocutions drive a retrieval-versus-semantic interpretation instead of a bare "impaired." The norm system is named with its exact inputs, and the interpretation stays inside the evaluation, with the non-diagnostic boundary stated. The change rules are set in advance: same form, set, procedure, and norms, reliable-change framing, and no cross-form arithmetic if a short form ever appears. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: requirements attach to the evaluation, not the test. Neuropsychological and psychological testing is billed under the 96130 to 96139 family, with CMS contractor policy requiring the record to support medical necessity and show the referral question, tests administered, scoring and interpretation work, findings, and recommendations (PAYER POLICY); speech-language evaluations run under 92521 to 92524 with their own documentation expectations, and no code names the BNT. In epilepsy surgery programs, presurgical neuropsychology is standard of care by professional guidance (CONVENTION), and naming carries specific weight: left temporal lobe epilepsy depresses BNT performance, postoperative decline after dominant-side resection is common (a systematic review found a mean 5.8-point decline exceeding reliable-change thresholds, and one series found decline in 56 percent of left-sided cases, correlated with fMRI language lateralization), so the defensible record fixes the baseline form, stimulus set, norms, language-dominance evidence, and the reliable-change method before surgery, and reports change only within that frame. In dementia work, CERAD-derived protocols use the separately permissioned CERAD 15-item naming form under CERAD norms, and the note names it as such rather than "BNT" (CONVENTION). Canada and Australia change the funding and the documentation registers, not the test. Canadian provincial plans generally do not cover outpatient psychologist-administered testing, which runs hospital-based, insurer-funded, or private (PAYER POLICY, provincial); the BNT is Canada's most used naming measure, Canadian norms exist (Tombaugh and Hubley's age-and-education strata; Quebec-French adaptation work), and Ontario's SLP college documentation standards (effective September 2025) expect records that identify the instrument, results, interpretation, and limitations (REGULATORY STANDARD, province-specific). A 2025 Canadian study of 525 multicultural older adults found immigration history and region of origin shaped BNT performance, which is exactly the confound analysis a Canadian chart should show. In Australia, the Better Access items are treatment items and cannot be used for formal cognitive or neuropsychological assessment, so BNT-based assessment is private, NDIS, or otherwise funded (PAYER POLICY); speech pathology services run through chronic-condition management and under-25 neurodevelopmental pathways with narrow eligibility, and Speech Pathology Australia's standards expect contemporaneous, sufficient records (CONVENTION). One dated transition worth noting: arrangements created under pre-July-2025 GP management plans remain usable only through June 2027, after which the current chronic-condition pathway applies. Rights, editions, and language are where BNT documentation is most often wrong. PRO-ED is the current US publisher and the permissions contact: reproduction of pictures, target words, record forms, or norm tables, in print, in an EHR, or on any public web tool, requires written permission, and the publisher's electronic-use terms (encrypted, access-restricted, non-downloadable) are incompatible with a public site, so a free web "BNT tool" is unauthorized on its face; the CERAD naming form is likewise excluded from the freely distributable CERAD packet and needs separate PRO-ED permission. What belongs in the chart is the patient's results and your interpretation, never scanned stimuli or completed copyrighted forms in portal-accessible locations. Edition facts to keep straight as of August 2026: the current edition is still the BNT-2 (no third edition exists), the discontinuation rule differs between the 1983 and 2001 editions, and the publisher's free stimulus upgrade replaced Item 48 while holding the maximum at 60, so legacy and upgraded administrations should be distinguished in longitudinal and forensic records. On language and culture, the test is not culture-fair and modern norms do not make it so: item-level differential functioning is documented, bilinguals underperform monolingual norms even in their dominant language, other-language responses are recorded separately absent a defined bilingual protocol, and US Alzheimer's research centers replaced the BNT with the Multilingual Naming Test in 2015 for exactly these reasons. The Boston Naming Test and the Boston Diagnostic Aphasia Examination are products of their publisher, PRO-ED, Inc., with regional distribution by PAR, Pearson Clinical, and others. BastionGPT is not affiliated with, or endorsed by, PRO-ED or any distributor. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Boston Naming Test documentation errors reviewers flag The numbers behind these errors are specific. In Ferman and colleagues' 1998 study , the two published interpretations of the discontinuation rule changed final scores in 31 percent of an Alzheimer's sample and by up to 16 points in normal older adults; Pedraza and colleagues found 12 items functioning differently between demographically matched groups; in matched young adults , 18 percent of one group versus 4 percent of the other fell in the "impaired" range on monolingual norms; and after dominant temporal resection, naming declined in 56 percent of left-sided cases . The BastionGPT Clinical Advisory Board sees the same errors most often in Boston Naming Test documentation reviews: A total with nothing around it. "BNT = 42, impaired." Without the form, edition, stimulus set, administration route, stop-rule convention, cue accounting, and named norms, the number cannot be interpreted, audited, or compared. Use the compact frame: spontaneous plus stimulus-cued equals the countable total, phonemic-cued recorded separately, norms named with their inputs. Phonemic-cued responses credited. "Answers after cues count." Only stimulus-cued corrects join the total; a phonemic cue supplies the word's sound and its successes are recorded but excluded. Collapsing the cue types inflates the score and erases the retrieval-versus-semantic information the cue profile carries. Short-form arithmetic. A CERAD 15 prorated to a 60-item estimate, a 30-item score doubled without a validated equation, or "decline" computed between a short form and a prior full form. The CERAD 15 is the documented non-extrapolable case; report each form as its own measure with its own norms and compare qualitatively. The stop rule and edition left unstated. Serial totals compared across administrations that used different editions (six- versus eight-failure rules) or different interpretations of the failure run. The discrepancies are largest in patients over 80 and in dementia samples; name the edition and convention every time, and hold them constant for change claims. Monolingual norms on the wrong examinee. Anomia diagnosed from an English administration in a bilingual or recently immigrated patient, or culture-fairness assumed. Bilinguals underperform monolingual norms even in their dominant language, item-level bias is documented, and research networks replaced the test for multilingual use; document language history, record other-language responses separately, and consider the MINT or a validated adaptation. Norms unnamed, mismatched, or misread. "Below cutoff" with no system, an 80-year-old scored on a sample that stops at 75, one system's strata mixed with another's corrections, or race-stratified historical norms treated as biological findings rather than sample calibrations. Name the system, year, and inputs, and confirm the examinee is represented in it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and neuropsychological evaluation reports. Give it the facts (form, edition, stimulus set, route and stop rule, spontaneous and cued counts, error types, norms and inputs, language history) and it drafts the documentation block or report paragraph: cue arithmetic explicit, norms named, retrieval-versus-semantic reasoning drafted from your error pattern, ready for your review. Cross-check a finished note or results section for the gaps reviewers flag: phonemic-cued responses in the total, a short form compared with a full form, an unstated stop rule, unnamed norms, or a missing language and culture analysis. Draft the presurgical baseline or serial-comparison paragraph: what was held constant, the reliable-change frame, and what the naming result does and does not establish, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Boston Naming Test scored, and which cues count? The countable total is spontaneous correct responses plus responses correct after a stimulus (often called semantic) cue, the clarification an examiner gives when the response shows the drawing was visually misperceived; it exists to restore a valid naming opportunity. Responses retrieved only after a phonemic cue, which supplies the beginning sound of the word, are recorded separately and never added to the total, because the cue hands over part of the answer. Self-corrections count when the correct name is the final response within the allowed interval. The defensible reporting frame makes the arithmetic visible: spontaneous correct, plus stimulus-cued correct, equals the countable total out of the form maximum, with phonemic-cued successes stated separately, because the cue profile itself is clinical information: strong phonemic-cue benefit suggests retrieval difficulty with the lexicon intact, while absent cue benefit with semantic errors points toward semantic degradation. Is there a Boston Naming Test cutoff for impairment? No universal cutoff exists, and "below cutoff" with no named system is not auditable. The same raw total means different things depending on the form, the norm system, and the corrections it applies: the publisher manual norms, Tombaugh and Hubley's age-and-education strata (adults 25 to 88), Mayo's older-adult systems (MOANS, and MOAANS for African American elders), Zec's large older-adult sample, Heaton's demographically corrected system, CERAD norms for the CERAD form only, and the 2024 Mayo regression norms built from 4,428 cognitively unimpaired adults aged 30 to 91 with age, sex, and education terms. A defensible note names the system and year, records the exact inputs used, and confirms the examinee's age, education, language, and background are actually represented in that sample; the freely circulating one-line cutoffs on scoring websites do none of this. Does it matter which edition was used and how the stop rule was applied? Materially. The classic route starts mid-list, establishes a basal, and stops after a run of consecutive failures, and that run's length differs by edition: six failures in the 1983 edition, eight in the 2001 BNT-2, a change made without published explanation. Worse, the six-failure rule itself has two published interpretations, turning on whether an item retrieved after a phonemic cue interrupts the failure run: in Ferman and colleagues' study the choice changed final scores in 3 percent of 655 normal older adults but 31 percent of 140 Alzheimer's patients, with differences up to 16 points, largest in people over 80. The all-items revised procedure avoids the problem by administering all 60 items with defined intervals. So the note names the edition, the route, and the convention as applied, and any change claim holds all three constant; the publisher's stimulus upgrade (the Item 48 replacement) gets named for the same reason. Are the 15-item and 30-item short forms interchangeable with the full BNT? No, and one of them is the documented trap. "BNT-15" is not a single form: the BNT-2 kit's own 15-item short form, the CERAD 15-item set used in dementia protocols, and several research-derived 15-item forms coexist, and the CERAD 15 is the specific case shown not to behave like the others and not to support extrapolation to a 60-item estimate. Thirty-item halves can be prorated with small error in some studies, but only with a validated equation for that exact half and population, labeled as an estimate. The safe rules: name the form's source every time, report each form against its own norms, never subtract a short-form score from a prior full-form score, and state in the record that the results are not directly numerically comparable, describing convergence qualitatively instead. Which norms should I use for an 80-year-old? Name a system that actually contains an 80-year-old with your patient's education and background, and say why you chose it. Candidates include Mayo's older-adult systems (MOANS above 55; MOAANS for African American elders aged 56 to 94), Tombaugh and Hubley through 88, Zec's sample through 95, and the 2024 Mayo regression norms through 91; Heaton's system tops out around 85, which puts the oldest patients at its boundary. Two cautions specific to this age band: the stop-rule discrepancies are largest in people over 80, so state the convention, and generational familiarity with the older stimulus set is a live research question, so an atypical error on a dated object is worth describing rather than just scoring. "Older-adult norms" with no source and year is not sufficient documentation. Can I use the Boston Naming Test with a bilingual patient? Only with the language analysis in the record, and often something else is better. Bilinguals underperform monolingual English norms even in their dominant language, item difficulty shifts across languages (frequency, cognate status, and familiarity all move), and a name produced correctly in the other language demonstrates intact conceptual knowledge but is not silently added to the English total; either-language scoring requires a predefined bilingual protocol. The Multilingual Naming Test was built for this problem, aligns better with bilingual dominance measures, and replaced the BNT in the US Alzheimer's centers' uniform dataset in 2015; validated adaptations and cross-cultural naming measures are the other options. If the BNT is used anyway, the note documents dominance, age of acquisition, education language, daily use, and immigration history, records other-language responses separately, and declines to diagnose anomia from monolingual norms the patient is not represented in. Can we photocopy BNT pictures, put them in our EHR, or build a naming web tool? Not without written permission from PRO-ED, the current US publisher and permissions contact. Reproduction of stimuli, target words, record forms, or norm tables is a case-by-case written-permission matter, and the publisher's electronic-use terms (encryption, access restriction, non-downloadable, non-printable) are incompatible with an ordinary EHR template or any public website, so a free web tool displaying BNT items or applying its norms is unauthorized on its face; the CERAD 15-item naming form likewise sits outside the freely distributable CERAD packet and needs separate PRO-ED permission, and the publisher no longer grants certain overseas translation requests. What belongs in the chart is the patient's scores, cue accounting, error pattern, and your interpretation; purchased materials stay in the authorized test-record workflow, not scanned into portal-accessible locations. Test names are used here for identification only, and this page reproduces no items or norms. Can BastionGPT draft my Boston Naming Test documentation? Yes. Give it the facts (form, edition, stimulus set, administration route and stop rule, spontaneous and cued counts, error types, norm system and inputs, language history, and prior testing) and it drafts the documentation block or report paragraph: cue arithmetic explicit, norms named, the retrieval-versus-semantic reasoning drafted from your error pattern, and the change rules stated, ready for your review. It can also cross-check a finished note for phonemic-cued responses in the total, short-form arithmetic, an unstated stop rule, unnamed norms, or a missing language analysis. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: PRO-ED, Inc., BNT-2 kit , stimulus book (Kaplan, Goodglass, and Weintraub authorship), no-cost stimulus upgrade kit (Item 48 replacement; 60-point maximum retained), and reprint and electronic permissions ; PAR, Inc., BDAE-3 package listing ; Pearson Clinical Australia, regional BNT-2 listing . Administration and scoring literature: Nicholas LE and colleagues, 1989, revised administration and scoring procedures (all-items route; cue and self-correction rules); Ferman TJ, Ivnik RJ, Lucas JA, 1998, Assessment, rigorous versus lenient stop-rule interpretations (score changes up to 16 points); NINDS common data elements, 30-item BNT protocol (cue accounting convention). Short forms: Mack WJ and colleagues, 1992, comparison of short forms (CERAD 15 not equivalent); Saxton J and colleagues, 2000, two 30-item forms with norms ; Graves RE and colleagues, 2004, IRT-derived forms ; estimated-score analysis, proration findings (CERAD 15 estimation not warranted); Katsumata Y and colleagues, 2015, short-form performance in an ADC cohort . Norm systems: Tombaugh TN, Hubley AM, 1997, age-and-education norms, ages 25 to 88 ; Ivnik RJ and colleagues, 1996, MOANS ; Lucas JA and colleagues, 2005, MOAANS ; Zec RF and colleagues, 2007, 1,026 adults aged 50 to 95 ; Karstens AJ and colleagues, 2024, updated Mayo regression norms (4,428 adults, ages 30 to 91); Heaton RK and colleagues, 2004 revised comprehensive norms via PAR . Bias, culture, and bilingual evidence: Pedraza O and colleagues, 2009, differential item functioning on 12 items , and 2011, item difficulty and discrimination ; Na S, King JH, 2019, impairment-range rates in matched young adults ; Gollan TH and colleagues, 2012, the Multilingual Naming Test and bilingual dominance ; Besser L and colleagues, 2018 (NACC UDS version 3 adopting the MINT, March 2015); 2025 Canadian multicultural cohort, immigration effects on BNT performance . Epilepsy and clinical validity: ILAE, neuropsychological assessment in epilepsy surgery ; naming outcomes after temporal resection, Neurology 2022 series (decline in 56 percent of left-sided cases; fMRI correlation); Zec RF and colleagues, 2005, test-retest stability in older adults ; 2026 MCI naming meta-analysis, 20 studies ; Harry A, Crowe SF, 2014, critique of continued use . Coverage and standards: CMS billing articles A57780 and A57481 (96130 to 96139 documentation); ASHA, SLP evaluation coding (92521 to 92524); NACC, CERAD forms and the separate naming-test permission ; CASLPO, Ontario documentation standards (effective September 2025); Services Australia, allied health MBS billing rules (Better Access excludes formal cognitive assessment). ## Related templates Neuropsychological Evaluation Report RBANS Report Write-Up NEPSY-II Report Write-Up D-KEFS Report Write-Up Stroop Test Score Documentation Trail Making Test Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/asq-3 All note templates # ASQ-3 Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The ASQ-3 (Ages & Stages Questionnaires, Third Edition) is a parent-completed developmental screen for ages 1 to 66 months, scored in five domains against interval-specific cutoffs. Pediatric, family, and early-childhood programs use it at well-child visits and program intakes. Each domain lands above cutoff, in the monitoring zone, or below cutoff, and each zone obligates something different. This page covers how to document ASQ-3 results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Pediatric and family practices, early intervention and Head Start programs, child care and public health screeners; parents complete it, trained staff score it, the clinician interprets and acts Audience Pediatricians and family physicians, early-intervention and Part C teams, Head Start and preschool programs, audiology and specialty referrals, payers auditing screening claims, families Typical length 4 to 10 chart lines plus the completed summary sheet · parent completion 10 to 15 minutes, scoring 2 to 3 minutes Format family Parent-completed developmental screen (five domains of six items each, scored against interval-specific cutoffs at 21 age intervals) When it's used 9-, 18-, and 30-month well-child screening, EPSDT checkups, Head Start 45-day screens, preterm follow-up, and any visit where surveillance or a parent concern warrants a standardized screen Standards context Published by Brookes (2009 edition current; ASQ-4 in development, undated); recommended by the AAP and named in many state Medicaid manuals, but no US law mandates ASQ-3 by name; no items or cutoffs reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the ASQ-3? The ASQ-3 (Ages & Stages Questionnaires, Third Edition; Squires and Bricker, Paul H. Brookes Publishing, 2009) is the most widely used parent-completed developmental screen in US pediatric practice. Twenty-one age-interval questionnaires cover children from 1 through 66 months; at each interval a parent answers six questions in each of five domains (communication, gross motor, fine motor, problem solving, personal-social), each response scored as demonstrated, emerging, or not yet, and summed within the domain. Every domain at every interval has its own empirically derived cutoff set at two standard deviations below that domain's mean, plus a monitoring zone between one and two standard deviations below, added in this edition. There is no valid global ASQ-3 total, and the closing Overall section, which gathers parent concerns about hearing, vision, talking, behavior, and possible loss of skills, is deliberately untotaled. A fourth edition is in development with no announced date; ASQ-3 remains the current edition as of August 2026. The load-bearing facts for documentation are the three-zone structure and the interval rule. A domain result is above cutoff, in the monitoring zone, or at or below cutoff, and each zone obligates something different: below cutoff is a positive screen that triggers referral, the monitoring zone requires an active, dated follow-up plan (it is neither a pass nor an automatic referral, and no published study isolates its predictive value, so its handling rests on publisher guidance), and even five above-cutoff domains do not make a negative screen when the Overall section holds a concern. The interval is selected by the child's exact age in months and days, corrected for prematurity when the child was born three or more weeks early and is under 24 months; the cutoffs were built on corrected ages, so an uncorrected interval choice can turn a true pass into a false positive. The ASQ:SE-2 is a separate social-emotional instrument, not a sixth domain, and a positive ASQ-3 screen leads toward evaluation instruments like those on the Bayley-4 and DAYC-2 pages, inside the framework on the developmental assessment page. ## Who uses ASQ-3 documentation and when Pediatric and family practices administer the ASQ-3 at the standardized screening ages the American Academy of Pediatrics recommends (9, 18, and 30 months, with autism-specific screening at 18 and 24 months via instruments like the M-CHAT-R/F ) and at any visit where surveillance or a parent concern warrants a standardized screen. Medicaid EPSDT programs make it a payer matter: several state manuals name the ASQ-3 among approved instruments and require the chart to show the instrument name, scores, and guidance provided, and Texas requires early-childhood-intervention referral within seven days of identifying a suspected delay. Early intervention, Head Start (which must screen within 45 days of entry), child care, and public health programs run it at scale, including the ASQ-TRAK adaptation in Australian Aboriginal and Torres Strait Islander contexts. Neonatal follow-up clinics use it with corrected ages in preterm cohorts, where a pass is highly reassuring but a positive screen carries a high false-positive rate. The documentation reader is usually downstream: the Part C intake team acting on the referral clock, the audiologist receiving a communication-domain referral, a payer auditing 96110 units, or the next screener judging whether a monitoring-zone domain has repeated. The PSC-17 page covers psychosocial screening in older children; it is not a developmental-domain alternative. ## How to document ASQ-3 results in the chart No US, Canadian, or Australian authority prescribes an ASQ-3 note format, and none mandates the instrument by name. What survives review is a record that justifies the interval, states each domain's zone rather than a bare number, treats the monitoring zone as an active category with a dated plan, gives the Overall section its veto power, and separates a positive screen from a diagnosis. Each element below carries the pitfall that most often undermines it. Interval selection and corrected age. Record the questionnaire interval administered and the age arithmetic that justified it: exact chronological age in months and days, and, for a child born three or more weeks early who is under 24 months, the corrected age that selected the form (with the gestational basis). At 24 months and older, chronological age governs. Pitfall: Choosing the nearest month label or the visit's billing age. The cutoffs were built on corrected ages, and a wrong interval samples the wrong skills against the wrong comparison: the commonest error that flips a true pass into a false positive. Respondent, language, and completeness. Chart who completed the questionnaire and their relationship, the language of the form actually used, any interpreter or interview assistance, and the completion date. Note whether every domain was scorable: one or two blank items in a domain permit the authorized adjusted score (identified as adjusted); more than two make that domain not scorable. Pitfall: An unscorable domain silently scored, an adjusted score presented as ordinary, or a translated administration charted without the language and assistance that shaped it. Domain results by zone, not bare sums. State each of the five domains relative to its interval-specific boundary: above cutoff, monitoring zone, or at or below cutoff. Raw sums alone carry no meaning across intervals, and there is no global total; the five-domain profile is the result. Pitfall: "ASQ score 45" or "passed." A sum without its interval and domain is uninterpretable, and pass language erases the zone structure the instrument is built on. The monitoring-zone plan, dated. For any monitoring-zone domain, document the contextual review (health, opportunity to practice the skill, language, culture), the parent discussion, the targeted activities provided for that domain, and a specific rescreen date or interval. Escalate to referral when a parent or provider is concerned, the same domain has sat in the zone before, several domains cluster low, or regression or clinical findings accompany the result. Pitfall: "Passed; watch fine motor." That entry erases the category, records no intervention, and creates no auditable follow-up commitment, which is exactly what the zone obligates. The Overall section, in words. Record the Overall section's answers in words: parent concerns about hearing, vision, talking, or behavior, and any reported loss of previously acquired skills. These untotaled items can override a reassuring domain profile; regression in particular is a red flag that acts regardless of scores. Pitfall: "Passed all domains" while the Overall section reports the child does not respond to quiet speech. The section exists to catch what the thirty scored items cannot, and skipping it is the most consequential omission in ASQ charting. Referrals and the clocks. For a below-cutoff result or an acted-on concern, document the referrals placed and when: early intervention (in the US, Part C child-find rules put a seven-day clock on referral after identification, and a 45-day clock on the evaluation and initial IFSP after referral), audiology for any communication concern, and the medical evaluation with the primary care clinician. Name who owns confirmation. Pitfall: A positive screen that ends at the score, or a referral charted as a diagnosis. The screen identifies suspected delay; eligibility requires a multidisciplinary evaluation that no single procedure, the ASQ-3 included, can constitute. Interpretation boundary and follow-up. State the result as a screening outcome: a positive screen for possible delay, not a diagnosis; a negative screen that does not rule out emerging concerns, especially milder ones the instrument detects less reliably. Close with the follow-up: the next routine screening age, the dated rescreen for monitored domains, or the referral confirmations pending. Pitfall: "Developmental delay" entered as a finding from a screen, or a declined screen charted (or billed) as completed. Offer, refusal, surveillance continued, and the re-offer plan are what a declined screen documents. ### Blank template (copy and adapt) ASQ-3 DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Screener: [ ] Interpreting clinician: [ ] Interval administered: [ ]-month questionnaire Chronological age: [ ] months [ ] days Born early: [N / Y: [ ] weeks] Corrected age (if under 24 months and 3+ weeks early): [ ] months [ ] days (corrected age selected the interval: Y/N) Respondent + relationship: [ ] Language of form: [ ] Interpreter or interview assistance: [ ] Completeness: [all domains scorable / domain [ ] adjusted for [1-2] blank items / domain [ ] NOT scorable (>2 blank)] Domain results (zone, not just sum): Communication: [above cutoff / monitoring zone / below cutoff] Gross motor: [ ] Fine motor: [ ] Problem solving: [ ] Personal-social: [ ] Overall section (in words): [hearing / vision / talking / behavior concerns; any loss of previously acquired skills; parent worries] Monitoring-zone plan (per domain): [contextual review, parent discussion, targeted activities provided, rescreen DATE, escalation criteria] Referrals (below cutoff or acted-on concern): [early intervention (date placed; 7-day clock met), audiology, medical evaluation] Owner: [ ] Interpretation: [positive screen for possible delay, not a diagnosis / negative screen; does not rule out milder or emerging delay] ASQ:SE-2 (separate instrument): [not administered / administered and documented separately] Follow-up: [next routine screening age / rescreen date / referral confirmation] Declined screen: [offered, declined, surveillance continued, re-offer plan; not billed as completed] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces questionnaire items, cutoff values, or summary-sheet layouts. ## Sample ASQ-3 documentation (fictional) Scenario: a 9-month well-child visit for a former preterm infant, where corrected age selects the interval and one domain lands in the monitoring zone, documented with the plan the zone obligates. All details are fictional. Patient: L.M., 9 mo (born 9 weeks early) · Setting: Pediatric primary care, 9-month well-child visit · Clinician: A. Rivera, MD · Note date: 08/20/2026 Screening: ASQ-3 completed today in Spanish by the child's mother without assistance. Chronological age 9 months 4 days; born at 31 weeks, 9 weeks before the due date, so corrected age (6 months 27 days) selected the questionnaire: the 6-month interval, within its administration window. Correction applies until 24 months of chronological age and is documented here so the next screener repeats the arithmetic. All thirty domain items answered; no adjusted scores needed. Results: Communication, fine motor, problem solving, and personal-social: above their interval-specific cutoffs. Gross motor: in the monitoring zone (between one and two standard deviations below the domain mean for this interval), not below cutoff. Overall section reviewed in words with the mother: no concern about hearing, vision, or use of both hands, no loss of previously acquired skills; she notes he spends most weekdays in a bouncer seat at his grandmother's home with limited floor time. Monitoring-zone plan: Contextual review suggests limited opportunity to practice trunk and floor skills rather than regression or an accompanying concern, and this is the first screen with any domain in the zone. Discussed with the mother: this result is neither a pass nor a referral; it calls for targeted practice and a scheduled recheck. Floor-time and supported-sitting activities for the gross motor domain were demonstrated and provided in Spanish, and the grandmother will be shown at pickup today. Rescreen scheduled for the last week of October (nine weeks out, appointment booked today), with the interval matched to corrected age at that date. Earlier referral criteria reviewed with the mother: any parent or provider concern, loss of skills, or a second screen with gross motor in the zone will prompt early-intervention referral without waiting. Interpretation: Screening result, not a diagnosis: four domains above cutoff and one monitoring-zone domain with a plausible opportunity explanation, a reassuring Overall section, and no red flags. In a former preterm infant a reassuring screen at corrected age carries strong negative predictive value, while any future positive result would still need evaluation rather than being read as a diagnosis. Preterm follow-up surveillance continues alongside screening rather than being replaced by it. Administrative: Completed information summary scanned to the record and results entered in the screening fields (instrument, interval, both ages, domain zones, plan), consistent with our Medicaid program's documentation requirements for developmental screening. One standardized developmental screen completed, scored, and documented this visit. Follow-up owner: Dr. Rivera; front desk to confirm the late-October rescreen appointment before the family leaves. This sample is fictional and for educational purposes. It does not describe a real patient or record; the ages, results, dates, and details are invented to show documentation structure and are not clinical guidance. No questionnaire items or cutoff values are reproduced. ↑ Back to the template and downloads ### Why this sample works The interval is justified by arithmetic on the record: both ages, the prematurity basis, and the statement that corrected age selected the form, which is the step most often skipped and most often wrong. Every domain is charted by zone against its own interval boundary, with no bare sums, no global total, and no "passed." The monitoring zone gets what it obligates: contextual review, a parent discussion, targeted activities actually provided, a dated rescreen, and explicit escalation criteria, instead of "watch." The Overall section is reviewed in words and given standing to override the domain profile, and the opportunity explanation (bouncer seat, limited floor time) is documented rather than assumed. The interpretation stays a screen: no diagnosis from a screen, preterm-specific reassurance framed correctly, and the administrative trail (summary sheet scanned, fields entered, one screen documented) matches what payers audit. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the screening duty comes from professional guidance and payer policy, and the referral duty from federal law. The American Academy of Pediatrics recommends developmental surveillance at every well visit, standardized screening at 9, 18, and 30 months, and autism-specific screening at 18 and 24 months (CONVENTION: the AAP recommends, it does not legislate). Medicaid EPSDT makes periodic screening a federal entitlement (LAW) whose instrument choices are state policy: Texas names the ASQ among accepted tools at designated ages and requires early-childhood-intervention referral within seven days of identifying suspected delay in a child under 36 months; Minnesota requires the chart to show the instrument name, the scores, and the guidance provided; North Dakota lists ASQ-3 under 96110 and ASQ:SE-2 separately under 96127 (PAYER POLICY, state by state). Billing follows the same logic: 96110 per completed, scored, documented standardized developmental screen, with same-day multiples and modifiers governed by the specific payer, and behavioral instruments under 96127; a declined screen generates no billable unit. IDEA Part C is the legal frame after a positive screen: primary referral sources, physicians included, must refer within seven days of identification, the lead agency then has 45 days to complete evaluation and the initial IFSP meeting, and no single procedure, an ASQ-3 score included, can be the sole basis of eligibility (LAW). Head Start programs must complete a research-based screen within 45 days of entry (LAW for grantees). The evidence posture is worth one honest sentence in counseling: the USPSTF finds insufficient evidence for universal speech-language screening in asymptomatic children, while the AAP recommends universal standardized screening; both positions are current. Canada and Australia run different screening architectures, and a defensible chart names the one it sits in. Canadian well-child care is organized around the Rourke Baby Record's surveillance framework, with Ontario funding an enhanced 18-month visit that requires an age-appropriate developmental review (commonly the Rourke plus the Looksee checklist) rather than mandating the ASQ-3 (PAYER POLICY, provincial); the Canadian Task Force on Preventive Health Care recommends against population-based standardized screening of asymptomatic children aged one to four while supporting surveillance and the assessment of identified concerns, a genuine divergence from the US model, so a Canadian ASQ-3 administration is usually concern-triggered and should say so (CONVENTION). Australia's states each run their own child health record and schedule (the NSW Blue Book, Victoria's ten Key Ages and Stages visits, Queensland's Red Book pathways using PEDS), none of which mandates the ASQ-3 universally; the ASQ-TRAK adaptation is used in Aboriginal and Torres Strait Islander contexts, and from November 2026 a Medicare-supported health assessment for 3-year-olds (through existing item families) adds a national touchpoint without prescribing an instrument (CONVENTION and state policy). In every jurisdiction the same two rules hold: the screen identifies, the evaluation determines; and prematurity correction follows the instrument's own rule (three or more weeks early, corrected until 24 months) rather than any local surveillance convention. Licensing and accuracy both reward precision. The ASQ-3 is a licensed instrument, not a free form: a purchased master set permits paper reproduction for that purchasing site only (each branch needs its own set), electronic reproduction is prohibited outright, building questionnaire content into an EHR or portal requires a separately priced license from the publisher's rights department, and posting or scoring the questionnaires on any website, free or not, password-protected or not, is unauthorized, which is why the third-party ASQ calculators that rank in search results should not be treated as authoritative or linked from clinical materials. What a practice may freely do is scan completed summary sheets and enter results into record fields, use the publisher's ASQ Online system for sanctioned digital administration, and bill usual screening reimbursement. On accuracy, the honest frame is high specificity with modest sensitivity for milder delay: the publisher's aggregate figures (sensitivity 86 percent, specificity 86 percent) sit above the best pooled independent estimate (about 0.77 and 0.81, with very low certainty in domain-level analyses), and a large primary-care study found sensitivity for any delay as low as 35 percent in younger children and 24 percent in older ones at specificity near 90 percent, with no competing screener superior overall. In preterm follow-up the pattern inverts usefully: at 24 months corrected age a pass carried 100 percent negative predictive value in a very-preterm cohort, while positive screens over-identify, and lower caregiver education measurably reduces classification accuracy, all of which belongs in interpretation rather than in a disclaimer. The Ages & Stages Questionnaires and ASQ are registered trademarks of Paul H. Brookes Publishing Co., Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common ASQ-3 documentation errors reviewers flag The numbers behind these errors are specific. In the largest independent primary-care study , ASQ-3 sensitivity for any delay was 35.2 percent in children up to 42 months and 23.5 percent in older children, at specificity near 90 percent; the 2022 meta-analysis pooled 0.77 sensitivity and 0.81 specificity overall but about 0.41 sensitivity for motor delay; in very preterm children at 24 months corrected age a pass carried 100 percent negative predictive value; and the publisher's own technical data rest some interval estimates on as few as 19 cases. The BastionGPT Clinical Advisory Board sees the same errors most often in ASQ-3 documentation reviews: The monitoring zone charted as a pass. "Passed; watch fine motor." The zone is a distinct result between one and two standard deviations below the domain mean, and it obligates contextual review, a parent discussion, targeted activities, and a dated rescreen, with referral when concern, repetition, clustering, or regression accompanies it. "Watch" records none of that and erases the category. The wrong interval, usually by skipping corrected age. The nearest month label chosen, or a 9-month form given to a former 31-week infant whose corrected age selects a younger interval. The cutoffs were built on corrected ages (three or more weeks early, corrected until 24 months), and exact age in months and days picks the window; chart both ages and the arithmetic, and re-administer rather than transpose when the wrong interval was completed. "Passed all domains" over a live Overall concern. Five above-cutoff domains charted as a clean negative while the Overall section records inconsistent response to sound, unclear speech, or lost skills. The section is untotaled precisely so those answers can override the profile; record them in words and act on them, with regression treated as a red flag regardless of scores. A global total, or ASQ:SE-2 folded in. Domain sums added into one "ASQ score," or the separate ASQ:SE-2 charted as a sixth domain. No global total exists, each domain reads only against its own interval boundary, and the ASQ:SE-2 is a different instrument with a single score, different cutoff logic, and the opposite direction (higher means concern), documented and billed separately. A screen charted as a diagnosis or an eligibility finding. "Developmental delay" entered as a finding, or an ASQ-3 result treated as Part C eligibility. Below cutoff is a positive screen: it starts the referral clocks (seven days to refer; 45 days from referral to evaluation and IFSP) and the medical workup (audiology for any communication concern), and eligibility requires a multidisciplinary evaluation no single procedure can constitute. Scoring what should not be scored. A domain with more than two blank items scored anyway, an adjusted score presented as ordinary, a wrong-interval result interpreted, or a declined screen billed as completed. Chart unscorable domains as unscorable, identify adjusted scores, re-administer the correct interval, and document a declined screen as offered, declined, surveillance continued, re-offer planned. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (interval, both ages and the prematurity basis, respondent and language, domain zones, Overall answers, plan) and it drafts the documentation block: interval arithmetic shown, zones stated, the monitoring-zone plan with its rescreen date and escalation criteria, and the referral clocks, ready for your review. Cross-check a finished note for the gaps reviewers flag: a monitoring-zone result charted as passed, a missing corrected-age calculation, an Overall concern left unaddressed, a global total, or screen language that reads as a diagnosis. Draft the follow-up documentation: the rescreen comparison (same domain, right interval for age at rescreen), the referral-confirmation note with the Part C clock status, or the declined-screen entry with surveillance and re-offer, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the ASQ-3 scored? By domain, against interval-specific boundaries, never as one number. The parent answers six questions in each of five domains (communication, gross motor, fine motor, problem solving, personal-social); responses of demonstrated, emerging, and not yet convert to descending point values and sum within the domain. Each domain total then reads against its own interval's empirically derived cutoff (two standard deviations below that domain's mean) and monitoring zone (one to two standard deviations below), producing one of three results: above cutoff, monitoring zone, or at or below cutoff. There is no valid global ASQ-3 total, and the closing Overall section, which asks about hearing, vision, talking, behavior, and loss of skills, is deliberately untotaled but can change the action regardless of the domain profile. A defensible note therefore reports five zones plus the Overall answers, not a sum. What does an ASQ-3 monitoring-zone score actually require? An active, documented plan; it is neither a pass nor an automatic referral. The publisher's guidance for a domain between one and two standard deviations below the mean is to review context (health, fatigue, opportunity to practice the skill, language, culture, blank items), discuss the result with the parent, provide targeted activities for that domain, and set a specific rescreen date, commonly within about one to four months. Referral instead of rescreening is appropriate when a parent or provider is concerned, the same domain has landed in the zone before, several domains cluster low, or regression or clinical findings accompany the result, and some state programs referral-trigger at broader thresholds. Two honesty points belong in the record's logic: no published study isolates the zone's predictive value, so its handling rests on publisher guidance, and "passed, watch" is the chart entry that erases all of it. How does prematurity correction work on the ASQ-3? It selects the questionnaire, not just a footnote. For a child born three or more weeks before the due date, subtract the weeks of prematurity from chronological age and use that corrected age, in months and days, to choose the interval, up until 24 months of chronological age; from age two, use chronological age for everyone. The rule is instrument-specific and consequential because the normative cutoffs were developed using corrected ages for preterm children: screening a former 31-week infant on the chronological 9-month form instead of the corrected-age interval compares the child against the wrong group and is the commonest way a true pass becomes a false monitoring-zone or below-cutoff result. The defensible chart shows both ages, the gestational basis, and the statement that corrected age selected the form, so the next screener can repeat the arithmetic. What if questionnaire items are blank, or the parent completed the wrong interval? Both have fixed rules. Blank items: a domain with one or two missing responses can be scored with the authorized adjustment (the completed items' average assigned to the blanks) and must be identified as adjusted in the record; a domain with more than two missing responses is not scorable and is charted that way, never as a zero, a pass, or a referral. Wrong interval: a questionnaire completed outside its authorized age window (wrong form sent home, corrected age not applied) does not produce a valid result in any zone; recalculate the age, administer the correct interval, and note that the earlier form was not interpretable, rather than transposing its numbers, because both the sampled skills and the cutoff structure differ between intervals. Each domain has only six items, so a single blank removes a sixth of that domain's data, which is why the rules are strict. Is a below-cutoff ASQ-3 score a diagnosis of developmental delay? No: it is a positive screen, and the chart should show what it starts rather than what it concludes. In the US the referral machinery is legal: physicians and other primary referral sources must refer a child with a suspected disability to early intervention as soon as possible and within seven days of identification, and the Part C lead agency then has 45 days from referral to complete the multidisciplinary evaluation and initial IFSP meeting; eligibility requires multiple procedures and sources, and no single instrument, the ASQ-3 included, can be the sole criterion. A below-cutoff communication domain also warrants a concurrent medical workup, audiology first, because hearing loss is the reversible cause the screen cannot see. Defensible language: "positive developmental screen; Part C and audiology referrals placed today; medical and developmental evaluation arranged," never "ASQ shows developmental delay." Is the ASQ:SE-2 part of the ASQ-3? No. The ASQ:SE-2 (Ages & Stages Questionnaires: Social-Emotional, Second Edition, 2015) is a separate instrument with a different architecture: one questionnaire score rather than five domains, cutoffs built on a different statistical basis, and the opposite direction, since higher ASQ:SE-2 scores indicate concern. The ASQ-3's personal-social domain samples related territory but is not a social-emotional screen, so "ASQ-3 social-emotional domain passed" is a documentation error, and the two instruments' results are never combined into one profile or total. Billing separates them the same way: developmental screening instruments under 96110 and behavioral or emotional instruments under 96127, each requiring its own completed, scored, documented administration; several state Medicaid manuals list the two instruments under exactly those separate codes. Can we put the ASQ-3 in our EHR or use a free online ASQ calculator? Results yes, content no, and the calculators are unauthorized. The purchased master set licenses paper reproduction for that purchasing site only (each branch buys its own), prohibits electronic reproduction outright, and bars revenue-generating reproduction; practices may freely scan completed summary sheets and key results into EHR fields, but building the questionnaire itself into an EHR or portal requires a separately priced license from the publisher's rights department, and the sanctioned digital administration channel is the publisher's own ASQ Online system with its Family Access add-on. The publisher states the questionnaires may not be posted on any website, even password-protected, and a free tool gets no exemption, so the third-party ASQ calculators ranking in search results are not authorized scoring sources and do not belong in clinical materials or patient instructions. How accurate is the ASQ-3? Better at ruling out than ruling in, and the note should use it that way. The publisher's aggregate figures are sensitivity 86.1 percent and specificity 85.6 percent, but they mask weaker age bands and some interval estimates built on under 20 cases. The best pooled independent estimate is about 0.77 sensitivity and 0.81 specificity for any delay, dropping to roughly 0.41 for motor and 0.44 for cognitive-language analyses with very low certainty, and the largest US primary-care comparison found sensitivity for any delay of 35 percent in younger and 24 percent in older children at specificity near 90, with no competing screener (PEDS, SWYC) superior overall. Preterm follow-up inverts the frame: at 24 months corrected age a pass carried 100 percent negative predictive value in a very-preterm cohort, while positives over-identify, and lower caregiver education measurably reduces classification accuracy. Practical reading: trust a negative screen but never let it silence a concern, and treat every positive as a referral trigger, not a finding. Can BastionGPT draft my ASQ-3 documentation? Yes. Give it the facts (interval, chronological and corrected ages with the gestational basis, respondent and language, domain zones, Overall answers, and your plan) and it drafts the documentation block: interval arithmetic shown, five zones stated, the monitoring-zone plan with targeted activities, a rescreen date, and escalation criteria, and the referral entries with the Part C clock language, ready for your review. It can also cross-check a finished note for a monitoring-zone result charted as passed, a missing corrected-age calculation, an unaddressed Overall concern, or diagnosis language a screen cannot support. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: Brookes Publishing, ASQ-3 product page (21 intervals, ages 1 to 66 months, 2009 edition), monitoring-zone guidance , prematurity correction rule , omitted-item rules , common screening mistakes , and ASQ-4 status FAQ (in development, no date). Licensing: Brookes Publishing, photocopying and use guidelines (site-bound license; electronic reproduction prohibited), EHR guidance (results permitted; content embedding separately licensed), and website-posting prohibition . Accuracy: ASQ-3 technical appendix (norming sample, aggregate and age-band estimates); Sheldrick RC and colleagues, 2020, JAMA Pediatrics, comparative accuracy of ASQ-3, PEDS, and SWYC ; Muthusamy S and colleagues, 2022, JAMA Pediatrics, systematic review and meta-analysis ; Kerstjens JM and colleagues, 2015, ASQ-3 in very preterm children at 24 months corrected ; caregiver-education effects, PLOS One preterm cohort . United States screening and referral law: Lipkin PH, Macias MM, AAP Council on Children with Disabilities, 2020, Pediatrics, promoting optimal development (surveillance and 9-, 18-, 30-month screening); CMS, EPSDT benefit ; eCFR, 34 CFR 303.303 (seven-day referral) and Part C evaluation rules ; 45 CFR 1302.33 (Head Start 45-day screening). State payer examples: Texas Medicaid, Texas Health Steps manual (named instruments; seven-day ECI referral); Minnesota, Child and Teen Checkups instruments at a glance ; North Dakota HHS, preventive services billing manual (ASQ-3 under 96110; ASQ:SE-2 under 96127); AAP, screening coding fact sheet . Evidence controversy: USPSTF, speech and language screening statement (insufficient evidence); Canadian Task Force on Preventive Health Care, developmental delay guideline (against screening asymptomatic children); Rourke Baby Record, Canadian well-child framework ; Ontario, enhanced 18-month visit . Australia: NSW Health, Blue Book child health record ; Children's Health Queensland, Red Book program ; Services Australia, 2026-27 budget: 3-year-old health assessment from November 2026 . ## Related templates Child & Adolescent Developmental Assessment Bayley-4 Report Write-Up M-CHAT-R/F Score Documentation PSC-17 Score Documentation DAYC-2 Report Write-Up Autism Spectrum Evaluation Report GFTA-3 Report Write-Up PLS-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/m-chat-r All note templates # M-CHAT-R/F Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The M-CHAT-R/F (Modified Checklist for Autism in Toddlers, Revised, with Follow-Up) is a two-stage autism screen for toddlers: a 20-item parent questionnaire, then a structured Follow-Up interview that resolves medium scores. Pediatric and family practices use it at 18- and 24-month visits. A medium score without the Follow-Up is an incomplete screen, not a result. This page covers how to document M-CHAT-R/F results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Parents complete the questionnaire; any trained staff member (nurse, medical assistant, clinician) administers the structured Follow-Up; the clinician interprets and refers; no certification is required Audience Pediatric and family practices, early-intervention and Part C programs, diagnostic and autism evaluation teams, audiology, payers auditing screening claims, families Typical length 4 to 10 chart lines · the questionnaire scores in under 2 minutes; the structured Follow-Up adds several minutes for the flagged items Format family Two-stage parent-report autism screen (20 yes/no items, three reverse-keyed, then a structured Follow-Up interview for medium scores) When it's used 18- and 24-month well-child visits in US practice, concern-triggered screening between 16 and 48 months, EPSDT schedules, rescreening after an early negative Standards context Copyright 2009 Robins, Fein, and Barton; free unmodified clinical use with licensed electronic distribution; AAP recommends universal 18- and 24-month screening; no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the M-CHAT-R/F? The M-CHAT-R/F (Modified Checklist for Autism in Toddlers, Revised, with Follow-Up) is the most widely used autism screen in toddler primary care, copyrighted in 2009 by Diana Robins, Deborah Fein, and Marianne Barton, validated in a 16,071-child study published in 2014, and revised most recently in March 2025, when the authors replaced "risk" with "likelihood" wording throughout while leaving the scoring untouched. It is a two-stage instrument by design. Stage one is a 20-item parent questionnaire (three items reverse-keyed; legacy shortcuts like critical-item and best-7 scoring are expressly retired) whose total falls into one of three bands: low likelihood (0 to 2), medium (3 to 7), or high (8 to 20). Stage two, the structured Follow-Up interview, exists for the medium band: a trained staff member re-visits only the flagged items with standardized probes, and the number of items still failed after the interview (2 or more positive, 0 or 1 negative) supersedes the questionnaire total as the final screen result. High scores skip the Follow-Up and refer immediately; low scores rescreen at the next visit. The load-bearing fact is what the second stage is worth: in the validation sample, the questionnaire threshold alone produced 700 false positives against 116 for the completed two-stage algorithm, and the Follow-Up resolved 63.2 percent of first-stage positives to negative while raising the positive predictive value for autism from 13.8 to 47.5 percent. A medium-range score charted without the Follow-Up is therefore not a positive result; it is an incomplete screen, and "M-CHAT positive, score 5" is the single most damaging entry a chart can carry. The instrument screens; it never diagnoses: about half of final positives are ultimately diagnosed with autism and roughly nine in ten have some developmental concern, which is why a positive screen triggers a two-track referral (diagnostic evaluation and early intervention, simultaneously) rather than a label. Broad developmental screening belongs to instruments like the ASQ-3 , and the diagnostic phase to the tools on the ADOS-2 and ADI-R pages. ## Who uses M-CHAT-R/F documentation and when US pediatric and family practices administer it at the 18- and 24-month well-child visits, where the American Academy of Pediatrics recommends universal autism-specific screening alongside general developmental screening (a position it has maintained against the US Preventive Services Task Force's insufficient-evidence statement, which is itself under active revision). The validated range is 16 to 30 months, with author guidance now permitting use through 48 months; children screened before age two get rescreened at 24 months because some presentations surface only later. Early-intervention and Part C intake teams read the documentation to act on referral clocks, diagnostic programs use the two-stage record to triage year-long waitlists, and audiologists receive the hearing referral that accompanies communication concerns. Canadian practice diverges deliberately: the Canadian Paediatric Society recommends surveillance for all children with autism-specific screening reserved for identified risk or concern, so a Canadian administration should record its indication. Australian systems lean on surveillance too, with Victoria and Tasmania running the clinician-observation SACS-R program through child health nurses rather than parent-report screening, and the national autism guideline governing what happens after concern arises. Results feed an autism evaluation report when the diagnostic phase begins; rating-scale measures like the GARS-3 and CARS-2 belong to that later phase, not to screening. ## How to document M-CHAT-R/F results in the chart No US, Canadian, or Australian authority prescribes an M-CHAT-R/F note format, and the authors publish an algorithm rather than a charting standard. What survives review is a record in which the two stages cannot be confused: the questionnaire total and band, the Follow-Up status and its own score, the final classification, and the referral chain with dates. Each element below carries the pitfall that most often undermines it. Instrument, version, and stage design. Name it in full: M-CHAT-R/F, not "M-CHAT" or "autism screen," using the current authorized form (the March 2025 revision uses likelihood wording; older reposted PDFs circulate with stale terms and contacts). The 20-item total is the only sanctioned first-stage score: legacy 23-item rules, critical items, and best-7 subsets are retired. Pitfall: Legacy scoring contaminating a current administration: 23-item cutoffs, critical-item counts, or a third-party form that was modified, which the license prohibits and which voids comparability. Respondent, language, and administration. Record who completed the questionnaire and their relationship to the child, the language version used (authorized translations exist in more than 65 languages and include the Follow-Up), interpreter participation, and whether it was self-completed or orally administered. Reconcile blank or double-marked items with the caregiver before scoring rather than guessing. Pitfall: "M-CHAT completed." Translation, literacy, and administration route all shape first-stage scores, and a silently corrected form is no longer the parent's report. First-stage total and band. Chart the questionnaire total out of 20 and its band: low likelihood (0 to 2), medium (3 to 7), or high (8 to 20), with the age in months at administration. The total stays in the record even after the Follow-Up; the two scores together are the screen. Pitfall: A band with no total, or a total with no band. The next reader needs both to see what the algorithm required next. Follow-Up status and score. For a medium total, chart the structured Follow-Up explicitly: completed (with the number of items still failed after the interview), bypassed under the high-score rule, pending, unavailable, or declined, and who administered it. The interview covers only the flagged items, uses the standardized probes, and its result supersedes the questionnaire for the final classification. Pitfall: A medium score with no Follow-Up field at all, the error an entire 2026 health-network study could not audit its way past because the EHR stored only the questionnaire. Build the second field, and never substitute an informal chat for the structured interview. Final classification, one of five. Close the screen with one status: negative on the questionnaire (low band); negative after Follow-Up (0 or 1 items remaining); positive after Follow-Up (2 or more remaining); positive on a high initial score (Follow-Up bypassed); or incomplete pending Follow-Up. The incomplete status is legitimate and honest when workflow interrupts the second stage. Pitfall: "M-CHAT positive, score 5." A medium questionnaire score is not a final result: in validation, first-stage-only scoring produced six times the false positives, and 63 percent of such positives resolved negative on interview. Referrals: two tracks plus audiology. For any positive screen, document simultaneous referral on two tracks: diagnostic evaluation and early-intervention or developmental services, neither waiting for the other, plus audiology when communication concerns exist. In the US, chart the Part C machinery: referral within seven days of identification, and the program's 45-day clock to evaluation and the initial IFSP; parent consent gates each step. Pitfall: A diagnostic waitlist as the only documented action. Early intervention requires no diagnosis, waitlists run a year in many systems, and the chart should show support starting while diagnosis is pending. Rescreen plan, surveillance, and refusal. A negative result is not clearance: chart the 24-month rescreen after any earlier screen, continued surveillance, and the rule that clinician or parent concern warrants evaluation regardless of score (misses concentrate in girls and at 18 months). For a declined referral after a positive screen, document informed refusal: what was recommended and explained, the family's decision and stated reason, alternatives offered, and the dated plan to revisit; the screen result itself stays positive. Pitfall: A negative screen closing the question a family keeps raising, or "parent refused" standing alone where an informed-refusal conversation belongs. ### Blank template (copy and adapt) M-CHAT-R/F DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Age: [ ] months Clinician: [ ] Instrument: M-CHAT-R/F, current authorized form [English / authorized translation: [ ]] Respondent + relationship: [ ] Administration: [self-completed / orally administered] Interpreter: [ ] Blank or ambiguous items reconciled with caregiver: [Y/N/none] First stage: total [ ]/20 Band: [low 0-2 / medium 3-7 / high 8-20] Follow-Up (structured interview, flagged items only): Status: [completed / bypassed (high band) / pending / declined] Administered by: [ ] Items remaining failed: [ ] Final classification: [negative on questionnaire / negative after Follow-Up / positive after Follow-Up / positive on high initial score / INCOMPLETE pending Follow-Up] Clinical context: [parent concerns, clinician observations, regression, hearing or language concerns, other screening results] Referrals (positive screens; two tracks, simultaneous, dated): Diagnostic evaluation: [date placed; waitlist status] Early intervention / Part C: [date placed; 7-day clock met; 45-day evaluation clock noted] Audiology: [ ] Interim plan while diagnosis pends: [developmental services, speech or hearing evaluation, parent coaching, follow-up visit date] Rescreen plan: [24-month rescreen after an early screen / next visit; concern warrants evaluation regardless of score] Declined recommendations: [what was recommended and explained, family's decision and reason, alternatives, dated plan to revisit; screen classification unchanged] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces questionnaire items, the Follow-Up probes, or scoring materials. ## Sample M-CHAT-R/F documentation (fictional) Scenario: a 24-month well-child visit where a medium questionnaire score goes through the structured Follow-Up, stays positive, and launches the two-track referral chain with an interim plan. All details are fictional. Patient: D.O., 24 mo · Setting: Pediatric primary care, 24-month well-child visit · Clinician: S. Patel, MD · Note date: 08/17/2026 Screening: M-CHAT-R/F completed at today's 24-month visit by the child's father using the authorized Spanish version, self-completed in the waiting room; two double-marked items were reconciled with him before scoring. First-stage total 4/20, medium likelihood band. Per the algorithm, our nurse administered the structured Follow-Up interview for the four flagged items in Spanish with a qualified medical interpreter supporting the probes; two items remained failed after the interview. Final classification: M-CHAT-R/F screen positive after Follow-Up (initial 4, Follow-Up 2). Context: Father reports the child uses about ten single words, points rarely, and responds inconsistently to his name, though he enjoys rough-and-tumble play and makes eye contact with family. No loss of previously acquired skills reported. In the room he did not respond to name over three trials, showed limited pointing, and lined up blocks repetitively. General developmental screening earlier this year flagged communication; today's result is convergent. Counseling: Reviewed with the father in plain terms through the interpreter: this is a positive screen, not a diagnosis; about half of children with this result are found to have autism and most of the rest have another developmental concern worth addressing, which is why evaluation matters either way. He was receptive and asked about next steps; written material in Spanish provided. Referrals and interim plan: Two-track referrals placed today, neither waiting on the other: comprehensive diagnostic evaluation through the regional developmental program (current waitlist quoted at nine months; waitlist date documented) and early-intervention referral to the state Part C program, placed the same day, within the seven-day referral window; the program's 45-day evaluation and IFSP clock was explained to the family, and eligibility requires no diagnosis. Audiology referral placed for the inconsistent response to speech. Interim plan while diagnosis pends: Part C developmental services on eligibility, speech-language evaluation through the program, parent-coaching resources provided, and a primary care follow-up visit in six weeks to confirm referral uptake and answer questions; return sooner for regression or new concerns. Administrative: Both stage scores, the final classification, the administering staff, and the language of administration entered in the screening fields; one standardized autism-specific screen completed, scored, and documented this visit alongside no other instrument. Rescreening is not scheduled because the diagnostic pathway is now active; surveillance continues at every visit. Follow-up owner: Dr. Patel; care coordinator to confirm receipt of both referrals within two weeks. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, observations, dates, and details are invented to show documentation structure and are not clinical guidance. No questionnaire items or Follow-Up probes are reproduced. ↑ Back to the template and downloads ### Why this sample works The two stages cannot be confused: the questionnaire total and band, the Follow-Up administration with its own score, and the final classification are all separately on the record. The language and administration facts (authorized Spanish version, self-completion, interpreter-supported Follow-Up, reconciled items) are documented, which is what makes the scores interpretable. The screen-not-diagnosis counseling is charted with the honest numbers behind it, so the referral urgency and the diagnostic uncertainty are both preserved. The referral chain is two-track and dated, with the Part C clocks named, audiology included, and an interim plan that starts support while the nine-month diagnostic waitlist runs. The closing administrative lines match what payers and auditors look for: instrument, both scores, staff, language, one screen documented, and a named owner for referral confirmation. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: recommendation, law, and payment sit at three different levels, and the chart should reflect all three. The American Academy of Pediatrics recommends universal autism-specific screening at 18 and 24 months alongside general developmental screening (CONVENTION; its 2020 clinical report was reaffirmed in October 2025), while the US Preventive Services Task Force's 2016 insufficient-evidence statement remains the published federal position, with an update actively in progress as of August 2026, and family medicine's national body aligns with the task force rather than the AAP, which partly explains practice variation. Medicaid EPSDT obligates periodic screening as a federal benefit (LAW) with instruments and schedules set by states (PAYER POLICY): Massachusetts, for example, schedules autism screening at 18 and 24 months and pays 96110 separately for a developmental and an autism screen at the same visit with state-specific result modifiers, and other states differ, so units, modifiers, and same-day edits are payer questions, never a substitute for charting the actual scores. IDEA Part C is the legal spine after a positive screen: primary referral sources must refer within seven days of identification, the program owes evaluation and an initial IFSP within 45 days of referral, parents may self-refer, and no diagnosis is required (LAW), while insurer coverage of autism-specific treatment typically gates on a completed diagnostic evaluation (PAYER POLICY), which is exactly why the two-track convention documents early intervention proceeding while the diagnostic waitlist runs. Canada and Australia diverge from US universalism, and the note should show which frame it sits in. The Canadian Paediatric Society recommends developmental surveillance for every child with autism-specific screening applied when risk or concern is identified rather than universally at fixed ages (CONVENTION), so a Canadian M-CHAT-R/F administration is usually concern-triggered and should record its indication; the national task force likewise found evidence insufficient for routine universal screening, and referral-to-assessment waits exceeding a year in many communities make the act-before-diagnosis interim plan more important, not less. Australia's NHMRC-approved national guideline for autism assessment and diagnosis (second edition, 2023) governs what happens after concern arises rather than mandating any screening instrument, Victoria and Tasmania run the clinician-observation SACS-R surveillance program through maternal and child health nurses in place of parent-report screening, and funded assessment flows through the MBS complex-neurodevelopmental pathways (items such as 135 and 289 with telehealth equivalents) and the NDIS early-childhood approach, which admits children under six on developmental concern without a formal diagnosis (PAYER POLICY). One equity finding travels across all three systems: real-world studies show screening completion, Follow-Up fidelity, referral uptake, and age at diagnosis differ by language, income, race, and setting even where the cutoff itself performs fairly, so the language of administration, interpreter use, and referral confirmation lines in the chart are equity documentation, not clerical detail. Version, licensing, and honesty about accuracy each need a line of care. The current instrument is the 2009-copyright M-CHAT-R/F as revised March 4, 2025 (terminology moved from risk to likelihood wording; cutoffs, reverse-keyed items, and the two-stage design unchanged), validated in 2014; stale third-party PDFs with legacy wording and old contact addresses circulate widely, so source the form from the authors' site. The license is generous and specific: clinical, research, and educational use is free with the instrument unmodified and the copyright line intact; a practice may build the unmodified questionnaire into its own internal EHR without a fee; but any distributed implementation, an EHR product, assessment platform, telehealth package, app, or any website presentation of the items, free or not, requires a license from the authors, and authorized translations must include the Follow-Up, not just the questionnaire. The authors' separate TASI is a diagnostic-support interview, not a successor screener, and no replacement edition is announced. On accuracy, chart-level honesty means holding two sets of numbers at once: validation performance (sensitivity around 0.85, specificity 0.99, autism PPV 47.5 percent after the Follow-Up) and real-world performance under imperfect implementation (sensitivity as low as 33 to 39 percent with PPV near 15 percent in large network studies, alongside earlier diagnoses for screen-positive children), with the gap driven by fidelity, Follow-Up completion, and rescreening, which is to say by exactly the things the documentation exists to secure. The M-CHAT-R/F is copyrighted by Diana Robins, Deborah Fein, and Marianne Barton. BastionGPT is not affiliated with, or endorsed by, the authors. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common M-CHAT-R/F documentation errors reviewers flag The numbers behind these errors are specific. In the 2014 validation study of 16,071 children, the questionnaire threshold alone produced 700 false positives against 116 for the two-stage algorithm, the Follow-Up resolved 63.2 percent of first-stage positives to negative, and autism PPV rose from 13.8 to 47.5 percent; in real-world network data sensitivity fell to 38.8 percent with screening and follow-through differing by demographic group; and a 2026 health-network study could not audit Follow-Up completion at all because the EHR stored only the questionnaire. The BastionGPT Clinical Advisory Board sees the same errors most often in M-CHAT-R/F documentation reviews: A medium score charted as positive without the Follow-Up. "M-CHAT positive, score 5." A total of 3 to 7 is an incomplete screen until the structured Follow-Up resolves it: first-stage-only scoring produced six times the false positives in validation, and nearly two thirds of such positives screen out on interview. Chart it as incomplete pending Follow-Up, schedule the interview, and never convert it to a final result for workflow convenience. Only one stage's score kept. A Follow-Up outcome with no questionnaire total, or a total with no Follow-Up field, often because the EHR has nowhere to put the second stage. The record needs both numbers plus the final classification (initial 5, Follow-Up 1, screen negative after Follow-Up), and an EHR that cannot hold the Follow-Up cannot produce an auditable screen; say so in the note and fix the build. Legacy M-CHAT scoring applied to the current instrument. 23-item rules, critical-item counts, best-7 subsets, or old cutoffs from reposted PDFs and secondary sites. The M-CHAT-R uses the full 20-item total with bands 0 to 2, 3 to 7, and 8 to 20, subsets are expressly retired, and the March 2025 revision changed wording only. Source the current form from the authors' site and score nothing else. A screen charted as a diagnosis, or referral gated on one. "Positive M-CHAT, consistent with autism," or early intervention deferred until the diagnostic evaluation returns. About half of final positives are diagnosed with autism and most others have another developmental concern; the convention is two simultaneous tracks (diagnostic evaluation and early intervention, plus audiology), with Part C requiring no diagnosis and the waitlist never the only documented action. A negative screen closing an open question. "M-CHAT negative, no concerns" at 18 months with no 24-month rescreen planned, or over a live parent concern. Misses concentrate at 18 months and in girls, some children screen negative early and positive at 24 months, and surveillance and concern override the score; chart the rescreen and the concern pathway explicitly. Administration facts missing, or unauthorized forms used. No respondent, language, or interpreter recorded; a modified or web-scraped version administered; items silently corrected by staff. The license requires the unmodified instrument with its copyright line, translations must include the Follow-Up, website and distributed electronic use require licensing, and reconciling ambiguous answers with the caregiver, documented, is the honest fix for a messy form. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (age, respondent, language, first-stage total and band, Follow-Up status and score, observations, referrals) and it drafts the documentation block: both stages separated, the final classification named, counseling language at screen level, and the two-track referral chain with the clocks, ready for your review. Cross-check a finished note for the gaps reviewers flag: a medium score standing as a final result, a missing Follow-Up field, legacy scoring language, a referral chain gated on diagnosis, or a negative screen with no rescreen plan. Draft the follow-up entries: the informed-refusal note that keeps the classification intact, the waitlist interim plan, or the 24-month rescreen documentation with the prior result carried forward, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the M-CHAT-R/F scored? In two stages, and both belong in the chart. Stage one: the parent answers 20 yes/no items (three are reverse-keyed), and the total of elevated-likelihood responses places the child in a band: low likelihood (0 to 2), medium (3 to 7), or high (8 to 20). Low scores are screen-negative, with rescreening at 24 months if the child is younger; high scores are screen-positive immediately, with the Follow-Up permissibly bypassed because it rarely changes the referral decision. Stage two applies to the medium band: a structured Follow-Up interview re-visits only the flagged items with standardized probes, and the count of items still failed decides the final result: 2 or more is screen-positive, 0 or 1 is screen-negative. The Follow-Up result supersedes the questionnaire total for classification, but the record keeps both numbers, and legacy shortcuts (critical items, best-7 subsets, 23-item rules) are expressly retired. Is an M-CHAT score of 3 to 7 a positive screen? Not yet: it is the band that requires the second stage, and treating it as positive is the instrument's most damaging misuse. In the 2014 validation of 16,071 children, applying the questionnaire threshold alone produced 700 false positives, against 116 after the completed two-stage algorithm, and the structured Follow-Up resolved 63.2 percent of first-stage positives to screen-negative while raising the positive predictive value for autism from 13.8 to 47.5 percent. The defensible chart entry for an unresolved medium score is "initial score in medium range; M-CHAT-R/F incomplete pending Follow-Up," with the interview scheduled; when the EHR's screening module has no Follow-Up field, say that too, because an un-auditable second stage is a workflow defect worth naming. A medium score becomes a result only when the Follow-Up says so. Who can administer the Follow-Up interview, and what does it involve? Any appropriately trained member of the team: the authors state the instrument requires no special certification, and the Follow-Up can be conducted by office staff, an early-intervention agency, a specialist, or a licensed electronic implementation. It is a structured branching interview, one flowchart per item, administered only for the items flagged on the questionnaire; the interviewer walks through standardized probes and concrete examples until each item resolves to pass or fail, and improvised general conversation does not count as the second stage. Practical honesty about time: the authors state the questionnaire scores in under two minutes but publish no minutes figure for the Follow-Up itself; the 5-to-10-minute estimates in circulation come from secondary sources. Chart who administered it, in what language, and with what interpreter support, because those facts govern the result's interpretability. What happens after a positive M-CHAT-R/F screen? Two referral tracks at once, plus hearing, with the chart showing dates and clocks. A positive screen (after Follow-Up, or on a high initial score) triggers referral for a comprehensive diagnostic evaluation and, simultaneously, referral to early intervention or developmental services, because intervention eligibility does not require a diagnosis and diagnostic waitlists commonly run months to a year. In the US, Part C law puts a seven-day clock on referral after identification and a 45-day clock on the program's evaluation and initial IFSP after referral, and parents may self-refer; audiology joins the chain whenever communication concerns exist, since hearing loss is the reversible mimic. The interim plan belongs in writing: developmental services on eligibility, speech-language evaluation, parent coaching, waitlist confirmation, and a dated primary care follow-up. What a positive screen never does is diagnose: roughly half of final positives are ultimately diagnosed with autism, and about nine in ten have some developmental concern worth evaluating either way. Does a negative M-CHAT-R/F rule out autism? No, and the chart should behave accordingly. Validation sensitivity around 0.85 fell to 33 to 39 percent in large real-world networks, misses concentrate at the 18-month screen and among girls, and longitudinal data show some children screen negative at 18 months and positive at 24, which is why the authors and the AAP direct rescreening at 24 months after any earlier screen and why surveillance continues at every visit regardless of results. A negative result with a live parent or clinician concern is not a closed question: current guidance says concern warrants evaluation whatever the score. Defensible chart language after an early negative: screen negative after Follow-Up, rescreen at the 24-month visit scheduled, parent advised to return sooner for loss of skills or new social-communication concerns. Can the M-CHAT-R/F be used after 30 months, or with a child who talks? Yes to both, with the nuance stated. The validation base is principally ages 16 to 30 months, and the authors' current guidance permits use through 48 months based on subsequent studies, while cautioning that older children with clear concerns may be better served by direct developmental or diagnostic evaluation than by another screening pass. Verbal ability changes neither rule: speaking in sentences does not invalidate a positive screen, autism occurs across the language spectrum, and a positive result at 30 months in a talkative child still warrants the two-track referral, interpreted within the full developmental picture. The defensible phrasing for the page and the chart alike: validated principally for 16 to 30 months, usable to 48 under current author guidance, with evidence less uniform outside the original window and concern always sufficient grounds for referral. Do the US, Canada, and Australia all recommend universal M-CHAT screening? No, and the difference is worth charting. The American Academy of Pediatrics recommends universal autism-specific screening at 18 and 24 months (reaffirmed October 2025), while the US Preventive Services Task Force's standing position is that evidence is insufficient for universal screening of children without raised concerns, an assessment currently under active update, and US family medicine's national body sides with the task force. The Canadian Paediatric Society recommends surveillance for all children and reserves standardized autism screening for identified risk or concern, so a Canadian administration should record its indication. Australia's national autism guideline governs assessment after concern arises rather than mandating screening, and Victoria and Tasmania run clinician-observation surveillance (SACS-R) through child health nurses instead of parent-report screening. The practical rule: state the frame your screen sits in, universal schedule or concern-triggered, and let the referral chain that follows a positive result look the same everywhere. Can we build the M-CHAT-R/F into our EHR or post it on our website? Internal EHR yes, website no, and the boundary is the authors' license. The instrument is free for clinical, research, and educational use provided it is unmodified (title, items, order, instructions, response format intact) and carries the copyright line, and the authors expressly permit a practice to build the unmodified questionnaire into its own internal EHR at no charge. Distribution is the licensed regime: an EHR vendor, assessment platform, telehealth product, app, or any website presentation of the items, free or paid, public or password-protected, requires a licensing agreement through the authors' current inquiry process, so a practice whose vendor displays the items should confirm the vendor holds the license. Authorized translations (more than 65 languages) must include the Follow-Up, and stale reposted PDFs with legacy wording should be replaced with the current form from the authors' site. What every implementation must preserve is the two-stage recordability: a build that stores only the questionnaire cannot produce an auditable screen. Can BastionGPT draft my M-CHAT-R/F documentation? Yes. Give it the facts (age, respondent and language, first-stage total and band, Follow-Up status with its score and administrator, observations, referrals and dates) and it drafts the full entry: both stages separated, the final classification named, screen-level counseling language, the two-track referral chain with the Part C clocks, and the rescreen or interim plan, ready for your review. It can also cross-check a finished note for a medium score standing as a final result, a missing Follow-Up score, legacy scoring language, or a negative screen with no rescreen plan, and it can draft the informed-refusal entry that keeps the classification intact. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Instrument and license, accessed August 2026: mchatscreen.com, official scoring page (bands, Follow-Up threshold, retired subsets), March 2025 instrument packet (likelihood terminology; 16-to-48-month guidance; internal-EHR permission), authorized translations , and the TASI (a separate diagnostic-support interview, not a successor). Validation and the value of the Follow-Up: Robins DL, Casagrande K, Barton M, Chen CA, Dumont-Mathieu T, Fein D, 2014, Pediatrics, validation of the M-CHAT-R/F (16,071 children; 700 versus 116 false positives; PPV 13.8 to 47.5 percent; 63.2 percent resolved); Kleinman JM and colleagues, 2008, original M-CHAT follow-up findings . Real-world performance and implementation: Guthrie W and colleagues, 2019, Pediatrics, universal screening in a primary care network (sensitivity 38.8 percent; demographic differences); Carbone PS and colleagues, 2020, screening in Utah primary care ; Harper K, Sarver DE, Sly J, 2026, rural network study (Follow-Up absent from the EHR; 29.5 percent dual-age screening). Meta-analyses and subgroup evidence: Wieckowski AT and colleagues, 2023, JAMA Pediatrics, 50-study meta-analysis (pooled sensitivity 0.83, specificity 0.94); Aishworiya R and colleagues, 2023, Pediatrics, pooled PPV analysis (57.7 percent autism, 89 percent any developmental disorder); Eldeeb S and colleagues, 2023, sex differences ; Dai YG and colleagues, 2021, PPV in Black and White children . United States policy: Hyman SL, Levy SE, Myers SM, AAP, 2020 (reaffirmed October 2025), Pediatrics, autism identification clinical report ; Lipkin PH, Macias MM, 2020, developmental surveillance and screening ; USPSTF, autism screening update in progress ; eCFR, 34 CFR 303.303 (seven-day referral) and 303.310 (45-day timeline); MassHealth, EPSDT screening codes appendix (96110 with state modifiers). Canada and Australia: Zwaigenbaum L, Brian JA, Ip A, Canadian Paediatric Society, 2019, early detection of autism spectrum disorder (surveillance for all; targeted screening); Autism CRC, national guideline for autism assessment and diagnosis (NHMRC-approved 2023 edition); SACS-R surveillance, developmental surveillance report ; MBS Online, complex neurodevelopmental assessment items . ## Related templates ASQ-3 Score Documentation GARS-3 Report Write-Up ADOS-2 Report Write-Up ADI-R Report Write-Up Autism Spectrum Evaluation Report Child & Adolescent Developmental Assessment PLS-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/psc-17 All note templates # PSC-17 Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The PSC-17 (Pediatric Symptom Checklist-17) is a free 17-item caregiver questionnaire screening children's psychosocial functioning, scored as one total plus internalizing, attention, and externalizing subscales, each with its own cutoff. Pediatric, family, and school-based clinicians use it at well visits and EPSDT checkups. A child can screen negative overall and positive on a subscale. This page covers how to document PSC-17 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Caregivers complete it (youth complete the self-report forms); trained staff score it; the pediatric, family medicine, or school-based clinician interprets, discusses, and acts Audience Pediatric and family practices, school-based health centers, integrated behavioral health teams, child welfare and care-management programs, payers auditing screening claims, families Typical length 4 to 10 chart lines · caregiver completion about 5 minutes, scoring 1 to 2 minutes Format family Brief broadband psychosocial screen (17 caregiver-report items rated 0 to 2; one total plus internalizing, attention, and externalizing subscales) When it's used Well-child visits from about age 4 through adolescence, EPSDT checkups, school-based health encounters, integrated-care intakes, rechecks after a positive screen Standards context Distributed free by the MGH team with its copyright line retained; named historically on some state Medicaid menus; no US, Canadian, or Australian authority mandates it; no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PSC-17? The PSC-17 (Pediatric Symptom Checklist-17) is the short form of American pediatrics' standard broadband psychosocial screen. The original 35-item Pediatric Symptom Checklist was validated by Michael Jellinek, Michael Murphy, and colleagues in 1988; the 17-item form was derived by William Gardner and colleagues in 1999 from parent reports on 18,045 primary-care children, and a 2016 restandardization of 80,680 outpatients replicated its structure. A caregiver rates 17 statements as never, sometimes, or often (0, 1, or 2 points), producing four concurrent signals: a total score (0 to 34, positive at 15 or higher) and three subscales, internalizing (0 to 10, positive at 5 or higher), attention (0 to 10, positive at 7 or higher), and externalizing (0 to 14, positive at 7 or higher). The thresholds are summed points, not counts of endorsed items, and "or higher" matters: a total of exactly 15 or an internalizing score of exactly 5 is positive. The family also includes the 35-item form (different cutoffs: 28, or 24 in the youngest children), the 35-item youth self-report (Y-PSC, cutoff 30, generally from age 11), a youth 17-item form with a thinner cutoff evidence base, and pictorial adaptations, and none of their cutoffs transfer between forms. The load-bearing facts are the four-signal structure and the form question. A child can screen negative on the total while positive on one subscale, or positive on the total with every subscale below threshold, and both patterns are real results that a chart entry of "PSC passed" erases; each subscale reads against its own cutoff every time. And "PSC positive, 20" is ambiguous on its face: 20 is positive on the PSC-17 and negative on the 35-item forms, so the exact form, respondent, and language belong in every note. A positive screen of any kind identifies possible psychosocial dysfunction and starts a conversation and a plan; it diagnoses nothing, and an elevated attention subscale in particular is the trigger for a proper multi-informant evaluation with instruments like the Vanderbilt scales , never an ADHD finding. Developmental screening in younger children belongs to the ASQ-3 ; symptom-severity measures like the PHQ-9 and GAD-7 are the targeted next step when an internalizing signal needs specification. ## Who uses PSC-17 documentation and when Pediatric and family practices use the PSC-17 to meet the psychosocial-screening expectation that runs through Bright Futures and Medicaid EPSDT: the AAP recommends behavioral, social, and emotional screening across preventive visits without endorsing one instrument, federal EPSDT law requires periodic screening without naming a tool, and state programs fill in the specifics (Massachusetts famously added the PSC-17 to its approved menu in 2017, then replaced the fixed list in 2023 with a requirement to use age-appropriate Bright Futures tools, a timeline outdated articles still get wrong). School-based health centers favor it because it is free, brief, and has youth self-report evidence; integrated-care teams use a positive screen as the warm-handoff trigger; child-welfare and care-management programs use it for monitoring. The Y-PSC and youth PSC-17 add the adolescent's own voice, and discordance is informative rather than noise: in the original school study, three quarters of the youth the self-report identified had not been identified by the parent form. Canadian primary care lists the PSC among usable tools without mandating it, and Australia's national outcomes architecture runs on the SDQ, with 2025 Australian normative data now supporting local PSC-17 use. Downstream readers include the behavioral health clinician receiving the handoff, the school team a collateral request goes to, and the payer auditing a 96127 claim; broader context sits on the developmental assessment page. ## How to document PSC-17 results in the chart No US, Canadian, or Australian authority prescribes a PSC-17 note format, and none mandates the instrument. What survives review is a record that names the exact form and respondent, reports all four score signals against their own cutoffs, handles missing responses by a stated rule, shows the follow-up conversation, and closes the loop with a dated disposition. Each element below carries the pitfall that most often undermines it. Exact form and family member. Name the instrument precisely: PSC-17 caregiver form, PSC-35, 35-item Y-PSC self-report, youth PSC-17 (with the cut-score source you applied), or a named pictorial adaptation. The cutoffs travel with the form: 15 for the PSC-17 total against 28 (or 24 in the youngest children) for the PSC-35 and 30 for the Y-PSC, so an unnamed "PSC positive, 20" cannot be interpreted. Pitfall: Cutoff cross-contamination: a 35-item threshold applied to a 17-item form, or a youth form scored with caregiver cutoffs nobody validated for it. Respondent, language, and administration. Record who completed it and their relationship (mother, father, foster parent, guardian; or youth self-report), the language of the form, interpreter or read-aloud assistance, and the setting and route (paper, portal, tablet). If a digital system regroups the items by subscale, score from that form's own key: grouped item numbers do not match the common interleaved paper form. Pitfall: "PSC completed." Respondent identity is central to interpreting results and discordance, translation availability is not validation, and a sum computed from the wrong form's numbering is a mathematically correct score of the wrong questions. All four scores, each against its own cutoff. Chart the total (out of 34, positive at 15 or higher) and all three subscales with their ranges and thresholds: internalizing (out of 10, positive at 5 or higher), attention (out of 10, positive at 7 or higher), externalizing (out of 14, positive at 7 or higher). The thresholds are points, not item counts, and boundary scores are positive. Pitfall: "PSC-17 negative" hiding a positive subscale. A subscale can be positive under a negative total (concentrated concern) and a total positive over negative subscales (diffuse concern); both are results, and the total is not a gatekeeper. Missing responses, by a stated rule. Ask the respondent to complete blanks when possible, then document the count. Blanks score 0, and circulating instructions genuinely conflict at exactly four blanks (four-or-more invalid versus more-than-four invalid), so adopt and cite one written rule for the practice and mark borderline forms provisional or repeat them. Pitfall: Missing information silently converted to "never," or a form with several blanks scored as if complete. The next reader cannot detect either without the count in the note. The follow-up conversation. For any positive signal, document the conversation the screen exists to start: duration and course, functioning at home, school, with friends, and in activities, strengths, stressors and context, the caregiver's (and youth's) own priorities, and safety questions where clinically indicated. Review the items marked often as conversation anchors, not as diagnoses. Pitfall: A score with no conversation. The screen's value is the structured discussion it triggers, and reviewers read a positive number followed by silence as screening performed and ignored. Disposition and the closed loop. Chart one of the recognized dispositions with a date: watchful waiting with a specific recheck, a targeted second-stage instrument, an integrated behavioral health warm handoff (say so when it happened; "referral placed" undersells it), or referral with the access plan: who places it, barriers discussed (waitlists, transportation, language, cost), interim supports, and who tracks completion. Pitfall: "Recheck at next annual visit" for a positive screen. In a longitudinal study roughly 30 percent of screened children did not return within 10 to 18 months, and the positives were less likely to return; the loop needs a nearer date and an owner. Interpretation boundary. State it as a screen: a positive result means further assessment is indicated, and a negative result does not rule out disorder, since subscale sensitivities ran 31 to 73 percent against diagnostic interviews, the total caught only 42 percent of any-diagnosis cases in an at-risk sample, and the instrument is comparatively insensitive to anxiety. An elevated attention subscale prompts a real ADHD evaluation across settings, not a diagnosis. Pitfall: "ADHD confirmed by PSC" or "anxiety ruled out." Neither direction of the screen supports diagnostic language, and youth-parent disagreement is clinical information, not an error to average away. ### Blank template (copy and adapt) PSC-17 DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Age: [ ] Clinician: [ ] Form: [PSC-17 caregiver / PSC-35 / Y-PSC (35-item youth) / youth PSC-17 (cut-score source named) / pictorial adaptation (named)] Respondent + relationship: [ ] Language: [ ] Assistance: [interpreter / read aloud / none] Route: [paper / portal / tablet] Missing responses: [count; practice rule applied; validity: valid / provisional / not interpretable] Scores (each against its own cutoff): Total: [ ]/34 (positive at 15 or higher): [negative / POSITIVE] Internalizing: [ ]/10 (positive at 5 or higher): [ ] Attention: [ ]/10 (positive at 7 or higher): [ ] Externalizing: [ ]/14 (positive at 7 or higher): [ ] Follow-up conversation (any positive signal): [duration and course; functioning at home, school, friends, activities; strengths; stressors; family priorities; safety questions as indicated] Interpretation: [screen result only; positive = further assessment indicated; negative does not rule out concern; attention elevation = ADHD evaluation trigger, not a diagnosis] Disposition: [watchful waiting with recheck DATE / targeted instrument (named) / integrated behavioral health warm handoff / referral] Access plan: [who places it, barriers discussed, interim supports, completion tracker] Owner: [ ] Youth self-report (if administered): [form named; scores kept separate; discordance described, never averaged] Declined: [what was offered and explained, decision, reason if volunteered, re-offer plan] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces checklist items or any form's layout. ## Sample PSC-17 documentation (fictional) Scenario: a well-child visit where the total screens negative but the internalizing subscale is positive, documented so the subscale result drives the plan instead of disappearing under a "passed." All details are fictional. Patient: T.W., 10 · Setting: Pediatric primary care, annual well-child visit · Clinician: L. Kim, MD · Note date: 08/13/2026 Screening: PSC-17 caregiver form completed in English by the patient's mother at today's visit, on paper, without assistance; no responses missing, so scoring is valid under our practice rule. Total 11/34, below the positive threshold of 15. Internalizing 5/10, positive at the threshold of 5. Attention 3/10 and externalizing 3/14, both below their thresholds of 7. Result charted as a negative broadband total with a positive internalizing subscale, not as a negative screen. Follow-up conversation: Reviewed the internalizing items her mother marked often and spoke with T.W. directly. Since changing schools in the spring she has been worrying most evenings, taking an hour or more to fall asleep, and has stopped asking to see friends; she still enjoys swim practice, grades are stable, and appetite is unchanged. No somatic pattern beyond occasional stomachaches on school mornings. Safety questions asked and answered without concern elicited. Her mother's priority is the sleep; T.W. says the worry is the harder part. Interpretation: A positive internalizing signal concentrated in worry and sleep, in the context of a school transition, without the breadth or intensity to raise the broadband total. This is a screening result, not a diagnosis, and the instrument's known insensitivity to anxiety cuts the other way here: the positive subscale earns follow-up even though the total is negative, and a negative subscale would not have ruled anxiety out. Plan: Targeted next step rather than watchful silence: an anxiety-specific measure will be completed at a recheck visit in four weeks (appointment booked before the family left), with sleep-hygiene coaching and a worry-time strategy reviewed today in plain terms. Thresholds for earlier contact reviewed with both: worsening mood, school refusal, functional decline, or any safety concern. If the recheck sustains the signal, referral to our integrated behavioral health clinician, introduced to the family today by name, with a warm handoff at that visit. Administrative: Form name, respondent, language, all four scores, and the disposition entered in the screening fields; one standardized instrument administered, scored, and documented this visit, billed as one unit per our payer's rules (subscales are outputs of the same instrument, never separate units). Follow-up owner: Dr. Kim; front desk confirmed the four-week recheck before departure. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, conversation, dates, and details are invented to show documentation structure and are not clinical guidance. No checklist items are reproduced. ↑ Back to the template and downloads ### Why this sample works The form, respondent, language, and completeness are named before any number, so the scores are interpretable and the youth-versus-caregiver question is answerable later. All four signals are charted against their own cutoffs, and the positive internalizing subscale survives the negative total instead of vanishing under "PSC passed." The follow-up conversation is documented as content (duration, functioning, strengths, priorities, safety), which is what turns a screening number into clinical reasoning. The interpretation states the boundary in both directions: positive is not a diagnosis, and negative would not have ruled out anxiety, the instrument's documented weak spot. The loop closes with a near-term dated recheck, named escalation thresholds, a warm-handoff path, and the one-unit billing note that matches how payers audit 96127. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the screening expectation is layered, and the chart should match the layer. Federal Medicaid EPSDT requires periodic and medically necessary interperiodic screening for enrollees under 21 and obligates states to arrange follow-up care, but names no instrument (LAW); CMS guidance adds that a concern raised by a clinician, school professional, or other qualified source can trigger an interperiodic screen, that states may not impose numerical limits or prior authorization on medically necessary screens, and that the initial screen need not come from a Medicaid-enrolled provider (LAW and federal guidance). Bright Futures and the AAP's 2025 mental and behavioral health report recommend routine broadband screening while expressly endorsing no single tool (CONVENTION). State layers vary and change: Massachusetts added the PSC-17 to its approved menu in 2017, eliminated the fixed menu effective January 2023 in favor of age-appropriate Bright Futures tools, and since January 2026 opens a preventive behavioral-health pathway after a positive screen without requiring a diagnosis (PAYER POLICY, dated); Alabama currently permits up to three different standardized tools per date under 96127 with the score documented; California's specialty mental-health outcomes program uses the PSC-35, not the PSC-17. Billing conventions follow the instrument, not the subscales: one unit per separately administered, scored, and documented instrument, with frequency and same-day rules payer-specific, and a denied unit never converts a positive result into a completed clinical pathway (PAYER POLICY). Canada and Australia treat the PSC-17 as an option inside SDQ-leaning systems, and equity findings travel everywhere. The Canadian Paediatric Society supports regular mental-health surveillance at well-child encounters and lists the PSC family among usable tools without mandating any instrument, with the SDQ the more common broadband choice and provincial pathways controlling access (CONVENTION); Canadian use is therefore charted as a clinical choice, not a program requirement. Australia's national outcomes architecture is SDQ-centered (the national collection uses the SDQ alongside other measures, and national wellbeing indicators prefer it), while a 2025 national Australian study of 2,097 caregivers supplied PSC-17 normative data with higher positive rates than US samples, supporting local use with local calibration in mind (CONVENTION and national infrastructure). On fairness, the three-factor structure has replicated across several populations, with measurement-invariance findings generally supportive but population-specific, item-level differences detected in some subgroups, a disadvantaged-population study that required cutoff modification, and clear evidence that respondent education and language shape results; none of that invalidates the screen, and all of it belongs in interpretation when the family in front of you is not the derivation sample. The respondent, language, and assistance lines in the note are the equity documentation. Rights and score-integrity both reward precision. The PSC family is free: the MGH team wrote in 2009, "The PSC is public domain, so can be used free of charge," and widely used forms state they may be freely reproduced, so printing clinical copies is well supported; the precise current framing is free reproduction for clinical use with the copyright line retained, because the forms carry a copyright notice, the authors ask to be notified of research and translations, current AAP materials caution that commercial use may require permission, and the old MGH web collection was retired in 2026, leaving no stable central rights page. A practice can enter results and build workflows in its EHR; a monetized or white-label implementation deserves a rights review rather than an assumption, and no vendor's hosted version should be treated as the scoring authority. The score-integrity hazards are quieter: digital systems that regroup items by subscale renumber them (their numbering must never be applied to the interleaved paper form), "PPSC" names both the pictorial PSC and the unrelated preschool checklist in the SWYC family, some circulating sheets misstate thresholds as item counts or use strict greater-than rules that misclassify boundary scores, and vendor severity bands, percentiles, and reliable-change values are interpretive overlays, not the canonical cutoffs (the externalizing subscale has no published reliable-change value at all). Score from the administered form's own key, and label overlays as overlays. The Pediatric Symptom Checklist and PSC-17 were created by their authors (Jellinek and Murphy; the 17-item form credited to Gardner and Kelleher) and are distributed through Massachusetts General Hospital. BastionGPT is not affiliated with, or endorsed by, the authors or Massachusetts General Hospital. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common PSC-17 documentation errors reviewers flag The numbers behind these errors are specific. Against structured diagnostic interviews in an at-risk sample, the PSC-17 total caught 42 percent of any-diagnosis cases and the attention subscale 58 percent of ADHD, with subscale sensitivities ranging 31 to 73 percent; in the original youth self-report study , three quarters of the youth identified were not identified by the parent form; in a longitudinal screening study about 30 percent of children did not return within 10 to 18 months, with positives less likely to return; and the construct-validity literature found the instrument comparatively insensitive to anxiety. The BastionGPT Clinical Advisory Board sees the same errors most often in PSC-17 documentation reviews: "PSC passed" over a positive subscale. A total of 11 with internalizing at 5 charted as a negative screen. The four signals are independent: each subscale reads against its own cutoff (5, 7, and 7 points) regardless of the total, and the defensible entry is "negative broadband total with a positive internalizing subscale," followed by the conversation and plan that subscale earns. Another form's cutoff, or an unnamed form. "PSC positive, 20" (positive on the PSC-17, negative on the PSC-35 and Y-PSC), a 17-item form scored against 28 or 30, or a youth form scored with caregiver thresholds nobody validated for it. Name the form, respondent, and language every time; the cutoffs travel with the form. Thresholds misread at the boundary. Subscale rules written as "5 or more items" when the thresholds are summed points, or strict greater-than rules that call a total of exactly 15, internalizing of exactly 5, or attention of exactly 7 negative. The published rules are at-or-above, in points; boundary scores are positive. A screen charted as a diagnosis. "ADHD confirmed by PSC" from an attention score of 8, or "anxiety ruled out" from a negative internalizing subscale. The attention elevation triggers a multi-informant ADHD evaluation (parent and teacher scales, settings, history); the anxiety insensitivity means a negative subscale cannot close the question; and positive language stays at "further assessment indicated." An open follow-up loop. "Recheck at next annual visit" for a positive screen, or "referral placed" when a warm handoff actually happened or when nobody owns confirmation. Positives are less likely to return, so the loop needs a near-term dated recheck, named escalation thresholds, barriers documented, interim supports, and an owner; for Medicaid children, EPSDT puts an affirmative access obligation behind that plan. Administration facts and form hazards ignored. No respondent or language recorded, blanks silently scored as never (document the count and your practice's written rule, since published instructions conflict at exactly four blanks), grouped digital item numbers applied to the interleaved paper form, parent and youth results averaged, or vendor severity bands charted as canonical. Score from the administered form's key and keep each respondent's result separate. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (form, respondent, language, all four scores, missing count, conversation content, disposition) and it drafts the documentation block: every signal against its own cutoff, the follow-up conversation structured, the boundary language in place, and the dated loop with an owner, ready for your review. Cross-check a finished note for the gaps reviewers flag: a positive subscale hidden under a negative total, an unnamed form, boundary scores misclassified, diagnostic language from a screen, or a positive result with no dated follow-up. Draft the follow-up entries: the recheck note comparing four signals over time on the same form, the warm-handoff documentation, or the youth-versus-caregiver discordance paragraph that keeps both results separate and reads the disagreement as information. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the PSC-17 scored? Seventeen caregiver-rated items, each scored 0 (never), 1 (sometimes), or 2 (often), producing four signals read independently. The total (0 to 34) is positive at 15 or higher. Three subscales sum their own items: internalizing (0 to 10) positive at 5 or higher, attention (0 to 10) positive at 7 or higher, and externalizing (0 to 14) positive at 7 or higher. Two precision points prevent most scoring errors: the thresholds are summed points, not counts of endorsed items, and the rules are at-or-above, so boundary scores (exactly 15, 5, or 7) are positive. Blank items score 0, the count of blanks belongs in the note, and circulating instructions conflict about whether exactly four blanks invalidate the form, so practices should adopt one written rule. The result is charted as four findings, never one word. What does it mean when a subscale is positive but the total is negative? It is a real positive result on that domain, and it drives the visit. The total and subscales use independent thresholds, so a child with concentrated worry can hit internalizing at 5 while the total sits at 11, and a child with moderate, spread-out concerns can hit a total of 16 with no subscale elevated; the first pattern points at a domain, the second at diffuse strain. The defensible entry names both findings ("negative broadband total with a positive internalizing subscale"), reviews the items marked often as conversation anchors, documents functioning in the relevant settings, and sets the plan the subscale earns: a targeted instrument, a dated recheck, or a handoff. What the negative total never does is erase the subscale: "PSC passed" over a positive domain is the instrument's most common documentation failure. Does a positive attention subscale mean the child has ADHD? No: it means an ADHD evaluation is indicated, and the note should say exactly that. Against structured diagnostic interviews, the attention subscale's sensitivity for ADHD was 58 percent in an at-risk sample, so it misses roughly four in ten cases and confirms none. A proper evaluation follows professional guidance: symptoms and impairment in more than one major setting, information from parents and teachers (the multi-informant Vanderbilt scales are the standard bridge), developmental and learning history, sleep, mood and anxiety, trauma, and medical factors. A strong bridge note reads: attention subscale positive; result supports further assessment and is not diagnostic; parent and teacher rating scales requested; school functioning and learning history to be reviewed; follow-up scheduled when informant forms return. The same logic runs in reverse: a negative attention subscale does not rule ADHD out when the history says otherwise. Can a negative PSC-17 rule out a mental health problem? No, and the note should never lean on it that way. Against diagnostic interviews in an at-risk sample, the total detected 42 percent of children with any diagnosis, subscale sensitivities ranged from 31 to 73 percent, and the construct-validity literature found the instrument comparatively insensitive to anxiety, its internalizing subscale notwithstanding. The youth-report evidence sharpens the point: three quarters of youth identified by the self-report were missed by the parent form, and among children whose parents screened positive, only half of youth agreed, with the disagreement itself carrying clinical meaning. So a negative screen coexisting with a worried teacher, a declining report card, a trauma history, or a youth's own disclosure is an open question, not a closed one: document the concern, act on the history, and use the negative score as one data point rather than a verdict. Should a 12-year-old complete the youth form, and what if it disagrees with the parent's? Both forms are defensible from about age 11, and administering both is often the most informative choice where workflow and confidentiality allow. The established youth instrument is the 35-item Y-PSC (cutoff 30, generally operationalized from age 11, originally validated at ages 9 to 14); a youth 17-item form also circulates and should be charted by name with the cut-score source you applied, never silently scored with caregiver thresholds. Chart each result by form and respondent, side by side, and never average them: discordance is expected and meaningful. Adolescents report internalizing symptoms parents do not observe (in one PTSD-symptom study, youth versions caught 75 to 78 percent of cases against 25 to 44 for parent versions), so a youth-positive, parent-negative pattern typically earns a confidential youth interview and prioritizes the adolescent's report for mood and anxiety follow-up. What are we obligated to do after a positive screen when behavioral health access is scarce? A documented clinical response, which scarce referral slots do not excuse. The defensible chain: assess severity, functioning, and clinically indicated safety questions in the visit; clarify the family's priorities and readiness; complete a targeted second-stage measure when it will sharpen the picture; then choose and date a disposition, integrated-care handoff, referral, or structured watchful waiting with a near-term recheck, and document the barriers (waitlists, transportation, language, cost), the interim supports, when to seek urgent help, and who tracks referral completion. The follow-up literature is blunt about why: about 30 percent of screened children did not return within 10 to 18 months, and the positives were less likely to return, so "recheck next year" is the riskiest plan available. For Medicaid-eligible children, EPSDT places an affirmative obligation on states and their contractors to arrange medically necessary follow-up care: a closed referral list does not close the obligation, and the chart should show the practice working the problem. Is the PSC-17 really free? Can we print it, put it in our EHR, or score it online? Free for ordinary clinical use, with one honest nuance. The MGH team wrote in 2009 that "The PSC is public domain, so can be used free of charge," widely used forms state they may be freely reproduced, and no license fee exists, so printing clinical copies and entering results in your EHR are well supported; keep the copyright line on reproductions, and the authors ask to be notified of research and translations. The nuance: the forms carry a copyright notice, current AAP materials caution that commercial use may require permission, and the old MGH web collection was retired in 2026, so a monetized tool, white-label product, or branded derivative deserves a rights review rather than an assumption, and third-party scoring sites are not authorities (several circulate with wrong thresholds, grouped item numbering, or vendor overlays presented as canonical). Score from the form you actually administered, and treat translation availability as availability, not validation. How is the PSC-17 billed? One unit per instrument, with everything else payer-specific. The convention is one 96127-family unit for one separately administered, scored, and documented standardized instrument: the total and three subscales are outputs of a single instrument and never justify multiple units. Same-day multiples (a broadband screen plus a targeted instrument, or parent plus youth forms when both are clinically indicated and separately documented) are allowed by some payers and capped or modifier-gated by others: Alabama Medicaid currently permits up to three different tools per date, Massachusetts runs its own modifier and same-day rules, and legacy state guides differ again, so the current member-specific manual governs. Two rules never bend: the clinical note records the actual form, respondent, scores, and plan regardless of what gets paid, and a payer denial does not convert a positive result into a completed pathway. Can BastionGPT draft my PSC-17 documentation? Yes. Give it the facts (form, respondent and relationship, language, all four scores, missing-item count, what the conversation covered, and your disposition) and it drafts the documentation block: every signal against its own cutoff, the follow-up conversation structured, screen-level interpretation in both directions, and the dated loop with an owner, ready for your review. It can also cross-check a finished note for a positive subscale hidden under a negative total, an unnamed form, a boundary score misclassified, diagnostic language, or a missing recheck date, and it can draft the youth-caregiver discordance paragraph that keeps both results separate. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Derivation and restandardization: Gardner W, Murphy M, Childs G, and colleagues, 1999, Ambulatory Child Health, the PSC-17 derivation (18,045 primary-care children; cutoffs 15, 5, 7, 7); Murphy JM and colleagues, 2016, Pediatrics, subscale scores, reliability, and factor structure in a new national sample (80,680 outpatients; positive rates and reliability); Jellinek MS and colleagues, 1988, Journal of Pediatrics (the original 35-item validation). Diagnostic-accuracy and informant evidence: Gardner W, Lucas A, Kolko DJ, Campo JV, 2007, PSC-17 against structured interviews (total sensitivity .42; attention .58 for ADHD; the University of Washington form reproduces the performance table); Kostanecka A and colleagues, 2008, construct validity in urban primary care (anxiety insensitivity); Pagano ME and colleagues, 2000, the youth self-report (75 percent of youth positives missed by parents); Duke N, Ireland M, Borowsky IW, 2005, parent-youth agreement ; follow-up attrition, longitudinal preventive-visit screening study . Rights record: MGH team in Contemporary Pediatrics, 2009, using the PSC in the pediatrician's office (the public-domain statement and clinical workflow); University of Washington, PSC-17 form (freely-reproduced statement; caregiver field); AAP Bright Futures and EQIPP screening materials (commercial-use caution; tool lists non-exhaustive). United States policy: CMS, EPSDT coverage guide (May 2026; interperiodic screening, no numerical limits, affirmative access obligation); AAP, 2025, Pediatrics, screening for mental, emotional, and behavioral problems ; AAP, 2019, ADHD clinical practice guideline (multi-informant, multi-setting evaluation). State payer examples: MassHealth, 2017 transmittal adding the PSC-17 , 2022 bulletin ending the fixed menu (effective January 2023), and 2026 preventive behavioral-health expansion ; Alabama Medicaid, April 2026 provider manual (96127 units and documentation). Canada, Australia, and local norms: Canadian Paediatric Society, mental health screening tools and rating scales and 2023 position statement ; AMHOCN, the SDQ in the national outcomes collection ; McLean RK, Tully LA, Dadds MR, and colleagues, 2025, Australian and New Zealand Journal of Psychiatry, Australian PSC-17 national normative data (2,097 caregivers). ## Related templates Vanderbilt ADHD Scale Documentation PHQ-9 Score Documentation GAD-7 Score Documentation ASQ-3 Score Documentation Child & Adolescent Developmental Assessment School & Teacher Collateral Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/dayc-2-report All note templates # DAYC-2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The DAYC-2 (Developmental Assessment of Young Children, Second Edition) is a norm-referenced battery for birth through 5 years 11 months, scoring five IDEA-aligned domains through observation, caregiver interview, and direct assessment. Early-intervention teams use it for Part C eligibility evaluations. The write-up must name each domain's method and the state's exact eligibility rule. This page covers how to write up DAYC-2 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Early-intervention evaluators and multidisciplinary teams (SLPs, OTs, PTs, special educators, psychologists); PRO-ED qualification level B Audience Part C eligibility teams and service coordinators, IFSP teams, referring physicians and NICU follow-up programs, receiving programs after interstate moves, families Typical length 400 to 800 words for the results section · administration 10 to 20 minutes per domain, plus history, observation, scoring, and integration Format family Norm-referenced developmental battery (five IDEA-aligned domains; observation, caregiver interview, and direct assessment permitted) When it's used Part C early-intervention eligibility evaluations, developmental monitoring from birth through 5 years 11 months, high-risk infant follow-up, preschool transition planning Standards context Published by PRO-ED (2013; DAYC-3 in field testing, unreleased as of August 2026); accepted for eligibility in many states; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the DAYC-2? The DAYC-2 (Developmental Assessment of Young Children, Second Edition; Voress and Maddox, PRO-ED, 2013, with Pearson's catalog additionally crediting Hammill) is a norm-referenced developmental battery for children from birth through 5 years 11 months. Its five domains (cognition, communication, social-emotional development, physical development, adaptive behavior) map one to one onto the five developmental areas IDEA Part C requires states to evaluate, which is why so many early-intervention systems accept it. Its signature is flexibility: each domain can be administered independently, by different disciplines in an arena model, using any mix of observation, caregiver interview, and direct assessment, in about 10 to 20 minutes per domain. Domains yield standard scores (mean 100, SD 15), percentile ranks, and age equivalents; communication splits into receptive and expressive subdomains and physical development into gross and fine motor; and the composite, available only when all five domains are administered, is the General Development Index (GDI). That name matters: the publisher's own materials say General Development Index, vendor pages loosely paraphrase it, and the "General Development Quotient" label belongs to first-edition-era literature, not this test. The load-bearing facts for the write-up are the source question and the eligibility arithmetic. Because the instrument permits interview-derived scores, and no published study establishes that interview-only scores are equivalent to directly assessed ones, a defensible report labels each domain's method and never launders a caregiver's report into observed performance. And because states define developmental delay in two different currencies, standard-deviation cuts (a standard score of 85 sits 1 SD below the mean, 77.5 sits at 1.5 SD, 70 at 2 SD) and age-equivalent percent delay, the report must apply the named state's actual rule: the two metrics are not interchangeable, and a 2026 analysis across six early-intervention tests found thousands of score combinations where they disagree. As of August 2026 the DAYC-2 remains the current edition, with DAYC-3 standardization research announced but no release date. The Bayley-4 page covers the structured direct-assessment alternative (against which the DAYC-2 has never been validated), and the BDI-3 page covers the other major eligibility battery. ## Who uses the DAYC-2 and when Part C early-intervention teams are the core users: multidisciplinary evaluators administering IDEA-aligned domains inside the federally required eligibility process, which must include an evaluation instrument, parent interview and history, functioning in all five developmental areas, records, and informed clinical opinion, all within 45 days of referral and at no cost to the family. State systems name it variously: Missouri builds DAYC-2 training into its First Steps program, Arizona lists it among approved comprehensive tools (latest edition required), New York and Illinois carry it on approved-instrument lists, and Texas currently allows teams to choose the DAYC-2 or the BDI-3 depending on the contract and service area. NICU and high-risk infant follow-up programs use it, including by telehealth, where a 2024 study found referral rates comparable to in-person Bayley-4 testing. Preschool teams meet it again at the transition to Part B services, where the receiving team applies its own criteria. The write-up's readers are exactly these downstream parties: a service coordinator applying the state rule, a receiving state re-determining eligibility after a move, an IFSP team turning results into outcomes, and a family entitled to a plain-language explanation. Adaptive behavior at depth belongs to the Vineland-3 and ABAS-3 pages; the broader evaluation frame lives on the developmental assessment page, and screening that precedes all of this on the ASQ-3 page. ## How to structure a DAYC-2 results section No authority prescribes a DAYC-2 report format, but the federal Part C rules and the instrument's own flexibility dictate what a defensible results section must contain: the ages and the governing rule, the method behind every domain score, uncertainty around every number, the scatter before the composite, and an eligibility statement written against the named rule. Each section below carries the pitfall that most often undermines it. Identification, ages, and the governing rule. Open with the full test name and edition, evaluation dates, chronological age, and, when applicable, the corrected age with the exact formula and the policy authorizing it (correction is state policy, not a manual rule: Missouri deducts half the prematurity; Washington stops correcting at 24 months). Name the state or program whose eligibility definition the report will apply. Pitfall: "DAYC completed" with one age and no rule. Another evaluator must be able to re-run the arithmetic, and "corrected for prematurity" without the formula makes that impossible. Measures and methods, per domain. State which domains were administered and, for each, the information source: directly elicited, observed in routines, caregiver interview, or records. The DAYC-2 expressly permits all three methods and arena administration; the report's job is to preserve which one produced each score, because no published study establishes interview-only scores as equivalent to direct assessment. Pitfall: Interview-derived numbers presented as observed performance. "The child demonstrated" when the only evidence is a parent's report is the write-up error reviewers flag most. Behavioral and validity conditions. Document engagement and regulation, fatigue, illness, or sensory factors, language exposure and interpreter use, motor, hearing, or vision limitations, the child's familiarity with the setting and examiner, opportunity to practice the skills sampled, and whether the caregiver considered the behavior representative. Pitfall: "Did not demonstrate today" converted into "cannot do," or an unrepresentative session scored without comment. State the conditions and let them bound the interpretation. Domain results with uncertainty. Report each administered domain's standard score (mean 100, SD 15), percentile rank, and the confidence band from the authorized scoring materials, with a plain-language functional line. Report subdomain scores (receptive and expressive; gross and fine motor) only when the referral or the state rule genuinely turns on them, remembering that some states, Pennsylvania among them, require the overall developmental-area score. Pitfall: Bare point estimates at an eligibility boundary. The publisher reports standard errors and intervals; a score of 79 with its band straddling 77.5 is a different sentence from one clearly below it. Scatter before the composite. Lead with the domain pattern. Report the General Development Index only when all five domains were administered, name it correctly, and say explicitly when it dilutes a focal delay (one very low domain averaged against four expected ones) or overstates uniformity (a mildly depressed profile with no qualifying domain). If a domain was not administered, say the GDI was not available rather than implying missing data. Pitfall: The composite doing eligibility work: an average GDI nullifying a qualifying communication score, or "GDQ" imported from first-edition-era habits. Eligibility analysis against the named rule. Quote the jurisdiction's actual criterion and apply it in its own currency: an SD-based rule reads off the standard score (85, 77.5, 70 map to 1, 1.5, and 2 SD, with the state's own rounding convention controlling the boundary), and a percent-delay rule uses age equivalents under the state's formula, without pretending the two are equivalent. Then situate the score inside the multisource process: no single procedure can determine eligibility, and informed clinical opinion can establish eligibility that scores miss but can never negate a qualifying score. Pitfall: "1.5 SD low, which equals a 25 percent delay." It does not, except where a state has administratively declared it so; the 2026 misclassification analysis is the receipts. Family summary and next steps. Close with a plain-language paragraph the family can use: what the child does well, what is harder, what came from the family's own report, whether the program's criterion was met, and what happens next (team review, IFSP development, transition planning near age three with its 90-day requirements). Part C requires results to be explained to parents and the family's priorities to shape the plan. Pitfall: A report that ends at score classifications. The family summary is a federal-process expectation, not a courtesy, and its absence reads as an evaluation done to the family rather than with it. ### Blank template (copy and adapt) DAYC-2 RESULTS SECTION SKELETON Child: [initials] Evaluation date(s): [ ] Evaluator(s) + discipline(s): [ ] Chronological age: [ ] Corrected age (if applicable): [ ] (formula: [ ]; authority: [state policy / program rule]) Referral question: [ ] Program + eligibility rule applied: [state/program; criterion quoted, e.g. 1.5 SD below the mean in one area / 25% delay] Measures + methods (per domain): [Cognition / Communication / Social-emotional / Physical / Adaptive: direct assessment, observation, caregiver interview, records; arena or discipline-specific] Behavioral + validity conditions: [engagement, regulation, health, language and interpreter, sensory or motor limits, representativeness] Domain results (each with uncertainty): [Domain]: SS [ ] (mean 100, SD 15), percentile [ ], confidence band [ ], functional description [ ] Source: [method] Subdomains (only when rule-relevant): [receptive/expressive; gross/fine motor] Composite: [GDI [ ] with confidence interval, all five domains given / GDI not available: fewer than five domains administered] Scatter statement: [profile even or uneven; what the composite hides] Eligibility analysis: [criterion in its own currency (SD from standard scores / percent delay from age equivalents per the state formula); result vs threshold; converging evidence; no single procedure; informed clinical opinion role] Family summary (plain language): [strengths, needs, family-reported information, criterion met or not, next steps] Team + next steps: [multidisciplinary determination, IFSP development, transition planning if approaching age three] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, forms, norms, or conversion tables. ## Sample DAYC-2 write-up (fictional) Scenario: a 22-month-old referred for limited spoken language, evaluated by a Part C team, with a focal communication delay documented against a named state rule and every score's source labeled. All details are fictional. Patient: N.R., 22 months · Setting: Part C early intervention eligibility evaluation, multidisciplinary team · Clinician: P. Sandoval, MS, CCC-SLP · Note date: 08/21/2026 Identification and rule: Developmental Assessment of Young Children, Second Edition (DAYC-2), administered 08/21/2026 across all five domains. N.R. was 22 months 4 days at evaluation, born at term, so chronological age anchored all scoring; no correction applied. Referral: family and childcare provider report limited spoken language. This evaluation applies our program's eligibility definition of a delay of at least 1.5 standard deviations below the mean in one developmental area, alongside the required multisource process. Measures and methods: Communication was assessed through direct interaction, observation during play and snack routines, and caregiver interview; receptive and expressive subdomains were both sampled directly. Cognition and physical development were directly assessed with observation support. Social-emotional and adaptive information came primarily from caregiver interview, corroborated by observation and the childcare record. Sources are attributed throughout: information the family reported is identified as reported, not as examiner-observed. Results: Cognition SS 96 (39th percentile), social-emotional SS 94 (34th), physical SS 98 (45th; gross and fine motor both age-expected), and adaptive SS 92 (30th) all fell within the average range, each reported with the scoring system's confidence band. Communication SS 70 (2nd percentile) fell well below expectations, with the confidence band entirely below the 1.5 SD threshold; receptive abilities were relatively stronger than expressive on subdomain testing, and the overall communication domain, not a subdomain, was used for the eligibility analysis. Behaviorally, N.R. engaged warmly, used consistent gestures, brought adults to desired objects, and followed familiar routine directions; spoken output observed by the examiner matched the caregiver's description and the childcare record. The five-domain profile is uneven, and although a General Development Index was available (all five domains administered), it is not the primary summary here because averaging four expected domains against one very low domain understates the focal communication delay. Eligibility analysis: Under the program's criterion, the communication standard score meets the 1.5 SD threshold in its own currency; no percent-delay conversion is offered because standard scores are available and the two metrics are not interchangeable. Parent history, direct observation, childcare records, and today's testing converge. Per Part C requirements, this instrument is one procedure among several, the team's informed clinical opinion supports the same conclusion, and final eligibility is the multidisciplinary team's determination, not the score's. Family summary and next steps: In plain terms, reviewed with N.R.'s parents today: he shows age-expected interest in people, play, movement, and everyday problem solving, and his strengths include gestures, warmth, and understanding familiar language. Talking is what is hard right now, and he met the program's definition of a communication delay. His parents' observations shaped this evaluation and matched what we saw. Next: the team meets with the family to confirm eligibility and develop the IFSP within the required timeline, building outcomes on his gestures, social engagement, and family routines; a hearing check was recommended and arranged, since hearing should always be verified when speech is delayed. This sample is fictional and for educational purposes. It does not describe a real child or record; the scores, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The ages, the edition, and the governing rule are stated before any score, so another evaluator (or a receiving state after a move) can re-run the arithmetic. Every domain score carries its source (direct, observed, interview) and its confidence band, and family-reported information is attributed rather than laundered into observed performance. The scatter is handled before the composite: the GDI is reported, correctly named, and explicitly declined as the primary summary because it would dilute the focal delay. The eligibility analysis stays in the rule's own currency (SD from standard scores), refuses the percent-delay conversion, and situates the score inside the multisource, no-single-procedure process. The family summary does the federally expected work in plain language, and the hearing referral shows the medical loop a communication delay always deserves. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the write-up sits inside a legal machine, and the machine's parts are citable. IDEA Part C defines the five developmental areas the DAYC-2 mirrors (34 CFR 303.21), requires each state to adopt its own rigorous definition of developmental delay, prohibits any single procedure as the sole eligibility criterion, and requires informed clinical opinion, which may independently establish eligibility that instruments miss but may never negate a qualifying score (34 CFR 303.321). Evaluation, assessment, and the initial IFSP meeting must finish within 45 days of referral (303.310), the evaluation is free to the family (303.521), and transition planning toward Part B preschool services carries its own 90-day requirements near the third birthday (303.209), with the receiving Part B team applying Part B criteria (LAW). The state layer is where write-ups go wrong: a 2026 federal technical-assistance review counted 19 jurisdictions using 2 SD in one area, 19 using 1.5 SD in two or more, 18 using a 25 percent delay, and a long tail of other rules, most states offering multiple routes; named examples include Missouri (half-prematurity age correction, organized DAYC-2 implementation), Arizona (approved-instrument list requiring the latest edition plus a two-discipline process), Illinois (instruments without age equivalents must be paired with one that produces them), Pennsylvania (overall developmental-area scores, not subdomains), and Texas (BDI-3 statewide with DAYC-2 permitted in specific contracts), so the report quotes the rule it applies (PROGRAM POLICY, state by state). Developmental testing billed to health payers uses the 96112 family, but Part C eligibility work is not billed to families (PAYER POLICY). Canada and Australia change the frame from score-versus-threshold to function-and-need. Canadian early intervention is provincial and program-based (Ontario's Infant and Child Development Program, British Columbia's infant development programs) with referral on developmental concern rather than a national numeric criterion; the DAYC-2 is commercially available in Canada but no province mandates it, so a Canadian report identifies the actual program framework instead of importing an IDEA threshold (PROGRAM POLICY, provincial). Australia's NDIS early childhood approach admits children under six on developmental delay or concern without a diagnosis (older children generally need one), and its evidence guidance centers on functional capacity relative to same-age peers, persistence of need, and coordinated multidisciplinary supports, expressly asking whether each ability was observed, assessed, or reported, the same source discipline this write-up teaches; NDIS processes are in active change through 2026, so current guidance should be checked at the time of writing (LAW and PROGRAM POLICY). In all three countries the portable artifact is the evidence, not the determination: a report that preserves domain scores, bands, sources, ages, and the originating rule lets a receiving team re-decide under its own definition, which is exactly what happens after an interstate or international move. Instrument facts, rights, and psychometric honesty round out a defensible page and report. The DAYC-2 (2013) remains the current edition as of August 2026, with PRO-ED publicly recruiting for DAYC-3 standardization since 2025 and no release date announced; the norms (1,832 children in 20 states, collected 2009 to 2011 against 2010 Census targets) are now about 15 years old, worth stating plainly. Publisher-reported reliability is strong but not uniform (internal consistency about .91 to .98, interscorer .99, test-retest about .70 to .91, with the low end worth flagging when a single domain sits at a threshold), the manual's validity studies ran against the BDI-2 and DOCS-2, the instrument has never been validated against any Bayley edition (a 2024 telehealth study found comparable referral rates against in-person Bayley-4, with only moderate score correlation in related work), and no published study establishes interview-only scoring as equivalent to direct assessment, which is why the method label is not optional. Rights are conventional PRO-ED: record forms are consumable and not reproducible, embedding scoring logic or forms in an EHR or IFSP platform requires a written license, the publisher's own online scorer is the sanctioned digital route, and a public web scorer of DAYC-2 conversions would be unauthorized; a report may always contain the child's derived scores and your interpretation. The DAYC-2 and Developmental Assessment of Young Children are trademarks or product names of PRO-ED, Inc. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common DAYC-2 write-up errors reviewers flag The numbers behind these errors are specific. A 2026 analysis across six early-intervention tests found 2,445 score combinations where age-equivalent percent delay could deny services to children whose standard scores showed delay, and 3,037 where it could qualify children within typical variation; the DAYC-2's norms rest on 1,832 children tested 2009 to 2011 ; publisher-reported test-retest reliability runs down to about .70 for one domain; and the federal rulebook itself, 34 CFR 303.321 , prohibits any single procedure from deciding eligibility. The BastionGPT Clinical Advisory Board sees the same errors most often in DAYC-2 documentation reviews: The composite misnamed or misused. "GDQ" (a first-edition-era label), a General Development Index reported when fewer than five domains were administered, or a middling GDI allowed to nullify a qualifying domain. Name it GDI, report it only with all five domains, lead with the domain pattern, and say out loud when the composite dilutes a focal delay. Percent delay treated as an SD translation. "1.5 SD below the mean, equal to a 25 percent delay." The metrics answer different questions: standard scores locate the child in the age distribution, while percent delay divides age-equivalent months as if development were an equal-interval ruler, and the two can disagree for the same child on the same test. Apply the state's rule in its own currency and label any state-adopted equivalence as administrative, not psychometric. Interview scores laundered into observed performance. "The child demonstrated" when the evidence is a caregiver's report. The instrument legitimately accepts interview information, no study establishes its equivalence to direct assessment, and Part C evaluation cannot be reduced to one interview-scored procedure; write the source per domain: reported, observed, directly elicited, or from records. Eligibility asserted from the score alone. "DAYC-2 qualifies the child" with no state rule quoted and no other sources cited. The score can satisfy a quantitative component, but no single procedure may be the sole criterion, informed clinical opinion can establish (never negate) eligibility, and the determination belongs to the multidisciplinary team applying the named jurisdiction's definition. Corrected age unstated or generic. "Corrected for prematurity" with no gestational age, formula, or authority. Correction is state policy, not a manual rule, and states genuinely differ (one deducts half the prematurity; others stop correcting at 24 months), so report both ages, the exact arithmetic, the policy relied on, and whether the eligibility conclusion changes under the other age. Age equivalents doing decisional work. Eligibility, severity, or progress read from age equivalents alone. They have unequal intervals, no error band, and instability near test floors; standard scores with confidence bands carry the decision where the rule permits, subdomain scores substitute for the overall developmental area only where the state expressly allows it, and a score at a boundary is reported with its band, not pre-rounded. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Give it the facts (ages and correction arithmetic, domains and methods, standard scores with bands, the state rule, observations) and it drafts the results section: sources labeled per domain, scatter handled before the composite, the eligibility analysis in the rule's own currency, and the family summary in plain language, ready for your review. Cross-check a finished report for the gaps reviewers flag: an interview score presented as observed, a GDI without five domains, a percent-delay conversion, an unquoted state rule, or a corrected age without its formula. Draft the companion paragraphs: the interstate-move version that preserves the evidence for re-determination, the informed-clinical-opinion rationale when scores and functioning disagree, or the transition summary as the third birthday approaches. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What scores does the DAYC-2 produce? Each administered domain (cognition, communication, social-emotional, physical, adaptive) yields a standard score on a mean-100, SD-15 scale, a percentile rank, and an age equivalent, with communication splitting into receptive and expressive subdomains and physical development into gross and fine motor. The publisher's materials report standard errors and confidence bands, so every score can and should be written with its uncertainty. The composite is the General Development Index (GDI), formed only when all five domains are administered: give some domains and there is no GDI, only domain scores, which the report says plainly. Two naming cautions from the source record: the manual term is General Development Index (vendor pages paraphrase it loosely), and "General Development Quotient" belongs to first-edition-era literature, not the DAYC-2. What DAYC-2 score qualifies a child for early intervention? Whatever the child's state says, and nothing else. Standard scores map cleanly onto SD-based rules (85 is 1 SD below the mean, 77.5 is 1.5 SD, 70 is 2 SD, with the state's own rounding convention deciding whether the 1.5 SD line is written 77 or 78), but the rules themselves vary enormously: a 2026 federal technical-assistance review counted 19 jurisdictions at 2 SD in one area, 19 at 1.5 SD in two or more areas, 18 at a 25 percent delay, and a tail of other criteria, with most states offering several routes. So there is no universal DAYC-2 cutoff, no fixed percentile that means delay everywhere, and the defensible write-up quotes the named jurisdiction's current criterion and applies it in its own currency, reporting the exact standard score with its confidence band and letting the state's rounding rule govern the boundary. Eligibility itself then belongs to the multisource, multidisciplinary process, not to the score. Is a 25 percent delay the same as 1.5 standard deviations below the mean? No, and treating them as interchangeable is the field's costliest write-up error. A standard score locates the child within the normative distribution for their exact age; percent delay divides an age-equivalent shortfall by chronological (or corrected) age, treating developmental months as an equal-interval ruler, which they are not: age equivalents have unequal intervals, no error band, and thin floors in infancy, so a 25 percent delay at 12 months and at 48 months describe very different standardized positions. A 2026 peer-reviewed analysis across six early-intervention tests found 2,445 score combinations where percent-delay rules could deny services despite qualifying standard scores and 3,037 where they could qualify children within typical variation. When a state's rule requires percent delay, apply it as written and label it as the state's administrative method; when standard scores exist, some states expressly prefer them. What a report never does is compute one metric and announce it as the other. Can the DAYC-2 be scored from caregiver interview alone? Mechanically yes, and the report must say so wherever it happened. The publisher expressly permits three information-gathering methods (observation, caregiver interview, direct assessment), domains can be administered separately, and arena or multidisciplinary use is part of the design, which is why a toddler who will not engage at a table does not automatically defeat the evaluation. But no published study establishes that interview-only scores are equivalent to directly assessed ones, so the defensible write-up labels every domain's source, corroborates reported skills through observation, routines, and records where possible, and never presents a caregiver's account as examiner-observed performance. The federal overlay adds the rest: a Part C eligibility evaluation must include parent history, functioning across all areas, other sources as needed, and records, so the process can never reduce to one interview-scored instrument, whatever the instrument permits. How should corrected age be reported on a DAYC-2? Completely, because the correction is state policy rather than a manual rule, and states genuinely differ. Report the gestational age at birth, the degree of prematurity, the chronological age, the corrected age, the exact formula used, the policy that authorizes it, which scores and decisions used which age, and whether the eligibility conclusion changes under the other age. The differences are not cosmetic: a common convention corrects until 24 months of chronological age, one state deducts only half the prematurity for up to 12 months (longer on physician recommendation), and others tie prematurity-based eligibility to referral timing, so "corrected for prematurity" standing alone leaves the next evaluator unable to reproduce the arithmetic. When the manual convention and the state rule differ, the state's eligibility policy governs the eligibility analysis, and the report says which rule it followed rather than silently choosing the age that produces the preferred outcome. Does a qualifying DAYC-2 score establish Part C eligibility by itself? No. Federal law prohibits any single procedure from serving as the sole criterion: the evaluation must include an instrument, child history and parent interview, functioning in all five developmental areas, information from other sources where needed, and record review, and qualified personnel must apply informed clinical opinion. That opinion runs one direction only: it can independently establish eligibility when scores miss a real delay (with the report documenting what the scores failed to capture and the functional evidence), but it can never be used to negate a qualifying instrument result because the team feels the child seems fine. The determination belongs to the multidisciplinary team applying the state's definition, inside the 45-day referral-to-IFSP window, at no cost to the family. The same logic governs moves: the evidence travels, the determination does not, and a receiving state re-decides under its own rule, which is why the write-up preserves scores, bands, sources, ages, and the originating criterion. When should a team use the DAYC-2 versus the Bayley-4 or BDI-3? By referral needs and program acceptance, not by a universal ranking. The DAYC-2's strengths are speed (10 to 20 minutes per domain), naturalistic and arena administration, domain-by-domain flexibility, caregiver-knowledge integration, and birth-through-5:11 coverage; its costs are 2009-to-2011 norms, permissive sourcing that demands method labeling, and the absence of validation against any Bayley edition. The Bayley-4 is the tightly standardized direct-assessment alternative for infants and toddlers through 42 months: slower, materials-bound, and only moderately correlated with the DAYC-2, with structured administration reading lower for some children. The BDI-3 is the other full eligibility battery, broader in age and often embedded in statewide infrastructure; program acceptance can decide the matter outright, as in Texas, where the state names the BDI-3 with the DAYC-2 permitted under specific contracts. Many strong evaluations pair a broad eligibility battery with discipline-specific direct assessment. Can BastionGPT draft my DAYC-2 report? Yes. Give it the facts (ages and any correction arithmetic, domains administered with each one's method, standard scores with confidence bands, the state or program rule, observations, and family-reported information) and it drafts the results section: sources labeled per domain, uncertainty reported, scatter handled before the composite, the eligibility analysis in the rule's own currency, and a plain-language family summary, ready for your review. It can also cross-check a finished report for interview scores presented as observed, a GDI without five domains, a percent-delay conversion, an unquoted rule, or a bare corrected-age claim, and it can draft the interstate or transition variants that preserve the evidence for the next team. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: PRO-ED, Inc., DAYC-2 product page (authors, domains, ages, methods, administration time, norm sample), online scoring and report system (General Development Index; five-domain requirement), qualifications policy (Level B), and reprint and electronic permissions ; DAYC-3 field-testing notice, April 2025 recruitment announcement . Federal law: eCFR, 34 CFR 303.21 (the five developmental areas), 303.321 (multisource evaluation; no single procedure; informed clinical opinion), 303.310 (45-day timeline), and 303.209 (transition). State variation and examples: ECTA Center, Part C eligibility criteria summary (February 2026 counts); Missouri DESE, First Steps eligibility and half-prematurity correction ; Arizona DES, AzEIP approved instruments ; Illinois DHS, approved assessment instruments ; Pennsylvania, overall developmental-area score guidance ; Texas HHS, ECI eligibility and instrument policy . Eligibility-metric evidence: Gladfelter A and colleagues, 2026, age-equivalent percent delay versus standard scores across six early-intervention tests (2,445 and 3,037 discordant combinations); Pearson, interpretation problems of age and grade equivalents . Independent studies: Ke JC and colleagues, 2024, Early Human Development, telehealth DAYC-2 versus in-person Bayley-4 in high-risk infant follow-up (comparable referral rates; no prior parallel study; never validated against any Bayley); Swartzmiller MD, 2014, Journal of Psychoeducational Assessment, DAYC-2 test review ; Teplitzky TB and colleagues, 2019, cognitive domain and behavioral audiometry . Canada and Australia: Ontario, Infant and Child Development Program ; NDIS, early intervention eligibility requirements and supporting-evidence guidance (observed, assessed, or reported). ## Related templates Bayley-4 Report Write-Up Battelle Developmental Inventory (BDI-3) Report Write-Up Vineland-3 Report Write-Up ABAS-3 Report Write-Up Child & Adolescent Developmental Assessment ASQ-3 Score Documentation PLS-5 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/battelle-report All note templates # Battelle (BDI-3) Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The BDI-3 (Battelle Developmental Inventory, Third Edition) is a norm-referenced battery for birth through 7 years 11 months, scoring five IDEA-aligned domains through structured tasks, observation, and interview. Early-intervention and preschool teams use it for eligibility and federal outcome reporting. It is unrelated to the Beck Depression Inventory (BDI-II). This page covers how to write up BDI-3 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Early-intervention and preschool evaluation teams (special educators, SLPs, OTs, PTs, psychologists); the publisher lists a low purchase-restriction level, with state programs adding their own training rules Audience Part C and Section 619 eligibility teams, IFSP and IEP teams, state data systems and federal outcome reporting, referring physicians and NICU follow-up, families Typical length 400 to 800 words for the results section · Screening Test 10 to 30 minutes; the full battery commonly spans sessions (roughly 60 to 90 minutes of testing time) Format family Norm-referenced developmental battery (five IDEA-aligned domains, 13 subdomains; structured tasks, observation, and informant interview) When it's used Part C eligibility and recurring assessment, preschool transition at age three, federal child-outcome reporting at entry and exit, re-evaluations documenting progress Standards context Third edition published by Riverside Insights (2020; the BDI-2 was sunset in 2023); named in many state systems; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Battelle Developmental Inventory (BDI-3)? The Battelle Developmental Inventory, Third Edition (BDI-3; Newborg; Riverside Insights, released 2020) is one of the two dominant early-childhood eligibility batteries in US practice. It spans birth through 7 years 11 months across five domains that mirror IDEA's developmental areas, Adaptive, Social-Emotional, Communication, Motor, and Cognitive, divided into 13 subdomains (several of which, including Peer Interaction and Perceptual Motor, begin at age 2). Items can be administered through structured tasks, observation, or caregiver and teacher interview, scored on a graded full, partial, or no-credit basis with three-consecutive-response basal and ceiling rules. Three separately sold components matter to reports: the Developmental Complete (the full battery), the BDI-3 Screening Test (about 100 items, 10 to 30 minutes, a flagging tool), and the Battelle Early Academic Survey, a literacy and mathematics add-on for ages 3:6 to 7:11. Scoring runs through licensed paper forms or the Riverside Score platform, which reports Developmental Quotients (mean 100, SD 15) with confidence intervals, subdomain scaled scores (mean 10, SD 3), percentiles, growth-oriented Change Sensitive Scores, and eligibility outputs including months, percentage, and standard-deviation delay. Two identity facts do heavy lifting on this page. First, the acronym: the BDI-3 is a developmental assessment for young children and has nothing to do with the Beck Depression Inventory (BDI-II) , Pearson's depression measure for ages 13 and up; records systems collide on "BDI," so the full name appears at first mention and bare "BDI" never appears in tables, problem lists, or export fields. Second, the edition: the BDI-2 was formally sunset in 2023 (final kits mid-2023, final record forms at year end), the third edition renamed the old Personal-Social domain to Social-Emotional (spotting "Personal-Social" in purported BDI-3 material is the reliable edition-drift detector), and the publisher itself warns that BDI-2 and BDI-3 scores are not directly comparable, so cross-edition change claims are measurement discontinuities, not progress. The DAYC-2 page covers the other major eligibility battery, and the Bayley-4 page the structured infant-toddler alternative. ## Who uses the BDI-3 and when State early-intervention and preschool systems are the BDI-3's home: Massachusetts moved its statewide early-intervention assessment to the BDI-3 in January 2025 for eligibility and outcomes, Tennessee uses it for initial eligibility, recurring assessment, and federal reporting through its state data system, New Jersey mandates the full Developmental Complete by two evaluators (results valid four months, telehealth administration prohibited), Texas transitioned from the BDI-2 in 2023 and now lets teams choose the BDI-3 or the DAYC-2, and South Dakota uses BDI-3 entry and exit data for its preschool outcomes indicator. Those examples show the same instrument doing different jobs, initial eligibility, program planning, annual reassessment, transition, and Part C Indicator 3 or Part B Indicator 7 outcome reporting, and a defensible report names which job this administration performed. The federal frame is constant: evaluation within 45 days of referral, no single measure as the sole eligibility criterion in either Part C or Part B, informed clinical opinion able to establish but never negate eligibility, and transition conferences at least 90 days before the third birthday. Clinical settings bill comprehensive developmental testing under the 96112 family where payer rules permit. Adaptive behavior at depth belongs to the Vineland-3 and ABAS-3 pages; the broader process frame lives on the developmental assessment page. ## How to structure a BDI-3 results section No authority prescribes a BDI-3 report format, but the instrument's flexibility and its legal contexts dictate the load-bearing elements: the full name with the Beck disambiguation, the exact component and edition, sources per domain, uncertainty around every quotient, adaptive integration without double-counting, an eligibility statement in the state rule's own currency, and a clean separation between eligibility and outcome reporting. Each section below carries the pitfall that most often undermines it. Identification, in full, with the disambiguation. First mention: Battelle Developmental Inventory, Third Edition, naming the component administered (Developmental Complete, Screening Test, or with the Early Academic Survey added) and the platform (paper or Riverside Score). Add one sentence that this developmental assessment is unrelated to the Beck Depression Inventory (BDI-II), then use BDI-3. Record chronological age, any corrected age with its state authority, testing dates, examiners and disciplines, setting, and language. Pitfall: Bare "BDI" in a table header, EHR field, or export. Two unrelated instruments share the acronym, records systems match on strings, and the Beck collision is a real chart-safety hazard, not pedantry. Edition and component boundaries. Confirm the write-up describes the third edition: Social-Emotional (not the BDI-2's Personal-Social), 2020 norms, current scoring criteria. Say explicitly whether the full battery or the Screening Test was given: the screener yields a flagging score, not a DQ profile, and cannot be dressed as comprehensive results. Where prior testing used the BDI-2, treat the edition change as a measurement discontinuity rather than computing change. Pitfall: Edition drift: Personal-Social terminology, BDI-2 crosswalks, or "gained eight points since 2022" across editions the publisher itself calls non-comparable. Administration and sources, per domain. Describe how information was obtained, structured administration, observation, caregiver or teacher interview, and how it varied by domain and subdomain; note multi-day or multi-examiner administration with every examiner and date (the platform records these per subdomain), accommodations, and which subdomains were age-inapplicable (several begin at age 2). Pitfall: A three-day, three-discipline arena assessment presented as one uninterrupted session, or interview-derived credit reading as elicited performance. The source labels are the evidence. Validity statement with content. Address engagement and regulation, hearing, vision, motor, language, and cultural factors, opportunity to perform the sampled skills, whether accommodations preserved standardization, and disagreements between informant report and observation, then conclude: valid, cautiously interpretable, or descriptive only. Pitfall: "Results are considered valid" as boilerplate. The sentence has to earn itself with what was actually observed. Domain results with intervals, scatter before the total. Report each domain Developmental Quotient with its confidence interval and percentile, then the subdomain scaled scores wherever they explain meaningful scatter (receptive versus expressive; gross versus fine motor; structured cognition versus functional application). Report the Total DQ, but when domains diverge, say plainly that the global score is not a sufficient summary and center the domain findings. Keep age equivalents secondary. Pitfall: A Total DQ doing eligibility work over an uneven profile, or bare quotients at a threshold: a DQ of 79 with an interval crossing the criterion is a different sentence from one clearly below it. Adaptive integration and the eligibility statement. When a Vineland-3 or ABAS-3 was also given, integrate rather than double-count: overlapping constructs, different respondents and methods, one adaptive conclusion. Then write eligibility in three separated statements: the normative interpretation (how many SDs below the mean), the state's quoted criterion in its own currency (an SD rule reads off the DQ; a percent-delay rule uses the state's own age-equivalent formula, and the two are not interchangeable), and the team's multisource determination. Pitfall: "Eligible because the BDI-3 was low." No single measure may decide eligibility, a DQ is a standard score that cannot be subtracted from 100 to yield percent delay, and informed clinical opinion can add but never subtract a qualifying result. Outcomes, progress, and the family summary. If the administration feeds federal outcome reporting, say which indicator, whether this is entry, interim, or exit, and how the crosswalk and team process were used (domain scores are the smallest crosswalkable unit, and the crosswalk is not a checklist); at re-evaluation, use the Change Sensitive Score to document absolute growth alongside the current DQ for normative standing. Close with a plain-language family summary: strengths, priority needs, examples from routines, and the next step. Pitfall: Outcome reporting conflated with eligibility, "scores went up" read as "no longer eligible" (peers advanced too), or a report that ends at classifications with nothing a family can use. ### Blank template (copy and adapt) BDI-3 RESULTS SECTION SKELETON Child: [initials] Evaluation date(s): [ ] Evaluator(s) + discipline(s): [ ] Instrument: Battelle Developmental Inventory, Third Edition (BDI-3), [Developmental Complete / Screening Test / + Early Academic Survey] Platform: [paper / Riverside Score / Mobile Data Solution] (Unrelated to the Beck Depression Inventory, BDI-II.) Chronological age: [ ] Corrected age (if used): [ ] (authority: [state policy]; effect on scoring stated) Referral question: [ ] Program + eligibility rule applied: [state/program; criterion quoted in its own currency] Administration + sources: [structured / observation / caregiver or teacher interview, by domain; examiners and dates per subdomain; accommodations; subdomains not age-applicable noted] Validity statement: [engagement, sensory and motor factors, language, opportunity, informant-observation disagreements; conclusion] Domain results (each with uncertainty): [Domain]: DQ [ ] (mean 100, SD 15), CI [ ], percentile [ ] Subdomain scaled scores (mean 10, SD 3) where scatter matters: [ ] Total DQ: [ ] (CI [ ]); [sufficient summary / not a sufficient summary because ...] Adaptive integration (if Vineland-3 / ABAS-3 given): [convergence or divergence; one adaptive conclusion, not two counts] Eligibility statement (three parts): [normative interpretation (SDs below mean) / quoted state criterion / team's multisource determination; informed clinical opinion role] Outcome reporting (if applicable): [Indicator 3 or 7; entry, interim, exit; crosswalk + team process; same-edition comparison only] Progress at re-evaluation: [Change Sensitive Score growth + current DQ standing; no cross-edition change claims] Family summary (plain language): [strengths, needs, routine examples, next step] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, record forms, norms, or conversion tables. ## Sample BDI-3 write-up (fictional) Scenario: an 18-month-old referred for limited words and social reciprocity, evaluated by a two-evaluator Part C team, qualifying on communication and social-emotional findings under a named rule. All details are fictional. Patient: J.M., 18 months · Setting: Part C early intervention eligibility evaluation, two-evaluator team · Clinician: R. Casteel, MEd (coordinating evaluator) · Note date: 08/24/2026 Identification: Battelle Developmental Inventory, Third Edition (BDI-3), Developmental Complete, administered on paper record forms with scoring through the licensed platform, 08/24/2026, by this evaluator and a speech-language pathologist as a two-evaluator team. This developmental assessment is unrelated to the Beck Depression Inventory (BDI-II). J.M. was 18 months 9 days, born at term; chronological age anchored all scoring. Referral: limited spoken words and reduced social reciprocity noted by the family and pediatrician. Eligibility rule applied: our program defines developmental delay as at least 1.5 standard deviations below the mean in two or more developmental areas. Administration and validity: Information came from structured activities, observation during floor play, and caregiver interview, varying by subdomain as the instrument permits; sources are attributed throughout, and interview-derived credit is identified as reported. Peer Interaction, Personal Responsibility, Perceptual Motor, and Reasoning and Academic Skills were not administered because they begin at age 2. J.M. regulated well with brief breaks, hearing screening earlier this month was passed, and the session was judged representative by his mother: results considered valid. Results: Motor DQ 94 (90 percent CI 88 to 100, 35th percentile) and Adaptive DQ 88 (CI 82 to 95, 21st percentile) fell within broad average limits. Cognitive DQ 84 (CI 78 to 91, 14th percentile) was low average, with the narrative noting that several cognitive items carry language demands, so this figure may partly reflect the communication picture. Communication DQ 68 (CI 63 to 75, 2nd percentile) was the profile's clearest weakness, receptive scaled score 6 sitting above expressive 4, both below expectations. Social-Emotional DQ 76 (CI 70 to 83, 5th percentile) was also below expectations, with Adult Interaction relatively spared and Self-Concept and Social Role items depressed. The Total DQ is reported in the score table but is not a sufficient summary of this uneven profile; the domain findings carry the interpretation. Eligibility analysis: Normatively, Communication fell more than 2 standard deviations below the mean and Social-Emotional about 1.6 below, each with its confidence interval below the criterion line. Under the program's quoted rule (1.5 SD in two or more areas), both areas meet the threshold in the rule's own currency; no percent-delay conversion is offered because the criterion is SD-based and the two metrics are not interchangeable. Caregiver report, direct observation, the pediatrician's referral concerns, and the speech-language pathologist's findings converge. Per Part C requirements, no single measure decides eligibility: the multidisciplinary team, applying informed clinical opinion, determined that J.M. meets the program's definition of developmental delay in communication and social-emotional development. Family summary and next steps: In plain terms, reviewed with J.M.'s parents: he moves well, feeds himself, solves little problems in play, and is warmly attached to familiar adults. Talking, and some of the back-and-forth of play, are what is hard right now, and he met the program's definition of delay in those two areas. What his parents described at home matched what the team saw. Next: the IFSP meeting is scheduled within the program's timeline to plan supports built into daily routines, with outcome-reporting entry data recorded from this same administration and any future re-evaluation using this same edition so growth can be measured honestly. This sample is fictional and for educational purposes. It does not describe a real child or record; the scores, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The instrument is identified in full with the component, platform, and the Beck disambiguation, so no records system or reader can collide the acronym. Sources are labeled per domain, age-inapplicable subdomains are named rather than silently skipped, and the validity statement earns its conclusion with observations. Every quotient carries its interval and percentile, the language-mediated cognitive caution is stated, and the Total DQ is explicitly declined as the summary of an uneven profile. The eligibility statement runs in three separated parts, normative interpretation, quoted rule in its own currency, team determination, with the no-single-measure and informed-clinical-opinion rules visible. The family summary is usable, and the closing lines set up honest measurement: same-edition re-evaluation and outcome-entry data from this administration. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the report answers to two laws and fifty-plus rulebooks. IDEA Part C requires evaluation, assessment, and the initial IFSP meeting within 45 days of referral, prohibits any single procedure as the sole eligibility criterion, and requires informed clinical opinion, which can establish eligibility the instruments miss but can never negate a qualifying result; Part B carries its own initial-evaluation timeline (commonly 60 days from consent), its own no-single-measure rule, and the transition machinery that puts a conference at least 90 days before the third birthday and an IEP in effect by it for eligible children (LAW). State definitions supply the arithmetic: a 2026 federal technical-assistance compilation counts 19 jurisdictions with a 2-SD-in-one-area criterion, 19 with 1.5 SD in two or more areas, 18 with a 25 percent delay, and overlapping others, so the report quotes the rule it applies. State Battelle practice is equally specific: Massachusetts moved statewide to the BDI-3 in January 2025 for eligibility and outcomes and warns against treating edition change as child change; Tennessee runs eligibility and recurring outcome assessment through its state data system without billing families; New Jersey mandates the Developmental Complete by two evaluators with four-month validity and no telehealth administration; Texas permits BDI-3 or DAYC-2; South Dakota's verified use is Indicator 7 outcome reporting (PROGRAM POLICY). Clinical billing under 96112 and 96113 follows payer rules and never doubles school or Part C work (PAYER POLICY). Canada and Australia buy the same kit into different systems. The BDI-3 launched with Canadian availability and is distributed through Nelson, but education and early-childhood identification are provincial matters with no national eligibility framework and no province-wide rule naming the Battelle, so a Canadian report identifies the actual program and criteria and weighs the fit of US-based norms (the English standardization sample is US-national; the separate Spanish sample is US bilingual) (PROGRAM POLICY, provincial). In Australia the BDI-3 is distributed through PAA and can contribute developmental evidence to the NDIS early childhood approach, which supports children under six on developmental delay or concern without a diagnosis and children under nine with disability, and whose evidence guidance centers on functional capacity, everyday participation, and support needs rather than any instrument cutoff; NDIS processes remain in active change through 2026, and the separate national Preschool Outcomes Measure pilot is a formative educator tool, not a standardized diagnostic to be conflated with Battelle outcome reporting (LAW and PROGRAM POLICY). Across all three countries the outcome-versus-eligibility distinction travels: federal US outcome reporting uses the ECTA crosswalk (updated February 2026) with domain scores as the smallest crosswalkable unit and an explicit warning that crosswalks are not checklists, and no jurisdiction anywhere reads a screener as a comprehensive evaluation. Editions, rights, and evidence honesty are where Battelle write-ups most often overreach. The third edition's English norms come from about 2,500 children in 20 age groups, gathered by 283 examiners in 44 states and weighted to 2020 census projections, with a separate roughly 1,000-child Spanish bilingual sample; publisher-reported reliability is strong (internal consistency about .92 to .97, test-retest about .79 to .87) but the publicly displayed interrater table covers only two domains (.82 and .97), so blanket interrater claims outrun the source. Independent peer-reviewed BDI-3 validation is essentially absent as of August 2026: the Buros review sits in the recently published Twenty-Second Mental Measurements Yearbook behind a paywall, screener accuracy ranges (sensitivity 0.80 to 0.94, specificity 0.76 to 0.84 in a state summary) derive from publisher documentation, no public infant floor-effect statistics exist, and the standing scholarly critique (the authentic-assessment literature) argues norm-referenced batteries under-represent young children's everyday functioning, which is exactly why observation, routines, and family evidence share the record. Rights are conventional: record forms are licensed consumables, no free public scorer exists (scoring runs through the subscription platform), EHR and state-system integration is a results-export matter rather than permission to recreate test content, and prematurity correction has no universal Battelle rule, so the state policy governs and the Bayley's correction convention must not be imported. Per the publisher's current legal notice, Battelle is a registered trademark of Battelle Memorial Institute, BDI-3 of LINC Associates, Inc., and Riverside Insights of Riverside Assessments, LLC. BastionGPT is not affiliated with, or endorsed by, any of these trademark owners or the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common BDI-3 write-up errors reviewers flag The numbers behind these errors are specific. The BDI-3's English norms rest on about 2,500 children across 44 states ; the publisher's public interrater table covers only two domains (.82 and .97); screener accuracy ranges (sensitivity 0.80 to 0.94) are publisher-derived, per a state evidence summary ; and a 2026 analysis found thousands of combinations where age-equivalent percent delay and standard scores disagree on eligibility. The BastionGPT Clinical Advisory Board sees the same errors most often in BDI-3 documentation reviews: Bare "BDI," and the Beck collision. "BDI score: 68" in a table, problem list, or export. The Battelle BDI-3 (a developmental battery for young children) and the Beck BDI-II (a depression measure for ages 13 and up) collide on the acronym, and Pearson's own pages style the Beck inconsistently, so punctuation cannot carry the distinction. Write the full name at first mention and "Battelle (BDI-3)" in every field. Edition drift. Personal-Social terminology, BDI-2 crosswalks or norms, or change computed across editions ("gained eight points since the 2022 evaluation") when the baseline was BDI-2. The second edition was sunset in 2023, the domain was renamed Social-Emotional, and the publisher warns the editions are not directly comparable: treat cross-edition differences as measurement discontinuities and re-baseline. Screener results dressed as the full battery. "BDI-3 results" from the Screening Test, or a screening score driving eligibility. The screener is a 10-to-30-minute flagging tool that yields a screening total, not a DQ profile; comprehensive language belongs only to the Developmental Complete, and some states expressly require the full battery (New Jersey: two evaluators, no telehealth). Name the component every time. The DQ treated as a ratio quotient. Percent delay computed as 100 minus the DQ, or an SD finding announced as its percent-delay "equivalent." The BDI-3 Developmental Quotient is a norm-referenced standard score, percent delay is an age-equivalent calculation under the state's own formula, and the two currencies disagree in thousands of score combinations; apply the quoted rule in its own metric and let the platform's authorized delay outputs do the arithmetic. Sources unlabeled, adaptive double-counted. Interview-derived credit reading as elicited performance, multi-day multi-examiner administration presented as one session, or the Battelle Adaptive domain and a Vineland-3 or ABAS-3 counted as two independent delays. Label the method per domain and examiner per date, and write one integrated adaptive conclusion that explains convergence or divergence. Eligibility, outcomes, and progress conflated. "Eligible because the score was low" (no single measure may decide, and informed clinical opinion can add but never subtract), outcome crosswalks used as eligibility checklists (the crosswalk's own documentation forbids it), or improved DQs read as "no longer delayed" without the Change Sensitive Score showing what actually grew while age peers advanced. Keep the three jobs in three labeled paragraphs. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Give it the facts (component and edition, ages, domains with sources, DQs with intervals, the state rule, adaptive measures given, outcome-reporting role) and it drafts the results section: disambiguated first mention, sources labeled, scatter handled before the total, the three-part eligibility statement, and a plain-language family summary, ready for your review. Cross-check a finished report for the gaps reviewers flag: a bare "BDI," Personal-Social terminology, a screener dressed as the full battery, a DQ-to-percent-delay conversion, an unlabeled source, or a double-counted adaptive finding. Draft the companion paragraphs: the re-evaluation section pairing Change Sensitive Score growth with current standing, the transition summary for the Part B team, or the outcome-reporting statement that keeps Indicator data separate from eligibility. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What scores does the BDI-3 report? Developmental Quotients (mean 100, SD 15) for each of the five domains and a Total DQ, each with a confidence interval (the publisher's sample reporting shows 95 percent intervals on domains and 90 percent on the total); subdomain scaled scores (mean 10, SD 3) beneath them; percentile ranks and related metrics; age equivalents, which stay secondary; a growth-purposed Change Sensitive Score centered so 500 represents a typical 36-month-old's level, built for comparing two administrations; and eligibility outputs through the scoring platform including months delay, percentage delay, and standard-deviation delay. Two structural rules govern reporting: a valid domain DQ requires all age-applicable subdomains in that domain, and a valid Total DQ requires all five domains, with the platform withholding scores (and issuing age-range warnings) when subdomains fall outside the child's age band, since several begin at age 2. Is the BDI-3 the same as the BDI-II? No, and the collision is worth engineering against. The BDI-3 is the Battelle Developmental Inventory, Third Edition (Riverside Insights, 2020), a developmental battery for children from birth through 7 years 11 months. The BDI-II is the Beck Depression Inventory, Second Edition (Pearson), a self-report depression measure for ages 13 and older. They share nothing but letters, yet EHR fields labeled "BDI," export tables that strip hyphens, search snippets, and clinical libraries holding both instruments create genuine chart-safety risk, and Pearson's own catalog styles the Beck as both "BDI-II" and "BDI-2," so punctuation alone cannot disambiguate. The working conventions: spell out the full name at first mention in every report, write "Battelle (BDI-3)" and "Beck (BDI-II)" wherever a field or table forces brevity, and never let bare "BDI" into a problem list. What changed from the BDI-2, and can scores be compared across editions? The third edition (2020) brought renormed scores weighted to 2020 census projections, repositioned and rescored content, the Riverside Score platform with mobile administration, a separate Spanish bilingual standardization sample, the new Early Academic Survey, and a renamed domain: the BDI-2's Personal-Social became the BDI-3's Social-Emotional, which makes stray "Personal-Social" wording the reliable detector of second-edition material masquerading as current. The BDI-2 was formally sunset in 2023 (final kits mid-year, final record forms at year end), and any post-transition administration needs a named local authority. On comparison, the publisher's own answer is no: BDI-2 and BDI-3 scores are not directly interchangeable, so longitudinal write-ups treat the edition change as a measurement discontinuity, re-baseline on the BDI-3, and use same-edition Change Sensitive Scores for genuine progress claims, exactly as state transition guidance (Massachusetts, notably) instructs. What BDI-3 score qualifies a child for early intervention? The state's rule decides, and the DQ is not a ratio you can turn into percent delay. Standard scores map onto SD-based criteria directly (a DQ of 70 sits 2 SD below the mean; 77.5 at 1.5 SD), and the state landscape genuinely varies: 19 jurisdictions carry a 2-SD-in-one-area criterion, 19 carry 1.5 SD in two or more areas, 18 carry a 25 percent delay, with most offering multiple routes. The trap is the label: the BDI-3 Developmental Quotient is a norm-referenced standard score, not developmental age divided by chronological age, so subtracting it from 100 yields nothing, and where a state requires percent delay the calculation runs through age equivalents under that state's exact formula, a psychometrically weaker metric that can disagree with the standard score for the same child. Then the law finishes the sentence: no single measure may determine eligibility, informed clinical opinion can establish what scores miss but never negate what they find, and the multidisciplinary team owns the decision. Is the BDI-3 Screening Test enough for an eligibility evaluation? As a general rule, no: it flags, the full battery evaluates. The Screening Test draws about 100 items across the five domains and thirteen subdomains, runs 10 to 30 minutes, and yields a screening total whose job is to identify children needing complete assessment; it does not produce the domain DQ profile that eligibility analysis and program planning require, and federal law's variety-of-tools requirement bars any single measure, screener included, from settling eligibility. State rules can be stricter still: New Jersey mandates the BDI-3 Developmental Complete administered by two evaluators, treats results as valid for four months, and prohibits telehealth administration. The defensible screening write-up therefore names the component ("BDI-3 Screening Test"), reports the broad areas of concern with sources and validity notes, and documents the follow-up evaluation plan, never borrowing full-battery language like "comprehensive developmental profile." Can BDI-3 domains be split across days, examiners, and methods? Yes, by design, provided the report shows it. Items may be obtained through structured administration, observation, or caregiver and teacher interview depending on the item; domains can be divided among disciplines in an arena model; and the scoring platform records a separate examiner and date per subdomain, with the publisher's own sample report documenting an assessment across three days. The write-up obligations follow directly: name every examiner, discipline, and date; label the method behind each domain's findings rather than presenting reported skills as elicited ones; follow the start-point, basal (three consecutive full-credit), and ceiling (three consecutive no-credit) rules consistently; note subdomains outside the child's age band (several begin at age 2); and keep the whole process inside the governing timeline, 45 days from referral to the initial IFSP in Part C. Platform flexibility is not a waiver of standardization, and a three-session battery presented as one sitting misstates the evidence. How does the BDI-3 fit federal outcome reporting, and how is progress documented? As a data source with its own crosswalk, kept separate from eligibility. Federal early-childhood outcome reporting (Part C Indicator 3; Part B preschool Indicator 7) rates three functional outcomes, and the ECTA Center's BDI-3 crosswalk (updated February 2026) maps domains and subdomains to them, with two explicit warnings: domain scores are the smallest unit that can be directly assigned, and the crosswalk is not a checklist, so family input and observation across routines remain required, whether a state computes ratings from entered data or runs the team-based summary process. The write-up states which indicator, whether the administration is entry, interim, or exit, and which edition was used at each point. For progress, the purpose-built metric is the Change Sensitive Score, which documents absolute skill growth between administrations, paired with the current DQ for normative standing: scores can rise while the child remains eligible because age peers advanced too, and only same-edition comparisons count. Can BastionGPT draft my BDI-3 report? Yes. Give it the facts (component and edition, ages and any state-authorized correction, domains with each one's sources and examiners, DQs with intervals and subdomain scatter, the quoted state rule, adaptive measures administered, and the outcome-reporting role) and it drafts the results section: disambiguated first mention, sources labeled, scatter handled before the total, the three-part eligibility statement, and a plain-language family summary, ready for your review. It can also cross-check a finished report for a bare "BDI," edition-drift wording, screener results in full-battery clothing, a DQ-to-percent-delay conversion, or a double-counted adaptive finding, and it can draft re-evaluation and transition variants. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: Riverside Insights, BDI-3 product page (edition, domains, ages), BDI-2 versus BDI-3 differences (non-comparability caution), BDI-2 sunset notice , standardized administration guidance (basal and ceiling logic), and the standardization brochure (norm sample; Spanish sample). Federal law: eCFR, 34 CFR 303.321 (multisource evaluation; informed clinical opinion), 303.310 (45 days), and 34 CFR 300.301 and related Part B evaluation rules (timelines; no single measure). State practice: Massachusetts, BDI-3 implementation (statewide from January 2025) ; Tennessee TEIS, evaluation and assessment guidance ; New Jersey, NJEIS-11 policy (two evaluators; four-month validity; no telehealth); Texas HHS, ECI instrument transition materials ; South Dakota DOE, Indicator 7 guide ; ECTA Center, state eligibility criteria summary . Outcome reporting: ECTA Center, BDI-3 crosswalk to the three child outcomes (updated February 2026) and crosswalk guidance (not a checklist; multiple sources required). Psychometrics and critique: Riverside, technical summary brochure (reliability ranges; two-domain interrater table); Minnesota Department of Health, BDI-3 Screening Test evidence summary (publisher-derived accuracy ranges); Gladfelter A and colleagues, 2026, percent delay versus standard scores ; Neisworth JT, Bagnato SJ, 2004, the authentic-assessment critique lineage ; Buros Center, Twenty-Second Mental Measurements Yearbook listing (review paywalled). Beck disambiguation and international context: Pearson, Beck Depression Inventory-II product page (inconsistent BDI-II and BDI-2 styling); Nelson, Canadian distribution ; PAA, Australian distribution ; NDIS, early childhood approach requirements . ## Related templates DAYC-2 Report Write-Up Bayley-4 Report Write-Up Vineland-3 Report Write-Up ABAS-3 Report Write-Up Child & Adolescent Developmental Assessment BDI-II Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/beery-vmi-report All note templates # Beery VMI Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Beery VMI (Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition) is a norm-referenced battery for ages 2 to 100: a core copying test plus optional Visual Perception and Motor Coordination supplemental tests whose contrast localizes a weakness. OTs, school psychologists, and evaluators use it in handwriting and developmental referrals. Reporting only one score discards the design. This page covers how to write up Beery VMI results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Occupational therapists, school psychologists, and psychologists; Pearson qualification level B; scoring is by hand against the manual's criteria Audience IEP and Section 504 teams, school OT services, psychoeducational and neuropsychological batteries, pediatric rehabilitation, payers reviewing evaluations, parents Typical length 200 to 500 words for the results section · core form 10 to 15 minutes; each supplemental about 5 minutes Format family Norm-referenced visual-motor battery (core copying test plus optional Visual Perception and Motor Coordination supplementals; ages 2 to 100) When it's used Handwriting and fine-motor referrals, psychoeducational and developmental evaluations, OT evaluations, rehabilitation and neuropsychological batteries across the lifespan Standards context Sixth edition (Pearson, 2010; adult norms collected 2006) remains current as of August 2026; no authority mandates it; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Beery VMI? The Beery VMI (Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition; Beery, Buktenica, and Beery; Pearson, 2010) is the most widely used visual-motor measure in school and pediatric practice, and it is a three-test system, not one score. The core test asks the examinee to copy geometric forms of increasing difficulty, a task that blends visual analysis, planning, and motor production; two optional supplemental tests use the same forms to pull those threads apart, Visual Perception (matching, with the motor demand stripped out) and Motor Coordination (controlled pencil work within boundaries, with the perceptual analysis reduced). There are four administration forms behind the three tests: the core exists as a Full Form (ages 2 to 100) and a Short Form (commonly used for ages 2 through 7), and when both supplementals are given the publisher fixes the sequence, core first, then Visual Perception, then Motor Coordination. Each test yields its own standard score (mean 100, SD 15) and percentile, scored by hand against developmental criteria on the examinee's first production; there is no combined three-test composite, no automated scoring platform, and no seventh edition announced as of August 2026. The load-bearing facts for the write-up are the contrast and its limits. The supplementals are what let a low copying score be localized, toward perception, motor execution, or the integration itself, and the strongest evidence in the literature says they carry independent value: in one achievement study the Visual Perception test out-predicted the core for math. But the contrast is a hypothesis generator, not a decomposition: across studies the two supplementals explain only about 22 to 36 percent of core-score variance, a normal core can coexist with a weak supplemental, and no authority publishes a universal discrepancy cutoff, so pattern language stays at "suggests," never "proves." A core-only administration is legitimate screening when its limits are stated; a core-only report claiming an "integration deficit" is not. Results usually land in a psychoeducational report beside cognitive measures like the WISC-V and achievement measures like the WIAT-4 , whose writing-fluency tasks approach the same referral question from the academic side. ## Who uses the Beery VMI and when School-based occupational therapists are the heaviest users, typically on handwriting and fine-motor referrals feeding IEP and Section 504 processes, where federal law requires a variety of tools and prohibits any single measure from deciding eligibility, so the Beery contributes evidence and never a verdict. School psychologists and psychologists fold it into psychoeducational and developmental batteries from age 2 (it is one of very few instruments normed that young); neuropsychologists use it across the lifespan in rehabilitation and dementia work, where the write-up should note that the adult norms were collected in 2006 and carried forward from the fifth edition. Pediatric evaluations pair it with sensory questionnaires like the Sensory Profile 2 (a caregiver-report instrument measuring a different construct) and with the NEPSY-II 's design-copying and visuomotor subtests inside fuller neuropsychological batteries. The report's readers are practical: an IEP team deciding services, a teacher looking for classroom implications, a payer reviewing an OT evaluation billed under the 97165 family (the test rides inside the evaluation, not as a separate charge), or a physician who referred for "messy handwriting" and needs to hear what the instrument can and cannot say about it. The broader evaluation frame lives on the developmental assessment page. ## How to structure a Beery VMI results section No authority prescribes a Beery VMI report format, but the instrument's three-test design dictates the write-up's logic: name exactly what was administered, report every administered test separately, put the behavior beside the scores, interpret the pattern as hypothesis, state the boundaries, and land on function. Each section below carries the pitfall that most often undermines it. Instrument identification, exactly. Name the edition and everything administered: Beery VMI Sixth Edition, core Full Form or Short Form, plus whichever supplementals were given, with the administration context (individual or group; the supplementals are individually administered) and any departure from standard procedure. When both supplementals were given, they followed the core in the publisher's fixed sequence. Pitfall: "Beery VMI: 82." Three tests and four forms live under one name, and a score without its test and form is as ambiguous as a Wechsler score without its index. Supplementals administered, or their absence justified. The publisher frames Visual Perception and Motor Coordination as optional follow-up tests, so a core-only screening is legitimate, but say so: state that without the supplementals the result cannot determine whether a weakness is predominantly perceptual, motor, or integrative. When the referral asks why performance is low, administer them; the Visual Perception test has out-predicted the core for academic outcomes in published work. Pitfall: A core-only score doing localization work ("a visual-motor integration deficit") or, the opposite error, supplementals described as universally mandatory. The precise rule is optional-but-sequenced, and the limits travel with the choice. Scores with the right labels. Report each administered test separately: standard score (mean 100, SD 15) and percentile, with the measurement-error framing the manual supports. Age equivalents stay secondary and descriptive. Scoring is by hand, on the first production (no erase-and-retry, no best-of-attempts), against the manual's developmental criteria; there is no tri-test composite to report. Pitfall: Age equivalents headlining ("functions like a 4-year-old"), an invented composite, or a score from a coached, retried administration presented as standardized. Observations beside the scores. Place the behavior next to the numbers: hand dominance and switching, grasp and pressure, page stabilization, posture and viewing distance, tremor or line control, pace and speed-accuracy trade-off, impulsive starts versus careful checking, fatigue, frustration, and comprehension of the task. These observations carry the functional interpretation and the validity call. Pitfall: Scores reported bare. A rushed one-minute administration is a behavioral finding on an untimed task, not a scoring artifact, and only the observations make that readable. Pattern interpretation, at hypothesis strength. Interpret the three-score contrast in plain words: low core with stronger supplementals suggests the combined demands were the difficulty; low core with weakest Motor Coordination supports a graphomotor hypothesis; weakest Visual Perception directs attention to form analysis and a vision check. Anchor any between-test gap to measurement error, and remember the supplementals explain only a minority (roughly 22 to 36 percent) of core variance. Pitfall: A contrast promoted to a diagnosis ("proves a motor deficit"), or a universal discrepancy cutoff applied: no public authority establishes one, and independent retest limits of agreement have spanned roughly 27 points. Boundaries: what the instrument does not establish. State them: the Beery VMI is not a handwriting assessment (in a developmental coordination disorder sample it did not correlate with handwriting product or process), it does not diagnose dysgraphia or determine eligibility (single measures cannot), and it is not an eye examination, though a weak Visual Perception result warrants excluding a near-vision problem, since induced blur demonstrably lowers all three scores. Pitfall: "Low VMI, consistent with dysgraphia." Handwriting conclusions require direct writing samples, and a low score with unchecked vision may be an optometry finding wearing an OT label. Functional linkage, goals, and progress plan. Connect the localized hypothesis to observed function: classroom writing samples, tool use, endurance, participation. Write goals on the activity (legibility, sustainable output, efficient grasp, accommodations), never on the score, and plan progress measurement through repeated authentic samples: retest reliability is moderate, no minimal clinically important difference has been established, and direct handwriting gains have occurred without Beery score movement. Pitfall: "Increase the Motor Coordination standard score to 90" as a goal, or re-administration every grading period as a progress measure. The instrument is poorly suited to frequent monitoring, and the publisher's own guidance sets a minimum retest interval. ### Blank template (copy and adapt) BEERY VMI RESULTS SECTION SKELETON Client: [initials] Evaluation date: [ ] Evaluator + credentials: [ ] Referral question: [why visual-motor assessment was selected] Instrument: Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition Tests administered: [core Full Form / core Short Form] [+ Visual Perception] [+ Motor Coordination] (sequence per manual when both supplementals given) Administration: [individual / group core] Deviations: [ ] If core-only: [stated limits: result does not determine whether a weakness is perceptual, motor, or integrative; reason supplementals omitted] Scores (each test separately; hand-scored, first production): Core VMI: SS [ ] (mean 100, SD 15), percentile [ ] Visual Perception: SS [ ], percentile [ ] Motor Coordination: SS [ ], percentile [ ] (Age equivalents secondary and descriptive only; no tri-test composite) Observations (beside the scores): [dominance, grasp, pressure, stabilization, posture, tremor or line control, pace, speed-accuracy trade-off, checking vs impulsivity, fatigue, comprehension] Vision status: [corrective lenses worn; recent screening; low Visual Perception = exclude near-vision problem before interpreting] Pattern interpretation (hypothesis strength): [what the contrast suggests; gaps anchored to measurement error; supplementals explain only a minority of core variance] Boundaries stated: [not a handwriting assessment; does not diagnose dysgraphia or determine eligibility; not an eye examination] Functional corroboration: [classroom or daily-task evidence: legibility, endurance, tool use, participation] Goals + progress plan: [activity-based goals; progress via repeated authentic samples, not score re-administration; retest interval respected if a retest is planned] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test forms, geometric stimuli, scoring criteria, or norms. ## Sample Beery VMI write-up (fictional) Scenario: a kindergarten handwriting referral where the supplemental contrast supports a graphomotor hypothesis, corroborated by classroom samples and kept at hypothesis strength. All details are fictional. Patient: A.K., 5 · Setting: School-based occupational therapy evaluation · Clinician: M. Duarte, OTR/L · Note date: 08/22/2026 Referral and instrument: A.K. was referred by his kindergarten team for effortful, poorly legible printing and avoidance of drawing tasks. As part of a broader occupational therapy evaluation, he was administered the Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition: the core Full Form followed by the Visual Perception and Motor Coordination supplemental tests in the publisher's sequence, individually, under standard conditions. He wears no corrective lenses and passed the school vision screening this month. Results: Core Visual-Motor Integration: standard score 82 (12th percentile), below the average range. Visual Perception: standard score 98 (45th percentile), squarely age-expected. Motor Coordination: standard score 76 (5th percentile), his weakest performance. Scores are reported separately by design; the instrument produces no combined composite, and each result carries the measurement-error framing of the manual. Age equivalents are omitted as secondary. Observations: A.K. showed consistent right-hand use with his left hand stabilizing the page. His grasp tightened visibly as precision demands increased, pencil pressure was heavy enough to indent the page, and he slowed markedly to stay within boundaries, trading speed for control until fatigue loosened both. On the matching task he scanned efficiently and answered promptly, with none of the effort that pencil work produced. He understood every task and persisted well with encouragement. Interpretation: The pattern, a below-average copying score beside age-expected visual matching and a clearly weaker constrained-pencil-control result, supports a working hypothesis that graphomotor precision and efficiency, more than visual analysis, are constraining A.K.'s visual-motor performance. This is stated as a hypothesis rather than proof: the supplemental tests account for only a minority of core-score variance, and score gaps of this size are interpreted alongside, not instead of, the observed behavior. The Beery VMI is not a handwriting assessment, does not diagnose dysgraphia, and is not an eye examination; his classroom writing samples, which show adequate letter recognition but declining legibility and endurance across longer tasks, are the functional evidence, and they corroborate the hypothesis. Recommendations: Intervention should target the activity, not the score: sustainable pencil control and writing endurance during authentic classroom tasks, an efficient grasp and page-stabilization strategy, and short-burst writing with planned breaks while stamina builds, with accommodations (reduced copying volume, pencil grip trial) as the team judges useful. Progress will be measured through repeated classroom writing samples rather than Beery re-administration, which is poorly suited to frequent monitoring; any future retest will respect the publisher's minimum interval and be interpreted against measurement error. These findings are one contribution to the team's evaluation: no single measure determines eligibility or services, and the full record, teacher input, and family perspective complete the picture. This sample is fictional and for educational purposes. It does not describe a real child or record; the scores, observations, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no test stimuli or norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works Everything administered is named: edition, core form, both supplementals, and their sequence, so "the VMI score" ambiguity never arises. All three tests are reported separately with the right labels, age equivalents are declined, and no invented composite appears. The observations sit beside the scores and do the functional work: the grasp, pressure, and speed-accuracy trade-off are what make the motor hypothesis readable. The interpretation stays at hypothesis strength with its evidence limits stated, and the boundaries (not handwriting, not dysgraphia, not an eye exam) are explicit, with vision checked before the perceptual score was trusted. The goals target classroom function with progress measured by authentic samples, and the eligibility frame keeps the instrument contributory, never decisive. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the legal frame is generic, and that is the point. IDEA's evaluation rules require a variety of assessment tools, prohibit any single measure or assessment as the sole criterion for disability determination or programming, require technically sound instruments, and require assessment in all areas of suspected disability, and no category (specific learning disability, developmental delay, orthopedic impairment, autism) names the Beery VMI; Section 504's evaluation rules likewise require multiple sources and suited tests (LAW). So the report contributes to a team decision and never announces one. In tiered handwriting support, direct repeated writing samples, legibility, speed, endurance, and participation, are the appropriate progress measures, with the Beery as an initial hypothesis-former at most (CONVENTION). Billing follows the container: an OT-administered Beery rides inside the occupational therapy evaluation codes (97165 to 97167, re-evaluation 97168) rather than generating a separate testing charge, while a psychologist folding it into a battery works under the 96130-series evaluation and administration codes, all subject to payer medical-necessity and documentation rules (PAYER POLICY). State OT practice acts add documentation requirements that vary by state and never prescribe the instrument (LAW or PROFESSIONAL REGULATION, state by state). Canada and Australia put the same instrument inside participation-first frameworks. Canadian school OT runs through provincial arrangements and regulatory colleges whose standards require suitable, evidence-informed assessment, competence, and clear documentation without mandating named tools, and Ontario's special-education guidance treats an OT report as one assessment source in the IEP record; Canadian school-practice models emphasize classroom participation, observation, and tiered support, with individual standardized testing reserved for the questions it actually answers (PROFESSIONAL STANDARD and CONVENTION, provincial). Australia's Disability Standards for Education require consultation and reasonable adjustments rather than named tests, the national data collection (NCCD) wants evidence of assessed need, implemented adjustments, and monitoring from multiple sources, and NDIS evidence turns on functional impact in everyday life, so a Beery profile helps only insofar as the report translates it into participation and support needs; the OT board's competency standards govern the professional side (LAW and POLICY). The cross-jurisdiction conclusion is identical everywhere: no law, payer, or professional body in the three countries makes the Beery VMI mandatory for any eligibility category, plan, or funding decision, and its role is evidentiary and hypothesis-generating. Instrument facts, rights, and evidence honesty complete the picture. The sixth edition (2010) remains current with no announced successor; its child norms date to roughly 2009 and 2010 (about 1,700 participants aged 2 to 18) while the adult norms (about 1,000 participants) were collected in 2006 and carried forward from the fifth edition, which a 2026 adult report should say plainly. The independent evidence urges restraint in three places: retest reliability was only moderate in an independent study (correlations around .54 to .58, with individual limits of agreement spanning roughly 27 points), no minimal clinically important difference could be established in an autism sample whose functional gains outran score movement, and the marketed "culture-free" framing overstates the case, since a 2025 Chinese preschool study found US norms flagged 1.4 percent of children as weak where local norms flagged 14.5 percent; "nonverbal and low in language demand" is the defensible phrasing, with a stated limitation whenever US norms meet a different cultural or educational background. Vision is the quiet confound: induced blur lowers all three scores, and a weak Visual Perception result warrants excluding a near-vision problem before any perceptual interpretation. Rights are Pearson-standard: record forms and stimuli may not be reproduced, scanned, or rebuilt into templates or EHRs (results and conclusions belong in the record; a narrow output-excerpt permission exists for licensed software, and electronic transfer of results is a permission-managed use), scoring is by hand with no authorized digital scorer, and no free web tool is legitimate. Beery VMI and the Beery-Buktenica Developmental Test of Visual-Motor Integration are trademarks used by Pearson (NCS Pearson, Inc. and affiliates). BastionGPT is not affiliated with, or endorsed by, Pearson. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Beery VMI write-up errors reviewers flag The numbers behind these errors are specific. Across published studies the two supplemental tests explained only about 22 to 36 percent of core-score variance ; independent test-retest correlations ran .54 to .58 with limits of agreement spanning roughly 27 points ; in a developmental coordination disorder sample the instrument did not correlate with handwriting product or process ; and a 2025 study found US norms flagged 1.4 percent of Chinese preschoolers against 14.5 percent by local norms . The BastionGPT Clinical Advisory Board sees the same errors most often in Beery VMI documentation reviews: "The VMI score," unidentified. A single number with no edition, no core form (Full or Short), and no statement of which of the three tests produced it. The instrument is a three-test system on four forms, scored separately with no combined composite; name everything administered, and when both supplementals were given, note they followed the core in the publisher's sequence. Localization from a core-only score. "A visual-motor integration deficit" concluded from the copying test alone. Without the supplementals the result cannot say whether the weakness is perceptual, motor, or integrative; core-only screening is legitimate only with that limit stated, and the supplementals earn their five minutes each: in published work Visual Perception out-predicted the core for academic achievement. A contrast promoted to a diagnosis. "Proves a motor deficit" from a Motor Coordination gap, or a universal discrepancy cutoff applied (a circulating 12-point rule has no located authority). The supplementals explain only a minority of core variance, gaps must clear measurement error, and the honest verbs are "suggests" and "raises the hypothesis," corroborated by observation and function. Handwriting and dysgraphia claimed from the score. "Low VMI, consistent with dysgraphia." The instrument was not designed to assess handwriting, correlations with legibility and speed are modest and sample-dependent, and one DCD study found none at all; handwriting conclusions require direct writing samples, instructional history, and the broader record, and no single measure determines eligibility anywhere. Progress monitored by re-administration. Beery scores as OT outcome measures, or goals written on the standard score. Retest reliability is moderate with wide individual limits, no minimal clinically important difference has been established, and direct handwriting gains have occurred without score movement; write goals on the activity, measure progress with authentic samples, and respect the publisher's minimum retest interval. Norms, culture, and vision left unexamined. Adult results on 2006 norms with no comment, "culture-free" repeated from marketing when cross-cultural work contradicts it, age equivalents presented as developmental ages, or a weak Visual Perception score interpreted without excluding a near-vision problem, though induced blur lowers all three scores. Each belongs in the validity and limitations lines. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Give it the facts (tests and forms administered, three standard scores with percentiles, observations, vision status, classroom or functional evidence) and it drafts the results section: everything named, scores separated, the pattern interpreted at hypothesis strength with boundaries stated, and goals written on function, ready for your review. Cross-check a finished report for the gaps reviewers flag: an unidentified "VMI score," localization from a core-only administration, a contrast promoted to a diagnosis, a dysgraphia claim without writing samples, or goals written on the score. Draft the companion pieces: the core-only limitation paragraph, the teacher-facing classroom summary, or the re-evaluation section that pairs authentic writing samples with an interval-respecting retest. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What are the three Beery VMI tests, and how are they scored? The core Visual-Motor Integration test asks the examinee to copy geometric forms of increasing difficulty; the Visual Perception supplemental presents the same forms as a matching task with the motor demand removed; and the Motor Coordination supplemental requires controlled pencil work within boundaries with the perceptual load reduced. The core comes in two alternative formats, a Full Form (ages 2 to 100) and a Short Form commonly used through age 7, so the system is three score-bearing tests on four forms. Each test is scored by hand against the manual's developmental criteria, on the examinee's first production (no erasing and retrying, no best-of-attempts), and yields its own standard score (mean 100, SD 15), percentile, and secondary age equivalent. There is no combined three-test composite and no authorized automated scoring: the design intent is that the three scores be compared, and the profile, not a sum, is the deliverable. Do the Visual Perception and Motor Coordination supplementals have to be administered? The publisher's precise position: they are optional follow-up tests, and when both are administered they follow the core in a fixed sequence (core, then Visual Perception, then Motor Coordination). The practical rule for reports is sharper: a core-only administration is legitimate screening, but it cannot support any claim about whether a weakness is perceptual, motor, or integrative, and the write-up must say so, in words like "the supplemental tests were not administered, so this result does not determine the source of the observed weakness." When the referral question is why performance is low, the supplementals are what answer it, they take about five minutes each, and the published evidence gives them independent standing: in one achievement study the Visual Perception test was the strongest predictor of math outcomes while the core copying score dropped out of the model. What does a low Beery VMI score mean? By itself, that reproducing visual material was hard, without saying why: the core task blends visual analysis, planning, and motor production, and a low score cannot apportion blame among them. The supplemental contrast generates the hypotheses: stronger matching with weaker constrained pencil control points toward graphomotor execution; weaker matching points toward form analysis and earns a near-vision check first, since experimentally induced blur lowers all three scores; all three low suggests a broader developmental picture rather than three separate deficits; and a low core with both supplementals intact leaves integration, planning, attention, or production strategy on the table. Every one of those readings stays at hypothesis strength: the supplementals explain only about 22 to 36 percent of core variance, gaps must clear measurement error, and corroboration comes from observations and real-task function, never from the numbers alone. Is the Beery VMI a handwriting test, and can it diagnose dysgraphia? No on both counts, and the evidence is blunter than the folklore. The instrument was not developed or intended to assess handwriting: correlations with legibility and speed are modest and sample-dependent, sensorimotor measures explained no more than about a quarter of handwriting variance in one school-age study, and in a developmental coordination disorder sample the Beery showed no significant correlation with handwriting product or process at all. Dysgraphia and written-expression disorder determinations require direct writing samples, instructional and intervention history, achievement and language evidence, and educational impact, and eligibility law in the US prohibits any single measure from deciding. The defensible role: a low Beery result can justify looking closely at written output, and an intact profile can usefully redirect the inquiry toward instruction, automaticity, language, attention, or endurance; the handwriting conclusion itself always rests on the writing. Can the Beery VMI measure OT progress? Poorly, and the write-up should plan around that. Independent test-retest correlations were only moderate (.54 to .58) with 95 percent limits of agreement spanning roughly 27 points, meaning an individual's score can swing substantially without any true change; a study attempting to establish a minimal clinically important difference in an autism sample could not, with scores flat across eleven months of therapist-documented functional improvement; and handwriting-intervention research has found direct writing outcomes improving while Beery scores stood still. The publisher's own guidance sets a minimum retest interval of about a month, which is an administration floor, not an invitation to monthly monitoring. The defensible plan: write goals on the functional activity, measure progress with repeated authentic samples (legibility, speed, endurance, participation), reserve any retest for a genuine re-evaluation at a respectful interval, and interpret change only beyond documented measurement error. Are the Beery VMI norms current, and is the test really culture-free? The norms are aging and the culture-free claim needs demotion. The sixth edition (2010) remains current with no announced successor, its child norms were collected around 2009 and 2010 (about 1,700 participants aged 2 through 18), and its roughly 1,000-person adult norms were collected in 2006 and carried forward from the fifth edition, so an adult evaluated in 2026 is being compared with a two-decade-old sample, which the report should say. On culture: "nonverbal and low in language demand" is defensible; "culture-free" is not, because drawing experience, pencil familiarity, preschool curriculum, and local developmental expectations all shape performance, and a 2025 study of 421 Chinese preschoolers found US norms flagged 1.4 percent of children as weak where locally derived norms flagged 14.5 percent, an order-of-magnitude difference. A clinician using US norms with a child from a different cultural or educational background states that limitation explicitly. Who can administer the Beery VMI, and can it go into an EHR or online tool? It sells at Pearson qualification level B (appropriate graduate training, licensure, or supervised assessment training), the core copying forms can be group-administered while the supplementals are individual, and scoring is by hand: the digital manual sold through the publisher's platform is view-only, and no authorized automated or free web scorer exists, so any online "Beery calculator" is unauthorized on its face. Rights follow Pearson's standard rules: record forms, stimuli, scoring criteria, and norms may not be photocopied, scanned, retyped, or rebuilt into report templates, EHR forms, or software, and permission processes govern electronic uses, with a narrow allowance for excerpting minimal output conclusions into a written evaluation and permission-managed transfer of results into records systems. What always belongs in the chart is yours: the scores, observations, interpretation, and recommendations, written in your own words. Can BastionGPT draft my Beery VMI report? Yes. Give it the facts (edition, core form, supplementals administered, the three standard scores with percentiles, your observations, vision status, and the classroom or functional evidence) and it drafts the results section: everything named, scores separated with the right labels, the pattern interpreted at hypothesis strength with boundaries stated, and recommendations written on function, ready for your review. It can also cross-check a finished report for an unidentified "VMI score," localization claimed from a core-only administration, a contrast promoted to a diagnosis, a dysgraphia claim without writing samples, or goals written on the score, and it can draft the core-only limitation paragraph or the teacher-facing summary. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: Pearson, Beery VMI Sixth Edition product page (ages, forms, times, manual scoring, qualification level B), forms and sequence support article (optional supplementals; fixed order), first-production scoring guidance , retesting interval guidance , and permissions and licensing . Supplemental-contrast evidence: Sortor JM, Kulp MT, 2003, Optometry and Vision Science, the supplementals and academic achievement (Visual Perception out-predicted the core for math); Kulp MT, Sortor JM, 2003, supplementals explain about 36 percent of core variance ; Avi-Itzhak T, Obler DR, 2008, about 22 percent in preschoolers, with intact cores over weak supplementals . Reliability and change: Harvey EM and colleagues, 2017, Optometry and Vision Science, interrater and test-retest study (retest correlations .54 to .58; limits of agreement about 27 points); Ohl AM, Schelly D, 2022, British Journal of Occupational Therapy, no estimable minimal clinically important difference in an ASD sample ; Pfeiffer B and colleagues, 2015, handwriting gains without corresponding VMI change . Handwriting boundary: Prunty M, Barnett AL, Wilmut K, Plumb M, 2016, no VMI-handwriting correlation in developmental coordination disorder ; Klein S and colleagues, 2011, sensorimotor measures and handwriting variance ; NINDS common data elements, Beery VMI catalog entry (not developed to assess handwriting; norm years). Culture and vision confounds: Tang X and colleagues, 2025, Frontiers in Pediatrics, Chinese preschool norms study (14.5 versus 1.4 percent flagged; supplementals about 25 percent of core variance); Findlay R and colleagues, 2020, PLOS ONE, induced blur lowers all three scores ; Pearson, age-equivalent interpretation problems . Law and billing: eCFR, IDEA evaluation procedures (34 CFR 300.304) (variety of tools; no single measure); Section 504 evaluation rules ; APA Services, testing billing and coding guide (96130-series families); CMS OT evaluation coding context for the 97165 family; NDIS, functional supporting-evidence guidance . ## Related templates WISC-V Report Write-Up WIAT-4 Report Write-Up Psychoeducational Evaluation Report NEPSY-II Report Write-Up Sensory Profile 2 Report Write-Up Child & Adolescent Developmental Assessment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/sensory-profile-2-report All note templates # Sensory Profile 2 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The Sensory Profile 2 (SP2) is a family of five caregiver and teacher questionnaires (birth through 14 years 11 months) measuring sensory processing patterns through Dunn's four quadrants. OTs and evaluation teams use it in autism, feeding, and school evaluations. The write-up must name the form and respondent, and report patterns without diagnosing. This page covers how to write up SP2 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Occupational therapists most centrally, plus school psychologists, developmental teams, and autism evaluators; Pearson qualification level B Audience Autism diagnostic teams, IEP and school-support teams, referring physicians and psychologists, OTs planning intervention, NDIS planners, families Typical length 300 to 700 words for the results section · the respondent completes the questionnaire independently; scoring (manual or Q-global) and integration add clinician time Format family Norm-referenced caregiver and teacher questionnaire family (five forms; Dunn's four-quadrant model plus sensory-system and behavioral sections) When it's used Autism evaluations (DSM-5 criterion B.4 evidence), feeding, dressing, and self-regulation referrals, school evaluations and IEP planning, birth through 14:11 Standards context Published by Pearson (2014; current edition as of August 2026); scored by hand or via Q-global; described here for write-up purposes, no items or cut values reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Sensory Profile 2? The Sensory Profile 2 (SP2; Winnie Dunn, Pearson, 2014) is not a single test but a family of five norm-referenced questionnaires: the Infant form (birth to 6 months, caregiver report, 25 items), Toddler form (7 to 35 months, caregiver, 54 items), Child form (3:0 to 14:11, caregiver, 86 items), Short form (3:0 to 14:11, a 34-item screening subset of the Child form), and School Companion (3:0 to 14:11, teacher report, 44 items). The 2014 revision cut items on every form except the Toddler form, so first-edition item counts still circulating online are wrong. Respondents rate how often behaviors occur on a five-point frequency scale, and results come back in two layers: quadrant scores from Dunn's model, in which neurological threshold (low to high) crossed with self-regulation strategy (passive to active) yields Seeking, Avoiding, Sensitivity, and Registration, plus sensory-system sections (auditory, visual, touch, movement, body position, oral) and behavioral sections, with the School Companion adding school factors. Every score is expressed as a classification band against same-age norms, in frequency language running from "Much Less Than Others" to "Much More Than Others." The upper age bound is 14 years 11 months, not 15; from age 11 the separate self-report Adolescent/Adult Sensory Profile (Brown and Dunn, 2002) becomes an alternative, and the evaluator's choice between them belongs in the report. Three facts carry most of the write-up's weight. First, the five forms have different items, respondents, and norms, so a results section that does not name the form and the respondent is uninterpretable; no reader can know what "the Sensory Profile 2" measured without it. Second, the bands describe frequency relative to peers, not severity and not disorder: "Much More Than Others" means the respondent reports that behavior much more often than the standardization sample's caregivers or teachers did, and an elevation is a pattern description, not a diagnosis. Third, when caregiver and teacher forms disagree, the disagreement is information about context rather than error, because each form measures behavior in its own setting and cross-informant convergence is only moderate. In autism evaluations the SP2 typically serves as supporting evidence for DSM-5 criterion B.4, the sensory-reactivity criterion, alongside instruments covered on the ADOS-2 page and the broader synthesis covered on the autism evaluation report page ; it never carries the diagnosis alone. ## Who uses the Sensory Profile 2 and when Occupational therapists are the core users, in outpatient clinics, early intervention, and schools, where the SP2 documents how sensory processing affects feeding, dressing, play, and classroom participation; Pearson gates materials at qualification level B. Autism diagnostic teams are the second major group: psychologists and developmental pediatricians fold a caregiver SP2, often paired with a teacher School Companion, into multidisciplinary evaluations as sensory-reactivity evidence, alongside rating scales covered on the SRS-2 and CARS-2 pages. School teams use the School Companion for IDEA evaluations and IEP planning, framed around educational participation rather than medical diagnosis. In Australia, SP2 findings feed the functional evidence inside an NDIS functional capacity assessment . The report's readers mirror those uses: an IEP team that must honor IDEA's prohibition on any single measure deciding eligibility, a diagnosing clinician weighing criterion B.4, a payer seeing the questionnaire inside OT evaluation codes 97165 to 97167, an NDIS planner looking for functional impact, and a family who deserves plain language about what the pattern means at the dinner table. Visual-motor assessment, the other staple of pediatric OT batteries, lives on the Beery VMI page , and the broader evaluation frame on the developmental assessment page . ## How to structure a Sensory Profile 2 results section No authority prescribes an SP2 report format, but the instrument's architecture dictates what a defensible results section must contain: which of the five forms and who answered it, the referral question, one plain sentence of the model so the labels mean something, classifications in honest frequency language tied to function, cross-rater differences read as context, and boundaries that keep a questionnaire from becoming a diagnosis. Each section below carries the pitfall that most often undermines it. Identification: form, respondent, ages. Open with the full instrument name and the specific form (Infant, Toddler, Child, Short, or School Companion), the respondent by role and relationship (which caregiver; which teacher, and how long they have known the child), the child's age against the form's age band, the administration date, platform (paper or Q-global), and language. The Spanish kit supplies Spanish Infant, Toddler, Child, and Short forms but an English School Companion, so language of administration is worth a clause. Pitfall: "Sensory Profile 2 completed" with no form and no respondent. The five forms have different items, respondents, and norms; unnamed, the results are uninterpretable and unreproducible. Referral question and battery context. State why sensory assessment was undertaken (feeding and dressing struggles, self-regulation concerns, an autism evaluation's sensory criterion, classroom participation) and what the SP2 sits alongside: observation, history, and any diagnostic instruments. A questionnaire only answers a question the report has actually posed. Pitfall: An SP2 dropped into a battery with no stated purpose, leaving the reader to guess whether it was screening, diagnosis support, or intervention planning. The model, briefly, before the numbers. Spend one or two sentences on Dunn's framework: neurological threshold crossed with self-regulation strategy yields four patterns (Seeking, Avoiding, Sensitivity, Registration), and scores describe how often the child's responses occur compared with same-age peers. Without this, classification language reads as jargon or, worse, as verdicts. Pitfall: Bare quadrant labels presented as self-explanatory. Readers meet "Avoiding: Much More Than Others" and hear a disorder unless the report has told them it is a frequency pattern. Quadrant results in frequency language, tied to function. Report each relevant quadrant's classification in Dunn's own band language and immediately anchor it to observed daily life: mealtime refusal, distress at clothing tags, missed instructions, movement during seatwork. Exact score values from the protocol belong in the clinical record; the interpretation rides on the band and the function, and combinations (high Seeking with high Avoiding) describe mixed patterns, not types. Pitfall: Bands inflated into severity ("severe sensory issues") or cut values reproduced from the manual. The bands describe reported frequency relative to peers; the cut ranges are proprietary. Section scores connected to participation. Bring in sensory-system sections (auditory, visual, touch, movement, body position, oral) and behavioral sections only where they inform the referral, and connect each to participation: what feeding, dressing, attention, or play actually looks like. On the School Companion, use the school factors to organize classroom observations. Pitfall: A bare list of section classifications with no participation link. A reader learns the child is "More Than Others" on touch and still knows nothing about the school day. Cross-rater comparison as context, never error. When caregiver and teacher forms both exist, name both respondents and compare by setting: home routines versus a structured classroom. Convergence between sensory questionnaires and raters is only moderate, so discordance is expected and informative. Interpret it (structure, predictability, demands), never average it, and never frame one rater as wrong. Pitfall: Averaging the two forms or writing "mother over-reported." Each form measures behavior in its own setting; the gap is the finding. Interpretation boundaries and recommendations. Close with the boundaries stated outright: the SP2 describes sensory processing patterns and is not diagnostic; sensory processing disorder is not a DSM-5-TR diagnosis; in autism evaluations the findings support, never establish, criterion B.4. Recommendations should be environmental and routine modifications matched to the pattern, with no outcome promises for sensory-integration therapy, and progress framed through function rather than re-administered bands. Pitfall: "Results confirm sensory processing disorder," or a sensory diet promised as treatment. Both claims outrun the instrument and the evidence. ### Blank template (copy and adapt) SENSORY PROFILE 2 RESULTS SECTION SKELETON Child: [initials] Age: [ ] Date(s): [ ] Form: [Infant / Toddler / Child / Short / School Companion] Age-band check: [child's age within the form's band] Respondent: [role + relationship; for teachers, time known child] Administration: [paper / Q-global; language; any accommodations] Referral question: [ ] Battery context: [observation, history, diagnostic instruments this feeds] Model statement: [1-2 sentences: threshold crossed with self-regulation strategy -> Seeking / Avoiding / Sensitivity / Registration; bands = frequency vs same-age peers] Quadrant results (band language, no cut values): [Quadrant]: [band, e.g. "Much More Than Others"] -> functional illustration: [mealtime, dressing, classroom, play] Section scores (only referral-relevant ones): [Sensory system / behavioral section]: [band] -> participation impact: [feeding, dressing, attention, transitions] Cross-rater comparison (if two forms): [respondent A setting vs respondent B setting; discordance interpreted by context, never averaged, no rater blamed] Interpretation boundaries: [patterns, not a diagnosis; SPD not a DSM-5-TR diagnosis; supports (not establishes) DSM-5 criterion B.4 when autism is the question] Recommendations: [environmental + routine modifications matched to the pattern; no sensory-integration outcome promises; progress monitored through function] Integration: [how findings sit with observation / history / other measures] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, forms, norms, or classification cut values. ## Sample Sensory Profile 2 write-up (fictional) Scenario: a 7-year-old in an autism diagnostic evaluation, with a caregiver Child form and a teacher School Companion that disagree, written up so the discordance becomes context rather than error. All details are fictional. Patient: L.M., 7 years · Setting: Autism diagnostic evaluation, outpatient developmental clinic · Clinician: R. Okafor, OTD, OTR/L · Note date: 08/18/2026 Identification and forms: Sensory Profile 2 administered as part of a multidisciplinary autism evaluation that also includes the ADOS-2, developmental history, and classroom observation. L.M., age 7 years 3 months, falls within the 3:0 to 14:11 band for both forms used: the Child Sensory Profile 2, completed by his mother on 08/12/2026, and the School Companion Sensory Profile 2, completed by his second-grade classroom teacher on 08/14/2026; the teacher has known him for six months. Both forms were administered in English through Q-global. Model and caregiver-form results: The SP2 describes how often a child's sensory responses occur compared with same-age peers, organized by Dunn's four patterns: Seeking, Avoiding, Sensitivity, and Registration. On his mother's Child form, Avoiding fell in the "Much More Than Others" range and Sensitivity in the "More Than Others" range, while Seeking and Registration fell in the "Just Like the Majority of Others" range. Touch and oral sensory sections drove the elevations, matching the referral concerns: refusal of most food textures with gagging, distress at clothing tags and seams that dictates the morning routine, and leaving the table at family meals. Exact score values accompany each classification in the clinical record; they are omitted here because the classification ranges are proprietary. Teacher-form results: On the School Companion, most patterns fell in the "Just Like the Majority of Others" range, with Seeking in the "More Than Others" range concentrated in unstructured times: he moves through the classroom during transitions, seeks movement at recess, and handles routine seatwork without incident in a highly structured classroom with predictable routines. School factors reflected good use of visual supports and advance warning of changes. Cross-rater interpretation: The caregiver and teacher forms disagree, and that disagreement is context, not error. Each form samples behavior in its own setting: home presents mealtimes, dressing, and unpredictable demands, while the classroom is structured, predictable, and low in the specific triggers his mother reports. Only moderate agreement between raters and settings is the empirical norm for sensory questionnaires, so the two forms were not averaged; together they suggest sensory reactivity that is most visible where tactile and oral demands are unavoidable, and better regulated where routines are predictable. Interpretation, boundaries, and recommendations: The SP2 describes sensory processing patterns relative to same-age peers; it is not diagnostic, and sensory processing disorder is not a DSM-5-TR diagnosis. Within this evaluation, the reported hyperreactivity to touch and texture provides supporting evidence relevant to DSM-5 autism criterion B.4, to be integrated with the ADOS-2, history, and observation; the questionnaire does not by itself establish that criterion or any diagnosis. Recommendations are matched to the pattern rather than to a label: a graded texture plan built with the family, clothing choices that remove daily flashpoints, preserved predictability around transitions at school, and movement opportunities before seated work. No claim is made that sensory-based therapy will resolve feeding difficulties; progress will be tracked through function, specifically foods accepted and mornings without distress, reviewed with the family in eight weeks at an appointment booked today. This sample is fictional and for educational purposes. It does not describe a real child or record; the details are invented to show write-up structure and are not clinical guidance. No score values, norms, or classification cut ranges are reproduced. ↑ Back to the template and downloads ### Why this sample works Both forms and both respondents are named, with dates, the age-band check, platform, and language, so every classification is interpretable against the right norms. The model gets one plain sentence before any result, so "Avoiding: Much More Than Others" reads as a frequency pattern rather than a verdict. Classifications stay in Dunn's band language tied to observed function, and the report says out loud why exact values and cut ranges are not reproduced. The caregiver-teacher gap is interpreted by setting as information, never averaged and never blamed on a rater. The boundaries are explicit: patterns not diagnosis, SPD not in DSM-5-TR, criterion B.4 supported rather than established, and recommendations carry no outcome promises. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the frames are legal and they are specific. In autism evaluations, DSM-5 added hyper- or hyporeactivity to sensory input as criterion B.4 in 2013, and DSM-5-TR retains it; SP2 findings serve as supporting evidence for that criterion, integrated with observation and history, never as the sole basis for it (CONVENTION anchored to the nosology). In schools, IDEA requires a variety of technically sound assessment tools and prohibits any single measure from serving as the sole criterion for eligibility (34 CFR 300.304), lists occupational therapy among related services (300.34), and expects IEP content to translate findings into present levels, goals, supports, and progress monitoring (300.320), so School Companion results appear as educational participation needs, not as a stand-alone sensory label (LAW). Billing runs through the OT evaluation codes 97165 to 97167; there is no SP2-specific code, and the questionnaire is reported inside the evaluation subject to payer rules (PAYER POLICY). The intervention side is where writers overreach: UnitedHealthcare's policy effective January 2026 classes sensory integration therapy broadly as unproven, Aetna considers it experimental for all indications, and Cigna's policy effective June 2026 conditionally covers fidelity-defined Ayres Sensory Integration for autism under stated criteria while excluding it otherwise, so a report should describe function and the proposed service precisely rather than converting an SP2 classification into a coverage conclusion (PAYER POLICY). The American Academy of Pediatrics' 2012 report advised against diagnosing sensory processing disorder and called the intervention evidence limited, while allowing a monitored trial inside a comprehensive plan (CONVENTION). Canada and Australia shift the frame from diagnosis to participation and function. Canadian education is provincial, with no national IDEA equivalent: in Ontario, Regulation 181/98 requires the IEP to contain educational expectations, the program and services to be provided, and progress-review methods, so SP2 findings enter as needs, services, and expectations for the named province rather than as a generic Canadian requirement, and Ontario classroom resources organize sensory programming around Dunn's four patterns (LAW and CONVENTION, provincial). Sensory-integration therapy is not a guaranteed insured service anywhere in Canada; coverage varies by province and plan. In Australia, the Disability Standards for Education require consultation and reasonable adjustments so students participate on the same basis, with obligations extending to covered early-childhood services from August 2026, and the SP2 neither is mandated nor itself determines adjustment entitlement: the report describes the functional barrier, the consultation, and the adjustment (LAW). For the NDIS, funding follows functional impact rather than diagnosis, and current guidance asks for recent professional evidence of day-to-day function, supports tried, and outcomes, expressly distinguishing what was observed, assessed, or reported; an SP2 classification supports that evidence but is not an access threshold, which is why findings are written into the functional capacity assessment as participation impact across self-care, learning, and social domains (PROGRAM POLICY). Instrument facts, psychometric honesty, and rights complete the picture. The 2014 SP2 remains the current edition as of August 2026, with no successor announced by Pearson; the STAR Institute's SP-3D work is a separate research program, not an SP2 revision. The strongest published critique, a 2023 COSMIN-based review of all five forms, credited internal consistency but rated structural validity and cross-cultural validity as weaknesses, found the manual thin on factor-analytic and Rasch detail, judged the Toddler form's evidence comparatively weak, and left responsiveness essentially unestablished, which is why change in classification bands should not be offered as proof of treatment response; the four-factor model's own confirmatory support comes from the standardization database, whose authors acknowledged no prior study had shown that model to be the best fit. Cross-informant and cross-instrument convergence is moderate at best, the empirical basis for treating rater discordance as expected. On rights: Pearson sells materials at qualification level B, items and norms are copyrighted, cut values circulating on flash-card and protocol sites are not citable content, Q-global is the sanctioned digital route, no authorized public web scorer exists, and the Spanish kit's School Companion is English. A report may always contain the child's classifications and your interpretation. Sensory Profile is a registered trademark of NCS Pearson, Inc. BastionGPT is not affiliated with, or endorsed by, Pearson. This page reproduces no test items, stimuli, norms, or classification cut values. ↑ Back to the template and downloads ## Common Sensory Profile 2 write-up errors reviewers flag The numbers behind these errors are specific. The five forms carry different item counts (25 to 86) and norms; a 2023 COSMIN review rated the SP2's structural and cross-cultural validity as weaknesses; convergence between the Child form and the SPM-2 runs only about 0.38 to 0.77 across subscales ; and an Australian chart audit found the Short form discriminates autism from ADHD well (AUC 0.82) but ADHD from no diagnosis poorly (AUC 0.59). The BastionGPT Clinical Advisory Board sees the same errors most often in SP2 documentation reviews: The form and respondent unnamed. "The Sensory Profile 2 was administered" with no statement of which of the five forms and who completed it. Different items, different respondents, different norms: unnamed, the classifications cannot be interpreted or reproduced, and an 11-to-14-year-old's report should also say why a caregiver SP2 form was chosen over the self-report Adolescent/Adult Sensory Profile. Frequency bands inflated into severity. "Much More Than Others" rewritten as "severe sensory dysfunction." The bands describe how often a respondent reports behaviors relative to same-age peers, nothing more; severity, distress, and impairment come from the functional evidence. The companion error is reproducing the classification cut values, which are proprietary and belong nowhere in a report or a public page. Quadrant elevations treated as diagnostic. "Results confirm sensory processing disorder." The SP2 describes patterns; SPD is not a DSM-5-TR or ICD-11 diagnosis (DC:0-5 lists it for infancy and early childhood, a nuance worth knowing before writing categorically), the AAP has advised against diagnosing it, and even the leading SPD research program has stepped back from its own 2007 diagnostic nosology. Describe the pattern and its functional impact. Caregiver-teacher discordance averaged or blamed. Two forms combined into one summary, or "mother appears to over-report." Each form measures behavior in its own setting, and only moderate cross-rater and cross-instrument agreement is the published norm. Name both respondents, interpret the gap by context (structure, predictability, demands), and let the discordance inform recommendations. Criterion B.4 overreached. An SP2 elevation offered as establishing DSM-5 criterion B.4, or the Short form standing in for a diagnostic workup. The questionnaire provides supporting evidence to integrate with observation, history, and diagnostic instruments; the Short form is a screener whose discrimination is good for autism versus ADHD (AUC 0.82) and poor for ADHD versus no diagnosis (AUC 0.59), numbers that argue for the full form and a proper battery when diagnosis is the question. Outcome promises the evidence will not carry. A sensory diet promised as treatment, or re-administered bands offered as proof of progress. Payer positions on sensory-integration therapy range from broad exclusion to narrow, fidelity-defined conditional coverage for autism, the AAP called the evidence limited, and the COSMIN review left the SP2's responsiveness unestablished: recommend environmental and routine modifications matched to the pattern, and monitor progress through function. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Give it the facts (which form and respondent, ages and dates, quadrant and section classifications, functional observations, the referral question, the rest of the battery) and it drafts the results section: the model explained in a sentence, bands in frequency language tied to participation, discordance interpreted by setting, boundaries stated, ready for your review. Cross-check a finished report for the gaps reviewers flag: an unnamed form or respondent, a band inflated into severity, an SPD diagnosis claim, averaged raters, an overreached criterion B.4, or an outcome promise. Draft the companion paragraphs: the IEP version framed around educational participation, the NDIS version framed around functional impact with observed, assessed, and reported evidence distinguished, or the plain-language summary for the family. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Which Sensory Profile 2 form do I name in the report? The one actually used, every time: Infant (birth to 6 months, caregiver, 25 items), Toddler (7 to 35 months, caregiver, 54 items), Child (3:0 to 14:11, caregiver, 86 items), Short (3:0 to 14:11, a 34-item screening subset), or School Companion (3:0 to 14:11, teacher, 44 items). The forms differ in items, respondents, and norms, so "the SP2 was administered" is not a reportable statement. Name the respondent by role and relationship (which caregiver; which teacher and how long they have known the child), confirm the child's age sits inside the form's band, and note the platform and language, remembering that the Spanish kit provides Spanish Infant, Toddler, Child, and Short forms but an English School Companion. For ages 11 to 14, say why a caregiver form was chosen over the separate self-report Adolescent/Adult Sensory Profile, which begins at 11. What do the four Sensory Profile 2 quadrants mean? They are the four intersections of Dunn's two axes: neurological threshold (how much input the nervous system needs before responding) and self-regulation strategy (passive or active). Seeking is high threshold with an active strategy, the degree to which a child obtains sensory input; Avoiding is low threshold with an active strategy, the degree to which a child is bothered by input and works to escape it; Sensitivity is low threshold with a passive strategy, the degree to which a child detects input; Registration is high threshold with a passive strategy, the degree to which a child misses input. They are dimensional patterns, not types: every child has a score in all four, combinations (high Seeking with high Avoiding) describe mixed self-regulation patterns, and an elevation is a description of reported frequency, not a disorder. The write-up should say this in a sentence before any classification appears. What does "Much More Than Others" mean on the SP2? That the respondent reported those behaviors much more frequently than the standardization sample's respondents reported them for same-age children. The SP2's classification bands run from "Much Less Than Others" through "Just Like the Majority of Others" to "Much More Than Others," defined by standard-deviation ranges around the normative mean; the exact cut values are proprietary and should not be reproduced in reports or on public pages. Two disciplines follow. First, the language is frequency, not severity: how often, compared with peers, not how bad. Severity and impairment come from the functional evidence the report pairs with each band. Second, the comparison is to the norms, filtered through one respondent's observation in one setting, which is why the band is always attributed ("his mother's ratings placed Avoiding in the Much More Than Others range") rather than stated as a property of the child. Can I diagnose sensory processing disorder from the SP2? No, on two independent grounds. The instrument ground: the SP2 describes sensory processing patterns relative to peers and its own framework treats elevations as pattern descriptions, not disorders. The nosology ground: sensory processing disorder is not a diagnosis in DSM-5-TR or ICD-11, where sensory features live inside the autism criteria; DC:0-5, the infant and early-childhood classification, does list it, which is the one nuance that keeps "SPD is not a recognized diagnosis" from being fully accurate. The American Academy of Pediatrics' 2012 report advised against diagnosing SPD (a statement that technically lapsed under the AAP's five-year rule yet is still cited as authority by payers), and the STAR Institute has stepped back from its own 2007 SPD nosology while awaiting new data. The defensible phrasing: describe the pattern, its functional impact, and what it supports, and let diagnoses come from diagnostic systems. How do Sensory Profile 2 results support an autism diagnosis? As evidence for one criterion, integrated with everything else. DSM-5 added hyper- or hyporeactivity to sensory input or unusual sensory interests as autism criterion B.4, and a caregiver SP2, often paired with a teacher School Companion, is a standard way to document that reactivity across settings. The findings support the criterion alongside direct observation (the ADOS-2 ), developmental history (the ADI-R ), and rating scales; they never establish the criterion or the diagnosis alone. The Short form deserves extra caution: in an Australian chart audit it discriminated autism from ADHD well (AUC 0.82) and autism from no diagnosis nearly as well (0.81), but ADHD from no diagnosis poorly (0.59), screener numbers, not diagnostic ones. Write the SP2 into the sensory-reactivity paragraph of the autism evaluation report as supporting evidence with the setting-by-setting pattern spelled out. How do I write up caregiver-teacher disagreement on the SP2? As context, never as error. The caregiver form measures behavior at home as that caregiver sees it; the School Companion measures behavior at school as that teacher sees it; and the published agreement between sensory-questionnaire raters and instruments is only moderate (Child SP2 to SPM-2 subscale correlations of roughly 0.38 to 0.77), so discordance is the expected finding, not a data problem. The write-up names both respondents, states each setting's results in band language, and then interprets the gap by what differs between the settings: structure, predictability, tactile and oral demands, escape options. Never average the two forms into a composite impression, and never write that a rater over- or under-reported; if response style genuinely concerns you, describe the specific inconsistency and gather more observation. Discordance handled this way usually sharpens recommendations, because it identifies the conditions under which the child regulates well. When should I use the SP2 versus the SPM-2? By the referral question, because the instruments answer different ones. The SP2 operationalizes Dunn's threshold-by-strategy quadrant model, uses strengths-neutral frequency language, and pairs a caregiver form with a teacher School Companion for a cross-setting pattern; it suits questions about a child's overall sensory pattern and home-school comparison. The SPM-2 (Sensory Processing Measure, Second Edition) is grounded in Ayres Sensory Integration, scores by sensory system with praxis and social participation, and offers environment-specific school forms (classroom, art, music, PE, recess, cafeteria, bus); it suits questions about where in the school day breakdowns occur and about praxis. Their convergence is only moderate (subscale correlations roughly 0.38 to 0.77), so they are not interchangeable and a report should not treat one as confirming the other. Questionnaires of either family remain respondent report, complemented by observation and, where praxis is the question, performance-based measures. Do insurers pay for the Sensory Profile 2? The questionnaire, usually, inside the evaluation; the therapy it often leads to, unevenly. There is no SP2-specific billing code: administration and interpretation are reported within the OT evaluation codes 97165 to 97167, subject to the payer's medical-necessity and documentation rules. The intervention side is where reports go wrong, because current policies genuinely contradict each other: UnitedHealthcare's 2026 policy classes sensory integration therapy broadly as unproven, Aetna considers it experimental for all indications, and Cigna's 2026 policy conditionally covers fidelity-defined Ayres Sensory Integration for autism under stated criteria while excluding it otherwise. So it is inaccurate to write either "insurance does not cover sensory integration" or "it is covered for autism." The defensible report names the actual proposed service, its functional goals, baseline, and expected change, and lets coverage be determined by the plan; in schools and the NDIS, the frame is educational participation and functional impact rather than payer coverage at all. Can BastionGPT draft my Sensory Profile 2 report? Yes. Give it the facts (which form and who completed it, ages and dates, quadrant and section classifications in band language, functional observations from home and school, the referral question, and the rest of the battery) and it drafts the results section: the model explained in a sentence, bands tied to participation, cross-rater differences interpreted by setting, boundaries stated, and recommendations matched to the pattern, ready for your review. It can also cross-check a finished report for an unnamed form or respondent, severity language, an SPD diagnosis claim, averaged raters, an overreached criterion B.4, or an outcome promise, and it can draft the IEP, NDIS, and family-summary variants of the same findings. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: Pearson, Sensory Profile 2 product page (five forms, age bands, qualification level B, Q-global, Spanish kit contents); ILOTA 2014 Pearson conference overview, item counts and revision changes . Reviews and psychometrics: Licciardi L and Brown T, 2023, Scandinavian Journal of Occupational Therapy, COSMIN-based critical review of the SP2 ; Dean E, Dunn W, and colleagues, 2016, American Journal of Occupational Therapy, confirmatory factor analysis of the quadrant model ; Jones L, Yu ML, and Brown T, 2024, Australian Occupational Therapy Journal, Child SP2 and SPM-2 convergence ; Marozza I, Hay K, and Frakking T, 2025, Short SP2 diagnostic discrimination chart audit . Nosology and professional guidance: American Psychiatric Association DSM-5 autism criteria including B.4, criteria text as republished by Autism Speaks ; Zimmer M and Desch L, American Academy of Pediatrics, 2012, Pediatrics, sensory integration therapies policy report . Payer and US legal frame: UnitedHealthcare, sensory integration therapy policy effective January 2026 ; Cigna, coverage policy effective June 2026 (fidelity-defined ASI conditionally covered for autism); Aetna, Clinical Policy Bulletin 0256 ; AOTA, 2026 OT evaluation code list ; eCFR, 34 CFR 300.304 . Canada and Australia: Ontario, Regulation 181/98 (IEP contents); CMEC, education in Canada overview (provincial governance); ConnectABILITY Ontario, four patterns of sensory processing ; Australian Department of Education, Disability Standards for Education update ; NDIS, supporting-evidence guidance (observed, assessed, or reported). ## Related templates ADOS-2 Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up SRS-2 Report Write-Up Autism Spectrum Evaluation Report GARS-3 Report Write-Up Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/gars-3-report All note templates # GARS-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The GARS-3 (Gilliam Autism Rating Scale, Third Edition) is an informant rating scale for ages 3 through 22: a parent, teacher, or clinician rates items across six subscales, producing an Autism Index and probability language. Schools and autism teams use it as supporting evidence, never as a diagnosis. The write-up must name the respondent and index used. This page covers how to write up GARS-3 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it School psychologists, educational diagnosticians, psychologists, and clinicians scoring parent or teacher ratings; PRO-ED test level B (WPS lists C) Audience IEP and autism evaluation teams, diagnosing clinicians, ABA reviewers and payers, referring physicians, families Typical length 250 to 600 words for the results section · the respondent completes the rating in about 5 to 10 minutes; scoring and integration add clinician time Format family Norm-referenced informant rating scale (six DSM-5-aligned subscales; Autism Index 4 or 6, mean 100, SD 15) When it's used School autism evaluations and re-evaluations, clinic batteries as standardized informant evidence, ages 3 through 22, alongside observation and interview measures Standards context Published by PRO-ED (©2014; no fourth edition as of August 2026); scored by manual or PRO-ED's online system; no items, anchors, or cut values reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the GARS-3? The GARS-3 (Gilliam Autism Rating Scale, Third Edition; James E. Gilliam; PRO-ED, ©2014, with some distributor listings dating the release 2013) is a norm-referenced informant rating scale for ages 3 through 22. A parent, teacher, or clinician who knows the examinee rates 56 items (current publisher pages say 56; some peer-reviewed papers report 58, a discrepancy worth checking against the licensed form) on a four-point frequency scale, across six DSM-5-aligned subscales: Restrictive/Repetitive Behaviors, Social Interaction, Social Communication, Emotional Responses, Cognitive Style, and Maladaptive Speech. Subscale raw scores convert to scaled scores (mean 10, SD 3) and combine into one of two composites: Autism Index 4, from the first four subscales, for individuals whose speech is absent or severely limited, or Autism Index 6, from all six, for individuals with sufficient functional speech. The index (mean 100, SD 15) maps to the manual's ordered probability-of-autism classifications and to a severity framing the publisher aligns with DSM-5 support levels. PRO-ED is the publisher; WPS, Pearson, and PAR distribute it (Pearson's own documentation says it distributes but does not publish), and a Spanish version, the GARS-3S, has its own forms and manual. The write-up's center of gravity is what this instrument is: one informant's standardized report. The rating quantifies how closely that respondent's account resembles the behavioral pattern in the instrument's norms, which were drawn entirely from individuals already diagnosed with autism, a fact a 2017 published review highlighted and almost no explainer page mentions. So the score is resemblance to an autism reference sample, filtered through one rater in one setting: evidence to integrate, never a verdict. The probability language is a classification label, not a literal percentage, and neither a diagnosis nor IDEA eligibility nor an insurance authorization follows from it alone. The instruments the GARS-3 must converge with have their own pages: structured observation on the ADOS-2 page , the caregiver developmental interview on the ADI-R page , and the full synthesis on the autism evaluation report page . ## Who uses the GARS-3 and when School teams are the heaviest users: the scale is brief (5 to 10 minutes for the respondent), inexpensive, and built for teacher and parent report, so it appears constantly in IDEA autism evaluations and re-evaluations, where federal law requires a variety of tools and prohibits any single measure from deciding eligibility. Clinics use it as the standardized informant component of diagnostic batteries, alongside clinician ratings covered on the CARS-2 page , trait measures covered on the SRS-2 page , and the observation and interview instruments payers actually name. Payer reviewers meet it inside psychological-testing claims (the 96130 to 96139 code family) and diagnostic packets for ABA authorization, where its role is strictly supporting: TRICARE's autism program lists the GARS among accepted instruments while stating a parent questionnaire alone is insufficient, and North Carolina Medicaid's policy amended August 2026 names other procedures entirely for a qualifying diagnosis. The report's readers are the IEP team applying the state's educational definition, the diagnosing clinician weighing convergence, the reviewer checking whether the packet contains more than a rating scale, and a family who was told a number and deserves to know what it does and does not mean. Toddler-age screening that precedes all of this lives on the M-CHAT-R/F page . ## How to structure a GARS-3 results section No authority prescribes a GARS-3 report format, but the instrument's nature as an informant rating dictates what a defensible results section must contain: who rated, under what conditions of contact, which index and why, the subscale pattern before the composite, the classification with its meaning attached, explicit convergence and divergence, and a bright line between diagnosis and eligibility. Each section below carries the pitfall that most often undermines it. Identification: instrument, respondent, and index. Open with the full title and edition, the date completed, the language or version (English GARS-3 or Spanish GARS-3S), and the respondent by name or role, relationship, setting, and approximate duration and frequency of contact. Then name the composite actually generated, Autism Index 4 or Autism Index 6, and give the communication-status reason it was the applicable one. Pitfall: "Parent completed the GARS-3." Which parent, how much contact, which index? An informant score without its informant context is not interpretable, and "the Autism Index" without a 4 or 6 is ambiguous. Rating conditions and informational validity. State whether the respondent had sufficient opportunity to observe the relevant behaviors, and record material limitations: a recent placement, contact limited to one highly structured setting, primarily virtual contact, interpreter use, or substantial unanswered items. This documents the conditions under which the informant formed the judgments, which is different from pronouncing the scores psychometrically valid. Pitfall: A bare "results are considered valid." The honest version describes the observation window and lets it bound the interpretation. Subscale pattern before the composite. Describe which of the six domains carried the elevation, which were comparatively unremarkable, and whether the pattern was broad or concentrated, in original clinical language tied to concrete observations. The pattern is the clinically useful content; the composite only summarizes it. Pitfall: A results section that reads as an item inventory. Paraphrasing proprietary items drifts into reproduction; describe domains and behavior themes instead. The index and classification, with meaning attached. Report the applicable index as a standard score (mean 100, SD 15) with the licensed classification from the scored protocol, then say immediately what it is: one respondent's standardized rating pattern resembling, or not resembling, the instrument's autism reference sample. The probability language is a category label, not a literal percentage chance. Pitfall: "The GARS-3 shows a 90 percent probability of autism." No such number exists; the classification is a band label, and presenting it as an individualized probability is the field's most repeated GARS-3 error. Convergence and divergence, written out. State where the rating agrees and disagrees with direct observation, developmental history, interview, records, and any other informant, and what the disagreement means: context-dependence, setting demands, compensatory behavior, or a better competing explanation. A discrepancy is a clinical finding to investigate, not an error to average away. Pitfall: The GARS-3, observation, and interview each in their own silo with no sentence connecting them, or the higher score quietly chosen as the truth. Diagnosis versus eligibility, kept separate. Write distinct conclusions: whether the full evidence supports DSM-5 autism (a clinical judgment integrating criteria, history, and observation) and whether the student meets the state's educational definition, which turns on adverse educational effect and need for specially designed instruction. The index satisfies neither by itself, and educational eligibility does not require a medical diagnosis in states like Texas, Minnesota, and Oregon. Pitfall: "The Autism Index supports eligibility." Eligibility is a team determination under the state rule; a score cannot make it, and a medical diagnosis is a different conclusion again. Recommendations and boundaries. Close with recommendations that are individualized rather than imported: publisher-generated instructional objectives are suggestions requiring clinical and educational judgment, not IEP content to paste. State the boundaries: the scale contributes evidence, re-administration has no established reliable-change standard, and any change in scores describes changed informant report, to be read alongside functional data. Pitfall: "Severity improved by one category since last year." No reliable-change index or minimal important difference exists for the GARS-3; that sentence outruns the evidence. ### Blank template (copy and adapt) GARS-3 RESULTS SECTION SKELETON Student/client: [initials] Age: [ ] Date completed: [ ] Version: [GARS-3 English / GARS-3S Spanish] Respondent: [name or role; relationship; setting; duration + frequency of contact] Index generated: [Autism Index 4 / Autism Index 6] because [communication status per the manual's rule] Rating conditions: [opportunity to observe; placement recency; settings known; interpreter; items unanswered] Subscale pattern (before the composite): Most prominent: [domains + concrete behavior themes] Comparatively unremarkable: [domains] Pattern: [broad / concentrated] Index result: [standard score, mean 100 SD 15] with the manual's classification: [licensed category from the scored protocol] Meaning: [one respondent's standardized rating pattern; a category label, not a percentage; not a diagnosis] Convergence / divergence: [agreement and disagreement with observation, history, interview, records, other informants; what the discrepancy means] Diagnostic conclusion (clinical): [supported / not supported / pending, from the full evidence] Educational eligibility: [team determination under the state definition; adverse effect + need documented separately] Recommendations: [individualized; publisher objectives adapted, never pasted; further data if divergent] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, rating anchors, norms, or cut values. ## Sample GARS-3 write-up (fictional) Scenario: a school autism evaluation with a teacher-completed GARS-3 that converges with observation and history, written so the classification supports the team's work without becoming a diagnosis. All details are fictional. Patient: D.K., 8 years · Setting: School multidisciplinary autism evaluation (initial) · Clinician: M. Reyes, PhD, NCSP, school psychologist · Note date: 08/20/2026 Identification and respondent context: Gilliam Autism Rating Scale, Third Edition (GARS-3), English version, completed 08/17/2026 by D.K.'s third-grade general-education teacher, who has taught him daily since the start of the school year and across structured instruction, small groups, and unstructured times. Because D.K. routinely uses functional phrase and sentence speech, all six subscales were applicable and Autism Index 6 was generated, per the manual's communication-status rule. The rating is one standardized informant measure within an evaluation that also includes structured observation, developmental history with his parents, a speech-language assessment, and record review. Rating conditions: The teacher reported ample opportunity to observe the rated behaviors across the school day and completed all items; no interpreter was needed. Her contact is limited to the school setting, so her ratings describe D.K. at school, and the parent history below supplies the home perspective. Subscale pattern and index: The teacher's ratings were most elevated on the Social Interaction, Social Communication, and Restrictive/Repetitive Behaviors subscales, with comparatively fewer concerns on Emotional Responses, Cognitive Style, and Maladaptive Speech: a concentrated pattern centered on reciprocal social behavior and sameness. The resulting Autism Index 6 fell within the manual's highest probability-of-autism classification. That classification means this teacher reported a pattern strongly resembling the behaviors represented in the GARS-3 autism reference sample; it is a category label from the scored protocol, not a percentage chance, and it does not by itself establish a diagnosis or eligibility. Exact values remain in the protocol; the manual's conversion tables are proprietary. Convergence and divergence: The rating converged with direct observation across two settings, which documented limited spontaneous peer engagement, reduced conversational reciprocity, distress at an unannounced schedule change, and repetitive organization of materials, and with the developmental history, in which his parents described longstanding sameness needs and reduced reciprocal interaction in unfamiliar settings, while reporting somewhat fewer difficulties within familiar home routines. That home-school difference reads as contextual (predictable routines and one-to-one attention at home) rather than as disagreement about the pattern itself. Conclusions and recommendations: Diagnostically, the converging evidence across informant rating, observation, history, and speech-language findings supports autism spectrum disorder as defined by DSM-5-TR; that conclusion rests on the full evidence, not on the GARS-3 score. Educationally, eligibility under the state's autism category is the team's determination and turns on adverse educational effect and need for specially designed instruction, documented in the eligibility section of this report. Recommendations were individualized from the evaluation as a whole; publisher-generated instructional objectives were reviewed as suggestions and adapted where they matched D.K.'s actual profile. Re-administration of the GARS-3 alone will not be used as a progress measure, because no reliable-change standard exists for it; progress will be monitored through IEP goal data and observation. This sample is fictional and for educational purposes. It does not describe a real student or record; the details are invented to show write-up structure and are not clinical guidance. No items, rating anchors, norms, or cut values are reproduced. ↑ Back to the template and downloads ### Why this sample works The respondent, relationship, setting, and contact time are stated before any score, so the reader knows exactly whose standardized report this is. Autism Index 6 is named with its communication-status rationale, removing the ambiguity that an unlabeled "Autism Index" always carries. The subscale pattern comes before the composite, in clinical language rather than item paraphrase, so the profile does the clinical work. The classification is quoted from the protocol and immediately defined as resemblance to the autism reference sample, never as a percentage or a verdict. Convergence and divergence are written out, and the diagnostic and eligibility conclusions are separated, each resting on its own full basis. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the governing frames are specific and they all say the same thing about single measures. IDEA requires a variety of assessment tools, expressly prohibits any single measure or assessment as the sole criterion for eligibility or programming, and requires technically sound instruments used by trained personnel per producer instructions (34 CFR 300.304); the educational autism definition (300.8(c)(1)) turns on characteristics that adversely affect educational performance, which no rating scale can establish alone; and public agencies must consider independent educational evaluations that meet agency criteria (300.502), where consider does not mean adopt (LAW). States make the eligibility-versus-diagnosis line explicit: Texas rule 19 TAC 89.1040(c)(1) states that eligibility "cannot require that the student meets the requirements for a medical/psychological diagnosis of autism," Minnesota requires multiple methods, settings, observations, and developmental history and makes a medical diagnosis neither required nor sufficient, and Oregon's multidomain rule (amended December 2023) treats rating-scale information as one component (LAW, state by state). On the payer side, no reviewed policy accepts a GARS-3-only evaluation: TRICARE's autism program lists the GARS among validated instruments while requiring a diagnostic evaluation beyond a parent questionnaire (the program is authorized through December 2028), North Carolina Medicaid's policy amended August 1, 2026 names BOSA, TAP, ADOS-2, or CARS-2 for a nonprovisional qualifying diagnosis, with the GARS-3 absent, Aetna's autism bulletin lists the Gilliam among informant instruments inside a comprehensive evaluation, and Cigna's autism policy (effective May 2026, next review December 2026) recognizes comprehensive assessment rather than any index threshold; testing itself bills through the 96130 to 96139 family, and EPSDT obliges states to cover medically necessary child services without converting any score into an authorization (PAYER POLICY). Canada and Australia put the same weight on multi-source assessment. The Canadian Paediatric Society's 2019 diagnostic guidance holds that no single tool makes the diagnosis and that standardized measures sit under clinical judgment; several jurisdictions add teeth, most concretely British Columbia, which requires both the ADI-R and the ADOS for publicly funded autism assessment, a pairing a GARS-type scale cannot substitute for (CONVENTION and provincial PAYER POLICY). Canada's earlier national best-practice guidelines went further and recommended against the original GARS on sensitivity grounds, part of why a Canadian report should present the GARS-3 strictly as supporting informant evidence with its respondent context intact. Australia's National Guideline for the Assessment and Diagnosis of Autism (2023 update, NHMRC-approved December 2023) prescribes comprehensive, needs-based assessment drawing on direct observation, information from people who know the person, developmental and medical history, and standardized tools where appropriate; a rating scale informs but never replaces that process, and report language that says so travels well across all three countries (CONVENTION). In every jurisdiction the portable artifact is the same: a write-up that preserves who rated, under what conditions, which index, the pattern, and how the evidence converged lets any downstream team, school, funding program, or diagnostician, reuse the evidence under its own rules. Instrument facts, psychometric honesty, and rights complete a defensible page. Facts: the GARS-3 remains the current edition as of August 2026 (no GARS-4 exists or is announced); the manual is ©2014 though some distributor metadata says 2013; current publisher pages specify 56 items while several peer-reviewed papers say 58, a discrepancy to resolve against the licensed form rather than a web page; and the same product is qualification level B at PRO-ED and Pearson but level C at WPS, vendor purchasing classifications rather than properties of the test. Psychometrics: the manual reports sensitivity .97, specificity .97, and AUC .93 from a standardization sample of 1,859 collected 2010 to 2011, drawn entirely from individuals already diagnosed with autism; the strongest independent test, a 2022 study of 186 clinically referred children published in the Journal of Autism and Developmental Disorders, found the recommended cutoff correctly classified about 47 percent with false-positive rates of roughly 83 to 88 percent, a 2024 short-form follow-up could not optimize sensitivity and specificity at any cutoff, and the earlier GARS editions drew two decades of similar critique, so the manual's figures are manual findings, not settled field performance, and elevations in complex referral populations demand differential thinking (anxiety, ADHD, language disorder, trauma). No reliable-change standard exists, so re-administration is not a validated outcome measure. Rights: record forms are copyrighted consumables, reproduction and EHR embedding are separately licensed uses under PRO-ED's permissions process, the publisher's online system is the sanctioned scorer, and no authorized free public scorer exists; a report may always contain the examinee's derived scores and your interpretation. The GARS-3 and Gilliam Autism Rating Scale are product names of PRO-ED, Inc., and the instrument is copyrighted material of its publisher. BastionGPT is not affiliated with, or endorsed by, PRO-ED. This page reproduces no test items, rating anchors, norms, or cut values. ↑ Back to the template and downloads ## Common GARS-3 write-up errors reviewers flag The numbers behind these errors are specific. The manual reports sensitivity and specificity of .97 from a 1,859-person standardization sample; an independent 2022 study of 186 clinically referred children found the recommended cutoff correctly classified about 47 percent, with false-positive rates near 83 to 88 percent; and North Carolina Medicaid's August 2026 policy names four qualifying diagnostic procedures, none of them the GARS-3. The BastionGPT Clinical Advisory Board sees the same errors most often in GARS-3 documentation reviews: Probability language read as a percentage. "The GARS-3 shows a 90 percent probability of autism." The manual's probability descriptors are ordered classification labels tied to index bands, not individualized probability estimates. Write the licensed category, then define it: this respondent's ratings resemble the instrument's autism reference pattern strongly enough to fall in that band, nothing more. The index unnamed. "Autism Index: elevated" without a 4 or a 6. The two composites are built from different subscales under the manual's communication-status rule, so an unlabeled index is ambiguous, and switching indexes to obtain a preferred number is indefensible. Name the index, give the communication-status reason, and use the same index for every rater of the same examinee. The respondent stripped from the result. A score reported with no relationship, setting, or contact time. An informant rating is one person's standardized report; the rater's opportunity to observe is part of the result. Search-ranking templates leave a blank for the score and none for the respondent, which is exactly backwards. Manual accuracy repeated as settled fact. ".97 sensitivity and specificity" quoted from marketing pages as field performance. Those are standardization findings against norms drawn only from already-diagnosed individuals; the independent clinic literature found roughly 47 percent correct classification and false-positive rates above 80 percent in referred samples. Report publisher figures as manual findings and let the independent evidence temper the interpretation. A score doing diagnostic or eligibility work. "Meets criteria for autism eligibility per GARS-3." Federal law prohibits any single measure as the sole criterion, educational eligibility turns on adverse effect and need rather than any score, and no reviewed payer accepts a GARS-3-only packet: TRICARE requires a diagnostic evaluation beyond the questionnaire and North Carolina Medicaid names other procedures entirely. The index is evidence for the team and the diagnostician, never the decision. Re-administration sold as progress. "Autism severity improved one category since last year." No reliable-change index, minimal clinically important difference, or treatment-response benchmark exists for the GARS-3, so a score difference documents changed informant report. Describe the change alongside observation, goal data, and functional evidence, and let those carry the progress claim. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychological and psychoeducational evaluation reports. Give it the facts (respondent and contact context, index used and why, subscale pattern, the licensed classification, observation and history findings, the applicable eligibility framework) and it drafts the results section: informant context first, pattern before composite, the classification defined rather than inflated, convergence written out, conclusions separated, ready for your review. Cross-check a finished report for the gaps reviewers flag: an unnamed index, a missing respondent context, probability language presented as a percentage, manual figures quoted as field performance, a score doing eligibility work, or a progress claim from re-administration. Draft the companion paragraphs: the parent-friendly explanation of what the classification does and does not mean, the divergence reconciliation when the observation was unremarkable, or the IEE integration paragraph that considers outside results under the district's criteria. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What scores does the GARS-3 produce? Six subscale scaled scores (mean 10, SD 3), one per domain: Restrictive/Repetitive Behaviors, Social Interaction, Social Communication, Emotional Responses, Cognitive Style, and Maladaptive Speech. The applicable subscales then combine into a composite Autism Index on a mean-100, SD-15 metric: Autism Index 4 from the first four subscales when speech is absent or severely limited, or Autism Index 6 from all six when the examinee has sufficient functional speech. The manual maps the index to ordered probability-of-autism classifications and to a severity framing the publisher aligns with DSM-5 support levels; the exact bands and conversion tables are proprietary, so reports quote the licensed category from the scored protocol rather than reproducing tables. Two cautions travel with the numbers: the classification is a category label, not a percentage, and the severity framing is the instrument's interpretive label, not a clinical support-level determination, which is made clinically and can differ across DSM-5's two domains. What does the GARS-3 probability language actually mean? It means the respondent's ratings resemble, to a stated degree, the behavioral pattern in the instrument's norms, and nothing more precise than that. The norms were drawn entirely from individuals already diagnosed with autism, so the index expresses closeness to an autism reference sample as reported by one rater in one setting. A classification in the manual's highest band is not a statement that the person has a specific percentage chance of autism, and a lower band does not rule autism out (the original GARS line was criticized for missing nearly half of clinically diagnosed cases). A serviceable parent explanation: this is the category assigned to the pattern this respondent reported; it tells us how closely those ratings resemble the profile in the instrument's autism sample; it is not a percentage and it does not make the diagnosis by itself. The report then shows what converged, what diverged, and what additional evidence carried the conclusion. Autism Index 4 versus Autism Index 6: which do I report? The one the manual's communication-status rule selects, named explicitly with its rationale. Autism Index 4 uses the four subscales that do not depend on speech and applies when the examinee is nonverbal or has severely limited speech; Autism Index 6 adds Cognitive Style and Maladaptive Speech and applies when functional speech is sufficient for all six domains to be rated meaningfully. They are not interchangeable options to report side by side, and they are not parent and teacher versions: when multiple informants rate the same examinee, every rating uses the index matching that examinee's communication status. The write-up error to avoid is the unlabeled "Autism Index," which leaves the reader unable to know which composite, and which subscales, produced the number; the indefensible error is switching indexes because one produces a preferred result. Is the GARS-3 enough to diagnose autism? No, by the publisher's own framing and by the independent evidence. PRO-ED describes the scale as assisting in identifying autism and estimating severity, and the manual builds in a diagnostic validation step mapping the presentation to DSM-5 criteria, both signals that the rating supports rather than makes the determination. The independent numbers close the question: while the manual reports .97 sensitivity and specificity, a 2022 study of 186 clinically referred children found the recommended cutoff correctly classified about 47 percent, with false-positive rates around 83 to 88 percent, because the behaviors it rates also occur in language disorder, intellectual disability, ADHD, anxiety, and trauma-related presentations. A diagnosis integrates direct observation (the ADOS-2 ), developmental history and interview (the ADI-R ), clinical judgment against DSM-5-TR criteria, and differential diagnosis; the GARS-3 contributes one informant's standardized evidence to that synthesis. Does an elevated GARS-3 make a student eligible under IDEA? No. Federal regulations require a variety of assessment tools and prohibit any single measure or assessment from serving as the sole criterion for determining eligibility, and the educational autism definition turns on characteristics that adversely affect educational performance, something no rating scale can establish. States say the rest plainly: Texas rule 19 TAC 89.1040(c)(1) states that eligibility "cannot require that the student meets the requirements for a medical/psychological diagnosis of autism," Minnesota requires multiple methods across settings plus systematic observation and developmental history, with a medical diagnosis neither required nor sufficient, and Oregon's rule makes rating-scale information one component of a multidomain evaluation. So the GARS-3 can help satisfy part of one method; the eligibility determination belongs to the team applying the state definition, documenting educational effect and the need for specially designed instruction. The same logic runs in reverse: a modest index does not defeat eligibility that the fuller evidence supports. Will insurers accept the GARS-3 for an autism diagnosis or ABA authorization? As a component, sometimes; alone, no policy reviewed for this page does. TRICARE's autism program lists the GARS among validated instruments a diagnosing provider may use, while stating that a parent questionnaire alone is insufficient and requiring a diagnostic evaluation from a qualified provider. North Carolina Medicaid's policy amended August 1, 2026 goes further for a nonprovisional diagnosis, naming BOSA, TAP, ADOS-2, or CARS-2 as qualifying diagnostic procedures, a list that does not include the GARS-3, and rejecting decisions based solely on screening tools or educational determinations. Aetna's autism bulletin lists the Gilliam among informant instruments within a comprehensive evaluation, and Cigna's policy recognizes comprehensive assessment rather than any index threshold. The practical rule for the write-up: present the GARS-3 as standardized informant evidence inside a comprehensive diagnostic evaluation, name the rest of the battery, and check the member's current policy rather than asserting the scale is insurance-accepted. How do I write up an elevated GARS-3 when the ADOS-2 was unremarkable? State the divergence directly, then reason through it instead of averaging it away. Neither result automatically invalidates the other: a brief structured observation cannot rule autism out, and an elevated informant scale cannot establish it. Work the plausible explanations in writing: behavior may be context-dependent (the structured session reduced the demands that trigger difficulties), the respondent may observe situations testing never reproduces, compensatory behavior or masking may operate in session, or the elevation may reflect difficulties better explained by anxiety, ADHD, language disorder, or trauma, which the false-positive literature makes a live possibility in referred populations. Then say what decides it: developmental history, additional settings and informants, language and cognitive findings, adaptive functioning, and differential diagnosis. The conclusion may be that autism is supported despite the observation, unsupported despite the rating, or unresolved pending more data; what makes the report defensible is showing the reconciliation rather than picking the score that matches a preferred answer. Can I use GARS-3 re-administration to show treatment progress? Not on its own. No independent reliable-change index, minimal clinically important difference, or treatment-response benchmark has been established for the GARS-3, so there is no defensible way to say a given score difference exceeds measurement noise, and "severity improved by one category" claims more than the instrument can support. A repeated administration documents change in one informant's report, which is worth describing: the same respondent, the same index, and the interval should all be stated, and the change should sit beside the evidence that can carry a progress claim, direct observation, IEP goal and service data, adaptive-behavior measures, and functional outcomes. If the respondent changed between administrations, say so prominently, because informant effects alone can move rating-scale scores. For payers and teams, frame re-administration as updated informant perception within a broader progress review, never as the outcome measure. Can BastionGPT draft my GARS-3 report? Yes. Give it the facts (respondent with relationship and contact time, the index used and its communication-status rationale, the subscale pattern, the licensed classification, observation and history findings, and the eligibility or diagnostic framework in play) and it drafts the results section: informant context first, pattern before composite, the classification defined rather than inflated, convergence and divergence written out, and the diagnostic and eligibility conclusions kept separate, ready for your review. It can also cross-check a finished report for an unnamed index, missing respondent context, percentage language, manual figures quoted as field performance, a score doing eligibility work, or an unsupported progress claim, and it can draft the parent explanation and divergence-reconciliation paragraphs. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Publisher record, accessed August 2026: PRO-ED, Inc., GARS-3 product page (ages, time, subscales, standardization figures, kit contents, ©2014), permissions and electronic-use policy , and GARS-3S Spanish version ; Pearson's Australian listing, distributor role statement . Independent psychometrics: diagnostic-utility study of the GARS-3 parent report in 186 clinically referred children (Journal of Autism and Developmental Disorders, 2022; about 47 percent correct classification at the recommended cutoff, false-positive rates 82.83 to 87.88 percent); Samadi, McConkey, and colleagues, 2022, Kurdish-language GARS-3 validation (weaker specificity against intellectual disability and communication disorders); Chinese-version study (2024 online); Karren, 2017, Journal of Psychoeducational Assessment (ASD-only norm sample; 16 items retained from prior editions), cited without a public link. US school law: eCFR, 34 CFR 300.304 (variety of tools; no single measure); Minnesota Rules, 3525.1325 (autism eligibility methods); Oregon, autism eligibility rule (amended December 2023); Texas, 19 TAC 89.1040(c)(1) in the Legal Framework (eligibility is not medical diagnosis). Payer policies: TRICARE Autism Care Demonstration, diagnosing-provider attestation (GARS listed; parent questionnaire alone insufficient); North Carolina Medicaid, Clinical Coverage Policy 8F (amended August 1, 2026; named qualifying diagnostic procedures); Cigna, autism coverage policy 0447 (effective May 2026); CMS, EPSDT coverage guide . ## Related templates ADOS-2 Report Write-Up ADI-R Report Write-Up CARS-2 Report Write-Up SRS-2 Report Write-Up Autism Spectrum Evaluation Report M-CHAT-R/F Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/caps-5 All note templates # CAPS-5 Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The CAPS-5 (Clinician-Administered PTSD Scale for DSM-5) is the structured-interview reference standard for diagnosing PTSD: a trained clinician rates 20 symptoms against an index trauma, applies the DSM-5 algorithm, and sums a 0-80 severity total. VA, research, and trauma clinics rely on it. The write-up must name the form, show the algorithm, and keep PCL-5 numbers separate. This page covers CAPS-5 documentation, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, psychiatrists, and trained clinicians or clinical researchers; distributed to credentialed professionals by the National Center for PTSD Audience Treating and referring clinicians, VA and veterans' claims examiners, disability and forensic reviewers, research teams, payers Typical length 250 to 600 words for the results summary · the interview itself runs about 45 to 60 minutes, plus scoring and integration Format family Clinician-administered structured diagnostic interview (30 items: 20 DSM-5 PTSD symptoms rated 0 to 4, plus onset, impairment, validity, and subtype items) When it's used PTSD diagnostic evaluations, VA and veterans' assessments, treatment-outcome measurement and research trials, whenever a self-report screen needs diagnostic confirmation Standards context Developed at the U.S. National Center for PTSD (request-gated, no listed fee); aligned with DSM-5 and DSM-5-TR adult criteria; no items or prompts reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CAPS-5? The CAPS-5 (Clinician-Administered PTSD Scale for DSM-5; Weathers, Blake, Schnurr, Kaloupek, Marx, and Keane; U.S. National Center for PTSD) is the structured interview the field treats as the reference standard for PTSD diagnosis. It is a 30-item interview: the 20 DSM-5 PTSD symptoms, each rated by the interviewer on a single 0 to 4 severity scale that combines frequency and intensity under prescribed rules (not multiplication), plus additional items covering onset and duration, distress and impairment, response validity, global severity, and the dissociative subtype. The 20 symptom ratings sum to a 0 to 80 total; a symptom counts toward diagnosis when rated 2 or higher. Three time-frame forms exist, and the difference is diagnostic, not cosmetic: Past Month supports a current diagnosis, Worst Month supports a lifetime determination, and Past Week measures short-interval change and cannot by itself establish a diagnosis, because DSM-5 requires symptoms lasting more than one month. All symptom inquiry is anchored to a defined index trauma, usually selected with the LEC-5. DSM-5-TR changed no adult PTSD criteria, so the CAPS-5 remains current; a revised CAPS-5-R with 0 to 10 item ratings and a 0 to 200 total has been published but is not yet distributed, and its totals must never be mixed with 0 to 80 CAPS-5 scores. Documentation is where the instrument's precision is most often thrown away. The diagnosis comes from an algorithm, never from the total: at least one intrusion and one avoidance symptom, two negative-cognition-and-mood symptoms, and two arousal symptoms at threshold, plus duration over one month, clinically significant distress or impairment, and the exclusion judgment that the presentation is not better attributed to substances or another medical condition. The 0 to 80 total is a dimensional severity index riding alongside that determination, and the self-report PCL-5 , which happens to share the 0 to 80 range, is a different instrument on a different metric whose 31 to 33 cut-points do not transfer. A note reading "CAPS/PCL score 38, positive for PTSD" fails on every one of these points at once. The interview feeds, and never replaces, the broader workup covered on the psychiatric diagnostic evaluation page . ## Who uses CAPS-5 documentation and when Trauma-focused clinicians and diagnosing psychologists and psychiatrists use the CAPS-5 when a definitive PTSD determination matters: confirming what a screen suggested, resolving a complicated differential, or establishing a defensible baseline before treatment. The VA system is its natural habitat, in clinical care and in compensation and pension examinations, where the interview may be used but is not required by the disability questionnaire, and where the 2023 VA/DoD practice guideline suggests a validated structured interview to confirm the diagnosis. Researchers make it the dominant primary endpoint in PTSD treatment trials, administered by trained, calibrated, often blinded raters. Forensic, disability, and personal-injury evaluators use it because its structure makes findings reproducible, while keeping diagnosis, causation, and work capacity as separate questions. The note's readers mirror those uses: a claims examiner checking that a DSM-based diagnosis is actually supported, a reviewer distinguishing interview findings from the self-report screen, an attorney or independent examiner re-running the logic, and the next clinician re-administering against the same index event. Broader exposure history and Criterion A pathways belong on the trauma assessment page ; the full diagnostic visit structure, including the mental status examination, lives on the MSE and biopsychosocial assessment pages. ## How to document CAPS-5 results in the chart No authority prescribes a CAPS-5 note format, but the instrument's own architecture dictates what a defensible summary must contain: which form and under what conditions, the index event at a structural level, the cluster arithmetic, the full algorithm, the total kept in its dimensional lane, the subtype and validity judgments, and change measured only against a true prior. Each section below carries the pitfall that most often undermines it. Instrument, form, and administration conditions. Name the exact form and time frame (CAPS-5 Past Month, Worst Month, or Past Week) with the version date printed on the form, the administration date, modality (in person, video, or telephone), interpreter use, approximate duration, and any breaks or split sessions. Official materials carry inconsistent citation years (the webpage says 2013, the PDFs 2015, the forms April 2018), so cite the form actually administered. Pitfall: "CAPS administered." Which form? A Past Week total presented as a current diagnosis is the version error reviewers catch first, because the one-month duration criterion cannot come from a one-week window. Index event, structurally. Record the event category and approximate period, the Criterion A pathway, and whether the index is one incident or a coherent set of closely related incidents, with the LEC-5 noted when used. Give enough to identify the same event at the next administration and no more: no graphic narrative, no scene-by-scene detail. The completed booklet can live in a restricted test-material section while the note carries this summary. Pitfall: Either extreme: "trauma history positive" with no identifiable index, or a graphic reconstruction that re-exposes the patient, inflates the record, and adds nothing to diagnosis, safety, or the referral question. Cluster results, not item content. Report the number of symptoms rated 2 or higher in each cluster, for example B=3 of 5, C=1 of 2, D=4 of 7, E=3 of 6, with cluster severity sums when useful. That one line makes the algorithm auditable. Individual symptoms earn discussion only when they matter to treatment, differential, or risk, and then conceptually, in your own words. Pitfall: Twenty miniature trauma narratives, or interview prompts and rating anchors paraphrased into the note. The instrument is request-gated; its content does not belong in an open chart. The full diagnostic determination. Make every element visible: Criterion A met, the B/C/D/E counts against their required minimums, duration over one month, clinically significant distress or impairment, and the exclusion judgment (no standalone score exists for it; the clinician documents that substances or another medical condition do not better explain the presentation), with symptoms attributed to the index event. Then state the determination: current PTSD met or not met. Pitfall: "CAPS-5 positive" or "met cutoff." The standard algorithm has no total-score cutoff, and a determination that does not show its elements cannot be audited or re-run. Total severity, kept dimensional. Report the 0 to 80 total as a severity index alongside, never instead of, the determination. Twelve is a mathematical floor (six threshold symptoms at 2), not a validated cutoff; the circulating severity bands come from trial registrations, not the developers; characterize severity clinically and say so. Pitfall: "Total 38, therefore PTSD," or "moderate per the 23-34 band." One VA trial registration states cut scores "have not yet been established," and no developer-published band exists. Dissociative subtype and validity judgments. State the subtype determination (present, absent, or not determinable): depersonalization and derealization are assessed by additional interview items that sit outside the 20 symptoms and contribute nothing to the 0 to 80 total. Record the overall response-validity judgment and any limits: fatigue, interruptions, interpreter constraints, inconsistent time framing. Pitfall: A "dissociation score" built by summing two of the 20 symptom items. Secondary pages, and even the National Center's own summary page in conflict with its booklet and training, misnumber these items; the booklet controls. Change, monitoring, and the PCL-5 boundary. Compare only true priors: same form, same time frame, same index event, and say so. Name any benchmark with its source: the suggested 10-point response indicator, the 12-to-13-point reliable-change estimates, the post-treatment 8 and below and remission below 12 conventions, all sample-derived. Document the PCL-5 separately as self-report with its own metric, and never combine, convert, or average across CAPS-5, PCL-5, CAPS-IV, or CAPS-5-R. Pitfall: "CAPS/PCL score 38," a CAPS-IV total treated as continuous with a CAPS-5, or "recovered" declared from a 10-point drop while the algorithm is still met. ### Blank template (copy and adapt) CAPS-5 RESULTS SUMMARY SKELETON Client: [initials] Age: [ ] Date(s): [ ] Interviewer: [ ] Form + time frame: [CAPS-5 Past Month / Worst Month / Past Week] (version date on form: [ ]) Modality: [in person / video / telephone] Interpreter: [ ] Duration: [ ] Breaks/split: [ ] Index event (structural): [category + approximate period; Criterion A pathway; single incident or coherent related series; LEC-5 administered: yes/no] No graphic narrative required. Cluster results (symptoms rated 2+): B (intrusion): [ ] of 5 C (avoidance): [ ] of 2 D (cognitions/mood): [ ] of 7 E (arousal): [ ] of 6 Cluster sums (optional): [ ] Diagnostic determination (all elements visible): A: [met/not] B >= 1: [ ] C >= 1: [ ] D >= 2: [ ] E >= 2: [ ] F duration > 1 month: [ ] G distress/impairment: [ ] H exclusion (substances / medical condition) addressed: [ ] Attribution to index event: [ ] Determination: [met / not met] Total severity: [ ]/80 (dimensional index; not the diagnosis; no invented cutoffs or bands) Dissociative subtype: [present / absent / not determinable] (additional items, outside the 20 and the total) Response validity + limitations: [ ] Change (only vs same form, time frame, index): [prior total, date, point change, named benchmark + source] Complementary measures: [PCL-5 self-report total, documented separately on its own metric] Plan: [treatment, monitoring instrument + interval, referrals] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-summary skeleton, ready to adapt. Neither reproduces interview items, prompts, rating anchors, or scoring materials. ## Sample CAPS-5 documentation (fictional) Scenario: an initial diagnostic evaluation in which the algorithm is met, written so every element is visible, the index event stays structural, and the PCL-5 stays in its own lane. All details are fictional. Patient: R.T., 34 years · Setting: Outpatient psychiatric diagnostic evaluation · Clinician: S. Delgado, PhD, licensed clinical psychologist · Note date: 08/19/2026 Instrument and administration: CAPS-5 Past Month, adult form (version date April 13, 2018), administered in person on 08/19/2026 as part of a psychiatric diagnostic evaluation, by a psychologist trained in the National Center for PTSD administration curriculum. The interview lasted approximately 55 minutes with one brief break. The LEC-5 was completed first to survey lifetime exposure and select the index event. Index event: The index trauma was a workplace vehicle collision in 2023 involving direct exposure to threatened death and serious injury, meeting Criterion A as a single incident. This structural description is sufficient to anchor today's ratings and to identify the same event at any future administration; a graphic narrative was not required for the referral question and is not recorded here. All symptom inquiry was anchored to this event. Findings: Symptoms rated at severity 2 or higher: intrusion B=3 of 5, avoidance C=1 of 2, negative alterations in cognitions and mood D=4 of 7, and arousal and reactivity E=3 of 6, satisfying the required minimums in every cluster. Onset followed the collision, with duration well beyond one month (approximately eight months of active symptoms). Symptoms are associated with clinically significant occupational impairment (missed shifts, a route change to avoid driving demands) and interpersonal strain. The presentation is not better attributed to substances or another medical condition; alcohol use is minimal by history and screening, and medical records show no competing explanation. Depersonalization and derealization, assessed by the interview's additional subtype items, did not reach the threshold: dissociative subtype absent. Overall response validity was judged good: engagement was consistent, and reported symptoms matched observed affect and collateral occupational records. Severity and complementary self-report: Total CAPS-5 severity was 32 of 80, the sum of the twenty 0 to 4 clinician ratings, recorded as a dimensional index of current symptom burden rather than as the diagnostic rule. Clinically, the burden is characterized as moderate; no published severity band is implied, because none is developer-endorsed. A PCL-5 completed the same day totaled 44 of 80 and is documented separately as self-reported symptom burden on its own metric: the two totals share a range, not a meaning, and neither replaces the other. Determination and plan: Current PTSD criteria are met under the standard CAPS-5 scoring rule: Criterion A met, cluster minimums satisfied, duration exceeded, impairment present, exclusions addressed. Plan: trauma-focused evidence-based psychotherapy was reviewed and scheduled; the PCL-5 will track symptoms between sessions; response will be re-evaluated with a CAPS-5 Past Month re-administration anchored to the same index event, read against the suggested 10-point response indicator and the published 12-to-13-point reliable-change estimates, both sample-derived benchmarks rather than universal rules. The completed interview booklet is retained in the restricted test-material section of the record; this summary serves the open chart. This sample is fictional and for educational purposes. It does not describe a real patient or record; the details are invented to show documentation structure and are not clinical guidance. No interview items, prompts, anchors, or scoring materials are reproduced, and the index event is described at a deliberately structural, non-graphic level. ↑ Back to the template and downloads ### Why this sample works The exact form, time frame, and version date are stated, so the reader knows a current diagnosis was even possible from this administration. The index event is structural: enough to satisfy Criterion A and re-anchor the next administration, with nothing graphic in the open chart. One line of cluster counts makes the whole algorithm auditable without reproducing a word of gated interview content. Every diagnostic element is visible, including the exclusion judgment that has no score of its own, so the determination can be re-run by the next reader. The total stays dimensional, the severity label is explicitly clinical rather than a band, and the PCL-5 sits in its own labeled lane on its own metric. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the highest-stakes reader is often a VA adjudicator, and the law is specific about what the interview does and does not decide. Service connection for PTSD requires a DSM-based diagnosis conforming to 38 CFR 4.125(a) (which still incorporates DSM-5), medical evidence linking current symptoms to an in-service stressor, and credible supporting evidence that the stressor occurred (38 CFR 3.304(f)); the disability percentage then flows from occupational and social impairment under the 38 CFR 4.130 general rating formula, not from any instrument score, so a CAPS-5 total maps to no percentage and establishes neither stressor, nexus, credibility, nor employability (LAW). The PTSD disability benefits questionnaire requires DSM-5 diagnostic information and impairment evidence but no named instrument; initial examinations are restricted to qualified examiners (psychiatrists meeting VA board requirements, doctorate-level psychologists, specified supervised trainees), and 2025 exam-ordering guidance advises that the treating clinician should not, where possible, perform the examination (VA POLICY). The 2023 VA/DoD practice guideline suggests a validated structured interview such as the CAPS-5 or PSS-I-5 to confirm the diagnosis, with validated measures for monitoring, and labels itself decision support rather than a legal standard (CONVENTION). In civilian billing there is no CAPS-specific code: the interview ordinarily lives inside the psychiatric diagnostic evaluation codes 90791 or 90792, with the 96130 to 96139 testing family reserved for genuine standardized-testing work, and Medicare guidance adds that testing is not separately covered when the clinical interview could reasonably yield the information (PAYER POLICY). In research the CAPS-5 is the dominant trial endpoint; the 2024 FDA complete response letter in the MDMA program concerned trial conduct, not the instrument (CONVENTION). Canada and Australia require the diagnosis, not the instrument. Veterans Affairs Canada's PTSD entitlement guideline (reviewed January 2025) works from DSM-5-TR, requires a clinical diagnosis from a qualified physician, nurse practitioner, or psychologist, and asks for supporting documentation as comprehensive as possible without naming the CAPS-5; the Table of Disabilities rates psychiatric impairment on functional domains, so the interview supports the diagnosis and severity picture while entitlement, service relationship, and assessment remain adjudicative questions (VAC POLICY). In Australia, veterans' claims run under legislative Statements of Principles built on DSM-5 definitions, with psychiatric compensation guidelines (amended July 2026) expecting a high-quality psychiatric report, usually from a psychiatrist, and departmental clinical resources name the CAPS-5 as the follow-up assessment after a positive PC-PTSD-5 screen, which is guidance rather than a statutory requirement (LAW and DVA POLICY). Phoenix Australia's national guideline prefers structured clinical interviewing over questionnaires for diagnosis; its NHMRC approval windows are time-limited (the updated recommendations reach their five-year mark in December 2026), so currency is worth checking at the time of writing (CONVENTION). Across all three countries the bottom line is identical and worth writing into reports: no authority mandates the CAPS-5 by name; what the law requires is a defensible DSM-based diagnosis with adequate supporting evidence, which is precisely what disciplined CAPS-5 documentation provides. Instrument facts, psychometrics, and rights complete the picture. Access is free but request-gated: non-VA clinicians email the National Center for PTSD and certify advanced psychodiagnostic training, a master's degree in a clinical discipline, or research credentials, and specific CAPS-5 training (a four-part curriculum with simulated practice) is strongly encouraged; there are no official National Center translations of the CAPS-5, so translated research versions must be documented by language, source, and validation. Psychometrics from the initial veteran validation: diagnostic interrater kappa .78 to 1.00 depending on scoring rule, test-retest kappa .83, total-score alpha .88, interrater ICC .91, test-retest ICC .78, correlation .83 with the CAPS-IV and .66 with the PCL-5; a 2024 meta-analysis across 15 studies put global internal consistency at .92 with the two-item avoidance cluster predictably weakest, and a VA centralized telephone model with trained, calibrated raters reached a total-score interrater ICC near .98, which supports remote administration under controlled conditions without proving every remote administration equivalent. Change anchors are sample-derived: the suggested 10-point response indicator, reliable change of 12 to 13 points, post-treatment totals of 8 and below, and remission below 12 all trace to specific veteran samples and study conventions. On rights: as U.S. government work the instrument likely sits outside federal copyright, but the qualification gate, professional test security, and confidentiality still control, so completed forms belong in restricted records, blank stock stays secured, and public reproduction, native EHR item banks, and public web scorers are unsupported without written clarification from the Center. The CAPS-5 was developed at the U.S. National Center for PTSD, a division of the Department of Veterans Affairs. BastionGPT is not affiliated with, or endorsed by, the National Center for PTSD or the VA. This page reproduces no interview items, prompts, rating anchors, or scoring materials. ↑ Back to the template and downloads ## Common CAPS-5 documentation errors reviewers flag The numbers behind these errors are specific. The National Center's own materials carry a 2013 webpage citation, 2015 PDF citations, and April 2018 form dates; the initial validation put the CAPS-5's correlation with the self-report PCL-5 at only .66; and the published reliable-change estimates are 12 to 13 points, derived from male veteran samples. The BastionGPT Clinical Advisory Board sees the same errors most often in CAPS-5 documentation reviews: CAPS-5 and PCL-5 conflated. "CAPS/PCL score 38, positive for PTSD." The shared 0 to 80 range is incidental: one is a clinician-administered diagnostic interview, the other a self-report screen whose 31 to 33 cut-points index probable PTSD on the PCL-5 only, and their totals correlate just .66. Name each instrument, its modality, its score, and what that score establishes, in separate sentences. The total treated as the diagnosis. "Met cutoff" or a severity band doing diagnostic work. The standard algorithm has no total cutoff: 12 is only the mathematical floor created by six threshold symptoms, the 26 figure was a study calibration rule, and the circulating bands come from trial registrations, one of which states plainly that cut scores "have not yet been established." Report the algorithm's elements and keep the total dimensional. The form and time frame unlabeled. "CAPS administered" with no version. Past Month supports a current diagnosis, Worst Month a lifetime determination, and Past Week is a monitoring window that cannot satisfy the one-month duration criterion by itself. Cite the form actually used with its printed version date, because even the official materials disagree on citation years (2013 webpage, 2015 PDFs, April 2018 forms). Scoring described as frequency times intensity. A ranking explainer teaches multiplication and composite ranges; that describes a system the CAPS-5 does not use. Each symptom receives one 0 to 4 severity rating combining frequency and intensity under prescribed combination rules, reflecting typical severity in the interval rather than the worst episode, and a symptom counts at 2 or higher. Notes that import the multiplication scheme misstate the instrument. A dissociation score summed from symptom items. The depersonalization and derealization items sit outside the 20 symptoms and contribute nothing to the 0 to 80 total; the last two symptom items are arousal-cluster symptoms. Secondary pages, and the National Center's own summary page in conflict with its booklet and training materials, misnumber them, so a chart that "sums items 19 and 20 for dissociation" has added two Criterion E ratings. State the subtype as present, absent, or not determinable from the subtype items. Graphic narratives and gated content in the open chart. A scene-by-scene trauma reconstruction, or interview prompts and anchors pasted into the note. The defensible record names the event category, period, and Criterion A pathway, holds detail to the minimum the diagnosis and referral require, and secures the completed booklet in a restricted section, mindful that portals expose open-chart content. The instrument is request-gated; reproducing its content publicly or in shared templates is unsupported. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric diagnostic evaluations and trauma assessments. Give it the facts (form and time frame, administration conditions, the structural index event, cluster counts, duration, impairment, exclusions, subtype and validity judgments, totals with any true prior) and it drafts the results summary: every algorithm element visible, the total kept dimensional, change benchmarked to a named source, and the PCL-5 in its own lane, ready for your review. Cross-check a finished note for the gaps reviewers flag: an unlabeled form, a Past Week total doing diagnostic work, a cutoff or band invented for the total, conflated CAPS and PCL numbers, a missing exclusion judgment, or trauma detail beyond the minimum necessary. Draft the companion paragraphs: the claims-ready summary that separates diagnosis from stressor, nexus, and impairment questions, the re-administration comparison against the same index event, or the plain-language explanation of what the interview found for the patient. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What are the CAPS-5 versions and which one do I document? Three time-frame forms of the same adult interview: Past Month, which supports a current PTSD diagnosis; Worst Month, which assesses the lifetime period when symptoms were worst and supports a lifetime determination; and Past Week, a short-interval symptom measure for treatment and research follow-up that cannot by itself establish a diagnosis, because DSM-5 requires duration beyond one month. The note names the exact form with the version date printed on it (the current adult forms are dated April 2018, while official citations vary between 2013 and 2015, so the form in hand controls). Two separate instruments complete the family and must not be blended into "the CAPS": the CAPS-CA-5 for children and adolescents 7 and older, and the published but not-yet-distributed CAPS-5-R, which rates items 0 to 10 and totals 0 to 200, a metric that can never be entered or compared as a 0 to 80 score. How is the CAPS-5 scored? The interviewer gathers frequency and intensity information for each of the 20 DSM-5 PTSD symptoms, anchored to the index trauma, and combines them under prescribed rules into a single 0 to 4 severity rating per symptom; the rating reflects typical severity during the interval, not the worst single episode, and the method is a structured combination, not frequency multiplied by intensity, whatever some explainer pages teach. A symptom counts toward diagnosis at severity 2 or higher. The 20 ratings sum to the 0 to 80 total, and cluster sums can be reported for intrusion (up to 20), avoidance (up to 8), negative cognitions and mood (up to 28), and arousal (up to 24). The additional interview items, covering onset and duration, distress and impairment, response validity, global severity, and the dissociative subtype, are separate clinical judgments that never add into the total. The scoring detail ships with the gated instrument; notes summarize results without reproducing anchors. What CAPS-5 score indicates PTSD? None, and any note built on a total cutoff is wrong on the instrument's own terms. The diagnosis comes from the algorithm: a qualifying Criterion A event, at least one intrusion symptom, one avoidance symptom, two negative-cognition-and-mood symptoms, and two arousal symptoms rated 2 or higher, duration beyond one month, clinically significant distress or impairment, and the exclusion judgment that substances or another medical condition do not better explain the presentation. Three numbers get misread into cutoffs: 12 is only the arithmetic floor six threshold symptoms create, 26 was a calibration rule inside the initial validation, and the tidy severity bands circulating online come from trial registrations rather than the developers, one of which states that cut scores "have not yet been established." A person can carry a fairly high total and fail the algorithm, or meet the diagnosis near the floor; the note reports the elements, the determination, and the total as a dimensional severity index beside it. What is the difference between the CAPS-5 and the PCL-5? Everything except the 0 to 80 range. The CAPS-5 is a clinician-administered structured interview that yields a diagnostic determination: the interviewer clarifies, probes, attributes symptoms to the index event, and weighs clinical significance. The PCL-5 is a 20-item self-report that takes 5 to 10 minutes and serves screening, provisional identification, and symptom monitoring; its widely cited 31 to 33 cut-points index probable PTSD on the PCL-5 only. The totals correlate moderately (.66 in the initial validation), PCL-5 totals commonly run higher than CAPS-5 totals for the same person, and longitudinal change tracks in parallel without being identical, which is why the developers say the scores are not interchangeable. Chart them as separate labeled entries ("CAPS-5 Past Month total 32/80, current PTSD criteria met; PCL-5 self-report 44/80 same date") and never as a blended "CAPS/PCL" number; each instrument also keeps its own change benchmarks. How do I document the index trauma without a graphic narrative? Structurally: the event category, the approximate date or period, the Criterion A pathway, and whether the index is a single incident or a coherent set of closely related incidents, for example "a workplace vehicle collision in 2023 involving direct exposure to threatened death and serious injury." That level lets the next administration re-anchor to the same event, supports the Criterion A conclusion, and explains the symptom relationship, which is everything the record needs. Scene-by-scene reconstruction, sensory detail, and intimate specifics that change nothing about diagnosis, safety, treatment, or the referral question add re-exposure risk and record bloat instead of clinical value, and trauma-informed record-keeping guidance runs minimum-necessary for exactly that reason. When several events qualify, name the one selected and why, keep it constant across items, and note the LEC-5 exposure survey; if the index changes at a later administration, say so, because the totals no longer form a clean same-event trajectory. The completed booklet can sit in a restricted test-material section while this summary serves the open, portal-visible chart. How do I document the dissociative subtype? As a determination from the interview's dedicated subtype items, stated as present, absent, or not determinable. Depersonalization and derealization are assessed by additional items that sit outside the 20 PTSD symptoms and contribute nothing to the 0 to 80 total; the subtype is specified when full PTSD criteria are met and clinically significant depersonalization or derealization is also established. The trap is numbering: because the 20 symptoms run intrusion 1 to 5, avoidance 6 to 7, cognitions and mood 8 to 14, and arousal 15 to 20, the last two symptom items are arousal symptoms, yet secondary pages, and even the National Center's own summary page in conflict with its booklet and training materials, describe a "dissociation score" summed from symptom items. A chart that follows that shortcut has added two Criterion E ratings and called them dissociation. Write the determination in words, from the subtype items, and skip item arithmetic entirely. How much CAPS-5 change counts as improvement? Benchmarks exist, and every one of them is sample-derived, so the note names its source. The National Center suggests roughly a 10-point drop as a response indicator while calling the evidence limited; published reliable-change estimates from two male veteran samples run 12 to 13 points; a post-treatment total of 8 or below marked clinically significant change in those samples; and totals below 12 have served as a remission convention, 12 being the arithmetic floor for a diagnosable configuration. Response, reliable change, loss of diagnosis, remission, and recovery are different claims: a 14-point drop can coexist with criteria still met, and losing the diagnosis is a statement about the algorithm, not the total. Valid comparison requires the same form, the same time frame, and the same index event, stated explicitly; CAPS-IV totals (a 0 to 136 metric), CAPS-5-R totals (0 to 200), and PCL-5 change all live on other scales and never mix. Pair the interview trajectory with function: work, sleep, avoidance behavior, and the patient's own goals. Does a CAPS-5 establish VA disability or service connection? No. It can establish the clinical diagnosis and make the severity picture transparent, which is real value in a claim, but the legal elements are separate. U.S. service connection for PTSD requires a DSM-conforming diagnosis, medical evidence linking current symptoms to an in-service stressor, and credible supporting evidence that the stressor occurred; the disability percentage then follows occupational and social impairment under the rating formula, so no CAPS-5 total maps to a percentage. The disability questionnaire requires DSM-5 diagnostic information but no named instrument (the CAPS-5 may be administered), initial examinations are restricted to qualified examiners, and current guidance advises against the treating clinician performing the exam where possible. Canada and Australia run parallel: Veterans Affairs Canada requires a qualified clinician's DSM-5-TR diagnosis with comprehensive documentation, and Australia's DVA works from legislative Statements of Principles with psychiatric reports, its clinical guidance naming the CAPS-5 only as recommended follow-up after screening. In every system the interview strengthens the evidence; the entitlement decisions belong to the adjudicator. Can BastionGPT draft my CAPS-5 documentation? Yes. Give it the facts (form and time frame, administration conditions, the structural index event, cluster counts at threshold, duration, impairment, the exclusion judgment, subtype and validity determinations, and totals with any true prior) and it drafts the results summary: every algorithm element visible, the total kept dimensional with no invented cutoffs, change read against named sample-derived benchmarks, and the PCL-5 documented separately on its own metric, ready for your review. It can also cross-check a finished note for an unlabeled form, a Past Week total doing diagnostic work, conflated CAPS and PCL numbers, a missing exclusion judgment, or trauma detail beyond the minimum necessary, and it can draft the claims-ready and patient-facing variants of the same findings. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: National Center for PTSD record, accessed August 2026: CAPS-5 measure page (structure, forms, access regime, training, change guidance), translations page (no official CAPS-5 translation), and LEC-5 page (index-event selection). Validation and change metrics: Weathers FW and colleagues, 2018, Psychological Assessment, initial CAPS-5 psychometric evaluation in military veterans ; Marx BP and colleagues, 2022, reliable and clinically significant change on the CAPS-5 and PCL-5 ; 2024 methodological exchange on the CAPS-PCL relationship, National Center article archive ; initial CAPS-5-R evaluation, Journal of Traumatic Stress . US law and VA policy: eCFR, 38 CFR 3.304(f) (service connection), 4.125 (DSM conformity), and 4.130 (rating formula); VA, Review PTSD disability benefits questionnaire ; VA/DoD, 2023 clinical practice guideline for PTSD . Billing context: CMS Medicare coverage database, psychiatric diagnostic evaluation and testing services article and psychological testing coverage determination (medical-necessity limits on separate testing claims). Canada and Australia: Veterans Affairs Canada, PTSD entitlement eligibility guideline ; Australian DVA, guidelines for psychiatric compensation claims (amended July 2026); Phoenix Australia, Australian PTSD guidelines . ## Related templates PCL-5 Score Documentation Trauma Assessment Psychiatric Diagnostic Evaluation Mental Status Exam (MSE) Note Biopsychosocial Assessment Treatment Plan Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/tova All note templates # T.O.V.A. Documentation: Score Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated August 2026 The T.O.V.A. (Test of Variables of Attention) is an FDA-cleared computerized continuous performance test measuring sustained attention and response inhibition across a 21.6-minute go/no-go task. ADHD evaluators use it as an aid to assessment, never as a standalone diagnostic. Documentation must name the version, conditions, and validity before scores. This page covers how to document T.O.V.A. results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychologists, neuropsychologists, psychiatrists, and pediatric clinicians; trained staff may administer, qualified professionals interpret Audience ADHD evaluation teams, prescribing clinicians monitoring treatment, schools and disability offices, payers and reviewers, families Typical length 200 to 500 words for the results section · the standard administration runs 21.6 minutes (10.8 for the preschool visual form), plus setup and interpretation Format family Computerized go/no-go continuous performance test (visual and auditory versions; standard scores, mean 100, SD 15, by quarter, half, and total) When it's used ADHD evaluations as adjunctive performance data, baseline and on-medication treatment monitoring (visual version), targeted differential questions about attention Standards context FDA 510(k)-cleared as an aid to assessment (the visual version also treatment evaluation); The TOVA Company; no stimuli, norms, or thresholds reproduced here On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the T.O.V.A.? The T.O.V.A. (Test of Variables of Attention; The TOVA Company; originated by Lawrence M. Greenberg, MD, with current manuals crediting Leark, Dupuy, Greenberg, Kindschi, and Hughes) is a computerized go/no-go continuous performance test delivered through licensed software and a dedicated microswitch the manufacturer specifies at roughly millisecond timing precision. The task runs in two contrasting halves: targets are infrequent in the first, taxing sustained attention, and frequent in the second, taxing response inhibition. The current product is T.O.V.A. 9 on the 9.1 software line, and the visual and auditory versions are distinct tests, not presentation options: separate norms, different age ranges (visual roughly 4 through 80, auditory 6 through 29), different FDA-cleared indications, and an Attention Comparison Score generated only for the visual test. Reports return standard scores (mean 100, SD 15, age- and sex-referenced) for omission errors, commission errors, response time, response-time variability, and d-prime response sensitivity, organized by quarter, half, and total, alongside anticipatory-response and validity information. The standard administration takes 21.6 minutes; a shorter preschool visual form takes 10.8. The load-bearing fact is regulatory. The system is FDA-cleared under 510(k) K173915 (March 2018, product code LQD, predicate QbTest) as an aid: the visual version aids assessment and treatment evaluation, the auditory version aids assessment only, and the cleared labeling restricts interpretation to qualified professionals, with the manual adding that the device does not diagnose any disorder. Clearance is a substantial-equivalence finding, not a validation of diagnostic accuracy, and professional guidance in the US, Canada, and Australia uniformly treats continuous performance tests as adjuncts: a recent meta-analysis of 19 commercial-CPT studies put pooled discrimination at an area under the curve of roughly 0.70 to 0.80, with sensitivity near .75 and specificity near .71. So the documentation job is to record a constrained laboratory performance sample honestly, aid-not-diagnosis framing intact, and feed it to the synthesis covered on the ADHD evaluation report page , alongside the rating-scale evidence covered on the Conners 4 and Vanderbilt pages. ## Who uses T.O.V.A. documentation and when Psychologists and neuropsychologists fold the T.O.V.A. into ADHD evaluations as performance-based evidence beside history, collateral report, and rating scales; psychiatrists and pediatric prescribers use the visual version's treatment-evaluation indication for baseline and on-medication comparisons during titration; and clinics with technician-administered testing use it because administration and interpretation can be split, provided the interpretation stays with a qualified professional. The note's readers are unusually varied. A payer reviewer may be applying a policy that classes computerized attention testing as experimental (several major US policies do), so the note must show the test as one component of a medically necessary evaluation rather than a standalone billable diagnostic. A school or testing agency reading an accommodations request needs multimethod evidence, because no single CPT score establishes the need for extended time. A prescriber reading a titration retest needs the medication name, dose, and dose-to-test interval before the score means anything. And a family often arrives having read that this is an "objective ADHD test," which the documentation must gently correct. Adult self-report screening lives on the ASRS page , adult observer ratings on the CAARS-2 page , and everyday executive-function ratings on the BRIEF-2 page ; this page owns the CPT session itself. ## How to document T.O.V.A. results in the chart No authority prescribes a T.O.V.A. note format, but the instrument's design dictates what a defensible entry must contain: exact identification, the conditions that move CPT scores, medication timing, validity before numbers, the pattern in prose, the comparison score framed as an aid, and an integration statement. Each section below carries the pitfall that most often undermines it. Instrument identification, exactly. Record T.O.V.A. 9, the modality (visual or auditory), the software or build identifier printed on the patient's report, the administration date and time, and the purpose: initial adjunctive assessment or treatment-evaluation retest. The modality line is not pedantry: the two versions carry different norms, age ranges, and FDA indications, and only the visual test generates the Attention Comparison Score. Pitfall: "TOVA administered" or "TOVA positive." Unversioned, unmodalitied results cannot be interpreted, compared at retest, or defended to a reviewer. Testing conditions. Chart time of day, the prior night's sleep, caffeine, nicotine, or other substance exposure, acute illness, pain, or mood state, vision or hearing correction, room interruptions and observed engagement, motor or hand limitations affecting the microswitch, and prior exposure to the T.O.V.A. or another CPT. These are the variables that move CPT scores without any attention disorder present. Pitfall: A score table with no conditions. Sleep deprivation or a 4 p.m. slot can manufacture a poor performance; a practice effect can manufacture an improvement. Medication status, precisely. State the medication name and dose, whether testing was on or off medication, the date and time of the last dose, and the elapsed interval, plus whether the status reflected usual treatment, a prescriber-directed hold, or a planned comparison. Whether to hold a stimulant is the prescriber's clinical and safety decision for the question at hand; no generic washout duration applies across formulations. Pitfall: "Tested unmedicated per protocol" with no prescriber plan, or a patient told to stop a stimulant because "the TOVA must be done off meds." Validity before scores. Resolve interpretability first: anticipatory responses, sufficient usable responses, interruptions or hardware warnings, instructions followed, and the unusual-pattern or performance-validity output where it applies (visual testing, ages 17 and older, and only under certain data conditions). A flag means caution and clinical adjudication. Pitfall: Scores interpreted past an unresolved validity problem, or a validity flag written up as malingering. The analysis identifies unusual patterns; it does not diagnose deception. The standard-score pattern, in prose. Describe the variables by the periods they came from: omissions concentrated in the low-target first half, variability worsening across later quarters, commissions rising under the inhibition demand, response speed slow but stable, d-prime discrimination reduced. Quarter, half, and total results are separate aggregations of the underlying responses, not averages of each other, and every variable has nonspecific causes. Pitfall: A bare number grid, or one outlying quarter promoted into a disorder. "Inattention" and "impulsivity" are shorthand for omission and commission patterns, not proven constructs. The Attention Comparison Score, framed as an aid. For visual administrations, report the ACS with its meaning attached: a composite of first-half response speed, second-half response sensitivity, and overall response-time variability, compared against the manufacturer's independently diagnosed ADHD sample and its normative sample. Below zero means more similar to the ADHD comparison group; at or above zero means more similar to the normative group and does not exclude ADHD. It was formerly called the ADHD Score, a name worth retiring. Pitfall: Zero treated as a diagnostic threshold, the ACS read as an ADHD probability, or an auditory report searched for an ACS it does not produce. Integration and retest documentation. Close with a synthesis: the T.O.V.A. finding was integrated with developmental history, cross-setting symptoms and impairment, informant ratings, interview, and observation, and it neither establishes nor excludes ADHD. For treatment-evaluation retests, document matched conditions (time of day, sleep, hardware), the interval, prior exposure and expected practice effects, and the medication timing, and conclude in performance terms: more consistent under the medicated condition. Pitfall: A CPT overruling the clinical picture in either direction, or a retest gain attributed wholly to medication with practice effects unmentioned. ### Blank template (copy and adapt) T.O.V.A. RESULTS DOCUMENTATION SKELETON Client: [initials] Age: [ ] Date + time: [ ] Version: [T.O.V.A. 9, visual / auditory] Build (from report): [ ] Purpose: [initial adjunctive assessment / treatment-evaluation retest] Administered by: [ ] Interpreted by: [ ] Conditions: [sleep hours; time of day; caffeine/nicotine/substances; illness, pain, mood; vision/hearing correction; interruptions; motor factors; prior T.O.V.A. or CPT exposure] Medication: [name + dose; on/off; last dose date + time; interval; usual regimen / prescriber-directed hold / planned comparison] Validity review (before any interpretation): [anticipatory responses; usable responses; unusual-pattern / performance- validity output if applicable (visual, 17+); interruptions; hardware; instructions followed] Interpretable: [yes / with caution / no] Score pattern (standard scores, prose by period): Omissions: [ ] Commissions: [ ] Response time: [ ] RT variability: [ ] d-prime: [ ] Pattern notes: [which half / quarters carried the findings] Attention Comparison Score (visual only): [value relative to zero + aid-to-assessment framing; not a probability or threshold] Integration: [convergent / discordant with history, ratings, observation, impairment; neither establishes nor excludes ADHD] Retest notes (if applicable): [interval; matched conditions; practice effects; medication timing; performance-level wording] Plan: [how the finding feeds the evaluation or titration] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank documentation skeleton, ready to adapt. Neither reproduces stimuli, timing parameters, norms, thresholds, or report content. ## Sample T.O.V.A. documentation (fictional) Scenario: an adult ADHD evaluation in which a valid off-medication visual administration converges with the clinical picture, written so the score aids the assessment without becoming the diagnosis. All details are fictional. Patient: J.H., 34 years · Setting: Outpatient ADHD evaluation, adjunctive CPT session · Clinician: K. Osei, PsyD, licensed psychologist · Note date: 08/21/2026 Identification and purpose: Visual T.O.V.A. 9 administered 08/21/2026 at 9:20 a.m. as an adjunct to a comprehensive adult ADHD evaluation; the software build identifier from the patient report is filed with the raw data. This was an initial assessment administration, not a treatment-evaluation retest. Administration was monitored by trained staff; interpretation is by the undersigned psychologist. Conditions and medication: J.H. reported approximately seven hours of sleep, one morning coffee two hours before testing, no nicotine, no acute illness or significant distress, and corrected vision in use. He has no prior exposure to the T.O.V.A. or any other continuous performance test. He had not taken stimulant medication that day: the hold was directed by the prescribing clinician as part of a planned baseline assessment, and the last dose, the prior morning, is documented in the medication record with the elapsed interval. Validity: The administration completed without interruption or hardware warning, instructions were followed, usable responses were sufficient, and anticipatory responding was not clinically significant. The performance-validity analysis applicable to visual administrations at this age raised no concern. The administration was judged interpretable. Results: Response speed was broadly adequate but became progressively less consistent: response-time variability worsened across the later quarters, omission errors increased late in the low-target first half, and d-prime target discrimination weakened under the second half's inhibition demand, while commissions remained only mildly elevated. Standard scores by quarter, half, and total are retained with the report. The visual Attention Comparison Score fell below zero, meaning this laboratory performance pattern was more similar to the manufacturer's independently diagnosed ADHD comparison sample than to its normative sample. The ACS is an aid to assessment; it is not a diagnostic probability, a severity rating, or a standalone determination. Integration and plan: The T.O.V.A. pattern converges with the evaluation's other evidence: documented childhood-onset disorganization and sustained-attention difficulty, current occupational impairment, collateral history, and elevated self- and observer ratings. The finding was integrated with that history, the interview, and observations, and was not used as the sole basis for any conclusion; a normal result would likewise not have excluded the diagnosis. The diagnostic formulation and recommendations appear in the full evaluation report. If stimulant treatment is initiated, a treatment-evaluation retest at a matched morning time, with dose-to-test interval documented and practice effects acknowledged, may inform titration alongside symptom ratings, function, and tolerability. This sample is fictional and for educational purposes. It does not describe a real patient or record; the details are invented to show documentation structure and are not clinical guidance. No stimuli, timing parameters, norms, thresholds, or proprietary report content are reproduced. ↑ Back to the template and downloads ### Why this sample works The modality, version, build, and purpose are recorded, so the result is comparable at retest and defensible to a reviewer. Conditions and medication timing are charted with the prescriber's baseline plan named, so the performance sample is interpretable and no one was told to stop medication by a testing rule. Validity is resolved before a single score is discussed, in caution-flag language rather than deception language. The pattern is described by period in prose, and the ACS is reported with its meaning attached: similarity to a comparison sample, below zero, aid to assessment. The integration statement runs both directions: the convergent result supports the picture, and a normal result would not have excluded it. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the regulatory frame is precise and routinely inflated, so the note should get it exactly right. The system is FDA-cleared, not approved, under 510(k) K173915 (March 2018, product code LQD, regulatory class listed as unclassified, predicate device QbTest): the visual version is cleared to aid assessment and treatment evaluation, the auditory version to aid assessment only, and the cleared labeling restricts interpretation to qualified professionals; clearance is a substantial-equivalence finding, not a validation of diagnostic accuracy (REGULATORY STATUS). Professional guidance points the same way: the AAP's ADHD guideline rests diagnosis on clinical criteria, its 2024 evidence review concluded sole reliance on continuous performance tests cannot be recommended, and the AACAP practice parameter states testing is not mandatory for routine diagnosis; no federal FDA or DEA rule requires any CPT before stimulant prescribing (prescribing law concerns controlled-substance registration, monitoring programs, and telemedicine rules, with the DEA's telemedicine flexibilities currently extended through December 2026), so a clinic's "objective test required" policy is a clinic, payer, or institutional rule, not law (CONVENTION and LAW). Payers are openly skeptical: Aetna's ADHD policy classes computerized attention and vigilance testing as experimental, Centene finds insufficient evidence for routine diagnostic use, Cigna's neuropsychological-testing policy (effective May 2026) does not establish uncomplicated-ADHD CPT testing as a covered indication, and a 2026 UnitedHealthcare policy names comparable technologies unproven, so coverage should be verified, never assumed (PAYER POLICY). Billing distinguishes the automated-instrument code 96146 from professional administration and scoring (96136 to 96137), technician administration (96138 to 96139), and evaluation services (96130 to 96133); the right code follows who did what, and no code creates medical necessity. For ADA accommodations, review is individualized and multimethod, and no CPT score is dispositive (LAW and CONVENTION). Canada and Australia register the device modestly and position it identically in practice. In Canada the system is supplied under a Health Canada Medical Device Establishment Licence (number 9655), an establishment-level licence for low-risk devices that should never be written up as a product approval, and revised federal establishment-licensing guidance takes effect in December 2026; clinically, the CADDRA Canadian ADHD practice guidelines (version 4.1) make the clinical interview and validated rating scales the mainstays, caution that standardized testing can be normal despite real impairment, and decline to let computerized results determine diagnosis, severity, disability, or accommodation eligibility, with coverage varying by province, plan, and whether testing sits inside an insured service (REGULATORY STATUS and CONVENTION). In Australia the device appears on the Australian Register of Therapeutic Goods as cognitive-assessment software (identifier 471105, Class IIa, entered November 2024), a registration rather than an efficacy endorsement; the national evidence-based ADHD guideline finds neuropsychological and computerized assessment not required for diagnosis and not shown to improve diagnostic accuracy enough for routine use, and the March 2026 Medicare Benefits Schedule contains no continuous-performance-test item, so any reimbursement rides on the practitioner's broader assessment service (REGULATORY STATUS, CONVENTION, and PAYER context). In all three countries the same sentence serves the reader: the T.O.V.A. contributes a standardized performance sample to a clinical evaluation that no jurisdiction requires and no instrument replaces. Instrument facts, psychometric honesty, and rights complete the page. Version discipline first: the current product is T.O.V.A. 9 on the 9.1 software line, official 2026 manuals reference different releases, and licensed installations vary, so the build identifier printed on the patient's report is the citable version. The visual and auditory versions keep separate norms (roughly 1,714 and 2,680 participants in the FDA filing) and the detailed manual describes historically regional, demographically homogeneous components in the visual norms, including child, adolescent, and adult samples from Minnesota described as approximately 99 percent White, so results deserve integration with the examinee's linguistic, cultural, educational, and sensory context. Reliability is stronger for response time and response-time variability than for individual error counts, commission performance shows practice effects on repeat administration, and diagnostic accuracy is genuinely moderate: an early clinical study reported sensitivity near 80 percent with specificity near 72 percent, a pediatric study found the test flagged about 30 percent of control children, response-time variability is the most reproducible ADHD-associated CPT finding while remaining nonspecific, and the pooled commercial-CPT meta-analysis landed at an area under the curve of 0.70 to 0.80; no current study supplies a T.O.V.A.-specific false-negative rate for high-ability or compensated presentations, a gap worth stating rather than filling with a guess. On rights: a licensed practice may print and save its own patients' reports into the chart, export is by file rather than any public API, and the licence bars redistribution of manuals, norms, report templates, and scoring logic, so public pages and tools may describe the paradigm but not clone the test, its stimuli, or its thresholds. T.O.V.A. is a registered trademark of The TOVA Company. BastionGPT is not affiliated with, or endorsed by, The TOVA Company. This page reproduces no stimuli, timing parameters, norms, thresholds, or report content. ↑ Back to the template and downloads ## Common T.O.V.A. documentation errors reviewers flag The numbers behind these errors are specific. The 510(k) clearance record limits the device to an aid role with asymmetric visual and auditory indications; a meta-analysis of 19 commercial-CPT studies pooled sensitivity near .75 and specificity near .71; and Aetna's ADHD policy classes computerized attention testing as experimental for the indication. The BastionGPT Clinical Advisory Board sees the same errors most often in T.O.V.A. documentation reviews: "FDA-approved objective ADHD test". Three inflations in four words. The T.O.V.A. is cleared, not approved: a 510(k) substantial-equivalence finding against a predicate device (the QbTest), which validates intended use, not diagnostic accuracy. The cleared language makes it an aid to assessment, interpretable only by qualified professionals, and the manufacturer's own manual states the device does not diagnose any disorder. Write "FDA-cleared aid to assessment" and the sentence survives review. Diagnosis or exclusion language. "TOVA confirms ADHD" or "TOVA ruled out ADHD." Guideline bodies in three countries treat CPTs as adjuncts, the AAP's evidence review concluded sole reliance cannot be recommended, and both error directions are empirically live: the test flagged roughly 30 percent of control children in one study, and people with ADHD, particularly high-ability adults, can perform normally under a short, novel, structured task. Document convergent, discordant, or noncontributory, never confirmed or excluded. The Attention Comparison Score overreached. Zero treated as a diagnostic boundary, the ACS quoted as an ADHD probability, or an "ADHD Score" narrative built from its former name. The ACS is a visual-only composite of three components (first-half response speed, second-half response sensitivity, overall response-time variability) expressing similarity to the manufacturer's ADHD comparison sample; below zero means more similar, above zero does not exclude, and a value near zero is a borderline similarity finding with limited meaning. It is absent from auditory reports. Modality and version blindness. "TOVA administered," with no visual-or-auditory statement and no build. The versions differ in norms, age ranges (visual about 4 through 80, auditory 6 through 29), FDA indication (treatment evaluation is visual-only), and ACS availability, and licensed installations run different 9.1 builds, so the report's printed identifier is part of the result. An auditory administration written up with visual conventions is wrong twice. Condition and medication blindness. A score table with no sleep, time of day, caffeine, prior CPT exposure, or medication timing. CPT performance moves with all of them, so the interpretable note records the conditions and states the medication name, dose, on-or-off status, last-dose time, and interval, under the prescriber's plan. "Must be tested unmedicated" is not a rule: whether to hold medication is a clinical decision tied to the question, and no generic washout applies across formulations. Billing shorthand that unbundles or upcodes. 96146 billed alongside time-based administration codes for the same automated result, or professional codes (96136 to 96137) used for technician work that belongs at 96138 to 96139, with interpretation services (96130 to 96133) claimed without a separately documented professional evaluation. Coding follows who administered, scored, and interpreted, and none of it creates coverage: several national payers class routine ADHD CPT testing as experimental or not medically necessary, so verify the member's policy first. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on ADHD evaluation reports. Give it the facts (modality and build, purpose, conditions, medication timing, validity findings, the standard-score pattern by period, the ACS, and the rest of the evaluation) and it drafts the results section: identification exact, validity first, the pattern in prose, the aid framing intact, and the integration statement running both directions, ready for your review. Cross-check a finished note for the gaps reviewers flag: approved-versus-cleared inflation, diagnosis or exclusion language, an ACS treated as a threshold, a missing modality or build, absent medication timing, or a retest gain with practice effects unmentioned. Draft the companion paragraphs: the family explanation of what the score does and does not mean, the titration-retest comparison with matched conditions, or the payer-facing sentence that places the CPT inside a medically necessary comprehensive evaluation. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does the T.O.V.A. actually measure? Performance on a 21.6-minute computerized go/no-go task (10.8 minutes for the preschool visual form), split into a low-target half that taxes sustained attention and a high-target half that taxes response inhibition. The report returns age- and sex-referenced standard scores (mean 100, SD 15) by quarter, half, and total for omission errors (missed targets), commission errors (responses to non-targets), response time, response-time variability (the most reproducible ADHD-associated CPT variable), and d-prime response sensitivity, plus anticipatory-response and validity information; halves and totals are separate aggregations of the underlying responses, not averages of quarter scores. Each variable has nonspecific causes: omissions can reflect fatigue or sensory access as well as inattention, commissions can reflect a speed-accuracy tradeoff as well as impulsivity, and variability moves with sleep, mood, pain, and engagement. It measures a constrained laboratory sample of attention and response control, never symptoms, impairment, onset, or the DSM criteria. What is the difference between the visual and auditory T.O.V.A.? They are distinct tests, not two skins on one instrument. The visual version is normed for roughly ages 4 through 80, carries the FDA indication to aid both assessment and treatment evaluation, generates the Attention Comparison Score, and hosts the performance-validity analysis for ages 17 and older. The auditory version is normed for ages 6 through 29, is cleared to aid assessment only, and produces no ACS. Their norms are separate, so their standard scores are not interchangeable, and a person can perform differently across modalities. Use the auditory version when a defined clinical question concerns auditory attention or when visual interpretation is limited, not routinely to generate more data: no strong evidence shows that administering both materially improves diagnostic accuracy over a good multimethod evaluation. Whichever is used, the note says which, because every downstream fact (norms, ages, indication, ACS) hangs on that word. What is the T.O.V.A. Attention Comparison Score? A visual-only composite, formerly called the ADHD Score and earlier the Attention Performance Index, built from three components: response speed from the first half, response sensitivity (d-prime) from the second half, and overall response-time variability. It compares the examinee's pattern with two manufacturer reference groups: a score below zero means the performance profile is more similar to the independently diagnosed ADHD comparison sample, and at or above zero more similar to the normative sample. What it is not: a DSM threshold, an ADHD probability, a severity rating, a medication rule, or a measure of real-world impairment, and a result above zero does not exclude ADHD, because many diagnosed people perform normatively on structured tasks. A value near zero is a borderline similarity finding with limited clinical meaning. Auditory reports mark it not applicable, and the proprietary equation stays in the manual. Defensible phrasing: "more similar to the manufacturer's ADHD comparison sample; an aid to assessment, not a diagnosis." Is the T.O.V.A. FDA-approved to diagnose ADHD? No, on both words. It is FDA-cleared, not approved: 510(k) K173915 (March 2018) found the system substantially equivalent to a predicate device, the QbTest, under product code LQD, with the regulatory class listed as unclassified; a 510(k) clearance addresses intended use and equivalence, not diagnostic accuracy. And the cleared indication is an aid: the visual version aids assessment and treatment evaluation, the auditory version aids assessment only, and results are to be interpreted by qualified professionals, with the manufacturer's manual stating the device does not diagnose any disorder. Abroad the registrations are similarly modest: a Health Canada establishment licence and an Australian ARTG software registration, neither an efficacy endorsement. Independent evidence matches the labeling: pooled commercial-CPT discrimination sits at an area under the curve of roughly 0.70 to 0.80, and guideline bodies conclude CPTs cannot carry a diagnosis alone. "FDA-cleared aid to the assessment of attention deficits" is the whole defensible claim. Should the T.O.V.A. be given on or off medication? It depends on the question, and the prescriber decides. For an untreated baseline, an off-medication administration can be appropriate when the prescriber has planned the hold and it is safe; for treatment monitoring, test on the usual regimen or at a deliberately chosen post-dose interval, commonly near expected peak effect. Either way the chart records the medication name and dose, on-or-off status, the last dose's date and time, the elapsed interval, and whether the status reflected usual treatment, a prescriber-directed hold, or a planned comparison. Two practices to avoid: instructing a patient to stop a stimulant because "the TOVA must be done unmedicated" (medication decisions belong to the prescriber and the clinical question), and applying a generic washout duration across formulations, which pharmacology does not support. A dose-to-test interval that goes unrecorded quietly invalidates the retest comparison the whole exercise was for. The history is classic ADHD but the T.O.V.A. was normal. Does that rule ADHD out? No. A normal result shows that performance stayed broadly age-referenced during one short, quiet, novel, externally structured session; it says nothing about regulation across prolonged, distracting, self-directed real-world demands, which is where ADHD lives. False negatives are an accepted feature of the instrument class: guideline bodies note that many people who meet criteria perform normally on standardized tasks, high-ability and compensated adults do so with particular frequency (no T.O.V.A.-specific false-negative rate exists, a gap to state rather than fill), and diagnosis rests on developmental history, cross-setting symptoms, impairment, and informant evidence. Defensible documentation: the result was discordant with, but does not invalidate, the cross-setting clinical evidence, and the T.O.V.A. was considered non-exclusionary rather than used to overrule the history, ratings, records, and observed impairment. The reverse discipline also holds: an abnormal result in someone without the clinical picture flags a performance difference to explain, not a diagnosis to award. What makes a T.O.V.A. administration invalid? Anything that breaks standardization or the response record: excessive anticipatory responses (responses too fast to be stimulus-driven, suggesting guessing, rhythmic responding, or misunderstanding), too few usable correct responses, room interruptions, hardware or administration errors, observed failure to follow instructions, or behavior that undermines the norms' assumptions. The current software adds an unusual-pattern or performance-validity analysis with deliberately restricted applicability: visual administrations, ages 17 and older, and only when the data conditions for it are met, so its absence from a report is often correct rather than an omission. Two rules govern the write-up. Validity is resolved before any score is interpreted, because a score table cannot outrank an unresolved standardization problem. And a flag is a caution requiring clinical adjudication, not a finding of malingering or deception; if effort or reporting style genuinely concerns you, that conclusion needs converging evidence from the broader evaluation, never a single CPT output. Can the T.O.V.A. monitor stimulant treatment? The visual version carries exactly that FDA-cleared indication, and the literature supports a group-level story: response-time variability, the variable most associated with ADHD, improves under stimulant treatment. The individual-level discipline is what the note must carry. Compare against a true baseline with matched conditions (time of day, sleep, hardware, setting), document the interval, prior exposure, and expected practice effects (commission performance in particular improves on repetition), and record the medication, dose, and dose-to-test timing, since a peak-effect test and a trough test are different experiments. Reliability is stronger for timing measures than for individual error counts, so small error-count changes prove little. Conclude in performance terms: "more consistent under the medicated condition," integrated with symptom ratings, classroom or work function, side effects, sleep, appetite, and vitals, and never as the sole basis for dose selection. There is no validated retest schedule; retest when the result could change a management decision and the comparison can be interpreted. Can BastionGPT draft my T.O.V.A. documentation? Yes. Give it the facts (modality and build, purpose, testing conditions, medication and last-dose timing, validity findings, the standard-score pattern by quarter and half, the ACS where applicable, and the surrounding evaluation) and it drafts the results section: identification exact, validity resolved first, the pattern described in prose, the Attention Comparison Score framed as an aid, and the integration statement running both directions, ready for your review. It can also cross-check a finished note for approved-versus-cleared inflation, diagnosis or exclusion language, an ACS treated as a threshold, a missing modality or medication interval, or a retest gain with practice effects unmentioned, and it can draft the family explanation and titration-retest comparisons. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified August 2026: Regulatory record: FDA, 510(k) K173915 clearance (March 2018; indications for use; product code LQD; predicate QbTest; norm totals; administration lengths); TGA, ARTG entry 471105 (Class IIa software registration, November 2024); Health Canada, medical device establishment licensing guidance (effective December 2026). Company record, accessed August 2026: The TOVA Company, current-version page (T.O.V.A. 9), Professional Manual (9.1 line; score architecture; ACS; validity analysis; norm histories), and licence terms (report printing; prohibited uses). Professional guidance: AAP, 2024 evidence review of ADHD diagnostic tools (sole CPT use cannot be recommended); AACAP, ADHD practice parameter ; CADDRA, Canadian ADHD practice guidelines 4.1 ; AADPA, Australian evidence-based ADHD guideline . Payer policies: Aetna, ADHD clinical policy bulletin 0426 (CPTs experimental for the indication); Centene, ADHD assessment and treatment policy ; Cigna, neuropsychological testing policy (effective May 2026). Psychometrics: meta-analysis of diagnostic accuracy across 19 commercial CPT studies (pooled AUC 0.70 to 0.80); Kofler MJ and colleagues, 2013, reaction-time variability meta-analysis of 319 studies ; Llorente AM and colleagues, 2001, T.O.V.A. test-retest reliability in children with ADHD ; Leark RA and colleagues, 2004, practice-effects study . ## Related templates Conners 4 Report Write-Up CAARS-2 Report Write-Up ASRS v1.1 Score Documentation Vanderbilt ADHD Scale Documentation BRIEF-2 Report Write-Up ADHD Evaluation Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cans All note templates # CANS Assessment Documentation: Ratings, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The CANS (Child and Adolescent Needs and Strengths) is a communimetric assessment, copyrighted by the Praed Foundation and free to use with annual certification, that rates a child's needs and strengths on four action levels rather than producing a score. Care coordinators, caseworkers, and behavioral health clinicians complete it at intake and reassessment in Medicaid and child welfare systems. This page covers how to document CANS ratings, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Care coordinators, child welfare caseworkers, therapists, and other assessors who hold current CANS certification for the version in use (annual vignette-based certification through TCOM Training at a reliability of 0.70 or higher); with training, bachelor's-level staff can rate reliably, though some versions and programs require a higher degree or a licensed clinician Audience Child and family teams, youth and caregivers, Medicaid managed care and Health Home reviewers, state child welfare and juvenile justice agencies, level-of-care and placement decision makers, auditors, and courts in placement and permanency proceedings Typical length 300 to 700 words of narrative beside the rating sheet (version, window, sources, actionable items with justification, strengths, consensus, plan linkage) · 30 to 60 minutes to write after the rating (clinical team estimate) Format family Communimetric needs-and-strengths assessment (core items grouped by domain, each rated on a four-level action scale, with extension modules opened by trigger items; state and program versions vary) When it's used Medicaid intensive care coordination, wraparound, and Health Home enrollment; behavioral health intake and periodic reassessment; child welfare case planning and placement decisions, including Family First qualified residential treatment reviews; juvenile justice and early childhood programs; level-of-care and eligibility algorithms Standards context Copyright Praed Foundation (Standard Comprehensive CANS 3.0 reference guide, January 2021), an open domain tool free to use with training and annual certification; required by no federal law, mandated by state Medicaid, child welfare, and managed care programs in their own versions; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CANS? The Child and Adolescent Needs and Strengths is a multipurpose information-integration tool developed by John S. Lyons and first released in 1999, built on the methodological approach of his earlier Childhood Severity of Psychiatric Illness scale and maintained by the Praed Foundation (the John Praed Foundation, Chicago) inside its Transformational Collaborative Outcomes Management framework, TCOM. The current core release is the Standard Comprehensive CANS 3.0, whose reference guide is dated January 13, 2021 and carries a copyright line of 2021, 2017, and 1999 for the 6 through 20 age form. Core items are grouped into domains covering life functioning, strengths, cultural factors, caregiver resources and needs, behavioral and emotional needs, and risk behaviors, and a rating on a trigger item opens an extension module (developmental needs, trauma, substance use, violence, sexually aggressive behavior, runaway, juvenile justice, and fire setting in the 3.0 Comprehensive). Praed reports versions in use in all 50 states across child welfare, mental health, juvenile justice, and early intervention, with early childhood forms in state implementations and an adult sibling, the Adult Needs and Strengths Assessment (ANSA). Training and annual certification run through TCOM Training, and the Foundation describes the tool as open domain and free to use, holding the copyright to keep it that way. Two facts carry the documentation. First, the CANS is a communimetric tool, designed on communication theory rather than psychometric theory: each item exists because it could change a planning pathway, and each is rated on four levels that translate into action rather than severity. For a need, the levels run from no evidence and no action, through a history or emerging concern that calls for watchful waiting, prevention, or further assessment, to a need that interferes with functioning and must be addressed in the plan, to a dangerous or disabling need that requires immediate or intensive action; for a strength the order reverses, from a centerpiece strength the plan can be built around, to a useful strength, to a strength that must be built, to none identified yet. There is no context-free total: Praed's guide monitors outcomes by following items that started at an action level or by summing items within a domain, and several states run decision-support algorithms on item patterns, but no universal global score exists and none of those summaries replaces the list of actionable items. Second, the version is part of the result. Illinois's IM+CANS, California's CANS-50, MassHealth's CANS, New York's CANS-NY, Texas's CANS 3.0, Ohio's Children's Initiative CANS, Michigan's MichiCANS, and Indiana's birth-to-five and 6-to-17 forms differ in items, modules, windows, cadence, and certification course, and a jurisdiction's version number is not the Praed core number (New York launched CANS-NY 2.0 in November 2023; Texas moved to its CANS 3.0 on January 1, 2025), so "CANS 2.0 completed" identifies nothing. Norm-referenced instruments such as the BASC-3 answer a different question, how unusual a behavior is against a reference population; the CANS answers what the team should do next, which is why its output belongs in the treatment plan rather than in a score report. ## Who uses CANS documentation and when The CANS is completed by whoever the mandating program names as the assessor, and the reader is almost never only the author. Care coordinators in Medicaid intensive care coordination, wraparound, and Health Home programs (OhioRISE, New York's children's Health Homes) rate it at enrollment and at each reassessment; community mental health clinicians in Illinois complete the IM+CANS as the state's integrated assessment and treatment planning instrument; MassHealth providers complete it for members under 21 at the initial behavioral health assessment and at intervals; child welfare caseworkers in Texas, Indiana, Wisconsin, and elsewhere complete it after removal, at case junctures, and before placement decisions; and juvenile justice and early childhood programs run their own versions. The write-up sits beside an intake note or biopsychosocial assessment , draws on collateral contacts and school collateral , and feeds the treatment plan , the care coordination note , and the case management note that record what the team then does. Neighbors win in three places: the BASC-3 or CBCL when the question is how a child compares with a reference population, the CAFAS or PECFAS when a program wants a functional impairment score, and the ANSA once the person is served as an adult; California pairs the CANS with the PSC-35, a member of the Pediatric Symptom Checklist family, so that a symptom measure and a planning tool travel together. ## How to document a CANS assessment in the chart No federal law and no Praed rule prescribes a CANS note format; the state or program version sets the form, the window, the cadence, and the certification course, and the Foundation's guide sets the logic. What survives review is a record that names the version and purpose, dates the window and lists the sources, identifies the rater and the certification behind the ratings, justifies every action-level item in original prose, keeps strengths and caregiver ratings as their own targets, records consensus and disagreement instead of a total, and carries each actionable item into the plan. Each element below carries the pitfall that most often undermines it. Version, form, and purpose. Write the exact instrument: the state or program version and the Praed core it descends from where known ("Standard Comprehensive CANS 3.0, 6 to 20 form, in the state's intensive care coordination version"), the age form, and the modules opened, plus the reason for this administration: initial assessment, periodic reassessment, reassessment after a significant change, transition, or discharge, and the program that requires it. If two programs with different cadences touch the same child (Illinois reviews every 180 days, New York's Health Homes annually), name the one that governs this record. Pitfall: "CANS 2.0 completed" with no state named: New York's 2.0 and Texas's retired 2.0 are different instruments, and a reviewer cannot tell which cadence, module set, or certification course applies. Rating window and information sources. State the window as dates. The Praed guide uses a 30-day window so that ratings stay current, allows the action levels to override it (a serious historical risk can remain actionable), and defers to each version's own rules, so name the window your version prescribes and say when historical information drove a rating. Then list the sources actually integrated: youth interview, caregiver interview, observation, collateral from school or prior providers, and the records reviewed, with dates. If the assessment was completed from records alone, say so, name the records, and state what the family's absence leaves uncertain and when direct participation will be sought. Pitfall: An undated window, or a chart-abstraction CANS presented as a family-engaged assessment in a program whose rules require the family at the table. Rater and certification. Name the rater, role, and credentials, then the certification that stands behind the ratings: the course or version, the date passed, and the expiry. Praed expects training and annual certification for appropriate use, most programs require the rater to hold current certification in their own version, and Praed's FAQ tells users to recertify in the same course as the original certification, with an Ohio exception, and deactivates accounts more than three months past expiry. Printing the certification in the note is an audit convention unless the program requires it; the underlying requirement to be current is the program's rule, and a rater whose certification lapsed or covers a different version has produced a rating the program may not accept. Pitfall: Portal access or last year's course mistaken for current certification in the version the program mandates, discovered at audit or when a claim is reviewed. Actionable needs, with a justification for each. For every need rated at an action-requiring level, write one to three sentences in your own words: what is happening, how it affects functioning or safety, whose account or which record supports it, and what already helps. Do not paste the manual's anchor wording into the chart; the anchor is a rating aid, the justification is the clinical evidence. Give watchful-waiting ratings a reason too (history, an emerging concern, or informant disagreement), because a 1 is a monitoring decision, not a null. Where the version opened a module, say which trigger opened it and summarize the module findings at the same level. Pitfall: Anchor text copied as justification, or a rating of 1 treated as nothing to do and dropped from the plan and the next review. Strengths and caregiver ratings as separate targets. Name the centerpiece and usable strengths and say how the plan will use them (a reliable adult relationship, a talent, a school or community connection), then the strengths the plan must build. Rate the caregiver domain on the caregiver's own needs and resources (supervision, knowledge, health, stress, housing, and the like as the version frames them), never as a restatement of the child's behavior, and keep the youth's and caregiver's perspectives visible as separate accounts. A caregiver-level need is a plan target for the caregiver, with its own goal or referral. Pitfall: Caregiver items rated from the child's symptoms, or strengths listed with no planning use, so the plan runs on deficits alone. Consensus, disagreement, and the summary (no total). Record how consensus was reached with the youth and family, whose account supports each actionable rating, and any disagreement as it stands: both perspectives, the contextual evidence, the action level the rater settled on, and when it will be revisited. Do not average two informants' views into one number. Summarize the assessment as its actionable pattern (which needs require action, which require immediate action, which strengths are usable) rather than a sum; if the program computes domain scores or runs an eligibility or level-of-care algorithm, report that output by name, as decision support, with the program cited. Pitfall: "CANS total 27, moderate," a number no version defines, or an algorithm tier entered as the clinical conclusion instead of the program's decision-support output. Plan linkage and the reassessment plan. Close the loop: each action-level need must be traceable to a plan response (a goal, an intervention, a referral, a safety response, a care-coordination task, or explicit monitoring), and the usable strengths should appear inside the interventions. Praed's model expects action-level needs to be addressed in planning, and Indiana's child welfare policy makes the same rule explicit for needs at the two action levels and strengths at the two usable levels. Then set the next CANS by the program's cadence or sooner on significant change, and at reassessment describe change in the evidence (what happened in the window, what the family reports, what the plan delivered), distinguishing real change from better information. Pitfall: Actionable items that never reach the treatment plan , or "improved from 3 to 1" with nothing to show what changed. ### Blank template (copy and adapt) CANS DOCUMENTATION BLOCK Date: [ ] Setting / program: [ ] Rater (name, role): [ ] Purpose: [initial / periodic reassessment / significant change / transition / discharge] Program requiring it: [ ] Version: [state or program version + Praed core and form, e.g. Standard Comprehensive 3.0, 6 to 20] Modules opened: [ ] Certification: [course or version, date passed, expiry] Rating window: [mm/dd/yyyy to mm/dd/yyyy; version rule; historical items noted] Sources integrated: [youth interview / caregiver interview / observation / school or provider collateral / records reviewed, with dates] Records-only: [no / yes: records used, why, limits, when family will be seen] Actionable needs (domain, item area, level, justification in own words): Immediate or intensive action: [ ] Action needed: [ ] Watchful waiting (reason: history / emerging / disagreement): [ ] Strengths: [centerpiece and usable, with planning use] To build: [ ] Caregiver domain: [caregiver's own needs and resources; plan target] Consensus and disagreement: [how reached; whose account supports what; both perspectives kept; revisit date] Summary: [actionable pattern; no total; program domain score or algorithm output named as decision support, if any] Plan linkage: [each action-level need to goal / intervention / referral / safety response / monitoring; strengths used] Next CANS: [program cadence or sooner on change] Prior CANS: [date, version] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces CANS items, anchor definitions, or any state version's item list. ## Sample CANS documentation (fictional) Scenario: an initial CANS at intake to a state Medicaid intensive care coordination (wraparound) program for a 14-year-old with mood and school-functioning needs, documented by a care coordinator who names the version, window, sources, and certification, justifies each actionable item, records a youth-caregiver disagreement, and carries every action-level need into the treatment plan. All details are fictional. Patient: K.R., 14 · Setting: Community behavioral health agency, Medicaid intensive care coordination (wraparound) intake · Clinician: M. Delgado, LCSW, care coordinator · Note date: 09/08/2026 Version, purpose, and rater: Initial CANS for enrollment in the state's Medicaid children's intensive care coordination program, completed in the program's required version (built on the Standard Comprehensive CANS 3.0, 6 to 20 form). No trigger item reached the level that opens an extension module, so none was completed. Rater: M. Delgado, LCSW, care coordinator; certification in this version current through May 2027 (TCOM Training); the program requires current certification for every enrollment CANS and reassessment, and this note records it for the audit file. Rating window and information sources: Window: 08/09/2026 to 09/08/2026, the version's 30-day period; one risk item is rated on history outside the window, as noted below. Sources: youth interview at home on 09/08/2026; interview with his mother, the sole caregiver, on 09/08/2026; telephone collateral with the school counselor on 09/04/2026; the outpatient therapy discharge summary dated 08/21/2026, reviewed with the mother's signed release; observation of the home. No prior CANS on file. Actionable needs, with justification: Immediate or intensive action, life functioning, school: K.R. attended 9 of 20 school days in the window, is failing three classes, and the district has scheduled an attendance review for late September; the counselor reports that he was engaged through last spring. Action needed, behavioral and emotional, mood: over the past month he has stopped seeing friends, sleeps until noon on missed school days, and told his mother and the counselor that school is pointless; both describe a change since a close friend moved away in July. Direct risk screening was completed today: he denies any thoughts of harming himself, no self-harm behavior is reported by any source, and the screening is documented in the risk screening note. Action needed, behavioral and emotional, anxiety: he describes stomach aches and dread on school mornings, confirmed by his mother, and links them to a class in which he was mocked in May; the discharge summary records the same pattern. Watchful waiting, life functioning, sleep: the reversed sleep schedule is rated for monitoring as part of the mood need rather than as a separate action. Watchful waiting, risk behaviors: one overnight absence from home without permission in April 2026, reported by the mother and recorded in the discharge summary; none in the window, so the item is held at a monitoring level on history and will be reviewed at reassessment. Strengths and caregiver ratings: Centerpiece strength, family: a reliable relationship with his maternal uncle, who lives nearby, sees him weekly, and is willing to take a role in the plan. Usable strength, talents and interests: he plays guitar daily and has asked about the school music program. Strength to build, community and school connection: no current activity or peer group outside the home. Caregiver domain, rated on the mother's own situation: supervision is an action-level need because her shift work leaves K.R. alone on three weekday mornings, which is when most absences occur; caregiver knowledge of adolescent mood is a watchful-waiting item, as she asks for guidance on how to respond; her own stress is rated for monitoring at her request, with no current need for her own treatment identified. Housing, income, and health are not actionable. Consensus and disagreement: The mother and the school counselor describe the school problem as the most urgent need; K.R. describes it as limited to one class and does not see it as requiring intervention. Both accounts are recorded. The immediate-action rating reflects the attendance record, the failing grades, and the pending district review rather than an average of opinions, and K.R.'s view remains visible in the plan and will be revisited at the first child and family team meeting on 09/15/2026. Youth and mother agreed on the mood and anxiety ratings and on the uncle and music as strengths. No total score is reported; the program's enrollment determination will use its own review of the actionable pattern, and this note records that the CANS informs that determination rather than making it. Plan linkage and reassessment: Every action-level need is carried into the treatment plan dated 09/08/2026: school attendance and the district review (goal 1, with the uncle providing morning transport on the three unsupervised mornings and the counselor as the school contact); mood, with the risk screening repeated at each visit (goal 2, referral for individual therapy within two weeks); school-morning anxiety (goal 3, folded into the therapy referral and a meeting with the class teacher); caregiver supervision and knowledge (goal 4, a caregiver support session and a parent guidance resource). Music program enrollment is an intervention under goal 1. The sleep and prior-absence items are monitoring tasks reviewed at each team meeting. Reassessment: the program's 90-day cadence, sooner on any significant change, on the same version and form. Ratings entered in the state CANS system with this narrative attached; the mother received a plain-language summary of the actionable items and the plan. This sample is fictional and for educational purposes. It does not describe a real child, family, or record; the ratings, dates, and details are invented to show documentation structure and are not clinical guidance. No CANS items, anchor definitions, or scoring materials are reproduced. ↑ Back to the template and downloads ### Why this sample works The version, form, purpose, window, sources, and certification are all named up front, so a reviewer can tell which program's rules apply, what the ratings rest on, and that the rater was authorized to make them. Every action-level need carries a justification in original prose (what is happening, its effect, whose account supports it) with no anchor text copied, and the watchful-waiting items carry their reasons instead of disappearing. Strengths are named with a planning use, and the caregiver domain is rated on the mother's own situation and turned into its own plan goal rather than restated from the child's behavior. The youth-caregiver disagreement is recorded as it stands, the rating is explained by evidence rather than an average, and the note reports the actionable pattern with no total score and no eligibility conclusion. Each actionable item is traceable to a numbered goal, intervention, or monitoring task in the dated treatment plan, and the reassessment is set by the program's cadence on the same version, which is the golden thread reviewers look for. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no federal law names the CANS, and the authorities that do name it are state programs. Under LAW, Medicaid's EPSDT benefit entitles children under 21 to medically necessary screening, diagnostic, and treatment services and leaves the assessment infrastructure to the states, and the Family First provisions of Title IV-E (42 U.S.C. 675a) require a qualified individual to assess a child's strengths and needs with an age-appropriate, evidence-based, validated functional assessment tool within 30 days of placement in a qualified residential treatment program, with family and permanency team involvement and court review within 60 days; the statute names no instrument, and New York's use of the CANS-NY inside that process is a state implementation choice. Under PAYER POLICY the mandates are specific and dated: Illinois HFS designates the IM+CANS as the approved instrument for its Integrated Assessment and Treatment Planning process, requires it for IATP reimbursement, requires review and update at least every 180 days, and requires annual certification through PATH (current materials effective April 1, 2024); Ohio bills the Ohio Children's Initiative CANS under H2000 for assessments on or after July 1, 2022, payable only to raters with current Praed certification in that version who have added the CANS Assessor specialty to their Medicaid enrollment; New York's children's Health Homes moved the CANS-NY from every six months to annually in May 2022, with significant life events triggering an earlier reassessment, and launched CANS-NY 2.0 in November 2023; MassHealth requires the CANS for members under 21 and moved its routine reassessment interval from 90 to 180 days in 2023; Texas DFPS moved to its CANS 3.0 on January 1, 2025 for children in conservatorship aged three and older, adding Medical Health and Exploitation modules, letting 2.0 certification stand until the rater's next annual cycle, and timing reassessment by age, acuity, and service rather than one interval; Indiana DCS Policy 5.19 (version 6, January 1, 2025) requires an initial CANS within five calendar days of specified case milestones, reassessment at least every 180 days and at case closure, current certification for the age form used, and the incorporation of needs at the two action levels and strengths at the two usable levels into the case or prevention plan, tied to outcomes and activities; California pairs the CANS with the PSC-35 as its statewide outcome tools for county behavioral health and, as of September 2026, is transitioning its core from the CANS-50 toward a cross-system IP-CANS, so check the current DHCS notice before citing a version. Under CONVENTION sit the rules clinicians assume are universal and no authority states: that a total CANS score exists and means something, that a rating of 1 needs no response, that the CANS diagnoses, that one certification covers every state version, and that federal law requires the tool. Medical necessity follows the same split: a CANS informs level-of-care and eligibility determinations through the program's own algorithm or review (OhioRISE tiers, CANS-NY acuity, Texas service packages, Illinois decision support, the Illinois child welfare placement algorithm), and the note should say the CANS informed the determination rather than that a rating established it; where a medical necessity letter is needed, the actionable pattern and its functional evidence are the content, not a sum. Canada and Australia use other instruments, and the evidence base rewards precision about what CANS reliability means. No Canadian federal or provincial mandate for the CANS was found for this page; Ontario's Child and Parent Resource Institute has supported implementation of the interRAI child and youth suite, whose Child and Youth Mental Health assessment carries embedded scales and decision support, across about 70 organizations (toolkit updated April 20, 2026), so a Canadian CANS is local practice and should be documented as such (CONVENTION). Australia's National Outcomes and Casemix Collection runs the public child and adolescent mental health outcome measures on the HoNOSCA and its companions, and no Australian jurisdiction adopts the CANS (PAYER POLICY through the national collection, not LAW). On the evidence: the Praed guide reports an average vignette reliability of 0.78 across more than 80,000 trainees, 0.84 with case records, and above 0.90 with live cases (Lyons, 2009), sets 0.70 on a test vignette as the certification threshold, and cites Anderson and colleagues' chart-review study (60 cases, 2003) for item-level reliability; Accomazzo and colleagues (2017) compared individual items, actionable counts, subscales, and a total composite for the strengths domain and found individual items the most interpretable for practice; and Chor and colleagues (2015), in 3,911 Illinois child welfare placement records over four years, found that placements concordant with the CANS-based algorithm consistently predicted improvement while discordant placements varied. No peer-reviewed study quantifies how much rater agreement decays between annual certifications; annual recertification is a fidelity policy, not a measured decay curve, and the developer-authored communimetric literature establishes the theory of use rather than independent validation. The qualification boundary is a version and program question: Praed states that with training anyone with a bachelor's degree can rate the tool reliably while some versions and applications require a higher degree or relevant experience, certification is course-specific in practice (recertify in the original course, Ohio excepted; accounts more than three months past expiry are deactivated), and a state transition (Texas 2.0 to 3.0, New York's 2.0 launch) sets its own migration rule, so the note names the version and the certification that matches it. The Child and Adolescent Needs and Strengths (CANS) is copyrighted by the Praed Foundation and made available as an open domain tool, free to use with training and annual certification, within its Transformational Collaborative Outcomes Management (TCOM) framework; the Foundation states that it holds the copyright to keep the tool free, its reference guide directs anyone seeking specific permission to use the tool to contact the Foundation, state and program versions are produced in collaboration with it, and no published authority establishes CANS or TCOM as registered trademarks, so reproducing the item set on a form, embedding it in an EHR, or building a public scoring tool are permission questions rather than automatic rights. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CANS documentation errors reviewers flag The numbers behind these errors are specific. The Praed guide reports average vignette reliability of 0.78 across more than 80,000 trainees , 0.84 with case records, and above 0.90 with live cases, and sets the certification threshold at 0.70 on a test vignette; item-level reliability rests on a 60-case chart-review study (Anderson and colleagues, 2003); Accomazzo and colleagues (2017) found individual items more interpretable than actionable counts, subscales, or a composite; and in 3,911 Illinois placement records , placements concordant with the CANS algorithm consistently predicted improvement while discordant ones varied (Chor and colleagues, 2015). The cadence facts age on their own: Illinois every 180 days, New York annually since May 2022, Texas on its 3.0 since January 1, 2025, Indiana every 180 days and at closure since January 1, 2025, and Praed deactivating accounts three months past certification expiry. No published audit counts how often a chart CANS omits its version, window, or certification; the program rules under compliance considerations are what turn those omissions into a payment or compliance problem. The BastionGPT Clinical Advisory Board sees the same errors most often in CANS documentation reviews: A total score that no version defines. "CANS score 27, moderate," or "CANS improved six points." The tool is item and action oriented; Praed monitors outcomes by following items that started at an action level or by summing within a defined domain, and states run their own algorithms, but no universal total exists and no severity band attaches to one. Report the actionable pattern, and name any domain score or algorithm output as the program's decision-support output. The version left unnamed, or a bare version number. "CANS completed" or "CANS 2.0" with no state or program. New York's CANS-NY 2.0 and Texas's retired 2.0 are different instruments with different modules, cadences, and certification courses, and a reviewer cannot apply the right rule to an unnamed version. Write the state or program version, the Praed core and form where known, and the modules opened. Anchor text as justification, or no justification. Manual wording pasted beside each rating, or a list of levels with nothing under them. The anchor is a rating aid; the record needs the evidence: what is happening, its effect on functioning or safety, whose account supports it, and what already helps, in original prose for every action-level item, and a stated reason for every watchful-waiting item. A rating of 1 treated as nothing to do. Watchful-waiting items dropped from the plan and never revisited. A 1 is a monitoring decision that can rest on history, an emerging concern, or unresolved informant disagreement; record the reason, assign the monitoring task, and review it at the next team meeting and reassessment. Caregiver and youth ratings blended, disagreement averaged. Caregiver items rated from the child's symptoms, or a youth's and a parent's views split into a middle number. The caregiver domain rates the caregiver's own needs and resources and becomes a caregiver-level plan target; disagreement is recorded as two accounts with the evidence and the rater's action-level judgment, and revisited, never averaged. Actionable items that never reach the plan, and change with no evidence. A CANS with five action-level needs beside a treatment plan with two goals, or "improved from 3 to 1" with nothing to show what happened in the window. Trace every action-level need to a goal, intervention, referral, safety response, or monitoring step, use the strengths inside the interventions, and describe change in the facts, distinguishing real change from better information, on the same version with a rater whose certification is current. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health intake assessments and treatment plans. Give it the facts (version and form, purpose, window dates, sources, rater and certification, the actionable items with what you observed and who reported it, strengths, caregiver findings, disagreements, and the plan goals) and it drafts the documentation block: version and window named, a justification in original prose for every action-level item, strengths and caregiver ratings as separate targets, consensus and disagreement recorded, no total, and each actionable item traced to the plan, ready for your review. Cross-check a finished note for the gaps reviewers flag: an unnamed version, an undated window, missing certification, anchor text used as justification, a watchful-waiting item with no reason, caregiver items rated from the child's behavior, a total score, or an action-level need with no plan response. Turn the entry into the next document: the treatment plan goals that carry each actionable item, the reassessment narrative that describes change in evidence, or the plain-language summary of actionable items and strengths for the youth and caregiver, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do CANS ratings of 0 to 3 mean, and is there a total score? They are action levels, not severity scores. For a need, the four levels run from no evidence and no action; to a history or emerging concern that calls for watchful waiting, prevention, or further assessment; to a need that interferes with functioning and must be addressed in the plan; to a dangerous or disabling need that requires immediate or intensive action. For a strength the order reverses: a centerpiece strength the plan can be built around, a useful strength, a strength that must be built, and no strength identified yet. The item-level anchors that guide a rater to one of those levels are the Praed Foundation's copyrighted text and belong in the manual, not the chart. There is no universal total: Praed's guide monitors outcomes by following items that started at an action level or by summing items within a domain, and some states run algorithms on item patterns, but no context-free global score exists, no severity band attaches to one, and Accomazzo and colleagues (2017) found individual items the most interpretable summary for practice. Report the actionable pattern, and name any program-defined domain score or algorithm output as decision support. Which CANS version do I document, and does one certification cover all of them? Document the version your program mandates, by its full name, and treat certification as course-specific until the program says otherwise. Versions in use include Illinois's IM+CANS, California's CANS-50 (transitioning toward an IP-CANS), MassHealth's CANS, New York's CANS-NY (2.0 since November 2023), Texas's CANS 3.0 (since January 1, 2025), Ohio's Children's Initiative CANS, Michigan's MichiCANS, and Indiana's birth-to-five and 6-to-17 child welfare forms, and each sets its own items, modules, window, cadence, and certification course. Praed's FAQ tells users to recertify in the same course as their original certification, with an Ohio exception, and deactivates accounts more than three months past expiry; Ohio pays H2000 only to raters with current certification in the Ohio Children's Initiative CANS; Texas let 2.0 certification stand only until the rater's next annual cycle after its 2025 transition. So a certificate from one state's course does not by itself satisfy another program's requirement, and a note that says only "CANS 2.0" leaves a reviewer unable to tell which instrument, cadence, or course applies. Write the program version, the Praed core and form where known, and the certification that matches it. How often does the CANS have to be repeated? As often as the mandating program says, and there is no single answer. Illinois requires the IM+CANS and treatment plan to be reviewed and updated at least every 180 days; Indiana DCS requires a CANS within five calendar days of specified case milestones, at least every 180 days, and at case closure (Policy 5.19, January 1, 2025); New York's children's Health Homes moved to an annual CANS-NY in May 2022, with significant life events requiring an earlier one; MassHealth moved its routine interval from 90 to 180 days in 2023, with intake and discharge administrations; Texas DFPS times reassessment by the child's age, acuity, and services under its periodicity schedule; Ohio and other programs set their own intervals. Layered on every calendar is the clinical rule: reassess after a significant change, at transition or discharge where required, and never carry a stale CANS into a new determination. Document the cadence that governs this record, when the next CANS is due, and the version it will use, and describe change at reassessment in the evidence rather than as arithmetic; the treatment plan review page covers the review documentation that sits beside it. Does every need rated 2 or 3 have to appear in the treatment plan? Every action-level need has to be addressed in the plan; it does not have to be its own goal. Praed's model links the two action-requiring levels directly to service planning, Indiana's child welfare policy requires needs at those levels and strengths at the two usable levels to be incorporated into the case or prevention plan and tied to outcomes and activities, and Illinois built the IM+CANS into an integrated assessment and treatment planning process for the same reason. The defensible test is traceability: a reviewer should be able to follow each action-level need to a goal, an intervention, a referral, a safety response, a care-coordination task, or an explicit monitoring decision with a rationale, and related needs can share one integrated goal unless a local policy requires a one-to-one structure. The reverse check matters as much: a plan goal with no actionable need behind it, or a strength the plan never uses, is the mismatch auditors flag first. The treatment plan page owns how goals, objectives, and interventions are written once the CANS has set the priorities. How do I rate and document when the youth and caregiver disagree? Record both accounts, decide the action level on the evidence, and keep the disagreement visible. TCOM's values make voice and choice and consensus central, but consensus is the outcome of a collaborative process, not a requirement that informants agree, and New York's OPWDD guidance states that each caregiver perspective is equally valid. A defensible entry says what the youth described, what the caregiver and any collateral described, what contextual evidence exists (attendance, records, observation), which action level the rater settled on and why, and when the team will revisit it; unresolved disagreement can itself justify a watchful-waiting rating. Do not average two views into a middle number, and do not resolve the conflict by omitting the youth's account. Bring it to the child and family team and record the discussion in the team meeting note ; if collateral sources supplied the deciding evidence, the collateral contact note holds the detail. Can I complete a CANS from records only when the family is unavailable? Sometimes, and only as the program allows. The CANS is designed to integrate information from interviews, observation, collateral, and records, and Anderson and colleagues (2003) showed that researchers rating 60 cases from chart review produced reliable item-level ratings, which is why record-based completion is legitimate for audit and research. Clinical programs are stricter: TCOM practice puts the youth and family at the center of the assessment, and many state workflows require family engagement, so a records-only CANS may not satisfy the program even if the state system accepts the data. If it is unavoidable and permitted, write that the assessment was completed from records, list the records used with dates, explain why direct informants were unavailable, state what remains uncertain, and set the date direct participation will be sought; treat the result as provisional at the next team meeting. Never cite the chart-review study as permission to bypass a program's engagement rule. Can we put the CANS items on our own form, in our EHR, or in a free web tool? Ask the Praed Foundation, because open domain is not public domain. The Foundation describes the CANS as an open domain tool that is free for anyone to use with training and certification, states that it holds the copyright to keep the tool free, and, in the Standard Comprehensive CANS 3.0 reference guide, directs anyone seeking specific permission to use it to contact the Foundation; its About page confirms that it maintains the copyrights on the CANS and its sibling tools. State and county versions are produced in collaboration with Praed, and state data systems and EHR vendors operate under arrangements with it, so a form that follows an approved state version inside that program's implementation is generally covered while redistribution outside it is a permission question. No published authorization covers a public web tool that displays or scores the items, and a scorer used by people without certification falls outside Praed's stated conditions for appropriate use, so treat that as needing written permission. No published authority establishes CANS or TCOM as registered trademarks; the right attribution is copyright, not a trademark symbol. This page and its templates stay on the safe side of that line: structure in words, no items, no anchors. Does a CANS rating establish medical necessity, a diagnosis, or a level of care? No. A CANS rating records what action a need requires; it is descriptive and agnostic as to cause, it is not a diagnosis, and a count of actionable items is not a severity band. Level-of-care and eligibility meaning comes from the program that uses the tool: OhioRISE uses the Ohio Children's Initiative CANS for eligibility and care-coordination tiers, New York's Health Homes derive acuity from the CANS-NY, Texas uses its CANS 3.0 to inform which of its service packages fits, Illinois has commissioned decision-support models on the IM+CANS, and Illinois child welfare has used a CANS-based placement algorithm whose concordant placements predicted better outcomes in Chor and colleagues' 2015 study; each of those is decision support inside a named program, and under Family First a qualified individual must synthesize the assessment with the family and permanency team rather than adopt an output. Write that the CANS informs the determination, name the program's algorithm or review if one applies, and put the functional evidence behind the actionable items into any medical necessity letter or authorization request; a bare rating proves nothing on its own, and no program reviewed for this page treats a CANS total as a test a child must pass. Can BastionGPT draft my CANS documentation? Yes. Give it the facts (version and form, purpose, window dates, sources, rater and certification, each actionable item with what you observed and who reported it, strengths, caregiver findings, disagreements, and the plan goals) and it drafts the full entry: version and window named, a justification in original prose for every action-level item, strengths and caregiver ratings as separate targets, consensus and disagreement recorded, no total, and each actionable item traced to the plan, ready for your review. It can also check a finished note for an unnamed version, an undated window, missing certification, anchor text used as justification, a watchful-waiting item with no reason, a total score, or an action-level need with no plan response. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Praed Foundation, the Child and Adolescent Needs and Strengths (CANS) (open domain and free to use; the action levels for needs and strengths; the rating window over-ridden by action levels; decision-support algorithms; outcome tracking by actionable items or dimension sums; versions used in 50 states; the bachelor's-level training statement; accessed September 2026), about the Foundation (copyrights maintained on the CANS, FAST, CAT, ANSA, SSIT, and RISE), frequently asked questions (recertify in the original course, Ohio excepted; deactivation three months after expiry; TCOM Training registration), TCOM (communimetric measures; collaboration, voice and choice, and consensus), and the Standard Comprehensive CANS 3.0 Reference Guide (January 13, 2021; copyright 2021, 2017, 1999 by the John Praed Foundation; ages 6 through 20; the six key principles; the 30-day window; reliability of 0.78, 0.84, and above 0.90; the 0.70 certification threshold; domain and module names; the permission-to-use sentence). Lyons JS, 2009, Communimetrics: A Communication Theory of Measurement in Human Service Settings, Springer (the theory of use); Anderson RL, Lyons JS, Giles DM, Price JA, Estle G, 2003, Journal of Child and Family Studies 12(3):279-289, reliability of the CANS-MH (60 cases; items reliable enough for individual use); Accomazzo S, Shapiro VB, Israel N, Kim BK, 2017, Journal of Behavioral Health Services and Research 44(2):274-288, the strengths of youth in a public behavioral health system (individual items, actionable counts, subscales, and a composite compared); Chor KHB, McClelland GM, Weiner DA, Jordan N, Lyons JS, 2015, Administration and Policy in Mental Health 42(1):70-86, out-of-home placement decision-making and outcomes in child welfare (3,911 placement records; concordant versus discordant decisions). Legal Information Institute, 42 U.S.C. 675a, additional case plan and case review system requirements (qualified individual assessment within 30 days of placement in a qualified residential treatment program; family and permanency team; court review within 60 days; no instrument named) and 42 U.S.C. 1396d(r) (the EPSDT benefit definition; no instrument named); Centers for Medicare and Medicaid Services, EPSDT guidance for children and youth in Medicaid and CHIP (September 26, 2024; medically necessary mental health and substance use services across the continuum; no instrument named). Illinois Department of Healthcare and Family Services, Illinois Medicaid Comprehensive Assessment of Needs and Strengths (IM+CANS) (approved IATP instrument; required for reimbursement; review and update every 180 days; annual certification through PATH; materials effective April 1, 2024); Ohio Department of Medicaid, CANS Ohio (Ohio Children's Initiative CANS; H2000 from July 1, 2022; current Praed certification and the CANS Assessor specialty); Texas Department of Family and Protective Services, CANS 3.0 (January 1, 2025; ages three and older; Medical Health and Exploitation modules; certification transition; periodicity schedule; T3C service packages), and Superior HealthPlan, new CANS assessment effective January 1, 2025 ; Indiana Department of Child Services, Child Welfare Policy 5.19, CANS Assessment (version 6, January 1, 2025; five-day and 180-day timing; certification; plan incorporation); New York State Department of Health, CANS-NY and guide to edits (annual CANS-NY unless a significant life event intervenes), and New York OPWDD, CANS assessment (0 to 5 and 6 to 21 manuals; yearly repetition; each caregiver perspective equally valid); MassHealth, CANS training and certification ; Wisconsin Child Welfare Professional Development System, CANS training and certification exam (score of .70 or more; annual). Ontario Child and Parent Resource Institute, interRAI child and youth (ChYMH) suite implementation toolkit (about 70 organizations; updated April 20, 2026); interRAI, Child and Youth Mental Health instrument ; Australian Institute of Health and Welfare, consumer outcomes in mental health care (the National Outcomes and Casemix Collection; HoNOSCA for children and adolescents). Praed Foundation, the Adult Needs and Strengths Assessment (ANSA) ; Multi-Health Systems, CAFAS and PECFAS ; Pearson, BASC-3 ; ASEBA, Achenbach System of Empirically Based Assessment . ## Related templates Treatment Plan Treatment Plan Review & Update Intake Note Biopsychosocial Assessment Care Coordination Note Case Management Note Parent & Caregiver Collateral Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cage All note templates # CAGE Questionnaire Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The CAGE questionnaire is a four-item, yes-or-no alcohol screen published by John A. Ewing in JAMA in 1984, worded for lifetime experience and built to detect dependence rather than current risky drinking. Primary care, emergency, and behavioral health clinicians use it, and the CAGE-AID variant for drugs, to decide who needs a fuller assessment. This page covers how to document and interpret CAGE results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Any clinician or trained staff member can ask and score it (verbally or on an intake form); no license, training requirement, or publisher qualification level applies; the treating clinician interprets and documents Audience Primary care and emergency physicians, nurses, addiction and behavioral health clinicians, trauma-center SBIRT teams, payers and auditors reviewing screening and brief-intervention claims Typical length 3 to 6 chart lines (version, time frame, item-level answers and total, current use, follow-up, intervention) · administration about 1 minute Format family Brief case-finding screen (4 yes-or-no items, one point each, total 0 to 4, lifetime wording; the CAGE-AID asks the same four topics about alcohol or other drugs) When it's used Intake packets and annual visits in primary care, emergency and trauma-center screening, addiction and behavioral health intakes, and any setting that wants a fast dependence-oriented probe; not the preferred routine screen for unhealthy drinking Standards context Ewing 1984 (JAMA) under a Bowles Center permission statement (cite the source; profit-making use needs permission); CAGE-AID Brown and Rounds 1995; USPSTF 2018 and RACGP 2024 name the AUDIT-C for routine screening; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CAGE questionnaire? The CAGE is a four-item alcohol screen developed by John A. Ewing and colleagues at the University of North Carolina at Chapel Hill, first presented at a 1970 meeting in Sydney, validated by Mayfield, McLeod, and Hall in the American Journal of Psychiatry in 1974, and given its canonical exposition by Ewing in JAMA in 1984, which is the citation every reproduction asks for. Each item is answered yes or no and scores one point, so the total runs 0 to 4. The acronym comes from the four topics the items cover: cutting down, annoyance at criticism, guilt, and morning drinking. The items are worded for lifetime experience, so the instrument finds people who have accumulated the consequences of dependence; it asks nothing about how much anyone drinks now. The CAGE-AID (CAGE Adapted to Include Drugs; Brown and Rounds, Wisconsin Medical Journal, 1995) asks the same four topics about alcohol or other drugs together, so its total is neither an alcohol score nor a drug score. Pregnancy-specific descendants, the T-ACE (1989) and the TWEAK (1994), exist because the CAGE performs poorly in prenatal populations. Three facts carry the documentation. First, a CAGE total is a screening result, never a diagnosis: two or more yes answers is the conventional positive threshold (Ewing 1984), a SAMHSA consensus panel recommended one in primary care to cast a wider net, and no authority maps 0 to 4 onto severity bands or onto DSM-5 alcohol use disorder, which needs its own criteria-based substance use assessment . Second, the time frame decides the meaning: a lifetime positive in a patient abstinent for ten years is a true positive for history and says nothing about current use, and a version reworded to the past 12 months is a modified instrument that must be charted as such. Third, the CAGE and the CAGE-AID are different instruments with different authors, rights holders, and conventions (the University of Washington ADAI overview treats one or more CAGE-AID yes answers as a positive screen), so the version belongs in every entry. For routine adult screening the USPSTF (2018) and Australia's RACGP (2024) now name the AUDIT-C or a single consumption question, because the CAGE detects dependence rather than the full spectrum of unhealthy drinking; the DAST-10 is the structured drug analogue. ## Who uses CAGE documentation and when The CAGE survives in three places. Primary care intake packets and annual-visit flowsheets still carry it as a legacy item, so family physicians, internists, and nurses chart CAGE results even where the practice's formal screen is now the AUDIT-C or the NIAAA single question. Emergency departments and verified trauma centers, which must screen admitted patients and deliver a documented brief intervention, use the CAGE or the CAGE-AID when a fast verbal probe fits the bedside. Addiction, psychiatric, and behavioral health intakes use it as a dependence-oriented history item beside the fuller substance use assessment , biopsychosocial assessment , or intake note , and the CAGE-AID is the choice when drug involvement is suspected or a program screens for alcohol and drugs in one pass. Outside those uses the neighbors win: the AUDIT-C for current hazardous drinking and quality-measure reporting, the DAST-10 for drug-problem severity, a pregnancy-specific screen in prenatal care, and the CIWA-Ar or COWS when the question is withdrawal severity, which no CAGE result can answer. ## How to document CAGE results in the chart No law, payer, or author prescribes a CAGE note format; what exists is a permission statement asking for the Ewing 1984 citation, and payer and accreditation rules that ask for a named validated tool, a result, the time spent, and an intervention. What survives review is an entry that names the version, states the time frame actually asked, records the four item-level answers with the total, charts current quantity and frequency beside the screen, labels the result a screen, names the follow-up assessment it triggered, and documents the intervention with its minutes. Each element below carries the pitfall that most often undermines it. Version: CAGE or CAGE-AID. Chart which instrument was used, in words a later reader cannot mistake: CAGE (original, alcohol only) or CAGE-AID (alcohol or other drugs, including non-medical use of prescription drugs). The two carry different authors, rights holders, validation figures, and positive conventions, and a CAGE-AID yes may concern alcohol on one item and a drug on another, so record which substance produced each positive answer when the patient can say. Pitfall: "CAGE 2/4" when the CAGE-AID was administered, or a CAGE-AID total copied into the problem list as an alcohol result. Time frame actually asked. The original items ask about lifetime experience. Write "lifetime frame" when the questions were asked as worded, and "modified, past 12 months" (or whatever window was used) when the clinician narrowed them; a narrowed version is a different instrument with different operating characteristics, not the original with a footnote. When lifetime positives refer to a resolved period, date them ("all endorsements precede the patient's sobriety date") and add the current-status line described below. Pitfall: A lifetime positive in a patient sober for years, charted so that the next reader believes the patient is drinking now. Item-level answers, then the total. Record the four answers in order with the total (C yes, A no, G yes, E no; 2 of 4) rather than the total alone. Which items were endorsed changes the picture: a morning-drinking yes points toward physiological dependence and a withdrawal history, guilt alone often reflects a remote episode, and a single yes is a prompt for follow-up under the SAMHSA wider-net convention rather than a negative. Keep the patient's explanation next to any yes, especially a morning-drinking answer from a shift worker whose "morning" is the end of a night shift and whose drink may have nothing to do with relieving withdrawal. Pitfall: "CAGE positive" or "CAGE 2" with no item pattern, so the entry can be neither interpreted nor reproduced. Current quantity and frequency, separately. The CAGE has no consumption item, so add the patient's own statement of drinks per typical week, heavy-drinking days, the last drink, and any drug use with route and frequency, on its own line and in their words. This is the line that separates a historical positive from present risk, that Medicare's misuse benefit and the SBIRT codes are built around, and that Canadian and Australian guidance reads against the national weekly limits. Pitfall: A positive screen with no consumption statement, leaving the reader unable to tell current hazardous drinking from remote history. Interpretation: a screen, in a named population. State the convention applied (positive at two or more yes answers, or at one under the primary-care wider-net convention) and write the conclusion as a screening result: "positive screen; diagnosis not established." Name the population when it changes the reading: at the conventional threshold, sensitivity is low for hazardous drinking in older adults, for women in several primary-care samples, and in pregnancy, so a negative CAGE in those groups rules little out, and a translated version does not import the English threshold's validity. Pitfall: "CAGE 3/4, alcohol use disorder" with no DSM-5 criteria assessed, or "CAGE negative, no alcohol concerns" in a 70-year-old who drinks daily. Follow-up assessment triggered. A positive screen starts a sequence, so name what happened next and its result: an AUDIT-C or full AUDIT to characterize current drinking, a DAST-10 or TAPS for a drug-related CAGE-AID yes, a DSM-5 criteria-based assessment when a disorder is suspected, and a withdrawal assessment (CIWA-Ar) when current heavy use or a morning-drinking history raises that risk. A negative screen in a patient reporting heavy use also needs a next step, because the CAGE misses risky drinking by design. Pitfall: A positive CAGE with no follow-up instrument or assessment named, or a diagnosis inferred from the screen alone. Intervention, minutes, referral, and record placement. Document the brief intervention (feedback given, the patient's goal, advice offered), the referral made or declined, or the reason no intervention was indicated, and the face-to-face time whenever a time-based code is claimed. Record where the result lives: a primary-care CAGE is an ordinary medical-record entry under HIPAA, while the same result created by a federally assisted Part 2 program is a Part 2 record with its own consent and redisclosure rules. Pitfall: "Counseled on alcohol" with no content, no minutes, and no goal, which supports neither the claim nor the next visit. ### Blank template (copy and adapt) CAGE / CAGE-AID DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [annual / intake / ED or trauma / prenatal / other] Version: [CAGE, alcohol only / CAGE-AID, alcohol or other drugs] Time frame asked: [lifetime, as worded / modified: state the window] Mode: [asked verbally / self-completed on intake form] Language: [ ] Item-level answers: C [ ] A [ ] G [ ] E [ ] Total: [ ] of 4 (CAGE-AID: substance behind each yes, when the patient can say) Timing of positives: [current / all precede (date or period); explanation] Current use (patient's words): [drinks per typical week; heavy days; last drink; drug use, route, frequency; or none, since (date)] Interpretation: [positive / negative under (convention named); a screen, not a diagnosis; population caveat: older adult, women, pregnancy, translation, or none] Follow-up triggered: [AUDIT-C / AUDIT / DAST-10 / TAPS / DSM-5 criteria assessment / withdrawal assessment; result and date] Intervention: [feedback given; patient's goal; referral made, declined, or not indicated; face-to-face minutes] Plan: [re-screen instrument and date; treatment or counseling; next visit] Record: [code claimed (bare number); storage location; Part 2 status if created by a Part 2 program; consent on file] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the questionnaire items. ## Sample CAGE documentation (fictional) Scenario: a US family medicine annual visit where the practice's intake packet still carries the CAGE; the lifetime-frame screen returns two yes answers, the physician adds a current consumption statement, follows with an AUDIT-C, delivers a brief intervention, and records the time, the claim, and the plan. All details are fictional. Patient: J.M., 58 · Setting: Family medicine, annual preventive visit · Clinician: A. Delgado, MD · Note date: 09/14/2026 Screen: CAGE, original alcohol-only version, self-completed in English on the practice intake form on 09/14/2026 and reviewed verbally during the visit; lifetime frame as worded. Item-level answers: C yes, A no, G yes, E no; total 2 of 4, positive at the conventional threshold. J.M. places both endorsements in the past two years and describes them as current: two attempts to cut back without lasting change, and regret about drinking more than intended on weekends. No morning drinking and no symptoms on alcohol-free days. No prior CAGE or AUDIT on file. Current use: In the patient's words: 3 to 4 standard drinks (beer, sometimes bourbon) on 4 or 5 evenings a week, 6 or more on about two Saturdays a month, last drink last night; about 16 to 20 drinks in a typical week, above the NIAAA weekly limit for men. No cannabis, no other drug use, no non-medical use of prescription medication. Denies driving after drinking, injuries, or work problems related to alcohol. Reports poor sleep; blood pressure 146/92 today, higher than at last year's visit. Interpretation: Positive CAGE in the lifetime frame with both endorsements current, plus a self-reported pattern above recommended limits: consistent with current unhealthy alcohol use. The CAGE is a screen and does not establish a diagnosis; DSM-5 alcohol use disorder criteria were not formally assessed today and will be reviewed at follow-up. No population caveat applies (58-year-old man, English, original wording). Follow-up assessment: AUDIT-C administered verbally in the same visit to characterize current drinking: 7 of 12, positive under the practice protocol for men; item results on the screening flowsheet. Withdrawal risk judged low (no morning drinking, no prior withdrawal, no tremor or sweating on alcohol-free days); CIWA-Ar not indicated. Hepatic panel and GGT added to today's labs. Brief intervention: Delivered by the physician: feedback on both screen results and where the reported pattern sits against the NIAAA limits, the likely contribution to poor sleep and the rise in blood pressure, and a menu of options. J.M. chose a goal of no more than 2 drinks on any day and two alcohol-free days each week starting this week, and accepted a printed drink-tracking sheet. Declined referral to counseling for now and knows the offer stands. Face-to-face time for the screen review and intervention: 18 minutes. Plan and record: Nurse phone check in 2 weeks; office visit in 6 weeks to review the tracking sheet, repeat the AUDIT-C, recheck blood pressure, and assess DSM-5 criteria if the pattern persists, with medication options for alcohol use disorder discussed then if criteria are met. Screening and brief intervention claimed under 99408, supported by the tool, result, intervention, and time documented above. Result recorded in the general medical record; this practice is not a Part 2 program, so no Part 2 consent applies. J.M. was told the screen is not a diagnosis and agreed with the plan. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version (CAGE, alcohol only) and the lifetime frame are named, and both endorsements are placed in time, so the result cannot be misread as remote history or as a CAGE-AID result. All four item-level answers appear with the total, and the patient's explanation sits beside them, so the entry can be interpreted and reproduced by the next reader. Current quantity and frequency are charted in the patient's words on their own line and read against a named limit, which is what turns a screen into a picture of present risk. The result is labeled a screen with the convention named, the diagnosis is left open until criteria are assessed, and the follow-up instrument, its result, and the withdrawal-risk judgment are documented. The brief intervention has content, a patient-chosen goal, a referral decision, and face-to-face minutes, so the 99408 claim is supported and the next visit has something to measure against. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the instrument is a convention; the rules around it are real. No federal or state LAW names the CAGE, and no law prescribes a screening note format. The USPSTF recommendation of 2018 (grade B; CONVENTION, with quality-measure teeth) covers screening adults 18 and older, including pregnant women, for unhealthy alcohol use, names the AUDIT-C and the NIAAA single question as the most accurate brief instruments, and passes over the CAGE because it detects dependence rather than the full spectrum of unhealthy use; NIAAA's clinician resource points to the same two tools, and the 2018 statement remains current as of September 2026 with no finalized replacement. The USPSTF's 2020 drug-use recommendation (grade B; adults 18 and older; screening by asking, not by testing specimens, and only where care can be offered) evaluated past-year single items, the NIDA Quick Screen, TAPS, and the DAST-10 rather than the CAGE-AID. Under PAYER POLICY, Medicare's alcohol misuse benefit (NCD 210.8, 2011) pays an annual screen under G0442 and up to four counseling sessions a year under G0443 for beneficiaries who misuse alcohol without meeting the decision's dependence criteria, which is the pattern the CAGE is least built to find; the SBIRT codes (99408 and 99409 for commercial plans, G0396 and G0397 for Medicare, H0049 and H0050 where a state Medicaid program adopts them) are time-based and need a named validated tool, the result, the face-to-face time, and, for the intervention codes, the intervention itself, so a bare "CAGE 2/4" supports none of them, and 99408 is not a Medicare substitute for G0442. Accreditation and quality CONVENTIONS ask for the same things: the Joint Commission SUB measure set (2026 specifications manual, as of September 2026) requires alcohol screening with a validated tool by the end of hospital day one for hospitals reporting it; MIPS quality measure 431 (2026) counts systematic screening for unhealthy alcohol use with brief counseling when it is found; and American College of Surgeons trauma verification standards (2022 standards, Q&A current in September 2026) require verified trauma centers to screen at least 80 percent of admitted patients older than 12 and to deliver a documented brief intervention, or a referral where inpatient intervention is unavailable, to at least 80 percent of those who screen positive, with the operative wording for your level in the current Resources for Optimal Care manual. Under LAW, 42 CFR Part 2 as amended by the February 2024 final rule (effective April 16, 2024; compliance date February 16, 2026, now passed, with HIPAA-style breach notification and civil enforcement) attaches to records created by a federally assisted Part 2 program, so a CAGE-AID recorded by such a program is a substance use disorder record with its own consent and redisclosure rules, while an ordinary primary-care CAGE result is a HIPAA record and nothing more; the Part 2 consent page covers the mechanics. Canada and Australia: no requirement, and reference limits a positive screen should be documented against. In Canada, no federal or provincial LAW or payer names the CAGE. Canada's Guidance on Alcohol and Health (Canadian Centre on Substance Use and Addiction, January 2023; CONVENTION) replaced the 2011 low-risk drinking guidelines with a continuum: 2 or fewer standard drinks a week is low risk, 3 to 6 is moderate, 7 or more is increasingly high, a Canadian standard drink holds 13.45 g of ethanol, and there is no known safe amount in pregnancy or when trying to conceive. CAMH's guidance for professionals asks for a recent alcohol history (drinks per week and the most in a day) and a validated tool such as the AUDIT-C or AUDIT, with assessment rather than diagnosis by screen after a positive result. A current, instrument-specific position from the Canadian Task Force on Preventive Health Care or the College of Family Physicians of Canada could not be confirmed for this page, so none is claimed. In Australia, the RACGP preventive-activities guidance (alcohol recommendation dated June 28, 2024; CONVENTION) recommends screening adults 18 and older, including in pregnancy, every two years and names the AUDIT-C; the NHMRC guidelines (December 2020) set the reference limits at no more than 10 standard drinks a week and 4 on any one day, with a standard drink at 10 g of ethanol and no alcohol under 18, in pregnancy, or while planning a pregnancy; no MBS item and no state law names the CAGE. Neither instrument carries a qualification level or purchase restriction: any clinician may ask the four questions, and the boundary is attribution and non-commercial use, covered in the next paragraph. In every jurisdiction a translated CAGE is charted with its language and the convention actually applied, because a Spanish primary-care study found a study-optimal threshold of one and a Brazilian Portuguese pregnancy study found the conventional threshold failed its construct-validity test. The CAGE questionnaire is the work of John A. Ewing, published in JAMA in 1984, and is made available under a Bowles Center for Alcohol Studies (University of North Carolina at Chapel Hill) permission statement, dated February 2004, that asks users to cite Ewing 1984 and reserves any profit-making use for negotiated permission; a 2014 Occupational Medicine questionnaire review states instead that reproduction requires permission from JAMA, and no current source resolves that conflict or grants blanket commercial, EHR-vendor, or public-web rights, so this page does not call the CAGE public domain, and a commercial product should clear rights before reproducing it. The CAGE-AID is copyrighted through the Wisconsin Medical Journal (Brown and Rounds, 1995) and is conventionally reproduced with the journal's permission and the authors' consent for non-commercial clinical use. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CAGE documentation errors reviewers flag The numbers behind these errors are specific. A diagnostic meta-analysis of 10 studies and 4,562 patients found pooled sensitivity at the conventional threshold of 0.87 in inpatients, 0.71 in primary care, and 0.60 in ambulatory patients and concluded the CAGE has limited value for screening at that threshold (Aertgeerts, Buntinx, and Kester, 2004); a 2007 review put average sensitivity at 0.71 and specificity at 0.90 for abuse or dependence and reported sensitivity of 0.38 in one primary-care sample of white women ( Dhalla and Kopec ); in older primary-care patients the conventional threshold identified 14 percent of hazardous drinkers with 97 percent specificity (Adams, Barry, and Fleming, JAMA, 1996), and in adults over 60 moving from two yes answers to one shifted sensitivity and specificity from 70 and 91 percent to 86 and 78 percent (Buchsbaum and colleagues, 1992). The 93 percent sensitivity that consumer pages repeat traces to the 1974 psychiatric-inpatient validation (366 admissions, 39 percent alcoholic), not to primary care. No published audit counts how often the version or the time frame is missing from CAGE chart entries; the SBIRT and Medicare documentation rules described under compliance considerations are what turn those omissions into a claim problem. The BastionGPT Clinical Advisory Board sees the same errors most often in CAGE documentation reviews: Version missing or wrong. "CAGE 2/4" when the CAGE-AID was used, or a CAGE-AID result carried into the problem list as an alcohol finding. The two instruments have different authors, rights holders, validation figures, and positive conventions, and a CAGE-AID yes can come from alcohol on one item and a drug on another. Name the version every time and record the substance behind each CAGE-AID positive. Lifetime positive read as current use. A patient abstinent for a decade still answers yes to the same items, so "CAGE 3/4" without a time frame reads as active drinking to the next clinician, the payer, and the court. Write "lifetime frame," date the endorsements, and add a current-status line ("no alcohol in 10 years; current consumption screen negative"). Total without the item pattern. A bare "CAGE positive" entry or a total alone cannot be interpreted or reproduced. Record the four answers in order with the total, keep the patient's explanation beside any yes, and treat a single yes as a prompt for follow-up under the SAMHSA wider-net convention rather than as a negative. No consumption statement. The CAGE has no quantity or frequency item, so a note that stops at the score cannot show current risk, cannot support Medicare's misuse benefit or an SBIRT claim, and cannot be read against the NIAAA, Canadian, or Australian limits. Chart drinks per week, heavy days, the last drink, and any drug use in the patient's words. Screen charted as a diagnosis, or population ignored. "CAGE 3/4, alcohol use disorder" with no criteria assessed, invented severity bands ("2 moderate, 3 severe"), or "CAGE negative, no alcohol concerns" in an older adult, a woman, or a pregnant patient, where the conventional threshold misses most hazardous drinking. Label the result a screen, name the convention used, assess criteria separately, and note the population caveat. Intervention and time undocumented. A note reading "counseled on alcohol" with no content, goal, referral decision, or minutes fails the time-based SBIRT codes and the trauma-center and Joint Commission expectation that the intervention itself be documented. Record the feedback given, the patient's goal, the referral made or declined, and the face-to-face time, and name the follow-up instrument the screen triggered. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on substance use assessments and screening documentation. Give it the facts (version, time frame as asked, the four item-level answers, the patient's consumption statement, any follow-up instrument and its result, the intervention delivered and its minutes, and where the result is stored) and it drafts the documentation block: the screen result with the convention named, current use on its own line, a screening-level interpretation with the population caveat, the follow-up, the intervention, the plan, and the record placement, ready for your review. Cross-check a finished note for the gaps reviewers flag: a missing version, a lifetime positive with no time frame or current-status line, a total without the item pattern, no consumption statement, a diagnosis inferred from the screen, or an intervention with no content or minutes. Turn the screen into the next document: the opening of a substance use assessment, the referral letter to an addiction program, or the SBIRT documentation a payer expects, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the CAGE scored, and what counts as a positive screen? Each of the four items is answered yes or no and each yes scores one point, so the total runs 0 to 4. Two or more yes answers is the conventional positive threshold (Ewing, JAMA, 1984; the 1974 validation found a threshold of two or three useful and four alone too insensitive). A SAMHSA consensus panel for primary care recommended lowering the threshold to one yes to cast a wider net, and in adults over 60 that move raised sensitivity from 70 to 86 percent while dropping specificity from 91 to 78 percent, so a single yes is a prompt for follow-up rather than a negative. There are no severity bands: pages that grade 2 as moderate and 3 as severe are inventing a scale no CAGE publication supports, and no score establishes DSM-5 alcohol use disorder, which needs its own criteria-based assessment. The CAGE-AID follows its own convention, with the University of Washington ADAI overview treating one or more yes answers as a positive screen. Chart the convention you applied and the item pattern, never the total alone. My patient has been sober for years but still answers yes to three items. Is that a positive screen, and how do I chart it? Under the lifetime frame, yes: the items ask about experience ever, so remote endorsements never expire and a lifetime CAGE can stay high indefinitely after recovery. That is a true positive for history and no evidence of current use. Chart it so the two cannot be confused: "CAGE, original alcohol-only version, lifetime frame; C yes, A yes, G yes, E no; 3 of 4. All endorsements precede the patient's sobriety date; reports no alcohol in nine years; current consumption screen negative. Lifetime result documents remote alcohol history and does not establish current drinking." Put the current status and its duration on its own line, keep the fuller history in the substance use assessment or a relapse prevention plan rather than in the screen line, and if you narrowed the questions to the past 12 months instead, chart the result as a modified instrument, because the operating characteristics change with the window. What is the difference between the CAGE and the CAGE-AID? The CAGE (Ewing, 1984) asks its four items about alcohol only. The CAGE-AID, CAGE Adapted to Include Drugs (Brown and Rounds, Wisconsin Medical Journal, 1995), asks the same four topics about alcohol or other drugs together, including non-medical use of prescription drugs. That single change means a CAGE-AID total is not substance-specific: two yes answers can reflect alcohol on one item and cannabis on another, so the chart needs the substance behind each positive when the patient can say. The instruments also differ in authorship, rights holders, validation (in the original primary-care study of 124 patients the conjoint version was more sensitive and less specific than the alcohol-only CAGE), and positive convention, with the CAGE-AID overview treating one or more yes answers as a positive screen. Name the version in every entry, never carry a CAGE-AID result into the problem list as an alcohol finding, and follow a drug-related yes with the DAST-10 or TAPS so the drug side of the result gets its own assessment. Should I be using the CAGE or the AUDIT-C for routine screening? For routine screening of adults for unhealthy alcohol use, the AUDIT-C or the NIAAA single question. The USPSTF's 2018 recommendation names those two among the most accurate brief instruments and passes over the CAGE because it detects dependence rather than the full spectrum of unhealthy use; NIAAA's clinician resource points to the same two; Australia's RACGP (2024) names the AUDIT-C for two-yearly screening; and CAMH's Canadian guidance asks for a recent consumption history plus the AUDIT-C or AUDIT. Medicare's misuse benefit and MIPS measure 431 are built around current unhealthy drinking short of dependence, which the CAGE is least designed to find. The CAGE keeps a role as a fast dependence-oriented probe, a legacy intake item, and historical case-finding information, and a positive CAGE still deserves a consumption screen so the note shows current risk rather than history alone. Can I bill SBIRT or Medicare G0442 on a CAGE alone? Not on the score alone. The SBIRT codes (99408 and 99409, G0396 and G0397, and H0049 and H0050 where a state Medicaid program adopts them) and Medicare's G0442 and G0443 are PAYER POLICY: they need a named validated instrument, the result, the face-to-face time for the time-based codes, and, for the intervention codes, the brief intervention itself documented with its content. The CAGE is a validated instrument, so a note that names it, records the item-level result and the time frame, adds the consumption statement, and documents the intervention with its minutes can support a claim; "CAGE 2/4, counseled" cannot. Two further cautions: Medicare's benefit under NCD 210.8 targets misuse that does not meet its dependence criteria, so an AUDIT-C or single-question result is the better documentation fit, and 99408 is not a Medicare substitute for G0442. Medicaid rules are state-specific, so check the state's SBIRT policy before choosing a code. Does a CAGE result fall under 42 CFR Part 2? Only if the record is created or received by a federally assisted Part 2 program. Part 2 follows the source of the record, not the word alcohol: an ordinary primary-care CAGE result is a HIPAA record and nothing more, while a CAGE-AID recorded at intake in a Part 2 program is a substance use disorder record with Part 2 consent and redisclosure rules. Information a patient conveys orally to a clinician outside a Part 2 program, who then documents it independently, does not become a Part 2 record for that reason alone. The February 2024 final rule (effective April 16, 2024; compliance date February 16, 2026, now passed) aligned Part 2 with HIPAA on breach notification and enforcement and allows a single consent for treatment, payment, and health care operations, while keeping special protection for SUD counseling notes. Document where the result lives and, in a Part 2 program, which consent is on file; the Part 2 consent page covers the forms and the workflow. How do I document a CAGE in an older adult, a woman, or a pregnant patient? With the population caveat written into the interpretation. At the conventional threshold the CAGE identified 14 percent of hazardous drinkers among older primary-care patients (Adams, Barry, and Fleming, 1996) and 48 percent of older medical outpatients with abuse or dependence (Jones and colleagues, 1993); in predominantly white women sensitivity ran 38 to 50 percent across the studies Bradley and colleagues reviewed in 1998; and in pregnancy the instrument performs poorly enough that the T-ACE (1989) and TWEAK (1994) were built to replace it, with ACOG endorsing the T-ACE. A negative CAGE in these groups therefore rules little out, and the note should say so and pair the screen with a consumption statement or an AUDIT-C. In prenatal care, document why a pregnancy-specific screen was used instead (the clinical question is current exposure, not lifetime dependence history), record that instrument's result, and note the counseling that no amount of alcohol is known to be safe in pregnancy. In older adults, interpret a single yes seriously under the wider-net convention rather than waiting for two. What do the four CAGE items ask, and can we reproduce the questionnaire? The four items cover, at the topic level, cutting down, annoyance at criticism, guilt, and morning drinking; that is where the acronym comes from. This page does not print or paraphrase the questions, because the rights position is contested: the Bowles Center for Alcohol Studies permission statement (2004) asks users to cite Ewing 1984 and requires negotiated permission for any profit-making use, a 2014 questionnaire review in Occupational Medicine states that reproduction requires JAMA's permission, and no current source grants blanket commercial, EHR-vendor, or public-web rights. For a non-commercial clinical form, the permission statement supports use with the citation, NIAAA and SAMHSA reproduce the instrument in free clinical materials, and the JAMA article is the source of record. For a commercial product, an ad-supported web tool, or an EHR vendor build, clear rights first. The CAGE-AID sits under separate copyright through the Wisconsin Medical Journal (1995) and is reproduced in toolkits with the journal's permission and the authors' consent. Can BastionGPT draft my CAGE documentation? Yes. Give it the facts (version, time frame as asked, the four item-level answers, the patient's consumption statement, any follow-up instrument and result, the intervention delivered and its minutes, and where the result is stored) and it drafts the full entry: the screen result with the convention named, current use on its own line, a screening-level interpretation with the population caveat, the follow-up, the intervention, the plan, and the record placement, ready for your review. It can also check a finished note for a missing version, a lifetime positive with no time frame or current-status line, a total without the item pattern, no consumption statement, a diagnosis inferred from the screen, or an intervention with no content or minutes. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Ewing JA, 1984, JAMA, Detecting alcoholism: the CAGE questionnaire (the canonical citation); Mayfield D, McLeod G, Hall P, 1974, American Journal of Psychiatry, validation of a new alcoholism screening instrument (366 psychiatric admissions); O'Brien CP, 2008, JAMA, JAMA Classics commentary on the CAGE (1970 Sydney presentation; interpretive claims); Bowles Center for Alcohol Studies, UNC Chapel Hill, CAGE permission statement (2004; hosted on the USPSTF site; cite Ewing 1984; profit-making use needs permission); UNC SABI instrument database entry (research and educational use). Brown RL, Rounds LA, 1995, Wisconsin Medical Journal, conjoint screening questionnaires for alcohol and other drug abuse (the CAGE-AID; 124 primary-care patients; more sensitive and less specific than the CAGE); University of Washington Alcohol and Drug Abuse Institute, CAGE-AID overview (one or more yes answers regarded as positive; does not discriminate active from inactive use; reprinted with permission). US Preventive Services Task Force, 2018, unhealthy alcohol use: screening and behavioral counseling (grade B; AUDIT-C and the single question; the CAGE detects only dependence), and 2020, unhealthy drug use: screening (grade B; screening by questions); AAFP News, June 2018, coverage of the draft recommendation (the CAGE discounted); NIAAA, Core Resource on Alcohol: screen and assess (AUDIT-C and single-question screening); SAMHSA, 1997, TIP 24, A Guide to Substance Abuse Services for Primary Care Clinicians (consensus panel: one positive answer as the primary-care threshold). CMS, 2011, decision memo on screening and behavioral counseling to reduce alcohol misuse (NCD 210.8; G0442 and G0443; misuse short of dependence); CMS Medicare Learning Network, SBIRT services fact sheet (structured tools; time and record requirements; bare code numbers only); CMS Quality Payment Program, 2026 MIPS quality measure 431 ; The Joint Commission, alcohol use status data element, 2026B ; American College of Surgeons, VRC 2022 standards Q&A (standards 5.30 and 5.31: 80 percent screening and a documented brief intervention). US Department of Health and Human Services, 2024, 42 CFR Part 2 final rule fact sheet (effective April 16, 2024; compliance February 16, 2026); eCFR, 42 CFR 2.11 definitions (records; Part 2 program). Aertgeerts B, Buntinx F, Kester A, 2004, Journal of Clinical Epidemiology, diagnostic meta-analysis of the CAGE (10 studies; sensitivity by setting; limited value at the conventional threshold); Dhalla S, Kopec JA, 2007, Clinical and Investigative Medicine, review of reliability and validity studies ; Adams WL, Barry KL, Fleming MF, 1996, JAMA, screening for problem drinking in older primary care patients (14 percent sensitivity for hazardous drinking); Buchsbaum DG and colleagues, 1992, Journal of the American Geriatrics Society, screening for drinking disorders in the elderly ; Jones TV and colleagues, 1993, Journal of General Internal Medicine, alcoholism screening questionnaires in elderly medical outpatients ; Fiellin DA, Reid MC, O'Connor PG, 2000, Archives of Internal Medicine, screening for alcohol problems in primary care (AUDIT for hazardous drinking; CAGE for abuse or dependence). Bradley KA and colleagues, 1998, JAMA, alcohol screening questionnaires in women ; Russell M, 1994, Alcoholism: Clinical and Experimental Research, screening for pregnancy risk-drinking (TWEAK); Midanik LT and colleagues, 1998, alcohol and drug CAGE screeners for pregnant, low-income women (modified 12-month window); Chang G and colleagues, Alcoholism: Clinical and Experimental Research, T-ACE and TWEAK screening in pregnancy (ACOG endorsement of the T-ACE). Canadian Centre on Substance Use and Addiction, January 2023, Canada's Guidance on Alcohol and Health (continuum of risk; 13.45 g standard drink; replaces the 2011 guidelines). RACGP, Guidelines for preventive activities in general practice: alcohol (recommendation dated June 28, 2024; AUDIT-C every two years); NHMRC, December 2020, Australian guidelines to reduce health risks from drinking alcohol (10 a week, 4 a day; 10 g standard drink). ## Related templates AUDIT-C Score Documentation DAST-10 Score Documentation Substance Use Assessment 42 CFR Part 2 Consent General Risk Screening Note CIWA-Ar Protocol Documentation Relapse Prevention Plan Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/audit-c All note templates # AUDIT-C Documentation: Scoring, Thresholds & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The AUDIT-C is a three-item alcohol consumption screen taken from the World Health Organization's ten-item AUDIT, validated as a standalone test by Bush and colleagues in 1998 and scored 0 to 12. Primary care, VA, and behavioral health clinicians use it to find unhealthy drinking and decide who needs a fuller assessment; it never diagnoses. This page covers how to document and interpret AUDIT-C results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report on paper, tablet, or portal, or asked verbally by any clinician or trained staff member; no license, training requirement, or publisher qualification level applies; the treating clinician interprets and documents Audience Primary care and emergency physicians, nurses, VA and DoD clinicians, perioperative teams, addiction and behavioral health clinicians, quality-measure abstractors, payers and auditors reviewing screening and brief-intervention claims Typical length 3 to 6 chart lines (version and standard drink, item pattern and total, threshold applied, patient's account, follow-up, intervention) · administration about 1 to 2 minutes Format family Brief consumption screen (3 items scored 0 to 4 each, total 0 to 12, past-year window; the first three items of the 10-item WHO AUDIT) When it's used Annual and preventive visits, intake packets, emergency and trauma-center screening, preoperative assessment, VA clinical reminders, behavioral health intakes, and repeated monitoring of drinking during treatment Standards context WHO AUDIT (Saunders and colleagues 1993; manual 2001) with the AUDIT-C validated by Bush and colleagues 1998; WHO copyright with free non-commercial use; named by the USPSTF, NIAAA, VA/DoD, and RACGP for routine screening, mandated by no law; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the AUDIT-C? The AUDIT-C is the consumption subscale of the Alcohol Use Disorders Identification Test, the ten-item screen the World Health Organization developed through a six-country collaborative project and published in Addiction in 1993 (Saunders, Aasland, Babor, de la Fuente, and Grant); the WHO manual for primary care reached its second edition in November 2001. The full AUDIT covers three domains: the first three items ask about consumption, the next three about dependence-related experiences, and the last four about harms. Bush, Kivlahan, McDonell, Fihn, and Bradley validated the three consumption items as a standalone screen in the Archives of Internal Medicine in 1998, in a sample of male veterans, and the abbreviation AUDIT-C has stuck. At the topic level the three items ask how often the person drinks, how many standard drinks they have on a typical drinking day, and how often they drink heavily on one occasion, each over the past year; each item scores 0 to 4, so the total runs 0 to 12, and a total of zero means no drinking was reported for the year. It can be asked in an interview or completed by the patient on paper, a tablet, or a portal, and NIAAA notes that self-report can be as accurate as direct questioning. Three facts carry the documentation. First, the score is not self-interpreting: the same three answers represent different amounts of ethanol depending on the standard drink the patient counted (14 g in the United States, 13.45 g in Canada, 10 g in Australia, about 8 g per UK unit) and on the wording of the heavy-occasion item, which the original instrument sets at six drinks for everyone, the US-adapted USAUDIT-C sets lower and by sex, and the VA clinical reminder tailors to sex assigned at birth. Second, the positive threshold is a setting convention rather than part of the instrument: 4 or more for men and 3 or more for women is the US primary-care convention from Bradley and colleagues (2007), the VA calls 5 or more positive for both sexes, Australia's treatment guideline uses 5 or more while the RACGP National Guide uses 4 and 3, and older-adult validations favor 5 for men and 4 for women, so the note has to say which convention it applied. Third, a positive AUDIT-C is a positive consumption screen, not a positive full AUDIT and not a diagnosis; the dependence and harm items, or a DSM-5 criteria assessment in a substance use assessment , come next. The CAGE asks about consequences and dependence rather than consumption, and the DAST-10 is the drug analogue. ## Who uses AUDIT-C documentation and when The AUDIT-C is the routine adult alcohol screen in US primary care because the USPSTF (2018, grade B) names it with the single-question screen, Medicare pays for an annual alcohol misuse screen, and MIPS measure 431 counts systematic screening with brief counseling, so family physicians, internists, and nurses chart it at wellness and annual visits and in intake packets. The VA has built it into a national clinical reminder with its own sex-tailored wording and threshold, so VA and DoD clinicians document it at nearly every primary care and mental health contact. Emergency departments and verified trauma centers use it or the single question to meet screening and brief-intervention standards; perioperative teams use a high total as the prompt for a withdrawal-risk assessment and a CIWA-Ar monitoring plan; addiction and behavioral health programs record it at the intake beside the substance use assessment and biopsychosocial assessment , and repeat it as a consumption measure during treatment under the conventions of the outcome measure note . Neighbors win in three places: the CAGE when the question is lifetime dependence history, the DAST-10 for drugs other than alcohol, and a pregnancy-specific or adolescent-specific screen where the USPSTF evidence for the adult instruments runs out. ## How to document AUDIT-C results in the chart No law, payer, or author prescribes an AUDIT-C note format; what exists is a WHO manual, several national adaptations with different wording and thresholds, and payer and quality rules that ask for a named validated tool, the result, the time spent, and the intervention. What survives review is an entry that names the version and the standard drink, records the three item scores with the total, states the threshold applied, adds the patient's own account of intake, labels the result a screen, names the follow-up it triggered, and documents the intervention with its minutes. Each element below carries the pitfall that most often undermines it. Version and standard drink. Write which wording was used (original international AUDIT-C, USAUDIT-C, the VA clinical reminder, or a local EHR build) and which standard drink the patient was told to count (US 14 g, Canada 13.45 g, Australia 10 g, UK unit). If the heavy-occasion item was adapted, say so, because a score of 5 on a form that asks about six drinks and a 5 on a form that asks about four drinks for women describe different drinking. Where the EHR carries the version as structured metadata, one phrase in the note still saves the next reader from guessing. Pitfall: "AUDIT-C 5" with no version or drink definition, so nobody can tell a VA reminder result from an original-wording result or what a drink meant. Mode, language, and completeness. Record whether the patient self-completed it (paper, tablet, portal) or a clinician asked the questions, the language or translation, that all three items were answered, and who computed the total (EHR logic or by hand). Mode affects disclosure, and translated forms may pertain to a version that differs from the original, so a translation is charted with its language and version rather than as plain AUDIT-C. Pitfall: An interviewer-asked score trended against a portal-completed baseline as if the conditions were the same, or a translation charted as plain AUDIT-C. Item-level pattern, then the total. Chart the three item scores in order with the total (frequency 3, typical quantity 1, heavy occasions 1; total 5 of 12) rather than the total alone. Two patients reach the same total by different routes, steady moderate drinking or intermittent heavy episodes, and the response differs. When every point comes from the frequency item and the quantity and heavy-occasion items score zero, the patient may be drinking within recommended limits, so review actual intake before calling the result a positive; Delaney and colleagues (2014) found score-based results inconsistent with the drinking reported on the same form in 13.8 to 21.1 percent of drinking men and 18.3 to 20.7 percent of drinking women across US and VA samples. Pitfall: "AUDIT-C 4" with no item pattern, or a positive built entirely from the frequency item charted as risky drinking without checking what the patient drinks. Threshold stated and applied. Name the convention and its source, then apply it: 4 or more for men and 3 or more for women (US primary care, Bradley and colleagues 2007, referenced by NIAAA), 5 or more for both sexes (VA clinical reminder, 2025 guide), 5 or more (Australian treatment guideline), or the higher older-adult cut points. Write the result under that convention, and when a score falls between conventions (a 4 in a man, a 3 or 4 in a woman over 65), say which rule decided it. Pitfall: "Positive" with no threshold, or the US primary-care rule applied inside a VA record or a treatment service that reads 5 or more. The patient's own account of intake. Add the patient's statement of drinks in a typical week, heavy days, the last drink, and pour size and strength converted to standard drinks (a 750 mL bottle of 12 percent wine is about five US standard drinks; a large pour or a strong craft beer is often two), plus pregnancy, medications that interact with alcohol, and any prior withdrawal. The screen bins intake into categories; the note needs the intake itself, read against the national limits, because that line supports the intervention, the claim, and any later comparison. Pitfall: "Two beers nightly" charted as two standard drinks, or a zero total charted as "no alcohol concerns" in a pregnant patient or someone in early recovery. Interpretation: a screen, then the follow-up. Write the conclusion as a screening result (positive consumption screen for unhealthy alcohol use; alcohol use disorder not established) and name what happened next: the full AUDIT or a DSM-5 criteria assessment when a disorder is possible, a withdrawal-risk assessment when the total is high and drinking is daily, and the population caveat (older adult, pregnancy, adolescent, translation) when it changes the reading. Higher totals raise the probability of a disorder; no primary authority converts the total into severity bands, and the VA's pairing of score ranges with escalating responses is a care pathway, not a diagnostic scale. Pitfall: "AUDIT-C 8, severe alcohol use disorder" with no criteria assessed, or a positive screen with nothing after it. Intervention, minutes, rescreen, and record. Document the brief intervention (feedback against the limits, the patient's goal, options offered), the referral made or declined, the face-to-face minutes for any time-based code, the rescreen instrument and date, and where the result lives. Annual screening is the routine convention (Medicare's benefit and MIPS 431 both run on a 12-month cycle); NIAAA advises following up at the next visit after a positive. A primary-care result is a HIPAA record, while the same result created by a federally assisted Part 2 program carries Part 2 consent and redisclosure rules. Pitfall: "Counseled on alcohol" with no content, goal, or minutes, or a bare score claimed under a counseling code. ### Blank template (copy and adapt) AUDIT-C DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [annual / intake / ED or trauma / preoperative / behavioral health / monitoring] Version: [original wording / USAUDIT-C / VA clinical reminder / local build] Standard drink used: [US 14 g / Canada 13.45 g / Australia 10 g / UK unit] Heavy-occasion item: [as published / adapted by sex; state it] Mode: [self-completed: paper, tablet, portal / asked verbally] Language: [ ] Item scores: frequency [ ] typical quantity [ ] heavy occasions [ ] Total: [ ] of 12 (all points from the frequency item? review intake) Threshold applied: [convention and source] Result: [positive / negative] Prior: [date, same version and drink] [ ] of 12 Change: [which items moved] Patient's account: [drinks per typical week; heavy days; last drink; pour size and strength converted; pregnancy; medications; prior withdrawal] Interpretation: [a consumption screen; disorder not established; pattern: steady use or heavy episodes; population caveat or none] Follow-up triggered: [full AUDIT / DSM-5 criteria assessment / withdrawal risk assessment (CIWA-Ar) / none indicated; result and date] Intervention: [feedback against limits; patient's goal; referral made, declined, or not indicated; face-to-face minutes] Plan: [rescreen instrument and date; treatment or counseling; next visit] Record: [code claimed (bare number); storage location; Part 2 status if created by a Part 2 program; consent on file] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the questionnaire items. ## Sample AUDIT-C documentation (fictional) Scenario: a US internal medicine practice's Medicare annual wellness visit, where a 67-year-old woman's portal-completed AUDIT-C comes back positive under the practice convention, the physician reviews the item pattern and her actual intake, assesses criteria, finds a medication interaction rather than a disorder, delivers timed counseling, and records the claim and the rescreen plan. All details are fictional. Patient: R.H., 67 · Setting: Internal medicine, Medicare annual wellness visit · Clinician: S. Lindqvist, MD · Note date: 09/16/2026 Screen: AUDIT-C, original three-item wording (heavy-occasion item at six drinks), US standard drink of 14 g shown on the form; self-completed in English on the patient portal on 09/15/2026, all three items answered, total computed by the EHR and confirmed by hand. Item scores in order: frequency 3, typical quantity 1, heavy occasions 1; total 5 of 12. Positive at the practice threshold of 3 or more for women (the US primary-care convention from Bradley and colleagues, 2007, as referenced by NIAAA). Prior: 4 of 12 at the 2025 wellness visit on the same portal form and wording (2, 1, 1); the change is in the frequency item. Patient's account: In her words: wine with dinner on two or three evenings a week, usually two generous pours from a 7-ounce glass, which we worked out together to about three standard drinks an evening; six or more drinks on a few evenings a year (holidays, a wedding in August); never before evening; last drink two nights ago. About 7 to 9 standard drinks in a typical week, above the NIAAA weekly limit for women and for adults over 65 and at the daily limit on drinking evenings. Takes zolpidem 5 mg most nights and reports two near-falls at night this summer. No prior withdrawal, no prior treatment, no concern voiced by family. Interpretation: Positive consumption screen for unhealthy alcohol use, with a self-reported pattern above recommended limits; the score is a screen and does not establish alcohol use disorder. Pattern is frequent moderate-quantity drinking rather than heavy episodes (four of five points from the frequency and quantity items). The result would also be positive under the VA convention (5 or more) and the older-adult convention for women (4 or more), so the threshold choice does not change today's reading; it is recorded because it will decide the reading when the score is 3 or 4. The clinical concern is alcohol on zolpidem nights in a patient with near-falls, not the total. Follow-up assessment: DSM-5 alcohol use disorder criteria reviewed in the visit: R.H. does not meet criteria (no loss of control, craving, tolerance, withdrawal, or role or social consequences). Full AUDIT not administered because the criteria review answered the dependence and harm question. Withdrawal risk low (no morning drinking, no prior withdrawal, no daily drinking); CIWA-Ar not indicated. GGT added to today's labs; fall risk assessment completed as part of the wellness visit. Brief intervention: Delivered by the physician: feedback on the score and where her weekly intake sits against the limits for women and adults over 65, the alcohol and zolpidem combination as the likely contributor to the near-falls, and a menu of options. R.H. chose a goal of no wine on zolpidem nights and no more than one standard drink on drinking evenings, and accepted a pour-size card. Zolpidem taper discussed and documented in the medication section. Referral not indicated. Face-to-face time: screening review 5 minutes, counseling 16 minutes, documented separately. Plan and record: Nurse phone check in 4 weeks on the goal and the near-falls; repeat the AUDIT-C on the same portal form at the three-month medication follow-up and at the next wellness visit, interpreted against the same 14 g drink. Annual screening claimed under G0442 and the counseling session under G0443, each supported by the tool, result, content, and time above; MIPS measure 431 met (systematic screen with brief counseling). Result recorded in the general medical record; this practice is not a Part 2 program. R.H. was told the screen is not a diagnosis and agreed with the plan. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, the heavy-occasion wording, and the standard drink are named, so the score can be reproduced and compared with the prior result on the same form. All three item scores appear with the total, and the prior pattern shows which item moved, so the entry can be interpreted rather than just counted. The threshold is stated with its source and applied, and the note says why the choice will matter at a lower score, which is what a VA or Australian reader needs to translate it. The patient's account converts pours to standard drinks and reads them against a named limit, and it surfaces the medication interaction that the total alone could not show. The result is labeled a screen, criteria are assessed and documented, the withdrawal-risk judgment is recorded, and the intervention has content, a patient-chosen goal, and separate minutes, so G0442, G0443, and measure 431 are all supported. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the instrument is a convention; the rules around it are real. No federal or state LAW names the AUDIT-C or prescribes a screening note format. The USPSTF recommendation of 2018 (grade B; CONVENTION, with quality-measure teeth) covers adults 18 and older including pregnant persons, names the AUDIT-C and the single-question screen among the most accurate brief instruments, and finds the evidence insufficient for adolescents; the topic is under update, with a draft statement posted August 5, 2025 that keeps both grades and names the AUDIT-C, the USAUDIT and USAUDIT-C, and the single question, and no final statement had published as of September 2, 2026. NIAAA's clinician resource shows the original wording with the six-drink heavy-occasion item, flags that the US adaptations use lower sex-specific amounts, and directs clinicians to obtain the weekly drinking pattern and assess for alcohol use disorder after a positive screen. The VA/DoD substance use disorder guideline (2021) recommends periodic screening with the AUDIT-C or a single item and a brief intervention for positives; the VA's 2025 quick reference guide states that the clinical reminder tailors the third item to sex assigned at birth, treats 5 or more as positive for both men and women, and requires documentation of brief alcohol counseling, and it pairs rising score ranges with escalating responses as a care pathway rather than as diagnostic bands. Under PAYER POLICY, Medicare's alcohol misuse benefit (NCD 210.8) pays one screen per 12 months under G0442 and up to four counseling sessions a year under G0443, no more than one a day, in a primary care setting; CMS's April 2026 booklet lists emergency departments, inpatient hospitals, ambulatory surgical centers, independent diagnostic testing facilities, skilled nursing facilities, inpatient rehabilitation facilities, and hospices as settings that do not qualify, so a clinically sound emergency-department AUDIT-C is not a Medicare preventive screen. The SBIRT codes (99408 and 99409 for commercial plans, G0396 and G0397 for Medicare, H0049 and H0050 where a state Medicaid program adopts them) are time-based and need a standardized tool, the result, the face-to-face time, and, for the intervention codes, the intervention itself, so a bare score supports none of them; MIPS quality measure 431 (2026) counts a systematic screen at least once in 12 months with brief counseling when unhealthy use is found. Under LAW, 42 CFR Part 2 as amended by the February 2024 final rule (compliance date February 16, 2026, now passed) attaches to records created by a federally assisted Part 2 program, not to the word alcohol: a primary-care AUDIT-C is a HIPAA record, a result recorded at intake in a Part 2 program is a Part 2 record, and the Part 2 consent page covers the mechanics. Canada and Australia: no requirement, and reference limits the score should be read against. In Canada, no federal or provincial LAW or payer names the AUDIT-C. Canada's Guidance on Alcohol and Health (Canadian Centre on Substance Use and Addiction, January 2023; CONVENTION) sets a continuum of risk: 2 or fewer standard drinks a week is low risk, 3 to 6 moderate, 7 or more increasingly high, with the Canadian standard drink at 13.45 g of ethanol and no known safe amount in pregnancy. A national clinical practice guideline update published in CMAJ in May 2026 (Wood and colleagues) states that existing screening tools have fallen behind that guidance and proposes a simpler approach that asks every patient about consumption and acts on the low-risk threshold, so a Canadian note should say whether an AUDIT-C total or the 2026 approach drove the counseling. In Australia, the RACGP preventive-activities guidance (alcohol recommendation dated June 28, 2024; CONVENTION) names the AUDIT-C for screening adults every two years, and the NHMRC guidelines (December 2020) set the limits at no more than 10 standard drinks a week and 4 on any one day, with a standard drink of 10 g and none under 18, in pregnancy, or while planning a pregnancy. The threshold splits inside the one country: the Guidelines for the Treatment of Alcohol Problems (2021 edition, current online in September 2026) read 5 or more as requiring further assessment, while the RACGP National Guide for Aboriginal and Torres Strait Islander people (4th edition) reads 4 or more in men and 3 or more in women as positive, which is why the convention belongs in the note. No MBS item names the AUDIT-C; screening sits inside attendance and health-assessment items. In every jurisdiction, an AUDIT-C in a pregnant patient is documented with the counseling that no amount is known to be safe rather than against an adult threshold, an adolescent gets an age-appropriate instrument, and a translated form is charted with its language and version, because the WHO manual lists English, Spanish, Hindi, Japanese, and Thai and other translations may not match the original wording. The AUDIT is copyrighted by the World Health Organization and made available under the non-commercial terms of its 2001 manual and the AUDIT custodial site: the questionnaire may be reproduced and used for any non-commercial purpose without permission, provided it is not materially changed, carries a note that it is a WHO-approved instrument, and no one completing it is charged a fee, and the custodial FAQ states that "as a WHO-approved instrument, the AUDIT is in the public domain" on that basis, while WHO's current permissions policy asks commercial users to seek permission, so this page does not treat that statement as a commercial or EHR-vendor license. The AUDIT-C consumption items were validated as a standalone screen by Bush and colleagues (Archives of Internal Medicine, 1998), and the USAUDIT-C and VA adaptations belong to their agencies. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common AUDIT-C documentation errors reviewers flag The numbers behind these errors are specific. In the original validation of 243 male veterans, the three consumption items detected heavy drinking slightly better than the full AUDIT (area under the curve 0.891 against 0.881 ) and active abuse or dependence slightly worse (0.786 against 0.811), which is the evidence that the AUDIT-C is a consumption screen and not the dependence instrument (Bush and colleagues, 1998). The familiar 4-for-men and 3-for-women convention comes from a later primary-care sample of 392 men and 927 women (sensitivity 0.86 and specificity 0.89 in men, 0.73 and 0.91 in women; Bradley and colleagues, 2007 ), the VA reads 5 or more as positive for both sexes, and a 2021 community study of 1,577 older adults found the best cut points at 5 or more for men and 4 or more for women (van Gils and colleagues). Delaney and colleagues (2014) found that score-based screening results disagreed with the drinking reported on the same form for 13.8 percent of drinking men and 18.3 percent of drinking women in a US general population sample and 21.1 and 20.7 percent in a VA sample. No published audit counts how often the version, the standard drink, or the threshold is missing from AUDIT-C chart entries; the Medicare, SBIRT, and measure 431 documentation rules described under compliance considerations are what turn those omissions into a claim problem. The BastionGPT Clinical Advisory Board sees the same errors most often in AUDIT-C documentation reviews: Version and standard drink missing. "AUDIT-C 5" means one thing on the VA reminder, another on the original wording, and a third on a UK-unit form, and a 5 built on 10 g Australian drinks is a different exposure from a 5 built on 14 g US drinks. Name the wording, the heavy-occasion item if adapted, and the drink definition every time, and keep them constant when you trend the score. Total without the item pattern. A bare total cannot show whether the patient drinks steadily at moderate quantities or rarely and heavily, and it hides the case where every point comes from the frequency item while the quantity and heavy-occasion items are zero. Chart the three scores in order, and when the pattern says the intake may be within limits, review the intake before calling the screen positive. Positive with no threshold stated. "AUDIT-C positive" is undefined: a 4 in a man is positive in a US primary-care clinic, negative in the VA, and a prompt for further assessment under the Australian treatment guideline only at 5. Write the convention and its source, and state that the reading would change under another convention when it would. Severity bands invented from the total. "AUDIT-C 8, likely dependence" or "6 to 7, harmful use" reads a consumption screen as a diagnostic scale. No primary authority publishes AUDIT-C severity bands; the VA pairs score ranges with escalating responses as a care pathway, and NIAAA requires a symptom assessment after a positive screen. Chart the probability language and the criteria assessment, never a band as a diagnosis. Containers counted as drinks, or a zero read as no concern. "Two beers nightly" can be four standard drinks when the beer is strong and the can is large; a 750 mL bottle of wine is about five US drinks. Convert by volume and strength, and read a zero as no past-year drinking reported, not as "no alcohol concerns," in a pregnant patient, a patient in recovery, or someone on interacting medication. Intervention, minutes, and follow-up undocumented. A note reading "counseled on alcohol" with no content, goal, referral decision, or minutes fails G0443, the time-based SBIRT codes, and measure 431, and a positive screen with no named next step (full AUDIT, criteria assessment, withdrawal check) cannot be defended. Record the feedback, the patient-chosen goal, the minutes, the follow-up instrument, and the rescreen date. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on substance use assessments and screening documentation. Give it the facts (the wording and drink definition used, mode and language, the three item scores, the threshold your setting applies, the patient's intake in their words, any criteria or withdrawal assessment and its result, the intervention and its minutes, and where the result is stored) and it drafts the documentation block: the screen result with the convention named, the item pattern, the patient's account converted to standard drinks, a screening-level interpretation, the follow-up, the intervention, the plan, and the record placement, ready for your review. Cross-check a finished note for the gaps reviewers flag: a missing version or drink definition, a total with no item pattern, a positive with no threshold, a severity band charted as a diagnosis, containers counted as drinks, or an intervention with no content or minutes. Turn the screen into the next document: the opening of a substance use assessment, the withdrawal-risk note for a preoperative team, or the SBIRT and G0443 documentation a payer expects, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the AUDIT-C scored, and what score is a positive screen? Each of the three items scores 0 to 4, so the total runs 0 to 12, and a zero means the patient reported no drinking in the past year. The positive threshold is a convention that depends on where you work: 4 or more for men and 3 or more for women is the US primary-care convention (Bradley and colleagues, 2007, referenced by NIAAA); the VA's 2025 guide treats 5 or more as positive for both men and women on its sex-tailored clinical reminder; Australia's treatment guideline reads 5 or more as needing further assessment while the RACGP National Guide uses 4 and 3; and older-adult validations favor 5 for men and 4 for women. None of these is part of the instrument, and none diagnoses anything. Chart the item scores in order, the total, and the convention you applied with its source, so a reader in another system can translate the result. When a score falls between conventions, say which rule decided it. Which threshold do I use for a woman, an older adult, or a VA patient, and what about pregnancy and adolescents? Use the convention your setting has adopted and write it down. In general US primary care that is 3 or more for women and 4 or more for men, from a validation in which sensitivity for women was 0.73 at that cut point, so a negative in a woman rules out less than a negative in a man. In the VA the clinical reminder decides: 5 or more is positive for both sexes, and the third item is already tailored to sex assigned at birth. For adults over 65, a 2021 community validation found the best cut points at 5 or more for men and 4 or more for women, and Australian older-person guidance discusses a lower cut point of 4 with at least annual review; document which you applied rather than silently changing the number because of age. In pregnancy, the USPSTF includes pregnant adults in the screening recommendation, but the counseling question is any use, not a threshold, so chart the counseling that no amount is known to be safe. For adolescents the USPSTF finds the evidence insufficient, and NIAAA and the AAP point to youth-specific instruments, so do not lower the adult cut point for a teenager. What is the difference between the AUDIT-C and the full AUDIT, and when do I use each? The full AUDIT has ten items in three domains: three about consumption, three about dependence-related experiences, and four about harms. The AUDIT-C is the first three alone. In the original validation the three items detected heavy drinking slightly better than the full instrument and active abuse or dependence slightly worse (Bush and colleagues, 1998), which is exactly the division of labor: use the AUDIT-C to find unhealthy drinking quickly, and use the full AUDIT, or a DSM-5 criteria assessment, when a positive screen needs the dependence and harm picture before any severity language enters the chart. Australia's treatment guideline reads a full AUDIT total of 8 or more as presumptive hazardous or harmful drinking and 15 or more as a prompt to assess for dependence; those are conventions for the ten-item instrument and do not transfer to the three-item total. Document "AUDIT-C positive" and "full AUDIT" as separate results with their own dates, and record the criteria assessment in the substance use assessment rather than in the screen line. Is a positive AUDIT-C a diagnosis of alcohol use disorder? No. It is a positive consumption screen, which raises the probability of unhealthy alcohol use and of a disorder but establishes neither; alcohol use disorder and its severity come from DSM-5 criteria, not from the total. The defensible chain in the chart is: the screen result under a named convention; the patient's actual intake; a criteria assessment or the full AUDIT when a disorder is possible; a withdrawal-risk assessment (and a CIWA-Ar plan where indicated) when the total is high and drinking is daily; then the diagnosis, if any, in its own line. Write "positive screen; alcohol use disorder not established" until the criteria are assessed, and never map the total onto severity bands: no primary authority publishes them, the VA's pairing of score ranges with escalating responses is a care pathway, and in specialty treatment populations heavy consumption is so common that the three items cannot grade disorder severity. Why does the standard drink matter, and how do I count a large pour or a strong beer? Because the items count drinks, and a drink is 14 g of ethanol in the United States, 13.45 g in Canada, 10 g in Australia, and about 8 g per UK unit. The same answer of three drinks a night is about 42 g in a US chart and 30 g on an Australian form, and a 750 mL bottle of 12 percent wine is about five US standard drinks, seven Australian ones, and nine UK units, so a score with no drink definition is not comparable across borders or between EHR builds. Convert by volume and strength rather than counting containers: a 7-ounce pour of wine is well over one US drink, a 16-ounce can of 8 percent beer is about two, and patients routinely under-count both. Chart the standard drink the patient was told to use, put the converted weekly total in the patient's account, and keep the definition constant when you trend the score. Can I bill SBIRT or Medicare G0442 and G0443 on an AUDIT-C alone? Not on the score alone. The AUDIT-C is an accepted standardized tool, so it can support the screening component, but every code is PAYER POLICY with its own conditions. Medicare pays one alcohol misuse screen per 12 months under G0442 and up to four counseling sessions a year under G0443 for beneficiaries who screen positive, no more than one a day, in a primary care setting; the April 2026 CMS booklet excludes emergency departments, inpatient hospitals, ambulatory surgical centers, skilled nursing facilities, inpatient rehabilitation facilities, and hospices, so an emergency-department screen is clinically right and still not a Medicare preventive service. The SBIRT codes (99408 and 99409 commercially, G0396 and G0397 for Medicare, H0049 and H0050 where a state Medicaid program adopts them) are time-based and need the named tool, the result, the face-to-face minutes, and, for the intervention codes, a documented intervention with content; a bare score meets none of that, and Medicaid rules are state-specific. MIPS measure 431 (2026) is a quality measure rather than a payment code: it counts a systematic screen within 12 months and brief counseling when unhealthy use is found. The sample above shows the minimum: tool, result, criteria review, counseling content and goal, and separate minutes. How often should I repeat the AUDIT-C, and can I trend the score? For routine prevention, annual screening is the working convention: Medicare's benefit and MIPS measure 431 run on a 12-month cycle, the WHO manual favored annual screening, and Australian guidance for older people asks for at least annual review; the USPSTF sets no interval. Repeat sooner when the result will monitor an intervention, after a positive, or when circumstances change (pregnancy, a new interacting medication, an injury, liver disease, preoperative planning). Trending is legitimate as a consumption measure, with one rule: keep the version, the heavy-occasion wording, the standard drink, the mode, and the scoring logic the same on both occasions and chart the item pattern each time, so a fall from 7 to 4 can be read as fewer heavy episodes rather than as a changed form. No validated change score defines treatment response, so describe the change; do not label it. Serial results follow the conventions of the outcome measure note . What do the three AUDIT-C items ask, and can we put the questionnaire in our EHR, on a form, or on our website? At the topic level the items ask how often the person drinks, how many standard drinks they have on a typical drinking day, and how often they drink heavily on one occasion, each over the past year. This page does not print or paraphrase the questions or response options; the WHO manual (2001) and the AUDIT custodial site are the sources of record. The rights position is more precise than the phrase public domain suggests: no permission is needed for any non-commercial use, the questionnaire may be reproduced on printed forms and in a health system's own records for administration, it must not be materially changed, it should carry a note that it is a WHO-approved instrument, and no one completing it may be charged a fee. On that basis the custodial FAQ calls the instrument public domain, but WHO's current permissions policy asks commercial users to seek permission, so an EHR vendor, an ad-supported web tool, or a paid app should clear rights rather than rely on that phrase. If you use a national adaptation, name it (USAUDIT-C, VA clinical reminder) and cite its agency as well as Saunders 1993 and Bush 1998. Can BastionGPT draft my AUDIT-C documentation? Yes. Give it the facts (the wording and drink definition used, mode and language, the three item scores, the threshold your setting applies, the patient's intake in their words, any criteria or withdrawal assessment and its result, the intervention and its minutes, and where the result is stored) and it drafts the full entry: the screen result with the convention named, the item pattern, the patient's account converted to standard drinks, a screening-level interpretation, the follow-up, the intervention, the plan, and the record placement, ready for your review. It can also check a finished note for a missing version or drink definition, a total with no item pattern, a positive with no threshold, a severity band charted as a diagnosis, containers counted as drinks, or an intervention with no content or minutes. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M, 1993, Addiction, development of the AUDIT (WHO collaborative project); World Health Organization, 2001, AUDIT: guidelines for use in primary health care, second edition (WHO/MSD/MSB/01.6a; English, Spanish, Hindi, Japanese, Thai); AUDIT custodial site, FAQ (non-commercial use without permission; not materially changed; WHO-approved note; no fee to the respondent); WHO, permissions and licensing . Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA, 1998, Archives of Internal Medicine, the AUDIT alcohol consumption questions (AUDIT-C) (243 male veterans; areas under the curve against the full AUDIT); Bradley KA and colleagues, 2007, Alcoholism: Clinical and Experimental Research, AUDIT-C as a brief screen for alcohol misuse in primary care (392 men, 927 women; 4 and 3 cut points); Delaney KE and colleagues, 2014, Addiction Science and Clinical Practice, inconsistencies between AUDIT-C scores and reported drinking ; van Gils Y and colleagues, 2021, AUDIT-C cut points in community-dwelling older adults . US Preventive Services Task Force, 2018, unhealthy alcohol use: screening and behavioral counseling (grade B adults; I adolescents), and 2025, draft recommendation statement (posted August 5, 2025; AUDIT-C, USAUDIT-C, single question); NIAAA, Core Resource on Alcohol: screen and assess (original wording; US adaptations; assess for AUD after a positive; self-report accuracy). US Department of Veterans Affairs, Pharmacy Benefits Management Academic Detailing Service, 2025, alcohol use disorder quick reference guide (sex-tailored third item; 5 or more positive for both sexes; brief alcohol counseling documentation); VA/DoD, 2021, substance use disorders clinical practice guideline provider summary . CMS, 2011, decision memo on screening and behavioral counseling to reduce alcohol misuse (NCD 210.8); CMS Medicare Learning Network, April 2026, substance use screenings and treatment booklet (G0442 and G0443; primary care setting and exclusions; SBIRT components; bare code numbers only); CMS Quality Payment Program, 2026 MIPS quality measure 431 ; SAMHSA, SBIRT (reimbursement pathways). US Department of Health and Human Services, 2024, 42 CFR Part 2 final rule fact sheet (compliance date February 16, 2026); eCFR, 42 CFR 2.11 definitions . Canadian Centre on Substance Use and Addiction, January 2023, Canada's Guidance on Alcohol and Health (continuum of risk; 13.45 g standard drink); Health Canada, alcohol and low-risk drinking guidance ; Wood E and colleagues, May 2026, CMAJ, screening for high-risk drinking and alcohol use disorder: update of the 2023 national clinical practice guideline . RACGP, Guidelines for preventive activities in general practice: alcohol (recommendation dated June 28, 2024; AUDIT-C every two years), and National Guide to preventive healthcare for Aboriginal and Torres Strait Islander people, 4th edition: alcohol (4 or more in men, 3 or more in women); Australian Government, Guidelines for the Treatment of Alcohol Problems: the AUDIT (three domains; AUDIT-C 5 or more for further assessment); NHMRC, December 2020, Australian guidelines to reduce health risks from drinking alcohol (10 a week, 4 a day; 10 g standard drink). ## Related templates CAGE Questionnaire Documentation DAST-10 Score Documentation Substance Use Assessment CIWA-Ar Protocol Documentation 42 CFR Part 2 Consent General Risk Screening Note Outcome Measure Review Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/geriatric-depression-scale All note templates # Geriatric Depression Scale (GDS-15) Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Geriatric Depression Scale (GDS) is a public-domain, yes-or-no self-report depression screen for older adults, with a 30-item original (Yesavage and colleagues, 1982) and the 15-item GDS-15 short form (Sheikh and Yesavage, 1986). Primary care, geriatric, and behavioral health clinicians use it to decide who needs a full mood assessment; it screens, never diagnoses. This page covers how to document and interpret GDS-15 results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report on paper, or read aloud by any clinician or trained staff member with the patient's own answers recorded; no license, training requirement, or publisher qualification level applies; the treating clinician interprets and documents Audience Primary care and geriatric clinicians, nurses and nurse practitioners, psychologists and behavioral health clinicians, assisted-living and home-health teams, quality-measure abstractors, payers and auditors reviewing wellness-visit and screening claims Typical length 3 to 6 chart lines (form and mode, cognitive status, score with the threshold applied, interpretation, separate risk note, follow-up plan) · patient completion about 5 to 7 minutes Format family Brief self-report screen (15 yes-or-no items, one point per answer in the depressive direction, total 0 to 15, past-week frame; the 30-item original and the 5- and 4-item derivatives are separate forms with their own ranges) When it's used Medicare annual wellness visits and annual depression screening, geriatric and primary care intakes, assisted-living and community reviews, comprehensive geriatric assessment in hospital and oncology, and repeat measurement during treatment; never as a suicide screen Standards context Yesavage and colleagues 1982 (GDS-30) and Sheikh and Yesavage 1986 (GDS-15); the original scale is public domain per the Stanford author site; an example tool in CMS Quality ID 134 (2026) and in CCSMH guidance (2021), mandated by no law; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Geriatric Depression Scale? The Geriatric Depression Scale is a self-report screen for depressive symptoms in older adults, developed by Jerome Yesavage, T. L. Brink, and colleagues at Stanford and the Palo Alto VA. The 30-item original appeared in the Journal of Psychiatric Research (volume 17, pages 37 to 49; PubMed dates it 1982 and the Stanford site 1983, so both years circulate), after an early report by Brink and colleagues in Clinical Gerontologist in 1982. The 15-item short form, the GDS-15, comes from a 1986 chapter by Javaid Sheikh and Jerome Yesavage in Clinical Gerontology: A Guide to Assessment and Intervention (Haworth Press), and it is the form most practices mean when they write GDS. Every item is answered yes or no about the past week and scores one point when answered in the depressive direction; some items count when answered yes and others when answered no, so the total is not a count of yes answers. The developers started from a 100-item pool and kept the items that best separated depressed from non-depressed elders, which is why complaints about sleep, appetite, and energy do not appear on it and why it is less confounded by medical illness than the PHQ-9 . Two shorter derivatives are common: the five-item GDS-5 of Hoyl and colleagues (Journal of the American Geriatrics Society, 1999) and the four-item set of D'Ath and colleagues (Family Practice, 1994), and a 2021 review counted five different four-item and seven different five-item versions in the literature, so a short form is identified by its author as well as its length. Stanford also hosts a 15-item informant version (Brown and Schinka, 2005) and more than 40 translations, with the note that the authors cannot vouch for the accuracy of the translations. Three facts carry the documentation. First, the threshold belongs to the form: the author site says that a GDS-15 total higher than 5 should prompt a careful interview, its scoring page reads totals above 10 as almost always depression, and the validation studies mostly evaluate a cut point of 5 or more, so the note has to name the form, the denominator, and the threshold it applied. A bare "GDS 7" is positive on 15 items, sits inside the original normal range on 30, and is impossible on a four- or five-item form. Second, a positive screen is not a diagnosis, and the informal severity labels that circulate for the GDS-15 are conventions rather than diagnostic strata: an individual participant data meta-analysis in 2024 found that a GDS-15 threshold of 5 or more identified 34.2 percent of older adults as depressed against 14.8 percent by structured interview. Third, the scale has no suicide item, so a positive result does not discharge the duty to ask about suicide risk, and a low score says nothing about it; the PHQ-9 carries a self-harm item and its own follow-up rules, and the suicide risk assessment page covers the direct assessment. A fourth fact governs who can be screened at all: the GDS is self-report, so every entry states the patient's cognitive status and ability to answer, using the GDS through mild and moderate impairment where the patient can follow yes-or-no questions and moving to an observer-rated instrument such as the Cornell Scale for Depression in Dementia when self-report is no longer reliable. ## Who uses GDS-15 documentation and when In the United States the GDS-15 lives in the Medicare annual wellness visit and the annual depression screening benefit, where CMS asks for a standardized tool recognized by national professional medical organizations and names none, and in the 2026 Quality ID 134 specification, which lists the GDS and the Cornell Scale among acceptable adult tools. Family physicians, internists, geriatricians, and nurse practitioners chart it at wellness visits and intakes, assisted-living and home-visit teams use it for community residents, and hospital and oncology teams place it inside a comprehensive geriatric assessment because it avoids the somatic items that fatigue, anorexia, and treatment effects would inflate on a PHQ-9 . Nursing homes and home health agencies are the exception: the MDS 3.0 resident mood interview and the OASIS mood section run on the PHQ family, so there the GDS is a clinical adjunct rather than the required record. In Canada the CCSMH guideline (2021) recommends the GDS or the PHQ-9 for older adults without significant cognitive impairment, and in Australia the 75-plus health assessment must cover psychological function including cognition and mood without naming a tool. The neighbors are fixed: a Mini-Cog , MoCA , MMSE , or SLUMS result is the cognitive statement every GDS entry references; the Cornell Scale takes over when self-report is unreliable; the suicide risk assessment or C-SSRS carries the risk question the GDS cannot ask; serial results follow the outcome measure note ; and a positive screen that leads to treatment continues in the medication management note or a psychotherapy progress note . ## How to document GDS-15 results in the chart No law, payer, or author prescribes a GDS note format. What exists is an author site with scoring guidance, a quality measure that requires the tool's name and a dated follow-up plan, and regulatory assessment systems in nursing homes and home health that use a different mood instrument. What survives review is an entry that names the form and its denominator, states how the scale was administered and whose answers it holds, records the patient's cognitive status, gives the total with the threshold applied, labels the result a screen, documents a separate suicide-risk question, and closes with a dated follow-up plan. Each element below carries the pitfall that most often undermines it. Form, denominator, and time frame. Write GDS-15, GDS-30, or the short derivative with its author (Hoyl five-item, D'Ath four-item), the denominator (7 of 15), the language or translation, and the past-week frame as worded. The forms have different item sets, ranges, and threshold conventions, and short forms of the same length are not one instrument, so the form is the first thing a later reader, an abstractor, or a payer needs. Pitfall: "GDS 7" with no form: positive on 15 items, inside the original normal range on 30, impossible on 4 or 5. Mode and answer source. Record whether the patient self-completed the form on paper or a tablet or a clinician read the items aloud and recorded the patient's own answers, whether an interpreter or a validated translation was used, and whether any workaround for vision, hearing, or aphasia was needed. A relative's answers are not the patient's GDS: if collateral report is the only option, use the informant version (GDSI-15) or an instrument built for informant input and label the result as informant-rated. Pitfall: A daughter's answers charted as the patient's GDS-15, or a translated form charted as plain GDS-15 with no language. Cognitive status and capacity to answer. Name the cognitive screen and its result (Mini-Cog 4 of 5, MoCA 22 of 30), the absence or presence of delirium, any established diagnosis with its stage, and whether the patient followed and answered yes-or-no questions. The evidence supports the GDS-15 through mild and moderate impairment and not in severe impairment, and completion falls sharply below an MMSE of about 10; no authority sets a MoCA or MMSE score at which validity stops, so document the observed capacity rather than a cutoff. When self-report is unreliable, say so and switch to the Cornell Scale. Pitfall: A GDS-15 in advanced dementia with no cognitive statement, or "valid" declared from a diagnosis label alone. Score, key, and completeness. Give the total over the denominator, confirm the key was applied to both answer directions, and say who computed the score. If items were missed, apply the author site's rule (prorate the completed items to the full denominator and round up) and chart the prorated figure as prorated; the same site suggests, as the author's own estimate, that more than five missing items invalidate the scale. Keep the item-level answers on the flowsheet rather than in the narrative. Pitfall: A count of yes answers charted as the score, or a prorated total with no note that items were missing. Threshold applied and the screening conclusion. State the convention and its source, then apply it: higher than 5 for a careful interview (author site), 5 or more as the practice protocol or the cut point most validation studies used, above 10 as almost always depression (author site scoring page). Write the result as a screen: "positive screen at the stated threshold; depression not established." The four-band severity labels that circulate for the GDS-15 have no primary-source authority and do not map onto DSM-5 severity, so they never enter the diagnosis line. Pitfall: "GDS-15 9, moderate depression" charted as a diagnosis, or "positive" with no threshold named. Suicide risk assessed separately. The GDS contains no suicide or self-harm item. Ask directly, record the question asked and the answer, name the tool if one was used (C-SSRS screener), and file the formulation and any safety plan in the risk section of the record. A low total documents nothing about ideation and a high total nothing about intent; the Hartford Institute guidance is that any positive short-form screen prompts a fuller assessment and an evaluation for suicidality. Pitfall: "GDS-15 3, no safety concerns" with no direct question asked. Follow-up plan, claim, and record. Document the plan a positive screen triggers (referral for evaluation, treatment, or another intervention for depression) on the date of the encounter or within two calendar days, and put the tool's name in the record; Quality ID 134 requires the name, not the score, and does not accept a repeat screen or a suicide risk assessment as the plan. Record the claim (G0444 with a subsequent wellness visit; it is bundled into the IPPE and the initial visit), the setting's other required mood instrument if any, and the repeat plan. Pitfall: A positive screen with no plan, or a repeat screen or suicide assessment logged as the plan. ### Blank template (copy and adapt) GDS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [wellness visit / intake / assisted living / hospital or oncology / monitoring] Form: [GDS-15 / GDS-30 / GDS-5 (author) / GDS-4 (author)] Language: [ ] Mode: [self-completed / read aloud, patient's own answers / interpreter / informant version, labeled as such] Time frame: [past week, as worded] Cognitive status: [screen and result; delirium absent or present; diagnosis and stage; followed yes-or-no questions: yes / no] Score: [ ] of [15] Key applied to both answer directions: [ ] Completeness: [all items answered / missing items; prorated, rounded up] Threshold applied: [convention and source] Result: [positive / negative] Prior: [date, same form and mode] [ ] of [15] Change: [ ] Interpretation: [a screen, not a diagnosis; congruence with interview; grief, medical, medication, substance contributors; function] Suicide risk (no GDS item): [direct question asked; tool; findings; formulation and safety plan filed at ...] Follow-up plan: [referral / treatment / other intervention; dated on the encounter date or within two calendar days] Record: [tool name in the record; code claimed (bare number); required setting instrument (MDS or OASIS mood interview) completed separately] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale items. ## Sample GDS-15 documentation (fictional) Scenario: a US internal medicine practice's Medicare subsequent annual wellness visit, where a 79-year-old widow's paper GDS-15 comes back positive, the physician documents the cognitive screen, the threshold, a direct suicide-risk question, the interview, and a same-day follow-up plan, and records the claim. All details are fictional. Patient: E.R., 79 · Setting: Internal medicine, Medicare subsequent annual wellness visit · Clinician: M. Okafor, MD · Note date: 09/15/2026 Screen: GDS-15 (15-item short form, English), self-completed on paper in the waiting room on 09/15/2026, past-week frame as worded, all 15 items answered, scored by the medical assistant with the key applied to both answer directions and confirmed by me. Total 7 of 15. Positive at the practice threshold of 5 or more, and above the author site's interview trigger of a total higher than 5. Item-level answers are on the screening flowsheet. Prior: 2 of 15 on the same form at last year's wellness visit. Cognitive status: Mini-Cog completed today as the wellness visit's cognitive assessment: 4 of 5 (recall 2 of 3, clock normal), negative under the practice protocol; no delirium, alert and oriented, followed the yes-or-no format without difficulty. No cognitive diagnosis. Self-report judged reliable, so the GDS-15 result is interpreted at face value. Interpretation: Positive depression screen; the GDS-15 is a screen and does not establish a diagnosis. In the interview E.R. describes low mood most days, tearfulness, and loss of interest in her garden club since her husband's death five months ago, with sleep and appetite preserved and no psychomotor change. No prior depressive episode, no alcohol use, medication list unchanged for two years. Working impression: depressive symptoms in the context of bereavement, major depressive disorder not established today; DSM-5 criteria and duration will be assessed at the follow-up visit. Result is congruent with the interview and with the change from last year's score. Risk: The GDS-15 contains no suicide item. Asked directly: denies thoughts of death, of being better off dead, or of harming herself, now or since her husband died; C-SSRS screener negative; no prior attempts; daughter lives nearby and visits daily. Formulation and the standing offer of the practice's crisis contact recorded in the risk section of the record. Follow-up plan and claim: Plan discussed with E.R. and documented today: referral placed to the practice's collaborative care behavioral health clinician for a diagnostic assessment and grief-focused counseling, information given on the hospital's bereavement group, TSH and B12 added to today's labs, nurse phone check on 09/22/2026, office visit in three weeks to assess criteria and discuss treatment options if symptoms persist. Screening reviewed and addressed on the encounter date; screening time 8 minutes. Annual depression screening claimed under G0444 with the subsequent wellness visit under G0439, supported by the named tool, result, and plan above; Quality ID 134 met (tool named, positive result, follow-up plan documented on the encounter date). E.R. was told the screen is not a diagnosis and agreed with the plan. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The form, denominator, language, mode, and time frame are named, so the 7 of 15 can be read, reproduced, and compared with the prior result on the same form. The cognitive screen and its result sit in the same note with a statement that the patient could follow the format, which is what makes a self-report score interpretable. The threshold is stated with its source, the result is written as a screen, and the diagnosis is left open until criteria are assessed at a dated visit. The missing suicide item is named and a direct question with a named tool is documented separately, so a positive or negative GDS is never mistaken for a risk finding. The follow-up plan is a referral and interventions, documented on the encounter date, and the claim and quality-measure elements are supported by the tool name, result, and plan. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the instrument is a convention; the rules around it are real. No federal or state LAW names the GDS or prescribes a screening note format. The Medicare annual wellness visit (42 CFR 410.15; PAYER POLICY) makes review of potential risk factors for depression a required element of the initial visit and tells clinicians to select from standardized screening tools recognized by national professional medical organizations, naming none; detection of cognitive impairment is a separate required element, which is why a GDS entry and a cognitive screen belong in the same note. Medicare's annual depression screening benefit (NCD 210.9, effective October 14, 2011; PAYER POLICY) covers one screen per year, up to 15 minutes, in a primary care setting with staff-assisted depression care supports, and states that CMS does not identify specific screening tools; it is billed under G0444, which is bundled into the IPPE and the initial wellness visit and separately payable with subsequent visits. CMS Quality ID 134 (2026 MIPS specification; PAYER POLICY with quality-measure teeth) requires an age-appropriate standardized tool, lists the Geriatric Depression Scale and the Cornell Scale for Depression in Dementia among the adult examples alongside the PHQ-9, BDI, CES-D, and others, requires the name of the tool in the medical record but not the score, and requires that a positive screen carry a follow-up plan (referral for evaluation, pharmacological intervention, or another intervention for depression) discussed during the encounter and documented on the date of the encounter or within two calendar days; a further screening or a suicide risk assessment does not count as the plan, positive results suggesting high self-harm risk call for more urgent action, and cognitive, functional, or motivational limitations that may affect accuracy are an allowed medical reason for not screening. Under LAW (the federal resident and patient assessment systems), the MDS 3.0 Resident Assessment Instrument (manual v1.20.1, effective October 1, 2025) runs Section D on the PHQ-2 to 9 resident interview and the PHQ-9 staff-observation version, validated for MDS 3.0 by Saliba and colleagues in 2012 (weighted kappa 0.69 against a gold-standard interview; 86 percent of residents could complete it), and the home health OASIS-E2 data set (effective April 1, 2026) carries its own mood section, so in nursing homes and home health a GDS is a clinical adjunct that never replaces the required mood items. HEDIS depression screening measures likewise accept only the instruments in their published value sets, so a practice reporting them should check the current specification before treating a GDS result as measure credit. The result is a HIPAA record; nothing about it invokes 42 CFR Part 2. Canada, Australia, and the evidence on cognition. In Canada no federal or provincial LAW or payer names the GDS. The Canadian Coalition for Seniors' Mental Health guideline (2021 update; CONVENTION, professional-society guidance) recommends the GDS or the PHQ-9 for older persons without significant cognitive impairment in general medical or geriatric settings, and recommends an observer-rated instrument such as the Cornell Scale for Depression in Dementia instead of the GDS for patients with moderate to severe cognitive impairment, both graded B and unchanged since 2006. British Columbia's cognitive-impairment guideline appendix lists the short GDS among tools for minor cognitive loss or early-stage dementia and the Cornell Scale for advanced impairment, and tells clinicians to choose by the person's remaining cognitive abilities and to follow any positive screen with assessment. Provincial long-term care runs on the interRAI RAI-MDS 2.0 in most provinces, whose Depression Rating Scale is an observer-rated score on a 0 to 14 range, not a GDS, so a "depression score" in a Canadian facility record needs its instrument named. In Australia, the older person's health assessment (MBS items 701, 703, 705, and 707; PAYER POLICY) is claimable once every 12 months for people 75 and over who are not inpatients or residential care residents and must include an assessment of psychological function including cognition and mood, with no instrument named, so a GDS-15 fits and is not required. In residential aged care the Aged Care Funding Instrument carried a modified Cornell Scale until the AN-ACC funding model replaced it in October 2022 with no depression measure, according to the Australian Institute of Health and Welfare, and Victoria's standardised care process for depression (October 2023; CONVENTION) recommends the Cornell Scale on admission and when symptoms emerge, paired with the Psychogeriatric Assessment Scales cognitive screen, and states that it is not diagnostic. The evidence behind the cognitive-status rule is now specific. In 24,674 US participants stratified by Clinical Dementia Rating, Bennett and Wang (2024) found negligible bias in GDS-15 scores at questionable, mild, and moderate impairment and significant bias at severe impairment, and concluded the scale can be used with mild or moderate impairment but not severe; a 2026 analysis of 14,077 people in the same national dataset found differential item functioning in 13 of 15 items, with cognitive status the strongest influence, which is the reason to record cognition rather than declare the scale valid; among 834 people aged 85 and over, Conradsson and colleagues (2013) found the GDS-15 completed by 1.5 percent of those with an MMSE of 0 to 4, 41.7 percent at 5 to 9, 65.5 percent at 10 to 14, and more than 90 percent at 20 and above; and in nursing homes the self-rated and the informant GDS both failed to detect depression adequately in residents with dementia (areas under the curve 0.57 to 0.70; Li and colleagues, 2015). Meta-analyses of medical settings found accuracy not significantly influenced by the presence of dementia (Mitchell and colleagues, 2010), so the boundary is functional, whether this patient can understand and reliably answer, and the chart should say what was observed. A translated GDS is charted with its language and version, because the author site hosts more than 40 translations it cannot vouch for and the cut point can shift with the version. The Geriatric Depression Scale is in the public domain; the authors are Jerome A. Yesavage, T. L. Brink, and colleagues for the 30-item scale (Journal of Psychiatric Research, 1982 to 1983) and Javaid I. Sheikh and Jerome A. Yesavage for the 15-item short form (Clinical Gerontology: A Guide to Assessment and Intervention, Haworth Press, 1986), and the Stanford author site states that the original scale is in the public domain because it was partly the result of federal support, hosts the forms, the informant version, the translations, and free scoring apps, and asks users to look for its versions because other versions circulate. That statement covers the original scale; the five-item form of Hoyl and colleagues (Journal of the American Geriatrics Society, 1999) and the four-item set of D'Ath and colleagues (Family Practice, 1994) were published in journals, and at least one instrument-licensing database lists rights conditions for the short forms, so a commercial product, app, or EHR vendor should confirm the rights of the exact version it embeds rather than assume every derivative shares the original's status. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common GDS-15 documentation errors reviewers flag The numbers behind these errors are specific. An individual participant data meta-analysis of 14 studies and 3,602 older adults found that a GDS-15 threshold of 5 or more classified 34.2 percent as depressed against 14.8 percent by structured interview, with 8 or more the closest match (Parsons and colleagues, 2024); at a cut point of 5 the pooled sensitivity is 0.89 and specificity 0.77 across 32 studies, with the authors warning of selective reporting of cut points (Pocklington and colleagues, 2016), and 86 and 79 percent across 53 studies and 17,018 participants (Krishnamoorthy and colleagues, 2020). A 2021 review found five four-item and seven five-item versions in use with different accuracy at the same numerical threshold and rated the evidence low certainty. In 24,674 adults the GDS-15 showed negligible bias through moderate cognitive impairment and significant bias in severe impairment ( Bennett and Wang, 2024 ), and in nursing homes neither the self-rated nor the informant version detected depression adequately in residents with dementia (Li and colleagues, 2015). No published audit counts how often the form, the mode, or the cognitive statement is missing from GDS chart entries; the Quality ID 134 and Medicare rules described under compliance considerations are what turn those omissions into a measure or claim problem. The BastionGPT Clinical Advisory Board sees the same errors most often in GDS-15 documentation reviews: Form and denominator missing. "GDS 7" cannot be read: it is a positive on the GDS-15, within the original normal range on the GDS-30, and impossible on a four- or five-item form, and two five-item forms with the same threshold are different instruments. Write the form, its denominator (7 of 15), the author of any short derivative, and the language, every time. Proxy answers or an unnamed mode. A relative completing the patient's form changes the measurement source; the informant version exists for that purpose and did not perform adequately in dementia when tested. Record self-completed or read aloud with the patient's own answers, name any interpreter or translation, and label an informant-rated result as such rather than charting it as the patient's GDS. No cognitive statement, or cognition used as a blanket rule. A GDS-15 in a patient with advanced dementia and no cognitive statement cannot be defended, and neither can a note that refuses the scale to everyone with a diagnosis label: the evidence supports use through mild and moderate impairment when the patient can follow yes-or-no questions. Chart the screen and result, delirium status, and observed capacity, and switch to the Cornell Scale when self-report is unreliable. Yes answers counted, or proration hidden. The GDS keys items in both directions, so a count of yes answers is not the score. Apply the key, state who computed the total, and when items are missing chart the prorated figure as prorated with the number of items missed; the author site treats more than five missing items as invalidating the scale. A severity band or the score charted as a diagnosis. "GDS-15 9, moderate depression" reads an informal band as a DSM-5 diagnosis. The bands have no primary-source authority, the threshold of 5 or more overestimates prevalence by 17.6 percentage points against structured interview, and the author site asks only for a careful interview above 5. Name the threshold and its source, write "positive screen; depression not established," and put the diagnosis, if any, on its own line after the interview. Risk inferred from the score, or no plan after a positive. The GDS has no suicide item, so a low total charted as no safety concern and a high total charted as risk are both unsupported; ask directly and document separately. A positive screen with no follow-up plan, or with a repeat screen or a suicide risk assessment logged as the plan, fails Quality ID 134; document the referral or intervention on the encounter date or within two calendar days and name the tool in the record. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on behavioral health progress notes and screening documentation. Give it the facts (form and language, mode, the cognitive screen and result, the total and denominator, any missing items, the threshold your practice applies, the prior score, the interview findings, the suicide-risk question and answer, and the plan) and it drafts the documentation block: the screen result with the threshold named, the cognitive statement, a screening-level interpretation, the separate risk line, the dated follow-up plan, and the claim, ready for your review. Cross-check a finished note for the gaps reviewers flag: a form or denominator missing, proxy answers charted as the patient's score, no cognitive statement, a yes-count or hidden proration, a band charted as a diagnosis, a risk conclusion drawn from the total, or a positive screen with no dated plan. Turn the screen into the next document: the referral to behavioral health or collaborative care, the outcome-measure entry for serial results, or the plan language a quality abstractor expects, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the GDS-15 scored, and what score is a positive screen? Each of the 15 items is answered yes or no about the past week and scores one point when answered in the depressive direction; some items count when answered yes and others when answered no, so the total (0 to 15) is not a count of yes answers and the key has to be applied. There is no single mandated cut point. The Stanford author site says patients with totals higher than 5 should be interviewed carefully, most validation studies evaluate a threshold of 5 or more (pooled sensitivity 0.89 and specificity 0.77 in a 2016 meta-analysis), and the author site's scoring page, echoed in secondary guidance, reads totals above 10 as almost always depression. The informal bands (normal, mild, moderate, severe) that circulate for the GDS-15 come from toolkits rather than a validation paper and do not map onto DSM-5 severity. Chart the form, the total over 15, the threshold you applied with its source, and the result as a screen; if items were missed, use the author site's proration rule (scale the completed items to 15 and round up) and say so. What is the difference between the GDS-15, GDS-30, GDS-5, and GDS-4? They are separate forms with different item sets, ranges, and threshold conventions, not one scale at different lengths. The GDS-30 is the 1982 original, scored 0 to 30, with the original validation treating totals up to about 10 as the normal range. The GDS-15 (Sheikh and Yesavage, 1986) is the form in general use, scored 0 to 15, with the interview trigger above 5 described above. The GDS-5 of Hoyl and colleagues (1999) selected the five GDS-15 items most associated with a clinical diagnosis in frail outpatients and is conventionally read as positive at 2 or more, with the development study's strong accuracy figures coming from a small veteran outpatient sample. The four-item set of D'Ath and colleagues (1994) has no stable threshold, because moving the boundary by one point trades most of its sensitivity for specificity. A 2021 systematic review counted five four-item and seven five-item versions in the literature, so a short form is charted with its author as well as its length. The practical test is the bare score: a 7 is positive on 15 items, normal on 30, and impossible on 4 or 5, which is why the form is the first thing in the entry. Can I use the GDS-15 in a patient with mild cognitive impairment or dementia? Through mild and moderate impairment, when the patient can understand and answer yes-or-no questions, and with the cognitive status written into the entry; not in severe impairment. In 24,674 US participants, Bennett and Wang (2024) found negligible bias in GDS-15 scores at questionable, mild, and moderate impairment (CDR 0.5 to 2) and significant bias at severe impairment (CDR 3), and a 2026 analysis of 14,077 people found that 13 of the 15 items behaved differently by cognitive status, which is why the chart records what was observed instead of declaring the scale valid. Completion collapses as cognition falls: among people aged 85 and over, 1.5 percent with an MMSE of 0 to 4 and 41.7 percent with 5 to 9 could complete it, against more than 90 percent at 20 and above. The CCSMH guideline (2021) recommends the GDS or PHQ-9 for older adults without significant cognitive impairment and an observer-rated instrument such as the Cornell Scale for Depression in Dementia instead of the GDS in moderate to severe impairment, and British Columbia's guideline places the short GDS with minor cognitive loss or early dementia and the Cornell with advanced impairment. No authority sets a MoCA or MMSE score at which validity stops, so document the screen and result, delirium status, the diagnosis and stage, and whether the patient followed the format, and when self-report is unreliable say so and use the Cornell. Can a family member fill out the GDS for the patient? Not as the patient's GDS. Reading the items aloud and recording the patient's own answers is an accepted administration mode (the author site even describes a point-board for aphasia), and that is charted as read aloud. A relative answering in the patient's place changes the measurement source, which is why Brown and Schinka (2005) built a separate 15-item informant version, the GDSI-15, hosted on the Stanford site. Its limits matter: in nursing homes the informant version performed worse than self-report in residents without dementia (area under the curve 0.67 against 0.75 to 0.79) and neither version detected depression adequately in residents with dementia (Li and colleagues, 2015), and proxy and self-ratings often disagree. So: use self-report while the patient can answer; if collateral report is the only option, use the informant version or an instrument designed around informant input such as the Cornell Scale, chart it as informant-rated with the informant's relationship, and keep the collateral history in the interview rather than inside the patient's score. GDS-15 or PHQ-9 for the Medicare annual wellness visit, and does the GDS count for Quality ID 134 and the MDS? Either satisfies the wellness visit. CMS asks for a standardized tool recognized by national professional medical organizations and names none, the annual screening benefit (G0444, once a year, in primary care with staff-assisted depression care supports) leaves the tool to the clinician, and the 2026 Quality ID 134 specification lists the Geriatric Depression Scale and the Cornell Scale among acceptable adult tools beside the PHQ-9, so a GDS-15 earns measure credit when the tool's name is in the record and a positive screen has a documented follow-up plan. Choose the GDS-15 when somatic illness would make sleep, appetite, and energy questions hard to read; choose the PHQ-9 when you want DSM-mapped symptom domains, a built-in self-harm item, and a validated severity scale for monitoring. The exceptions are regulatory: nursing homes complete the MDS 3.0 Section D mood interview (PHQ-2 to 9, or the staff-observation version) and home health agencies complete the OASIS-E2 mood section, so in those settings a GDS is an adjunct and does not replace the required items. On repetition, G0444 is limited to once per 12 months as a benefit rule, not a psychometric one; repeat the GDS-15 when it will monitor treatment or after a clinical change, on the same form and mode, following the outcome measure note conventions. The GDS has no suicide question. What do I document? A separate, direct assessment, every time the screen is positive and whenever the interview or history raises concern. Unlike the PHQ-9, the GDS carries no item about death or self-harm, so a positive total does not discharge the duty to ask and a low total does not rule anything out; the Hartford Institute's guidance is that any positive short-form screen should prompt an in-depth assessment and an evaluation for suicidality. Document the question asked in plain terms, the answer, the tool if one was used (a C-SSRS screener, for example), protective factors, and where the formulation and any safety plan are filed, using the structure on the suicide risk assessment page. One quality-measure detail: Quality ID 134 encourages a suicide risk assessment when appropriate but does not count it as the follow-up plan, so the plan still needs a referral or intervention for depression documented in its own right. The screen is positive but the patient says they are not depressed. What goes in the chart? Both facts, then the interview that reconciles them. Write the result as a screen ("GDS-15 7 of 15, positive at the stated threshold; patient does not identify herself as depressed"), record what you observed (affect, engagement, tearfulness, psychomotor change), and document the interview: duration, loss of interest, function, grief, medical contributors, medications, alcohol, and sleep, with a working impression that leaves the diagnosis open until criteria are assessed. Older adults often endorse the mood and outlook items while rejecting the word depression, and the GDS-15 threshold of 5 or more classified 34.2 percent of older adults as depressed against 14.8 percent by structured interview in a 2024 meta-analysis, so a positive screen with a denial is common and neither fact wins on its own. Then document the same-day plan the measure requires (a return visit for a fuller assessment, a referral, or an intervention), the separate suicide-risk question, and any labs; the follow-up conversation belongs in a narrative progress note or the psychiatric diagnostic evaluation if a diagnosis is made. What do the GDS-15 items ask, and can we put the scale in our EHR, in an app, or on our website? At the topic level the items ask about mood, satisfaction with life, interest and activity, outlook and hopelessness, energy and concentration as the person experiences them, and self-worth, over the past week; by design there are no items about sleep, appetite, or physical symptoms and none about suicide. This page does not print or paraphrase the items; the Stanford author site is the source of record and hosts the forms, the scoring guidance, the informant version, more than 40 translations it cannot vouch for, and free scoring apps. The site states that the original scale is in the public domain because it was partly the result of federal support, so printing it, building it into a health system's own record for administration, and citing the authors are all ordinary uses. Two cautions: that statement covers the original scale, and the short derivatives (Hoyl's five-item form, D'Ath's four-item set) were published in journals, with at least one licensing database listing rights conditions for the short forms, so a commercial app, an EHR vendor, or a public website should confirm the rights of the exact version it embeds; and a translation posted on the site is not a validated version, so chart the language and, where one exists, the validation study. Can BastionGPT draft my GDS-15 documentation? Yes. Give it the facts (form and language, mode, the cognitive screen and result, the total and denominator, any missing items, the threshold your practice applies, the prior score, the interview findings, the suicide-risk question and answer, and the plan) and it drafts the full entry: the screen result with the threshold named, the cognitive statement, a screening-level interpretation, the separate risk line, the dated follow-up plan, and the claim, ready for your review. It can also check a finished note for a form or denominator missing, proxy answers charted as the patient's score, no cognitive statement, a yes-count or hidden proration, a band charted as a diagnosis, a risk conclusion drawn from the total, or a positive screen with no dated plan. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Stanford University, Yesavage JA, Geriatric Depression Scale author site (public-domain statement; forms, informant version, translations and disclaimer, proration guidance, scoring guidance, free apps; accessed September 2026); Yesavage JA, Brink TL, Rose TL, and colleagues, Journal of Psychiatric Research, development and validation of a geriatric depression screening scale (volume 17, pages 37 to 49; dated 1982 on PubMed and 1983 on the author site); Sheikh JI, Yesavage JA, 1986, Geriatric Depression Scale (GDS): recent evidence and development of a shorter version, in Brink TL (ed.), Clinical Gerontology: A Guide to Assessment and Intervention, Haworth Press, pages 165 to 173, as cited on the author site. Hoyl MT, Alessi CA, Harker JO, and colleagues, 1999, Journal of the American Geriatrics Society, development and testing of a five-item version of the Geriatric Depression Scale ; D'Ath P, Katona P, Mullan E, Evans S, Katona C, 1994, Family Practice, the 15-item GDS in elderly primary care attenders and the development of short versions ; Brañez-Condorena A and colleagues, 2021, PLOS One, accuracy of the GDS-4 and GDS-5 for depression screening in older adults (five four-item and seven five-item versions); Brown LM, Schinka JA, 2005, International Journal of Geriatric Psychiatry, development and initial validation of a 15-item informant version (GDSI-15). Wancata J and colleagues, 2006, Acta Psychiatrica Scandinavica, the criterion validity of the Geriatric Depression Scale: a systematic review ; Mitchell AJ, Bird V, Rizzo M, Meader N, 2010, Journal of Affective Disorders, GDS-30 and GDS-15 in primary care , and American Journal of Geriatric Psychiatry, which version is most useful in medical settings and nursing homes (results not significantly influenced by the presence of dementia); Pocklington C, Gilbody S, Manea L, McMillan D, 2016, International Journal of Geriatric Psychiatry, diagnostic accuracy of brief versions of the GDS (GDS-15 at cut point 5: sensitivity 0.89, specificity 0.77); Krishnamoorthy Y, Rajaa S, Rehman T, 2020, Archives of Gerontology and Geriatrics, diagnostic accuracy of various forms of the GDS (53 studies, 17,018 participants); Parsons M and colleagues for the DEPRESSD GDS Group, 2024, Scientific Reports, GDS-15 prevalence compared with the Structured Clinical Interview (34.2 percent against 14.8 percent; 8 or more closest). Bennett AV, Wang M, 2024, Quality of Life Research, mild to moderate cognitive impairment does not bias the Geriatric Depression Scale (24,674 participants by CDR); Renn BN and colleagues, 2026, Alzheimer's and Dementia, differential item functioning of the GDS short form in the NACC dataset (14,077 participants; 13 of 15 items); Conradsson M and colleagues, 2013, Aging and Mental Health, usefulness of the GDS-15 among very old people with and without cognitive impairment (completion by MMSE stratum); Li Z, Jeon YH, Low LF, and colleagues, 2015, International Psychogeriatrics, validity of the GDS and the collateral source GDS in nursing homes ; Alexopoulos GS, Abrams RC, Young RC, Shamoian CA, 1988, Biological Psychiatry, Cornell Scale for Depression in Dementia . CMS, annual wellness visit (depression risk review and cognitive assessment as required elements; standardized tools recognized by national professional medical organizations; G0438 and G0439); CMS, NCD 210.9, screening for depression in adults (annual; primary care with staff-assisted supports; no specific tool identified; 15 minutes); CMS, Medicare preventive services quick reference (G0444 bundled into the IPPE and the initial wellness visit; bare code numbers only); CMS Quality Payment Program, 2026 Quality ID 134 MIPS CQM specification (GDS and Cornell among adult examples; tool name required, score not required; follow-up plan within two calendar days; suicide risk assessment not a follow-up plan). CMS, MDS 3.0 RAI manual (v1.20.1, effective October 1, 2025); Saliba D, DiFilippo S, Edelen MO, and colleagues, 2012, Journal of the American Medical Directors Association, testing the PHQ-9 interview and observational versions for MDS 3.0 (weighted kappa 0.69; 86 percent completion); CMS, OASIS data sets (OASIS-E2 effective April 1, 2026); Kroenke K, Spitzer RL, Williams JB, 2001, the PHQ-9 (the DSM-mapped comparator with a self-harm item); Hartford Institute for Geriatric Nursing, Try This series issue 4 (revised 2019), the Geriatric Depression Scale (does not assess suicidality; a positive short-form score prompts assessment and evaluation for suicidality). Canadian Coalition for Seniors' Mental Health, 2021, Canadian Guidelines on Prevention, Assessment and Treatment of Depression Among Older Adults (GDS or PHQ-9 without significant cognitive impairment; observer-rated instrument such as the Cornell in moderate to severe impairment; both graded B); British Columbia Guidelines and Protocols Advisory Committee, cognitive impairment guideline, appendix D: depression screening tools ; Canadian Institute for Health Information, RAI-MDS 2.0 outcome scales (Depression Rating Scale). MBS Online, explanatory note AN.0.39, health assessments (items 701, 703, 705, 707; once every 12 months at 75 and over; psychological function including cognition and mood; bare item numbers only); Victorian Department of Health, October 2023, depression: standardised care process (Cornell Scale on admission and when symptoms emerge; PAS cognitive screen; not diagnostic); Australian Institute of Health and Welfare, mental health in aged care: symptoms of depression in residential care (modified Cornell Scale under the ACFI; not collected under AN-ACC). ## Related templates PHQ-9 Score Documentation Mini-Cog Score Documentation MoCA Score Documentation Suicide Risk Assessment MMSE Score Documentation BDI-II Score Documentation Outcome Measure Review Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/gaf-score All note templates # GAF Score Documentation: Rating Conventions, Legacy Use & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The GAF (Global Assessment of Functioning) is a clinician-rated 1 to 100 score of psychiatric symptom severity and functioning, published as Axis V of DSM-IV-TR and dropped from DSM-5 in 2013. Psychiatrists, psychologists, and community mental health clinicians still enter one when a payer form, legacy EHR field, disability file, or court asks for it. This page covers how to document a defensible legacy GAF, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Psychiatrists, psychologists, psychiatric nurse practitioners, and licensed mental health clinicians who assign the rating from the interview and the record; no publisher qualification level, license, or training requirement applies, and the APA offers no current training or license for it Audience Payers and utilization reviewers with legacy forms, state Medicaid and community mental health systems, SSA and VA adjudicators reading historical records, courts and attorneys, disability and workers' compensation evaluators, later treating clinicians Typical length 2 to 6 chart lines (edition frame and reason, period rated, basis, supporting facts, companion measure) · rating takes 1 to 3 minutes after the interview Format family Clinician-rated single global rating (1 to 100 in ten conceptual bands, 0 for inadequate information; symptom severity and functioning rated together, with the lower of the two governing) When it's used Legacy Axis V fields in payer authorization and utilization review forms, state Medicaid and community mental health level-of-care records, disability, workers' compensation, and forensic reports, historical record review, and EHR templates never migrated to DSM-5 Standards context DSM-IV-TR Axis V (American Psychiatric Association, 2000), copyrighted; removed from DSM-5 in 2013 with the WHODAS 2.0 placed in Section III; no federal authority requires it, some state rules and payer forms still do; described here for documentation, no anchor text reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the GAF score? The Global Assessment of Functioning is a single clinician-rated number, 1 to 100, that summarizes a person's psychiatric symptom severity and psychological, social, and occupational functioning, with 0 reserved for inadequate information. It descends from the Global Assessment Scale of Endicott, Spitzer, Fleiss, and Cohen (1976), entered the DSM as Axis V in DSM-III-R (1987), and was carried through DSM-IV (1994) and DSM-IV-TR (2000), which is the last edition to contain it. The scale is divided into ten conceptual ten-point bands, each pairing a level of symptom severity with a parallel level of functioning, running from a symptom-free, high-functioning top of the scale to bottom bands where safety or basic self-care cannot be maintained; the exact wording of each band is the American Psychiatric Association's copyrighted text and is not reproduced here. Three rating instructions matter for documentation: the rating describes the current period unless another is specified, and clinicians have operationalized current as today, the past week, or the past month; when symptom severity and functioning fall in different bands, the lower of the two governs rather than an average; and impairment caused by physical or environmental limitations is excluded from the rating. DSM-5 (2013) removed the multiaxial system and the GAF, citing its lack of conceptual clarity (one number mixing symptoms, suicide risk, and disability) and questionable psychometrics in routine practice, and placed the WHO Disability Assessment Schedule 2.0 in Section III as an emerging measure rather than a mandated replacement; DSM-5-TR (2022) kept that arrangement. The load-bearing fact for documentation is that the GAF is retired from diagnosis but not from administration. No clinical body requires it in 2026, the VA stopped using it for disability examinations in 2014, and the Social Security Administration treats it as one piece of opinion evidence, yet a Rhode Island Medicaid rehabilitative-services guideline updated in April 2026 still asks for a GAF on Axis V, Texas level-of-care guidelines still carry a GAF intake marker beside the ANSA, a Washington regulation effective July 2025 still hard-codes a DSM-IV GAF screening rule for one counseling credential, Wisconsin still hosts a 2010 prior-authorization form with a current-GAF field, and payer utilization-review forms, legacy EHR templates, disability files, and court records keep asking for a number. That gap is what a defensible entry has to bridge: the note must say which edition's conventions produced the number, what period it covers, whether symptoms or functioning drove it, what facts support it, and what current functional measure sits beside it, usually the WHODAS 2.0 , which measures disability across six domains and does not convert to a GAF. The modified GAF (Hall, 1995), the split GAF-S and GAF-F (Karterud and colleagues, 1998), the Kennedy Axis V (2003), the VA's MIRECC GAF (Niv and colleagues, 2007), and the Children's Global Assessment Scale (Shaffer and colleagues, 1983) are separate instruments with their own provenance, and none is a DSM-IV-TR GAF. The diagnostic structure around the entry belongs to the psychiatric diagnostic evaluation ; the goals it is sometimes asked to justify belong to the treatment plan . ## Who uses GAF documentation and when Community mental health and public-sector clinicians meet the GAF most often: psychiatrists, psychiatric nurse practitioners, psychologists, and licensed clinical social workers completing state Medicaid or managed-care authorization forms that never lost their Axis V field, level-of-care records that still carry a GAF marker, and EHR intake templates with a mandatory numeric field. Utilization reviewers and prior-authorization staff read those entries, so the number and its explanation travel into the utilization review summary and the prior authorization request . Disability, workers' compensation, and forensic evaluators supply one when a referral form, a state schedule, or a tribunal asks, and then explain its limits in the medical source statement or the court report , where the legally useful content is the concrete functional limitation, not the number. VA and SSA adjudicators encounter GAFs in older records and weigh them as history. In every one of those settings the GAF loses to its neighbors as the functional measure: the WHODAS 2.0 for a current, domain-level disability score, the mental status exam for the observed symptom picture the number is supposed to summarize, and the system's own instrument (ANSA, CANS, LOCUS, HoNOS) where one exists. The GAF page owns one narrow job: how to write the retired number so that it cannot be misread. ## How to document a GAF score in the chart No authority publishes a 2026 GAF documentation standard, because the scale is no longer part of any current diagnostic manual; what exists is the DSM-IV-TR rating instructions, the SSA's 2013 caution that evaluators rarely say whether a score reflects symptoms or functioning, and the forms that still demand a number. What survives review is an entry that names the edition frame and the reason for the entry, states the period rated, discloses which component drove the number, gives the supporting facts in a sentence, enters one integer with no threshold claim, records a current functional measure beside it, and treats prior scores as dated opinions. Each element below carries the pitfall that most often undermines it. Edition frame and reason for the entry. Open with the convention and the trigger: "GAF rated per DSM-IV-TR conventions (APA, 2000); scale removed from DSM-5 in 2013; entered because the payer's continued-stay form requires an Axis V value." That one sentence tells the next reader that the number is a legacy rating supplied on request, not a current DSM-5-TR measure, and it names the authority that asked, which is the first thing a reviewer or attorney will want to know. If the field is a software validation rule rather than a policy requirement, say so. Pitfall: A bare "GAF: 55" in a 2026 note, read by the next clinician, the payer, or the court as a current standardized measure that the DSM still endorses. Period rated. State the window the number describes: at today's evaluation, the past week, the past month, or a defined date range. If the form or your practice also records the highest level sustained in the past year, label it separately; DSM-IV practice treated current and highest-past-year as different questions, and SSA's guidance notes that the higher sustained level is the better prognostic signal. A retrospective number for a period you did not examine is an inference, and the note should say so or omit it. Pitfall: Current and past-year values mixed in one field, or a present-tense number assigned to a period the rater never observed. Basis: symptoms, functioning, or both. Under DSM-IV-TR conventions the overall number reflects the worse of the two components, never their average. Write which side governed: "symptom severity is the limiting component; occupational and self-care functioning are comparatively preserved," or the reverse, or "symptoms and functioning fall in the same band." This is the disclosure the SSA found evaluators rarely make, and it is what stops a symptom-driven number from being read as global incapacity. If a split rating (GAF-S and GAF-F) or a modified version was used, name it; a split score is not a DSM-IV-TR GAF and should not be entered as one. Pitfall: Serious symptoms and an intact job averaged into the middle of the scale, or a number with no indication of which dimension produced it. Supporting facts, in one sentence. Give the observations a reviewer can weigh: work or school attendance, independent living and self-care, relationships and withdrawal, psychotic symptoms, cognitive organization, and suicidal ideation with a pointer to the documented risk assessment. Keep limitations from physical illness, pain, transportation, housing, or the labor market out of the rating and note them elsewhere; the GAF excludes them by instruction, and a low psychiatric number built from an orthopedic or housing problem is the error disability reviewers look for. Pitfall: A number with no facts under it, or a physical or environmental limitation silently pulling the psychiatric rating down. The number itself. Enter one integer, not a range, and describe the level in your own words rather than copying anchor text into the chart. Do not attach a level-of-care or medical-necessity conclusion to the number unless a named policy actually sets that threshold, and then cite the policy; unofficial guides that map GAF values to inpatient, intensive outpatient, or outpatient care tie to no authority. Remember the reliability record: routine-practice raters agree with trained raters only moderately, so the number is an estimate with error around it, and the note should read that way. Pitfall: "GAF 45 to 50," or "GAF 48, meets criteria for intensive outpatient" with no policy behind the threshold. Companion current measure. Record a present-day functional measure beside the GAF with its own date and result: the WHODAS 2.0 (36- or 12-item, self, proxy, or interviewer version, 30-day recall) for a DSM-oriented adult evaluation, or the instrument the system already runs (ANSA, CANS, LOCUS, HoNOS, the K10 in Australian services). Keep the two analytically separate: there is no validated conversion between a WHODAS total and a GAF value, and the note should say the GAF is supplied for the form while the companion measure carries the functional evidence. Pitfall: A WHODAS domain profile translated into a GAF number, or the two scales charted as one trend line. Historical scores and change. Carry prior GAFs forward as what they are: a dated opinion by a named rater about a stated period. Quote them with the date and author, do not re-date them as current, and do not present the difference between two raters' numbers as treatment response; the SSA's own guidance warns that differences between clinicians' GAFs cannot be read as reliable change, and the published inter-rater figures explain why. Describe change in the facts (attendance, symptoms, risk, self-care) and let the number follow. Pitfall: "Improved from 42 to 55 since intake" across two raters, or a 2011 GAF re-entered in a 2026 field without its date. ### Blank template (copy and adapt) GAF DOCUMENTATION BLOCK (legacy Axis V entry) Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [intake / continued-stay authorization / disability or forensic report / review] Edition frame: [rated per DSM-IV-TR conventions (APA, 2000); scale removed from DSM-5 (2013)] Reason for entry: [form / field / order that requires it; policy or software validation rule] Period rated: [at today's evaluation / past week / past month / dates] Highest sustained in past year (if requested, separate): [ ] GAF value: [one integer, 1 to 100; 0 only for inadequate information] Version if not standard: [split GAF-S / GAF-F; modified; MIRECC; CGAS] Basis: [symptom severity / functioning / both in the same band is the limiting component] Supporting facts (one sentence): [work or school; self-care and living situation; relationships; psychotic symptoms; cognition; suicidal ideation with risk assessment reference] Excluded from the rating: [physical, pain, transportation, housing, labor-market limits, noted at ...] Companion current measure: [WHODAS 2.0 version and date, or ANSA / CANS / LOCUS / HoNOS / K10; result; no conversion to GAF] Historical GAFs: [value, date, rater, period rated; kept as dated opinions] Limits statement: [not a current DSM-5-TR measure; not normed; weigh with the narrative and the companion measure] Plan: [next review; what the entry supports] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's anchor descriptions. ## Sample GAF documentation (fictional) Scenario: a community mental health center continued-stay authorization for an adult with recurrent major depressive disorder, on a state Medicaid form that still carries an Axis V field; the psychiatric nurse practitioner enters a symptom-driven legacy GAF, explains the basis, records a WHODAS 2.0 beside it, and carries the intake score forward as a dated opinion. All details are fictional. Patient: A.K., 41 · Setting: Community mental health center, continued-stay authorization · Clinician: D. Reyes, PMHNP-BC · Note date: 09/10/2026 Frame and reason for entry: GAF rated per DSM-IV-TR conventions (APA, 2000); the scale was removed from DSM-5 in 2013 and is entered here only because the state Medicaid continued-stay authorization form requires an Axis V value. The rating is not a current DSM-5-TR measure and is not normed; the functional evidence for this request is the WHODAS 2.0 and the narrative below. Rating: Current GAF, rated for functioning at today's evaluation on 09/10/2026: 45. No past-year value is requested by the form and none is estimated. Standard single rating; no split or modified version used. Basis and supporting facts: Symptom severity is the limiting component. A.K. reports persistent depressed mood, early waking, and recurrent passive thoughts of death without plan or intent (risk assessment documented separately today, low acute risk, safety plan reviewed), with marked withdrawal from friends and family over the past month. Functioning is comparatively preserved: she has kept her part-time retail job with two late arrivals this month, lives alone, manages meals and medication, and attends appointments. Under the lower-of-the-two convention the symptom picture, not occupational capacity, produced the number. Chronic knee pain that limits standing is noted in the medical history and is excluded from this rating. Companion current measure: WHODAS 2.0, 12-item, self-administered, 30-day recall, completed in the waiting room on 09/10/2026 and entered in the outcome-measure record: greatest difficulty reported in participation and life activities, moderate difficulty getting along with people, little difficulty with cognition, mobility, or self-care, consistent with the interview. No numerical conversion between the WHODAS and the GAF was made. Historical scores: Intake on 08/18/2026, rated by the intake clinician (R. Nakamura, LCSW) for the preceding week: GAF 42. The two ratings come from different raters and are recorded as dated opinions, not as measured change; the clinical change since intake is described in the facts above (sleep and attendance slightly better, withdrawal and passive death thoughts unchanged). Plan and record: Continued weekly psychotherapy and medication management requested for 90 days; treatment plan goals and measures are in the treatment plan dated 08/18/2026. Repeat the WHODAS 2.0 at the December review; the GAF will be re-entered only if the form still requires it, with the same frame. Entry stored in the authorization section of the record with the WHODAS in the outcome-measure record; A.K. was told the number is a legacy rating supplied for the payer's form. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No scale anchor text is reproduced. ↑ Back to the template and downloads ### Why this sample works The edition frame, the 2013 removal, and the reason for the entry are stated first, so no reader can mistake the number for a current DSM-5-TR measure or wonder who asked for it. The period rated is explicit, no past-year value is invented, and the version is named as the standard single rating, which is exactly what a reviewer needs to know what question the number answers. The basis is disclosed (symptom severity limiting, functioning preserved) with the facts under it, so a symptom-driven number cannot be read as global incapacity, and the physical limitation is kept out by instruction. A dated WHODAS 2.0 sits beside the GAF with an explicit statement that no conversion was made, so the current functional evidence carries the request instead of the retired number. The intake score is carried forward with its date, rater, and period and labeled a dated opinion rather than a baseline for change, which is how VA and SSA adjudicators treat historical GAFs. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the authority map is specific, and it points in two directions. Under LAW, the Department of Veterans Affairs adopted DSM-5 for its rating schedule by an interim final rule effective August 4, 2014 (79 FR 45093), stating that the APA had found the GAF of "limited usefulness in the assessment of the level of disability" and that examiners would stop reporting it without any change to the rating criteria, and the Court of Appeals for Veterans Claims held in Golden v. Shulkin, 29 Vet. App. 221 (February 23, 2018), that "the Board errs when it uses GAF scores to assign a psychiatric rating" where DSM-5 applies, while accepting that older scores remain minimally relevant as history. Under ADJUDICATIVE POLICY, the Social Security Administration's Administrative Message AM-13066 (effective July 22, 2013) treats a GAF as opinion evidence that needs supporting detail and a stated period, calls it a snapshot, warns that differences between clinicians' scores cannot be read as reliable change, notes that evaluators rarely say whether a score reflects symptoms or functioning, and states that "a GAF score is never dispositive of impairment severity"; for claims filed on or after March 27, 2017, medical opinions are weighed for supportability and consistency with no controlling weight, and adult mental impairments are rated on SSA's four functional areas, so no GAF value equals a listing or a residual functional capacity. At the state level the picture reverses. Rhode Island's Medicaid Rehabilitative Services Coverage Guidelines (page updated April 28, 2026) still require a functional narrative that supports "the GAF score entered on Axis V" for authorization (PAYER POLICY); Texas Resilience and Recovery adult utilization-management guidelines (effective April 2017, still posted in September 2026) carry a GAF at or below 50 at intake as a marker of significant functional impairment inside level-of-care criteria that otherwise run on the ANSA (PAYER POLICY, legacy marker); Washington's WAC 246-810-021 (effective July 1, 2025) requires the state's limited-scope adviser credential to screen functional impairment with the DSM-IV GAF and to refer any client scoring 60 or less to a licensed practitioner (LAW, for that credential only); and Wisconsin ForwardHealth still hosts prior-authorization form F-00212 (2010) with a current-GAF field while the form itself says its use is voluntary and optional (CONVENTION: the field exists, the requirement does not). Nationally, CMS's 2026 Behavioral Health Core Set contains no GAF measure, and the circulating charts that map GAF ranges to inpatient, intensive outpatient, or "medical necessity not met" tie to no named payer or regulation (CONVENTION, habit rather than policy). In civil, family, and workers' compensation litigation the number appears because the underlying records are old or the form is, and its weight turns on foundation, not on a cutoff (CONVENTION); California's pre-2013 permanent disability rating schedule, which converted a GAF to a whole-person impairment, is the kind of legacy schedule an evaluator should cite by name if it governs the case. Canada and Australia have moved on, and the reliability record explains why the edition line matters everywhere. No Canadian federal or provincial law or payer names the GAF; Ontario's Workplace Safety and Insurance Board renamed its psychology program the mental health program of care on April 27, 2026 and asks for history, testing, functional status, a current DSM diagnosis, treatment planning, and a separate return-to-work form describing abilities and restrictions, with no GAF field (PAYER POLICY). In Australia the national outcomes and casemix collection is built on the HoNOS family with consumer measures such as the K10+ and the LSP-16, and the CGAS for children; the adult GAF is not part of that set (CONVENTION), NDIS access evidence is organized around functional capacity in everyday activities rather than an Axis V number (LAW), and no MBS item names a GAF (PAYER POLICY). On the evidence: Vatnaland and colleagues (2007) found routine clinical GAF scores agreed with research ratings only moderately (intraclass correlations of 0.39 to 0.59) while two trained researchers agreed at 0.81 and 0.85; Soderberg and colleagues (2005) reached an intraclass correlation of .81 when 81 staff rated first-time outpatient vignettes; Abbo and colleagues (2013) raised agreement on current ratings from 0.59 to 0.83 with one hour of training; Hall's modified GAF (1995) reported admission agreement of 0.81 against 0.62 for the original scale. Moos, McCoy, and Moos (2000) found in 1,688 patients that diagnoses and symptoms predicted GAF ratings more strongly than social and occupational functioning and that the ratings were only minimally associated with one-year outcomes; Pedersen and Karterud (2012) found substantial symptom-versus-function disagreement in about 10 percent of 2,695 patients, with functioning usually the worse of the two; and Suzuki and colleagues (2015) found a modest cross-sectional relationship with the PANSS across 40 schizophrenia trials (r of about -0.40, roughly 8,000 participants). Those numbers are the reason to write "per DSM-IV-TR conventions" with the date, disclose the basis, and treat any single value as an estimate with error around it; DSM-IV-TR (2000) is the only edition whose conventions a rater can cite, DSM-5 (2013) and DSM-5-TR (2022) contain no GAF, and the split, modified, MIRECC, Kennedy, and children's versions are separate instruments that must be named if used. The GAF is copyrighted by the American Psychiatric Association as part of DSM-IV-TR (2000), and it is not made available under any general license: the scale was dropped from DSM-5 in 2013, the APA declined a documented request to reproduce it in March 2013 (recorded in the CamCOPS task documentation, which for that reason collects the number without the scale), and as of September 2026 no APA grant covering forms, EHR builds, or public web tools has been published, so the anchor descriptions should be treated as unlicensed text and paraphrased in original words wherever they are needed; the concept of a 1 to 100 rating is not protected, the APA's wording of the bands is. The WHODAS 2.0 is copyrighted by the World Health Organization, which licenses its inclusion in electronic records and any reproduction separately. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common GAF documentation errors reviewers flag The numbers behind these errors are specific. Routine clinical GAF scores agreed with trained research ratings at intraclass correlations of only 0.39 to 0.59 , against 0.81 to 0.85 between two trained raters (Vatnaland and colleagues, 2007); one hour of training moved agreement on current ratings from 0.59 to 0.83 (Abbo and colleagues, 2013); in 1,688 patients, diagnoses and symptoms predicted the rating more strongly than social and occupational functioning, and the rating barely predicted one-year outcomes (Moos, McCoy, and Moos, 2000); symptom and function ratings diverged substantially in about 10 percent of 2,695 patients, usually with functioning worse (Pedersen and Karterud, 2012); and the Social Security Administration's 2013 guidance recorded that evaluators rarely note whether a score reflects symptoms, functioning, or both. No published audit counts how often a 2026 chart entry omits the edition frame or the period; the VA, SSA, and state policies described under compliance considerations are what turn those omissions into a weight problem. The BastionGPT Clinical Advisory Board sees the same errors most often in GAF documentation reviews: A bare number in a current note. "GAF: 55" with no edition frame, no reason for the entry, and no period. The next reader takes it for a current DSM measure; the payer or court cannot tell whether it answers today, the past month, or the best of the past year. Write the DSM-IV-TR frame, the 2013 removal, the trigger, and the window every time. Symptoms and functioning averaged, or the driver hidden. A patient with active suicidal ideation and an intact job charted in the middle of the scale, or a low number with nothing to show which dimension produced it. The convention is the lower of the two, and the disclosure is the sentence SSA found missing: name the limiting component and the preserved one. Period mixed, or a number for a period never examined. Current and highest-past-year values in one field, a present-tense number assigned to a 2019 record the evaluator did not observe, or an old score re-dated as current. State the window, label any past-year value separately, and carry prior scores forward with their original date and rater. Physical or environmental limits pulled into the rating. Inability to work from orthopedic disease, no transportation, or unstable housing turned into a low psychiatric number. The DSM-IV-TR instruction excludes them, and disability reviewers look for exactly this conflation. Note those limits where they belong and keep the rating to psychiatric symptoms and functioning. Change or level of care inferred from the number. "Improved from 42 to 55" across two raters read as treatment response, "GAF 48, meets intensive outpatient criteria," or "below 50, disabled." Inter-rater agreement in routine practice does not support reading differences as change, no federal rule sets a GAF threshold, and unofficial level-of-care charts cite no authority. Describe change in the facts and cite a named policy or none. Conversion or substitution across instruments. A WHODAS 2.0 total translated into a GAF, a split GAF-S and GAF-F or a MIRECC rating entered in a field that asks for a DSM-IV-TR GAF, or a CGAS presented as an adult score. There is no validated crosswalk; record the companion measure with its own date and result, name any variant, and state that no conversion was made. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric diagnostic evaluations and treatment plans. Give it the facts (the form or field that requires the number, the period rated, the symptom picture, the functional facts, the risk assessment reference, the companion measure and its result, and any prior scores with their dates and raters) and it drafts the documentation block: the DSM-IV-TR frame and reason, the period, the basis with the supporting sentence, the value, the companion measure, the historical scores as dated opinions, and the limits statement, ready for your review. Cross-check a finished note for the gaps reviewers flag: a bare number with no frame or period, an undisclosed driver, a physical limitation inside the rating, a threshold or level-of-care claim with no policy behind it, or a WHODAS converted into a GAF. Turn the entry into the next document: the functional narrative for a continued-stay authorization, the historical-scores paragraph of a disability or forensic report, or the plain-language explanation for a patient who asks what the number means, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does a GAF score mean, and what do the numbers represent? A GAF is a clinician's single judgment, on a 1 to 100 scale, of how severe a person's psychiatric symptoms are and how well they function psychologically, socially, and occupationally; 0 means there was not enough information to rate. Under DSM-IV-TR the scale is divided into ten-point bands, each pairing a description of symptom severity with a parallel description of functioning, stepping from a symptom-free, high-functioning top of the scale down to bands where safety or basic self-care cannot be maintained. When the two descriptions point to different bands the lower one governs, the rating covers the current period unless another is stated, and impairment from physical or environmental causes is left out. The band wording is the American Psychiatric Association's copyrighted text, so use the manual itself when you rate and describe the level in your own words in the chart. The number was never normed, and routine-practice raters agree only moderately, so read any single value as an estimate with error around it. Is the GAF still used in 2026, and do I have to give one? Only when something specific asks for it. DSM-5 (2013) and DSM-5-TR (2022) do not contain the scale, no clinical body requires it, the VA stopped using it for disability examinations in August 2014, and Social Security treats it as one non-dispositive opinion. What still asks: some state Medicaid and community mental health forms (Rhode Island's rehabilitative-services guideline, updated April 2026, still calls for a GAF on Axis V; Texas adult level-of-care guidelines still carry a GAF intake marker), one Washington regulation for a limited-scope counseling credential (effective July 2025), legacy prior-authorization forms such as Wisconsin's 2010 F-00212 (which says its own use is optional), payer utilization-review workflows, EHR templates never migrated to DSM-5, and disability or court files built on older records. The first documentation step is to find out which of those you are dealing with: a policy requirement gets a properly framed legacy entry; a software validation rule gets the same entry plus a note that the field is a system artifact; and where text is allowed instead of a number, a statement that the GAF is not a current DSM-5-TR measure, with a pointer to the functional assessment, is the better answer. How do I write a defensible GAF entry when a form requires one? Answer five questions in one short paragraph. Which conventions: "rated per DSM-IV-TR conventions (APA, 2000); scale removed from DSM-5 in 2013; entered because [form or field] requires it." What period: today's evaluation, the past week, the past month, or a date range, with any highest-past-year value labeled separately. What drove the number: symptom severity, functioning, or both in the same band, never an average. What supports it: one sentence of facts (work, self-care, relationships, psychosis, cognition, suicidal ideation with the risk assessment referenced), with physical and environmental limits excluded. What current measure sits beside it: a dated WHODAS 2.0 or the system's own instrument, with an explicit statement that no conversion was made. Close with a limits line: not a current DSM-5-TR measure, not normed, weigh with the narrative. The sample on this page shows the five parts in a continued-stay authorization; the blank template carries them as fields. My patient has severe symptoms but works full time. Which side wins? Under DSM-IV-TR conventions the symptoms do: when symptom severity and functioning fall in different bands, the overall rating reflects the lower of the two, so a patient with active suicidal ideation or psychotic symptoms and an intact job receives a symptom-driven number, and a patient with modest symptoms and a collapsed work and social life receives a functioning-driven one. Neither is averaged. The documentation fix is to say so: "symptom severity is the limiting component; occupational and self-care functioning are comparatively preserved," followed by the facts, so a reviewer does not read the low number as global incapacity or the high one as absence of illness. If your setting uses the split GAF (a separate symptom score and functioning score), record both under their own labels and do not enter either in a field that asks for a standard DSM-IV-TR GAF without saying which it is. Pedersen and Karterud found the two dimensions diverged substantially in about one patient in ten, and functioning was usually the worse of the two, which is why the disclosure is worth a sentence every time. Can I report a GAF from an old record, or compare two GAFs as change? Report it as history, never as a current opinion, and do not compute change from it. Quote the prior score with its date, its rater, and the period it described, keep it in a historical-scores line, and let your own current rating stand on its own frame. Two numbers from two raters are not a measurement of change: routine clinical raters agreed with trained raters at intraclass correlations of only 0.39 to 0.59 in Vatnaland and colleagues' 2007 study, and SSA's 2013 guidance tells adjudicators not to assume that differences between clinicians' scores reflect real change. Even two ratings by the same clinician describe two snapshots of possibly different periods. VA case law treats older GAFs as minimally relevant psychiatric history that cannot drive a DSM-5-era rating, and SSA treats them as opinion evidence that needs supporting detail. Describe change in the facts (attendance, self-care, symptoms, risk) and, if the form insists on a comparison, say the two values come from different raters or periods and are not a trend. GAF or WHODAS 2.0: are they interchangeable, and is there a conversion? No and no. The GAF is a clinician's single global judgment that blends symptom severity with functioning and lets the worse component dominate. The WHODAS 2.0 is the World Health Organization's disability measure, built on the International Classification of Functioning, that asks about difficulty over the past 30 days across six domains (cognition, mobility, self-care, getting along, life activities, participation) in 36- or 12-item versions, self, proxy, or interviewer administered; DSM-5 placed it in Section III as an emerging measure, not as a mandated replacement. The two answer different questions, no validated crosswalk between a WHODAS total and a GAF value exists, and a note that converts one into the other invents data. The defensible pattern is side by side: the WHODAS carries the current, domain-level functional evidence; the GAF, when a form demands it, is entered with its frame, period, basis, and limits; and the entry says no conversion was made. In systems that run their own instrument (ANSA, CANS, LOCUS, HoNOS) the same rule applies with that instrument in the WHODAS's place. Do the VA or Social Security still use GAF scores? Not to decide anything. The VA replaced DSM-IV references in its rating schedule with DSM-5 by an interim final rule effective August 4, 2014 (79 FR 45093), stating that examiners would no longer report the GAF and that the change affected only the form of their findings, not the rating criteria; in Golden v. Shulkin (2018) the Court of Appeals for Veterans Claims held that the Board errs when it uses GAF scores to assign a psychiatric rating in cases where DSM-5 applies, while accepting that older scores remain minimally relevant as history. Social Security's AM-13066 (2013) treats a GAF as opinion evidence that needs supporting detail and a stated period, calls it a snapshot, and says it is never dispositive of impairment severity; for claims filed on or after March 27, 2017, medical opinions are weighed for supportability and consistency with no controlling weight, and adult mental impairments are rated on SSA's four functional areas (understanding, remembering, and applying information; interacting with others; concentrating, persisting, and maintaining pace; adapting and managing oneself). So no GAF value equals a VA percentage, an SSA listing, or a residual functional capacity. For a disability report, document the concrete limitations in those terms in the medical source statement , and if a form still demands a GAF, enter it with its frame and limits rather than as the opinion itself. Can we put the GAF scale in our EHR, on an intake form, or in a free web calculator? Not the anchor descriptions, without written permission. The GAF is copyrighted by the American Psychiatric Association as part of DSM-IV-TR; the APA declined a documented request to reproduce the scale in March 2013 (the CamCOPS project records the refusal and collects the number without the scale for that reason), and no general APA grant covering forms, EHR builds, or web tools has been published as of September 2026. Being free, non-commercial, or attributed does not create a license, and the widespread claim that the GAF is public domain is not supported by anything the APA has published. What you may do is implement the concept: a numeric 1 to 100 field, your own independently written prompts about symptoms and functioning, and the documentation frame on this page; copyright protects the APA's wording of the bands, not the idea of a global rating. Two cautions: the WHODAS 2.0 is not a rights-free substitute (the WHO licenses its inclusion in electronic records and any reproduction), and a tool that stores a bare GAF number should also capture the period, basis, and frame, or it will generate exactly the undocumented numbers this page exists to prevent. Can BastionGPT draft my GAF documentation? Yes. Give it the facts (the form or field that requires the number, the period rated, the symptom picture, the functional facts, the risk assessment reference, the companion measure and its result, and any prior scores with their dates and raters) and it drafts the full entry: the DSM-IV-TR frame and reason, the period, the basis with its supporting sentence, the value, the companion measure with the no-conversion statement, the historical scores as dated opinions, and the limits line, ready for your review. It can also check a finished note for a bare number with no frame or period, an undisclosed driver, a physical limitation inside the rating, a threshold claim with no policy behind it, or a WHODAS converted into a GAF. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: American Psychiatric Association, 2000, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision, Axis V (the controlling text; not reproduced), and DSM-5-TR overview (no multiaxial system; WHODAS 2.0 among Section III measures); Endicott J, Spitzer RL, Fleiss JL, Cohen J, 1976, Archives of General Psychiatry, the Global Assessment Scale ; Gold LH, 2014, Journal of the American Academy of Psychiatry and the Law 42(2):173-181, DSM-5 and the assessment of functioning (why the GAF was dropped; WHODAS 2.0 in Section III); CamCOPS project documentation, GAF task page (APA copyright; reproduction permission declined March 21, 2013). US Department of Veterans Affairs, 2014, interim final rule, Schedule for Rating Disabilities: Mental Disorders (79 FR 45093, effective August 4, 2014) ; US Court of Appeals for Veterans Claims, 2018, Golden v. Shulkin, No. 16-1208 (29 Vet. App. 221) (decided February 23, 2018). Social Security Administration, Administrative Message AM-13066, Global Assessment of Functioning (GAF) Evidence in Disability Adjudication, effective July 22, 2013 (opinion evidence; supporting detail and period required; never dispositive; not an SSA-hosted public document, cite as historical guidance); SSA, Listing 12.00, mental disorders, adult (the four functional areas); SSA, revisions to rules regarding the evaluation of medical evidence (claims filed on or after March 27, 2017). State authorities: Washington State Legislature, WAC 246-810-021 (effective July 1, 2025; DSM-IV GAF screening and referral rule for one counseling credential); Rhode Island EOHHS, Rehabilitative Services Coverage Guidelines (page updated April 28, 2026; GAF on Axis V); Texas Health and Human Services, Texas Resilience and Recovery utilization management guidelines, adult (effective April 2017; GAF intake marker beside the ANSA); Wisconsin Department of Health Services, ForwardHealth form F-00212 (2010; current-GAF field; use voluntary and optional); CMS, 2026 Behavioral Health Core Set (no GAF measure); California Division of Workers' Compensation, 2005 permanent disability rating schedule (legacy GAF conversion). Canada and Australia: Ontario Workplace Safety and Insurance Board, mental health program of care overview (renamed April 27, 2026; assessment, functional status, current DSM diagnosis, return-to-work form; no GAF); Australian Mental Health Outcomes and Classification Network, NOCC measures (HoNOS family, K10+, LSP-16, CGAS); AIHW, consumer outcomes in mental health services ; NDIS, access and functional capacity evidence . Reliability and validity: Vatnaland T and colleagues, 2007, Acta Psychiatrica Scandinavica 115(4):326-330, are GAF scores reliable in routine clinical use? ; Soderberg P, Tungstrom S, Armelius BA, 2005, Psychiatric Services 56(4):434-438, reliability of GAF ratings by clinical staff , and Tungstrom S and colleagues, 2005, the GAF and other DSM axes in routine work ; Abbo C and colleagues, 2013, African Health Sciences, effect of brief training on GAF reliability ; Store-Valen J and colleagues, 2015, Psychological Assessment, web-based feedback training and GAF reliability ; Startup M, Jackson MC, Bendix S, 2002, British Journal of Clinical Psychology, concurrent validity of the GAF ; Aas IH, 2010, Annals of General Psychiatry 9:20, GAF properties and frontier of current knowledge , and 2011, 10:2, guidelines for rating the GAF . Symptom and function conflation: Moos RH, McCoy L, Moos BS, 2000, Journal of Clinical Psychology, GAF ratings: determinants and role as predictors of one-year outcomes (1,688 patients); Pedersen G, Karterud S, 2012, Comprehensive Psychiatry 53(3):292-298, the symptom and function dimensions of the GAF (2,695 patients); Pedersen G, Hagtvet KA, Karterud S, 2007, Comprehensive Psychiatry, generalizability studies of the GAF split version ; Hall RC, 1995, Psychosomatics 36(3):267-275, a modified GAF scale ; Suzuki T and colleagues, 2015, Psychiatry Research, GAF and other rating scales in schizophrenia trials . Variants and the replacement: Shaffer D and colleagues, 1983, Archives of General Psychiatry, a Children's Global Assessment Scale ; VA VISN 22 MIRECC, MIRECC version of the GAF (Niv and colleagues, 2007; three subscales); American Psychiatric Association Publishing, Mastering the Kennedy Axis V (2003); World Health Organization, WHODAS 2.0 (36- and 12-item versions; six domains; licensing required for electronic records and reproduction); Sjonnesen K and colleagues, 2016, Canadian Journal of Psychiatry, the 12-item WHODAS 2.0 in Canadians with mental disorders (23,757 respondents). ## Related templates WHODAS 2.0 Score Documentation Psychiatric Diagnostic Evaluation Treatment Plan Utilization Review Summary Disability Evaluation & Medical Source Statement Mental Status Exam (MSE) Note Court Report & Treatment Progress Update Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/celf-5-report All note templates # CELF-5 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The CELF-5 (Clinical Evaluation of Language Fundamentals, Fifth Edition) is a norm-referenced language battery for ages 5:0 through 21:11 that yields a Core Language Score and index scores from age-selected tests. Speech-language pathologists use it in school eligibility and clinical language evaluations. The write-up must name the tests behind each composite and the eligibility rule applied. This page covers how to write up CELF-5 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Speech-language pathologists (SLPs) in schools, clinics, and hospitals; publisher qualification level B Audience IEP and eligibility teams, school psychologists and psychoeducational evaluators, referring physicians and pediatricians, payers, NDIS and provincial program reviewers, families Typical length 500 to 900 words for the results section · administration 30 to 45 minutes for the Core Language Score, 90 to 120 minutes for a full battery Format family Norm-referenced omnibus language battery (16 standalone tests, about 10 per age band; observational and pragmatic components) When it's used School speech or language impairment eligibility evaluations and reevaluations, clinical evaluations for developmental language disorder, reading and written-expression referrals needing an oral-language profile, bilingual evaluations with converging evidence, NDIS and school-support applications Standards context Published by NCS Pearson (2013; CELF-6 standardization research announced 2025, unreleased as of September 2026); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CELF-5? The CELF-5 (Clinical Evaluation of Language Fundamentals, Fifth Edition; Wiig, Semel, and Secord; NCS Pearson, fall 2013) is a norm-referenced battery of spoken and written language for ages 5:0 through 21:11. It is built as 16 standalone tests, about 10 of which apply at any age, each scored as a scaled score (mean 10, SD 3). Four age-selected tests form the Core Language Score, and other combinations form the Receptive Language, Expressive Language, and Language Content indexes across the range, with a Language Structure Index at ages 5 to 8 replaced by a Language Memory Index from age 9; every composite is a standard score (mean 100, SD 15) reported with a confidence interval, a percentile rank, a Growth Scale Value for tracking change, and an age equivalent. Around the norm-referenced core sit the components that make a report defensible: the Observational Rating Scale (parent, teacher, and student observations of language at home and in class), the Pragmatics Profile (which in this edition yields a norm-referenced scaled score), the criterion-based Pragmatics Activities Checklist, and optional Reading Comprehension and Structured Writing tests from age 9. The family also includes the CELF Preschool-3 (2020; ages 3:0 to 6:11), CELF-5 Metalinguistics (2014; ages 9:0 to 21:11; a revision of the Test of Language Competence-Expanded), the CELF-5 Screening Test, a Spain-normed Spanish edition (2018; ages 5:0 to 15:11), an Australian and New Zealand edition (2017) with local census-based norms, and a Canadian French version on Q-interactive. As of September 2026 the CELF-5 remains the marketed edition; Pearson announced CELF-6 standardization field research in 2025 for ages 5 to 21, with no release date published. The load-bearing fact for the write-up is that three different decision systems read the same number. Pearson's own severity guidance describes composite scores within 1 SD of the mean (86 to 114) as average, reports its best classification balance at a standard score of 80 (1.33 SD below the mean, with sensitivity and specificity of .97 in the publisher's study), and in the same document acknowledges that agencies qualify students at 1, 1.5, or 2 SD below the mean and that a student with real deficits may not meet a program's criterion. Federal law sets no CELF-5 cutoff at all and prohibits any single measure from being the sole criterion for a disability determination (34 CFR 300.304(b)(2)); states then write their own rules, which differ in threshold, in the number of measures required, and in whether a language sample can stand in for a second test. A second fact hides inside the score labels: the Core Language Score and the indexes are composed of different tests at ages 5 to 8, 9 to 12, and 13 to 21, so an index name without its constituent tests tells the next reader nothing about the evidence. Speech sound production belongs to the GFTA-3 page , children below the CELF-5 floor to the PLS-5 page , and the whole-report architecture to the psychoeducational report page . ## Who uses the CELF-5 and when School-based speech-language pathologists are the core users, administering the CELF-5 inside IDEA evaluations for the speech or language impairment category (34 CFR 300.8(c)(11)) and at reevaluation, where the write-up feeds the eligibility team, the IEP input , and any Section 504 statement . Clinic and hospital SLPs use it to characterize developmental language disorder and to support medical necessity for treatment, billing the untimed evaluation codes 92523 or 92522 depending on whether speech sound production was evaluated alongside language. Psychoeducational and dyslexia teams use it for the oral-language side of a reading or written-expression referral, alongside the achievement measures covered on the WIAT-4 page , and autism teams use its pragmatic components as one input to a social-communication question that the autism evaluation report owns. Outside the United States, Canadian school boards use the US-normed English edition or the Canadian French version within provincial identification processes, and Australian clinicians use the Australian and New Zealand edition to supply age-standardized evidence for NDIS and school-support applications that are judged on functional impact. The readers are therefore a team applying a rule, a psychologist integrating the profile, a payer or funder checking necessity and function, and a family entitled to a plain-language account; the write-up has to survive all four. ## How to structure a CELF-5 results section No regulation prescribes a CELF-5 report format. What the federal evaluation rules, the state criteria, and the instrument's own design dictate is the content: the edition and norm set, the tests behind every composite, uncertainty around every number compared with a threshold, the observational and pragmatic evidence in its correct category, and an eligibility statement written against the named rule. Each section below carries the pitfall that most often undermines it. Identification, norm set, and tests administered. Open with the full test name and edition, the norm set (US, Australian and New Zealand, or Spain), the platform (paper, Q-global scoring, or Q-interactive administration), the language of testing, the evaluation dates and the student's age in years and months, and the age band that governs which tests compose each composite. List every test administered and, briefly, why: the CELF-5 is a flexible battery of standalone tests, so the report explains the selection rather than implying the whole kit was given. Pitfall: "CELF-5 administered" with no norm set and no test list. A Canadian reader assumes Canadian norms that do not exist for the English edition; a psychologist cannot tell what produced the index. Behavioral observations and validity conditions. Document attention, effort, and fatigue; hearing and vision status; language history, English exposure, and dialect; interpreter use; any extension testing or dynamic assessment procedures; and telepractice conditions if applicable. State plainly whether the normative comparison is appropriate for this student. When it is not, say that the standardized scores are reported descriptively or not at all, and shift the evidence to language samples and converging measures. Pitfall: Second-language or dialect patterns scored as disorder. Federal rules require nondiscriminatory assessment in the language most likely to yield accurate information, and state rules (Georgia's, for one) exclude normal second-language acquisition from the impairment definition. Composite results with confidence intervals. Report the Core Language Score and each relevant index as a standard score (mean 100, SD 15) with its 95 percent confidence interval, its percentile rank, the publisher's descriptor in prose (within 1 SD of the mean, 86 to 114, is average), and the tests that composed it at this age band. Report the indexes that answer the referral question; an index built from tests that were not administered does not exist for this student and is not estimated. Pitfall: A bare 78 compared with a 1.5 SD threshold of 77.5. The interval is the honest sentence, and the publisher itself ties confidence intervals to eligibility and placement decisions. Test-level pattern. Describe scaled-score patterns (mean 10, SD 3) that converge, in prose, rather than listing every test as high or low. When an index discrepancy will be interpreted, use the manual's significance and base-rate procedures and say so; the same test contributes to more than one index, so a visible numerical gap is not automatically a clinical finding. Age equivalents, if an agency requires them, appear here with an explicit limitation and never as the headline. Pitfall: A low Language Memory Index written up as a memory disorder, or ordinary scatter narrated as a diagnostic profile. The indexes are descriptive groupings of overlapping tasks, not separate disorders. Contextual and pragmatic evidence. Summarize the Observational Rating Scale by rater and by context (where communication breaks down and with what classroom consequence), report the Pragmatics Profile as the norm-referenced scaled score it is in this edition, and report the Pragmatics Activities Checklist, when given, as a criterion-based observation of authentic interaction. Add the language sample (many US rules expect at least 50 utterances) and any dynamic assessment response, because these are what carry the report when an average composite meets a struggling student. Pitfall: The Pragmatics Profile called criterion-referenced, or the three components treated as interchangeable measures of pragmatics. They are three kinds of evidence, and a rule that counts measures may treat them differently. Eligibility or diagnostic statement against the named rule. Quote the governing criterion (state regulation, board policy, funder guidance, or payer policy), compare each element of the evidence with it, and state whether the criteria are met, whether adverse educational effect or functional impact is documented, and that the team makes the determination. In a clinic, the same paragraph states the diagnosis and the functional limitations that establish medical necessity. Pitfall: "A Core Language Score below 85 qualifies the student." No federal rule, no state rule quoted, and not the publisher's own cut, which is 80 in its classification study. Recommendations and progress plan. Link each documented weakness to a service, accommodation, or classroom strategy, and name how progress will be measured: Growth Scale Values for change over repeated CELF-5 administrations, language-sample measures, and classroom data, with a reevaluation point. Close with the referrals the profile implies (audiology, reading evaluation, psychology) and a plain-language summary the family can use. Pitfall: Recommendations copied from a template with no thread back to the profile, or age-normed standard scores used to claim progress when the question was within-student growth. ### Blank template (copy and adapt) CELF-5 RESULTS SECTION SKELETON Student: [initials] Age: [y:m] Grade: [ ] Evaluation date(s): [ ] Evaluator: [name, credentials] Referral question: [ ] Instrument: Clinical Evaluation of Language Fundamentals, Fifth Edition Norm set: [US / Australian and NZ / Spain] Platform: [paper / Q-global scoring / Q-interactive] Language of testing: [ ] Age band: [5 to 8 / 9 to 12 / 13 to 21] Tests administered + rationale: [each test by name; why selected] Other measures: [language sample: context, utterance count, analysis; dynamic assessment; records; second standardized measure if required] Behavioral observations + validity: [attention, effort, hearing and vision, language history and dialect, interpreter, extension testing; normative comparison appropriate: yes / no, and why] Composite results (each with uncertainty): Core Language Score: SS [ ] (95% CI [ ] to [ ]), percentile [ ], descriptor [ ]; composed of [tests at this age band] [Index]: SS [ ], CI [ ], percentile [ ], descriptor [ ]; composed of [ ] Test-level pattern: [converging strengths and weaknesses in prose; discrepancy analysis per the manual; age equivalents only if required, with limitation stated] Contextual + pragmatic evidence: [ORS by rater and context; Pragmatics Profile scaled score; PAC criterion result if given; language sample findings; dynamic assessment response] Eligibility / diagnostic statement: [governing rule quoted; each element compared; adverse effect or functional impact; team determination] Recommendations + progress plan: [weakness to service or accommodation; progress measure (GSVs, sample measures, classroom data); reevaluation; referrals; family summary] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, stimuli, record forms, norms, or conversion tables. ## Sample CELF-5 write-up (fictional) Scenario: a second-grader referred for reading difficulty and trouble following classroom directions, evaluated by a school SLP, with composites reported with intervals and constituent tests, a language sample as the second measure, and the eligibility statement written against a named state-style criterion. All details are fictional. Patient: J.M., 7 years 4 months, grade 2 · Setting: Public school speech-language eligibility evaluation · Clinician: A. Okafor, MS, CCC-SLP · Note date: 09/16/2026 Measures and conditions: Clinical Evaluation of Language Fundamentals, Fifth Edition (CELF-5), US edition and norms, paper administration scored on Q-global, in English, across two sessions on 09/10/2026 and 09/14/2026. J.M. was 7 years 4 months at testing, so the 5 to 8 age band governs test composition. Tests administered: Sentence Comprehension, Linguistic Concepts, Word Structure, Word Classes, Following Directions, Formulated Sentences, and Recalling Sentences, selected to obtain the Core Language Score and the Receptive Language, Expressive Language, Language Content, and Language Structure indexes; the Pragmatics Profile was completed by the classroom teacher, and Observational Rating Scale forms were returned by the teacher and a parent on 09/11/2026. A 62-utterance language sample (conversation and a story retell) was recorded and transcribed on 09/14/2026 as the second measure. J.M. is a monolingual English speaker of the local community dialect; hearing was screened and passed at school on 09/08/2026, and vision is corrected with glasses that were worn. Attention was adequate with one movement break per session, and the normative comparison is judged appropriate. Composite results: The Core Language Score was 76 (95 percent confidence interval 70 to 82; 5th percentile), below the average range, composed at this age of Sentence Comprehension, Word Structure, Formulated Sentences, and Recalling Sentences. The Expressive Language Index was 74 (68 to 80; 4th percentile), from Word Structure, Formulated Sentences, and Recalling Sentences. The Receptive Language Index was 84 (77 to 91; 14th percentile), from Sentence Comprehension, Word Classes, and Following Directions, at the lower edge of average with an interval that spans it. The Language Content Index was 88 (81 to 95; 21st percentile), from Linguistic Concepts, Word Classes, and Following Directions, within the average range. The Language Structure Index, which at ages 5 to 8 draws on the same four tests as the Core Language Score, was 76 (70 to 82; 5th percentile). Descriptors follow the publisher's convention that composite scores within 1 SD of the mean are average. Test-level pattern: Scaled scores (mean 10, SD 3) converged on a structural-language weakness: Word Structure 5, Recalling Sentences 5, Formulated Sentences 6, and Following Directions 6 were all more than 1 SD below the mean, while Linguistic Concepts 9, Word Classes 8, and Sentence Comprehension 7 sat at or near the average range. The content-versus-structure contrast (Language Content Index 88 against Language Structure Index 76) was tested with the manual's index-comparison procedure and met its significance criterion; it is interpreted as weak morphosyntax and sentence-level formulation and recall with relatively preserved vocabulary knowledge, not as a discrete disorder of any single index. Age equivalents are not reported; standard scores, intervals, and percentiles carry the comparison. Contextual and pragmatic evidence: On the Observational Rating Scale, the teacher and the parent both marked frequent difficulty following multistep spoken directions, retelling events in order, and producing complete sentences in writing; the teacher tied these to missed instructions during whole-class lessons and incomplete written responses. The Pragmatics Profile, a norm-referenced scaled score in this edition, was 9, within the average range: conversational turn-taking, topic maintenance, and nonverbal communication are age-appropriate, so the concern is structural language, not social communication. The 62-utterance sample showed a mean length of utterance below expectations for age, inconsistent regular past-tense and plural marking, and few complex sentences; in a brief dynamic teaching trial, J.M. followed two-step directions accurately when they were paired with a visual sequence and repeated once. Eligibility analysis: Our state criterion requires a score at least 1.5 standard deviations below the mean, or below the 7th percentile, on two or more standardized language tests in morphology, syntax, semantics, or pragmatics, or on one such test together with a representative language sample of at least 50 utterances. The CELF-5 is one standardized test; on it the Expressive Language Index (74; interval 68 to 80) sits entirely below the 1.5 SD line of 77.5, and the Core Language Score (76) sits below it with an interval whose upper bound crosses it, which is why this statement rests on the convergence of the expressive index, the language sample, and the classroom evidence rather than on the Core score alone. The 62-utterance sample documents morphological and syntactic use below age expectations, satisfying the second route. Adverse educational effect is documented by the teacher's observations and by the reading evaluation in progress with the school psychologist; these language findings contribute to that evaluation but do not by themselves establish a specific learning disability. No single measure has been used as the sole criterion, and eligibility under the speech or language impairment category is the team's determination. Recommendations and progress plan: Recommended: direct language intervention targeting sentence formulation, grammatical morphology, and recall of spoken directions, coordinated with the reading evaluation so that oral-language and decoding goals are written together; classroom accommodations of chunked and visually sequenced directions, a repeat-and-check routine before independent work, and sentence starters for written responses. Progress will be measured with CELF-5 Growth Scale Values at reevaluation, quarterly language-sample measures on the same tasks, and teacher data on direction-following, with reevaluation planned within the state timeline. Reviewed with J.M.'s parents on 09/16/2026 in plain language: J.M. understands words and ideas well, and what is hard right now is putting sentences together, using word endings, and holding on to long spoken directions, all of which are teachable. This sample is fictional and for educational purposes. It does not describe a real student or record; the scores, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The edition, norm set, platform, age band, and every test administered are stated before any score, so a psychologist, a receiving district, or a Canadian or Australian reader can tell exactly what produced each number. Every composite carries its confidence interval, percentile, and constituent tests, and the one interval that crosses the threshold is named rather than hidden behind the integer. The Pragmatics Profile is reported as the norm-referenced scaled score it is, the Observational Rating Scale as rater-by-context observation, and the language sample as the second measure, each in its own evidentiary category. The eligibility statement quotes the rule, walks through each element, keeps the reading question with the reading evaluation, and leaves the determination to the team, which is what 34 CFR 300.304 requires. Recommendations trace back to the documented pattern, and the progress plan names Growth Scale Values for within-student change instead of re-comparing age-normed scores. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the write-up sits inside three decision systems, and the report should say which one each sentence serves. Under IDEA, speech or language impairment is a disability category defined as a communication disorder that adversely affects educational performance (34 CFR 300.8(c)(11)), and the evaluation rules require a variety of tools and strategies including parent information, prohibit any single measure as the sole criterion, require nondiscriminatory instruments administered in the child's native language or the mode most likely to yield accurate information, and require assessment in all areas related to the suspected disability (34 CFR 300.304(b) and (c)); no federal CELF-5 cutoff exists (LAW). States supply the operational rule and they differ: California requires at least 1.5 SD below the mean or below the 7th percentile on two or more standardized tests in morphology, syntax, semantics, or pragmatics, or on one test plus a representative language sample of at least 50 utterances (5 CCR 3030(b)(11)(D)); New Jersey requires functional assessment outside the testing situation plus performance below 1.5 SD or the 10th percentile on at least two standardized language tests, one a comprehensive receptive and expressive measure, and its April 17, 2024 clarification states that index, subtest, and standard scores may all be considered (N.J.A.C. 6A:14-3.5(c)4 and 3.6(a)); Georgia requires at least two measures or procedures, at least one formal, plus documented adverse effect, and excludes normal second-language acquisition and dialect difference from the impairment definition (Rule 160-4-7-.05) (LAW, state by state). The publisher's guidance is a third voice: its best classification balance sits at a standard score of 80, with sensitivity and specificity of .97 in its own study, and it acknowledges that programs use 1, 1.5, or 2 SD (publisher guidance), while the familiar "below 85" threshold is a local CONVENTION wherever it is not written into a rule. In clinics, coverage turns on documented clinical history, objective findings, functional limitation, and medical necessity under Medicare's therapy documentation policy and plan-specific commercial rules, billed through the untimed evaluation codes 92523 or 92522; a standard score supports the case and never makes it (PAYER POLICY). Canada and Australia change both the norm set and the frame. Pearson Canada sells the English CELF-5 with US norms (based on the 2010 US Census) and a Canadian French version on Q-interactive, so a Canadian report names which one was used and, for the English edition, discloses the normative population; identification of a communication exceptionality is a provincial and board process (Ontario's identification, placement, and review committee route, British Columbia's designation and IEP orders) with no national numeric criterion, so any local threshold is labeled a board CONVENTION unless a provincial policy source is cited (LAW and PROGRAM POLICY). Australia has its own edition, the CELF-5 Australian and New Zealand (2017), normed on the 2011 Australian and 2013 New Zealand censuses, and NDIS evidence is judged on functional impact across communication, social interaction, and learning rather than on a cutoff (PAYER POLICY); state school-support criteria could not be verified from primary sources for this page and should be taken from the funder's current document. Edition and norms: the CELF-5 (2013) is the current marketed edition as of September 2026, Pearson has publicly recruited for CELF-6 standardization since 2025, and no release date is published, so a report names the edition and platform and dates itself. The Spanish product on the US store is the Spain-normed edition (2018; ages 5:0 to 15:11; Spanish, not Catalan, as the home language), not a US Hispanic norm set, and English-edition standard scores are not diagnostic for a student the norming sample does not represent. The publisher classifies the CELF-5 at qualification level B; record forms and stimulus materials are protected and are not scanned, photographed, or reproduced in a report, transferring results into an electronic record and placing scoring on another platform are permission and license matters with Pearson, and a free public web scorer has no authorization. A report may always contain the student's derived scores and your interpretation in your own words. CELF-5 is a trademark of NCS Pearson, Inc. (the CELF word mark is US registration 1776508, held by NCS Pearson, Inc., of Bloomington, Minnesota, and renewed in 2023). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CELF-5 write-up errors reviewers flag The numbers behind these errors are specific. Pearson's own severity guidance puts its best classification balance at a standard score of 80 (sensitivity and specificity of .97 in the publisher's study) while acknowledging that programs qualify students at 1, 1.5, or 2 SD below the mean; California requires 1.5 SD or the 7th percentile on two standardized tests, or one plus a 50-utterance language sample; New Jersey requires 1.5 SD or the 10th percentile on two tests, one of them a comprehensive receptive and expressive measure; Georgia requires at least two measures or procedures, at least one formal; the battery itself is 16 tests of which about 10 apply at any age, with the Core Language Score built from four different tests in each of three age bands; and 34 CFR 300.304(b)(2) prohibits any single measure from deciding. The BastionGPT Clinical Advisory Board sees the same errors most often in CELF-5 documentation reviews: The 85 rule written as the test's rule. "A Core Language Score below 85 indicates a language disorder and qualifies the student." No federal rule says so, the publisher's own studied cut is 80, and the state may require 1.5 or 2 SD on two measures. Write the number, its interval, and the named rule, and let the team apply it. Composites without their constituent tests. "Receptive Language Index 84" with no tests named. The index is built from different tests at ages 5 to 8, 9 to 12, and 13 to 21, and the Language Structure Index gives way to a Language Memory Index at 9, so the label alone hides what evidence produced the number. Name the tests behind every composite at the student's age band. Point scores at a threshold with no interval. A 78 written against a 77.5 line as if the integer were exact. The publisher reports confidence intervals and ties them to classification and placement decisions; a report says where the interval falls relative to the criterion and, when it straddles the line, rests the conclusion on converging evidence. Index scatter narrated as a diagnosis. A low Language Memory Index becomes a "memory disorder," and a receptive-expressive gap becomes "the diagnosis." The same test feeds more than one index, gaps need the manual's significance and base-rate procedures before they mean anything, and the indexes are descriptive groupings, not separate conditions. The Pragmatics Profile mislabeled. Called criterion-referenced, or lumped with the Observational Rating Scale and the Pragmatics Activities Checklist as one "pragmatics measure." In this edition the Profile yields a norm-referenced scaled score, the Checklist is the criterion-based authentic observation, and the Rating Scale is rater-by-context observation; a rule that counts measures may treat each differently, so the report labels each correctly. Language age doing decisional work. "J. is functioning at the level of a five-year-old." Age equivalents show no rank among peers, move sharply with small raw-score changes, and are not comparable across tests, and the publisher advises against using them as the basis for diagnosis or placement. Report them only where a form requires them, with the limitation stated. English norms applied to an unrepresented student. Standard scores from the English edition reported as diagnostic for a bilingual or dialect-speaking student, or the Spain-normed Spanish edition treated as a US norm set. Federal rules require assessment in the language most likely to yield accurate information, some states exclude normal second-language acquisition from the definition outright, and the defensible report presents such tasks descriptively while resting the conclusion on the dominant language, a language sample, and converging evidence. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on speech-language evaluation reports. Give it the facts (age band and tests administered, scaled and standard scores with intervals and percentiles, Observational Rating Scale and Pragmatics Profile results, language-sample findings, the governing rule) and it drafts the results section: constituent tests named per composite, intervals compared with the threshold, the pragmatic components in their correct categories, and the eligibility statement written against the rule, ready for your review. Cross-check a finished report for the gaps reviewers flag: a composite without its tests, a bare score at a threshold, the Pragmatics Profile called criterion-referenced, an age equivalent doing decisional work, or an eligibility sentence that names no rule. Draft the companion paragraphs: the plain-language family summary, the bilingual-evaluation validity statement, or the progress paragraph that uses Growth Scale Values instead of re-comparing age-normed scores. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do CELF-5 scores mean? Each test yields a scaled score with a mean of 10 and SD of 3, and the composites (the Core Language Score and the Receptive Language, Expressive Language, Language Content, and Language Structure or Language Memory indexes) are standard scores with a mean of 100 and SD of 15, each reported with a confidence interval, a percentile rank, a Growth Scale Value, and an age equivalent. In the publisher's own prose, composites within 1 SD of the mean (86 to 114) are average and scores below that are below average to very low relative to age peers, which may or may not affect classroom participation; a scaled score of 7 sits 1 SD below the test mean. The interval matters more than the label whenever a score is compared with a rule, and the constituent tests matter more than the index name, because composition changes at ages 9 and 13. Growth Scale Values answer a different question (how much the student's own ability changed) and are not peer rankings. What CELF-5 score qualifies a student for speech-language services? Whatever the governing rule says, and there are three voices to keep apart. Federal law sets no cutoff and prohibits any single measure from being the sole criterion (34 CFR 300.304(b)(2)) (LAW). States set the operational rule: California uses 1.5 SD or the 7th percentile on two standardized tests, or one plus a 50-utterance language sample; New Jersey uses 1.5 SD or the 10th percentile on two tests, one comprehensive; Georgia requires two measures or procedures, one formal, plus adverse effect (LAW, state by state). The publisher reports its best classification balance at a standard score of 80 and acknowledges that programs use 1, 1.5, or 2 SD (publisher guidance), and the familiar 85 is a CONVENTION unless a rule adopts it. So the report quotes the rule, compares the interval with it, documents adverse educational effect, and leaves the determination to the team. Is the CELF-5 Pragmatics Profile norm-referenced or criterion-referenced? Norm-referenced. Pearson lists norm-referenced Pragmatics Profile scores among the fifth-edition changes, and the product page states that the Profile now produces scaled scores. The criterion-based pragmatic component is the Pragmatics Activities Checklist, an authentic interaction observation, and the Observational Rating Scale is a rater-by-context observation of listening, speaking, reading, and writing at home and school. Getting the labels right matters when a state rule counts measures or formal measures: a norm-referenced Profile score may count differently from a checklist result, and the report should not present the three as interchangeable measures of pragmatics. Should a CELF-5 report include age equivalents or a language age? Not as an interpretive result. The publisher's own guidance is that age equivalents show no rank among age peers, can shift substantially with small raw-score differences, are not comparable across tests, and do not mean the student functions overall like a child of that age, and it advises against using them as the primary basis for diagnosis or placement. Standard scores with confidence intervals and percentiles carry the peer comparison; Growth Scale Values carry change over time. If a form or district requires an age equivalent, report it with that limitation written next to it, and never write that the student's language is at the level of a younger child. How do I write up the CELF-5 for a bilingual or dialect-speaking student? Start with language history and the normative question: the English edition's norms assume English-dominant experience, so if the student is not represented, the standard scores are reported descriptively or withheld, and the conclusion rests on assessment in the dominant language, a language sample in each language where feasible, dynamic assessment, and converging parent and teacher evidence. Federal rules require instruments that are not racially or culturally discriminatory and administration in the native language or mode most likely to yield accurate information (34 CFR 300.304(c)(1)), and states such as Georgia exclude normal second-language acquisition and dialect difference from the impairment definition. Do not reach for the Spanish CELF-5 on the US store as a fix: it is the Spain-normed edition (2018; ages 5:0 to 15:11), a different normative population rather than a US bilingual norm set. Is the CELF-5 still the current edition, and is a CELF-6 coming? As of September 2026 the CELF-5 (2013) is the edition Pearson markets in the US, Canada, and Australia. Pearson published a CELF-6 standardization field-research notice in 2025 recruiting English-speaking participants aged 5 to 21, which means a sixth edition is in development, and no release date has been published. A report therefore names the edition, the norm set, and the platform, and carries its own date. When a new edition arrives, the defensible practice is to state which edition and norms were used and, for reevaluations, to note that scores across editions are not directly comparable and that Growth Scale Values are edition-specific. CELF-5, CELF Preschool-3, PLS-5, or GFTA-3: which one does the referral need? By referral question and developmental level, not birthday. The CELF Preschool-3 (2020) covers 3:0 to 6:11 and the CELF-5 begins at 5:0, so five- and six-year-olds fall in both: a child already facing school-age eligibility questions usually matches the CELF-5's Core and index architecture, while a developmentally younger child with broad delay matches the Preschool-3 or the play-based PLS-5, covered on the PLS-5 page . The CELF-5 assesses language, not speech sound production; an articulation or phonology question needs the GFTA-3, covered on the GFTA-3 page , and the two are written up in separate sections even when billed in one evaluation. Reading and written-expression questions add achievement measures such as the WIAT-4 ; the CELF-5 explains the oral-language substrate but does not establish a specific learning disability. Can I include CELF-5 test questions or the record form in my report? No, and the report does not need them. Record forms, stimulus materials, scoring keys, and norms tables are the publisher's protected content at qualification level B; a report contains the student's derived scores, the tests administered by name, your observations, and your interpretation in your own words, which is exactly the content downstream readers use. Pearson's permissions process covers transfer of results into an electronic record and any placement of an assessment on another platform, and describing a task at the topic level (for example, recalling sentences of increasing length) is fine while quoting an item is not. Test security is also a credibility matter: a team that sees leaked stimuli in a report has reason to doubt the scores that followed. Can BastionGPT draft my CELF-5 report? Yes. Give it the facts (age band and tests administered, scaled and standard scores with intervals and percentiles, Observational Rating Scale and Pragmatics Profile results, language-sample findings, observations, and the governing rule) and it drafts the results section: constituent tests named per composite, intervals compared with the threshold, the pragmatic components labeled correctly, the eligibility statement written against the rule, and a plain-language family summary, ready for your review. It can also cross-check a finished report for a composite without its tests, a bare score at a threshold, a mislabeled Pragmatics Profile, an age equivalent doing decisional work, or an eligibility sentence with no rule. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Publisher record, accessed September 2026: NCS Pearson, CELF-5 product page (fall 2013; ages 5:0 to 21:11; 30 to 45 minutes for the Core Language Score; qualification level B; 16 tests, about 10 per age; scores and platforms; Pragmatics Profile scaled scores), What's New About CELF-5 (norm-referenced Pragmatics Profile; Pragmatics Activities Checklist; Growth Scale Values; optional tests), Determining the Severity of a Language Disorder (average range; optimal cut of 80; agency criteria of 1, 1.5, or 2 SD), and CELF-6 standardization field-research notice (2025; ages 5 to 21). Family and international editions, accessed September 2026: Pearson, CELF Preschool-3 (2020; ages 3:0 to 6:11), CELF-5 Metalinguistics (2014; ages 9:0 to 21:11), and CELF-5 Spain version (2018; ages 5:0 to 15:11; Spain norms); Pearson Canada, CELF-5 (US norms; French Canadian version on Q-interactive); Pearson Australia and New Zealand, CELF-5 Australian and New Zealand edition (2017; 2011 Australian and 2013 New Zealand census-based norms). Federal law: 34 CFR 300.8(c)(11) (speech or language impairment definition) and 34 CFR 300.304 (variety of tools; no single measure as sole criterion; nondiscriminatory assessment in the native language; all areas of suspected disability), via Cornell LII. State rules: California, 5 CCR 3030(b)(11)(D) ; New Jersey Department of Education, Speech-Language Services Eligibility broadcast , April 17, 2024 (N.J.A.C. 6A:14-3.5(c)4 and 3.6(a)); Georgia Department of Education, Rule 160-4-7-.05, Appendix (j) . Independent reviews: Coret MC and McCrimmon AW, 2015, Journal of Psychoeducational Assessment 33(5), 495 to 500, CELF-5 test review ( doi 10.1177/0734282914557616 ; 30 to 45 minutes for the core, 90 to 120 minutes for a full battery); Columbia University LEADERS Project, CELF-5 test review (2014; classification-sample and cut-score critique; second-language misidentification risk). Payer and professional context: CMS, Medicare Benefit Policy Manual, Chapter 15 , section 220.3 (therapy documentation requirements); ASHA Practice Portal, Spoken Language Disorders (assessment components including language sampling and dynamic assessment). Rights and trademark: Pearson, Permissions and Licensing (electronic-record transfer and commercial platform licensing); USPTO TSDR, CELF, US registration 1776508 (NCS Pearson, Inc.; renewed 2023). Canada and Australia: Ontario Ministry of Education, Special Education in Ontario, Kindergarten to Grade 12: Policy and Resource Guide ; NDIS, supporting-evidence guidance for health professionals (functional impact across communication, social interaction, and learning). ## Related templates GFTA-3 Report Write-Up PLS-5 Report Write-Up Psychoeducational Evaluation Report IEP Psychoeducational Input WIAT-4 Report Write-Up Autism Spectrum Evaluation Report Section 504 Clinician Statement Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/gfta-3-report All note templates # GFTA-3 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The GFTA-3 (Goldman-Fristoe Test of Articulation, Third Edition) is a norm-referenced single-word articulation test for ages 2:0 through 21:11 that scores consonant production against same-age, same-sex norms. Speech-language pathologists use it in school eligibility and clinical speech sound evaluations. The write-up must pair the score with pattern, stimulability, and connected-speech evidence. This page covers how to write up GFTA-3 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Speech-language pathologists (SLPs) in schools, clinics, hospitals, and early-intervention programs; publisher qualification level B Audience IEP and eligibility teams, school psychologists and psychoeducational evaluators, referring physicians and pediatricians, payers, NDIS and provincial program reviewers, families Typical length 400 to 800 words for the results section · administration 5 to 15 minutes for Sounds-in-Words, longer with Sounds-in-Sentences, stimulability, and a connected-speech sample Format family Norm-referenced single-word articulation test (Sounds-in-Words core; separately normed Sounds-in-Sentences; criterion-referenced stimulability; KLPA-3 phonological-process companion) When it's used School speech or language impairment eligibility evaluations and reevaluations for articulation and phonology concerns, clinical speech sound disorder evaluations, preschool and early-intervention speech referrals, bilingual evaluations alongside the Spanish edition, progress reevaluations Standards context Published by NCS Pearson (2015; Spanish edition 2017; no fourth edition announced as of September 2026); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the GFTA-3? The GFTA-3 (Goldman-Fristoe Test of Articulation, Third Edition; Goldman and Fristoe; NCS Pearson, 2015) is a norm-referenced test of consonant articulation for ages 2:0 through 21:11. Its core, Sounds-in-Words, elicits single consonants in initial, medial, and final positions plus consonant clusters through picture naming in 5 to 15 minutes; the raw score is the number of errors (substitutions, omissions, and distortions count; recognized dialectal variations do not), converted through separate male and female norm tables into a standard score (mean 100, SD 15) with a 95 percent confidence interval, a percentile rank, a test-age equivalent, and a Growth Scale Value. Sounds-in-Sentences is a structured sentence-imitation task (the examiner reads a short story, then the child repeats each sentence) with its own standard score on its own scale from age 4, and the examiner's rating of each repeated sentence produces the record-form intelligibility rating. The stimulability section records whether error sounds can be imitated in syllables, words, and sentences; the third edition added vowel and R error analyses, two picture sets (ages 2 to 6 and 7 to 21), and dialect-sensitive scoring for American English dialects and English influenced by other languages; and it runs on paper, Q-global, and Q-interactive. The GFTA-3 Spanish (2017) is an adaptation with its own norms based on Spanish speakers living in the US and Puerto Rico, not a translation. As of September 2026 the 2015 edition remains current on Pearson's US, Canadian, and Australian sites, with no fourth edition announced. The load-bearing distinction for the write-up is articulation versus phonology, and the GFTA-3 sits on one side of it. Its Sounds-in-Words score counts how many consonants were wrong in single words against same-age, same-sex norms; it does not say whether the errors are organized. Whether they form rule-governed patterns (velar fronting, cluster simplification, deletion of final consonants) is a phonological question, and Pearson answers it with a companion product, the KLPA-3 (Khan-Lewis Phonological Analysis, Third Edition; Khan and Lewis, 2015), which re-analyzes the same Sounds-in-Words responses into core developmental and supplemental atypical processes, each by percent of occurrence, with its own standard score and percentile by age and sex; no second administration is involved, and the companion is English only. So a report that gives a GFTA-3 score and then diagnoses a phonological disorder has skipped the construct that defines the diagnosis, and a report that treats the KLPA-3 as a second test has double-counted one set of responses. The defensible write-up keeps two sections with two jobs: the articulation section (sounds and positions in error, error type, consistency, stimulability, and the norm-referenced score) and the phonology section (which patterns, whether they are developmental at this age, and how much of the intelligibility problem they explain), and it derives the treatment approach from whichever mechanism the evidence supports. Two further things the GFTA-3 does not measure need their own sections: intelligibility (the record-form rating is the examiner's judgment of imitated sentences summarized as a percentage, not a norm-referenced score) and spontaneous connected speech (the sentence task is imitation, and Pearson states its raw score is not comparable with Sounds-in-Words). Language belongs to the CELF-5 page and the PLS-5 page , and the whole-report architecture to the psychoeducational report page . ## Who uses the GFTA-3 and when School-based speech-language pathologists are the core users, giving the GFTA-3 inside IDEA evaluations for the speech or language impairment category, whose federal definition expressly includes impaired articulation (34 CFR 300.8(c)(11)), and at reevaluation, where the write-up feeds the eligibility team, the IEP input , and any Section 504 statement . Clinic, hospital, and private-practice SLPs use it to characterize a speech sound disorder and to support medical necessity for treatment, billing the untimed evaluation codes 92522 (speech sound production) or 92523 (speech sound production with language). Early-intervention and preschool teams use it from age 2:0 alongside the broader picture covered on the developmental assessment and ASQ-3 pages, and bilingual evaluators pair the English edition with the GFTA-3 Spanish. Outside the United States, Canadian clinicians use the same US-normed edition inside provincial preschool programs and school-board processes, and Australian clinicians use it, with no Australian normative study, to supply standardized evidence for NDIS and school-support applications that are judged on functional impact. Each of those readers asks a different question of the same paragraph: does the criterion hold, does the speech section agree with the language section, is treatment necessary, and what does it mean for this child. The write-up is defensible when it answers all four without changing the facts. ## How to structure a GFTA-3 results section No regulation prescribes a GFTA-3 report format. What the federal evaluation rules, the state criteria, and the instrument's own design dictate is the content: the sections administered and the norm group, uncertainty around every number compared with a threshold, an articulation section and a phonology section built from one set of responses, intelligibility by a named method, a developmental comparison against a named source, and an eligibility statement written against the named rule. Each section below carries the pitfall that most often undermines it. Identification, sections administered, and norm group. Open with the full test name and edition, the sections administered (Sounds-in-Words, Sounds-in-Sentences, stimulability, the vowel analysis) and whether the KLPA-3 analysis was run, the platform (paper, Q-global scoring, or Q-interactive administration) and mode (in person or telepractice, following Pearson's guidance), the evaluation dates, the child's age in years and months, the norm table used (male or female) with the reason, and the language and dialect of testing. Name the rule the eligibility statement will apply. Pitfall: "GFTA-3 administered" with no sections, no norm group, and no rule. A reader cannot tell whether the sentence task was given, which table produced the percentile, or what criterion the conclusion answers. Validity conditions, language background, and oral mechanism. Document attention and effort, hearing status (screened and passed, or referred), oral structure and function relevant to speech, any motor-speech signs, the language history (languages, dominance, simultaneous or sequential learning, community dialect), interpreter use, and whether the normative comparison is appropriate: the English norm sample's bilingual members were simultaneous learners who used English most, so a sequential or English-learning child is not represented, and ASHA's position is that standard scores are not reported for a child the norm group does not represent. Pitfall: Dialect or transfer features scored as errors. Pearson's scoring counts recognized dialectal variations as correct, and its own FAQ retains the correction of a 2016 webinar slide that said otherwise. Standard scores with uncertainty. Report the Sounds-in-Words standard score (mean 100, SD 15) with its 95 percent confidence interval, percentile rank, and named norm group, and describe the standing in prose (within 1 SD of the mean is conventionally average). Report Sounds-in-Sentences, when given from age 4, as a separate standard score on its own scale, and never compare the two raw scores, which Pearson states are not comparable. Age equivalents, if a form requires them, carry their limitation and never lead. Pitfall: A bare 78 written against a threshold as if the integer were exact, or a Sounds-in-Sentences raw score set beside the Sounds-in-Words raw score as evidence of a context effect. Error inventory: the articulation section. Describe the errors by sound class and position (which consonants and clusters; initial, medial, final), by type (substitution, omission, distortion, including lateral or dentalized sibilants), and by consistency, adding the vowel and R analyses when relevant as descriptive findings. This is where a motor-based error such as a lateral lisp lives. Never list target words or item-level responses; the pattern description says more and respects test security. Pitfall: An item-by-item word list copied from the record form. It leaks protected content, reads like a score report, and still does not say whether the errors are organized. Pattern analysis: the phonology section (KLPA-3). Report the KLPA-3 as a re-analysis of the same Sounds-in-Words responses: its standard score with interval and percentile, then each core or supplemental process by percent of occurrence in prose, stating whether the pattern is developmental or atypical at the child's age and how much of the intelligibility problem it explains. This section, not the articulation score, carries a phonological diagnosis, and it decides between a motor-based and a pattern-based treatment approach. Pitfall: "Phonological disorder" diagnosed from an articulation standard score with no pattern analysis, or the KLPA-3 written up as a second test that double-counts one set of responses. Stimulability, intelligibility, and connected speech. Report stimulability descriptively (which error sounds became accurate, at which level, with what support) as prognostic and target-selection information, never as a score. Report intelligibility by a named method: the Intelligibility in Context Scale, a Percentage of Consonants Correct from a connected-speech sample of stated size and context, or a listener-based percentage of words understood; the record-form rating is the examiner's judgment of imitated sentences and is reported as descriptive information. Then compare single-word accuracy with the conversational sample, because the two diverge in children with speech delay. Pitfall: "The GFTA-3 indicates 60 percent intelligibility." It produces no norm-referenced intelligibility score, and an average single-word score does not rule out a child nobody can understand. Developmental comparison, eligibility statement, and recommendations. Compare each error sound or pattern with a named contemporary source (Crowe and McLeod, 2020, for US English; McLeod and Crowe, 2018, cross-linguistically), quote the governing criterion, walk through each of its elements with the evidence, document adverse educational effect or functional impact, and leave the determination to the team. Then match recommendations to the mechanism: pattern-based intervention for rule-governed processes, motor-based work for isolated distortions, targets ordered by stimulability and developmental expectation, a language evaluation when the profile suggests one, and progress measures that repeat the same probes and sample. Pitfall: "A standard score below 85 qualifies." No federal rule says so, California's articulation subsection names no number, and Wisconsin's turns on intelligibility and stimulability. ### Blank template (copy and adapt) GFTA-3 RESULTS SECTION SKELETON Child: [initials] Age: [y:m] Grade: [ ] Evaluation date(s): [ ] Evaluator: [name, credentials] Referral question: [ ] Instrument: Goldman-Fristoe Test of Articulation, Third Edition Sections given: [Sounds-in-Words / Sounds-in-Sentences / stimulability / vowel analysis] Platform + mode: [paper / Q-global / Q-interactive; in person / telepractice] Norm group: [age; male or female table] Language + dialect of testing: [ ] KLPA-3 analysis: [yes / no] Other measures: [connected-speech sample: context, size, PCC; intelligibility measure (ICS, listener estimate); oral mechanism; hearing; language] Validity conditions: [attention, hearing status, language history and bilingual status, interpreter; norm sample representative: yes / no, why] Standard scores (each with uncertainty): Sounds-in-Words: SS [ ] (95% CI [ ] to [ ]), percentile [ ], norm group [ ] Sounds-in-Sentences (own scale, from age 4): SS [ ], CI [ ], percentile [ ] Age equivalent: [only if required, with limitation stated] Error inventory (articulation): [sounds and positions in error; substitution, omission, distortion; consistency; vowel or R findings; no target words] Pattern analysis (phonology): [KLPA-3 SS [ ], CI [ ], percentile [ ]; each process by percent of occurrence; developmental or atypical for age] Stimulability: [sound; level (syllable, word, sentence); support needed] Intelligibility + connected speech: [method named; sample size and context; PCC [ ]%; single-word versus conversational accuracy compared] Developmental comparison: [each error against Crowe and McLeod (2020)] Eligibility / diagnostic statement: [governing rule quoted; each element compared; adverse effect or functional impact; team determination] Recommendations + progress plan: [approach matched to mechanism; targets ordered by stimulability; language referral if needed; progress measures] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, target words, stimuli, record forms, norms, or conversion tables. ## Sample GFTA-3 write-up (fictional) Scenario: a first-grader referred by his teacher because classmates cannot understand him, evaluated by a school SLP, with articulation and phonology written as separate sections from one set of responses, intelligibility measured by named methods, and the eligibility statement written against a state-style criterion that names no cutoff. All details are fictional. Patient: D.R., 6 years 2 months, grade 1 · Setting: Public school speech-language eligibility evaluation · Clinician: M. Delgado, MS, CCC-SLP · Note date: 09/21/2026 Measures and conditions: Goldman-Fristoe Test of Articulation, Third Edition (GFTA-3), English edition with US norms, administered in person on paper and scored on Q-global on 09/14/2026; Sounds-in-Words, Sounds-in-Sentences, and the stimulability section were given, and the Sounds-in-Words responses were analyzed with the Khan-Lewis Phonological Analysis, Third Edition (KLPA-3). D.R. was 6 years 2 months at testing and was scored against the male norm table. He is a monolingual English speaker of the local community variety; no dialectal productions were counted as errors, and the normative comparison is judged appropriate. Hearing was screened and passed at school on 09/09/2026. Oral structure and function were adequate for speech, diadochokinetic rates were age-appropriate, and no signs of a motor speech disorder were observed. A 118-word conversational and story-retell sample was recorded on 09/17/2026 and transcribed for a Percentage of Consonants Correct; the Intelligibility in Context Scale was completed by his mother on 09/14/2026, and the classroom teacher returned an observation form on 09/15/2026. Language was evaluated separately with the CELF-5 and is reported in the language section, where the Core Language Score fell within the average range. Attention was good across both sessions. Standard scores: On Sounds-in-Words, D.R. obtained a standard score of 72 (95 percent confidence interval 66 to 78; 3rd percentile) compared with same-age peers in the male norm group, well below the average range. On Sounds-in-Sentences, scored on its own scale, the standard score was 70 (63 to 77; 2nd percentile); the two raw scores are not compared because Pearson states they are not comparable across tasks. The KLPA-3 standard score, derived from the same Sounds-in-Words responses, was 67 (60 to 74; 1st percentile). The test-age equivalent is not reported; standard scores, intervals, and percentiles carry the comparison. The sentence-task intelligibility rating (the examiner's rating of each imitated sentence, summarized as a percentage) was 40 percent and is reported here as descriptive information, not as a normed score. Error inventory and pattern analysis: Articulation: the velar stops /k/ and /g/ were replaced by alveolar stops in initial, medial, and final positions on every opportunity; clusters beginning with /s/ and clusters ending in /r/ or /l/ were reduced to a single consonant in most opportunities; and /s/ and /z/ were produced with lateral airflow in all positions, a distortion rather than a substitution. Singleton /r/ and /l/, the non-sibilant fricatives, and all vowels were produced correctly, so the phonetic inventory is otherwise complete for age. Phonology (KLPA-3): velar fronting occurred in roughly four of five opportunities and cluster simplification in about two thirds, both core processes; deletion of final consonants was occasional in single words but frequent in conversation; no supplemental (atypical) process was observed. Interpretation: the fronting and cluster patterns are rule-governed simplifications that are expected to have resolved well before this age, while the lateralized sibilants are a separate, motor-based articulation error that is not a developmental pattern at any age. Stimulability, intelligibility, and connected speech: Stimulability: /k/ was produced accurately in syllables and words with a model and a light tactile cue, /g/ in syllables only, and the reduced clusters when both consonants were modeled slowly; the lateral /s/ was not stimulable at any level on this date. Intelligibility: on the Intelligibility in Context Scale his mother's ratings averaged 2.7 of 5, describing speech usually understood by parents, sometimes by his teacher, and rarely by unfamiliar adults; the teacher estimated that she understood about half of his classroom speech. The conversational sample yielded a Percentage of Consonants Correct of 61 percent, which the Shriberg and Kwiatkowski convention describes as moderate to severe; single-word accuracy was clearly higher than conversational accuracy, and final consonants present in single words were often omitted in connected speech. Developmental comparison: Crowe and McLeod (2020) place plosives at the 90 percent criterion by 3;11, so the velar substitutions are more than two years past expectation; the cluster and final-consonant patterns are likewise well past the ages at which they typically resolve; the lateral sibilant distortion is not developmental. Eligibility analysis: Our state's articulation criterion requires reduced intelligibility or an inability to use the speech mechanism that significantly interferes with communication and attracts adverse attention, speech sound production below developmental expectation for chronological age, and an adverse effect on educational performance; it names no standard-score cutoff, and the 1.5 SD language rule in the same regulation does not apply to articulation. Element one is met by the conversational PCC, the parent and teacher ratings, and the peer reactions recorded by the teacher; element two by the pattern analysis against the 2020 acquisition data, corroborated by the norm-referenced scores; element three by the teacher's report of 09/15/2026 that classmates ask him to repeat, that he shortens his answers, and that he avoids reading aloud. Dialect and language background were considered and do not account for the errors. No single measure has been used as the sole criterion (34 CFR 300.304(b)(2)); eligibility under speech or language impairment is the team's determination. Recommendations and progress plan: Recommended: direct speech intervention using a pattern-based phonological approach for velar fronting and cluster simplification, starting with /k/ because it is stimulable, with final-consonant production targeted in connected speech; the lateral sibilants addressed as a separate motor-based articulation target once the velar pattern is established; classroom supports of a repair routine (repeat, rephrase, show), teacher checks for understanding before independent work, and no cold-call reading aloud until intelligibility improves. Progress will be measured by percent of occurrence of each pattern on the same monthly probe set, a Percentage of Consonants Correct from a repeated conversational sample each quarter, and the Intelligibility in Context Scale at reevaluation; the GFTA-3 will not be re-administered for progress. Reviewed with D.R.'s parents on 09/21/2026 in plain language: he knows and uses almost all of the sounds of English, a few sounds follow a simplifying pattern that most children outgrow earlier, and one sound is made with the air going out the sides; both are teachable, and he can already make the first target with help. This sample is fictional and for educational purposes. It does not describe a real child or record; the scores, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The sections administered, the platform, the norm table, and the language of testing are stated before any score, so a reader knows which tasks were given and which normative group produced each percentile. Articulation and phonology sit in separate sections built from one set of responses: the error inventory names sounds, positions, and error types without target words, and the KLPA-3 patterns carry the phonological interpretation and the treatment approach. Intelligibility is measured by named methods (a parent-report scale, a percentage of consonants correct from a sized conversational sample, and teacher report), the record-form rating is reported as descriptive, and single-word accuracy is compared with connected speech. Each error is compared with a named 2020 acquisition source, and the eligibility statement quotes the criterion, walks through its elements, documents adverse effect, and leaves the determination to the team, as 34 CFR 300.304 requires. Recommendations follow the mechanism (pattern-based intervention for the processes, motor-based work for the distortion), targets are ordered by stimulability, and progress is measured by repeating the same probes and sample rather than re-administering the norm-referenced test. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the write-up sits inside three decision systems, and the report should say which one each sentence serves. Under IDEA, speech or language impairment is a communication disorder, expressly including impaired articulation, that adversely affects educational performance (34 CFR 300.8(c)(11)); the evaluation rules require a variety of tools and strategies, prohibit any single measure or assessment as the sole criterion, require instruments that are not racially or culturally discriminatory and are administered in the child's native language or mode of communication, and require assessment in all areas of suspected disability (34 CFR 300.304(b) and (c)); no federal GFTA-3 cutoff exists (LAW). States supply the operational rule, and for articulation the rules are mostly functional rather than numeric: California defines an articulation disorder by reduced intelligibility or an inability to use the speech mechanism that significantly interferes with communication and attracts adverse attention, with sound production below developmental expectation and adverse educational effect, and it reserves its 1.5 SD or 7th percentile rule for the language-disorder subsection (5 CCR 3030(b)(11)(A) and (D)); Wisconsin requires, after consideration of age, culture, language background, and dialect, delayed production documented in a natural environment and on a criterion-referenced or norm-referenced measure, intelligibility below the expected range that is not due to home language or dialect and is rated across environments, and production less than 30 percent stimulable for incorrect sounds (PI 11.36(5)(b)1) (LAW, state by state). Texas has no state number either: the Texas Speech-Language-Hearing Association's 2020 articulation guidelines, which carry no regulatory authority, build the determination on a communication-disorder finding plus adverse effect, draw their acquisition table from McLeod and Crowe (2018), use a 100-consecutive-word intelligibility procedure, a 50 to 100 utterance connected-speech sample, and stimulability, and warn that the skewed distribution of articulation errors across ages should not be used to determine eligibility (CONVENTION). The familiar threshold of 85 (1 SD below the mean) is a descriptive convention; where a publisher reports classification statistics at a cutpoint, that statistic describes the test against a clinical reference group, not any agency's definition of disability, so "below 85 qualifies" is a local CONVENTION wherever no rule adopts it. In clinics, coverage turns on documented history, objective findings, functional limitation, and medical necessity under the payer's policy, billed through the untimed evaluation codes 92522 or 92523; a standard score supports the case and never makes it (PAYER POLICY). Canada and Australia change the norm question and the frame. The English GFTA-3 carries US norms only: Pearson's Canadian and Australian catalogs sell the same edition, and no Canadian or Australian normative study exists, so a report written in either country names the American-English normative frame instead of writing "same-age peers" as if they were local. Canadian service is provincial and program-based: Ontario's Preschool Speech and Language Program, funded by the Ministry of Children, Community and Social Services, serves children from birth until they start school on self-referral and developmental concern with no test-score gate, school-aged children are served through school boards, and other provinces run their own pathways, so any local threshold is labeled a board or agency CONVENTION unless a provincial policy source is cited (PROGRAM POLICY). Australia's NDIS early childhood approach admits young children with developmental delay without a diagnosis and judges evidence on functional impact across communication, social interaction, learning, and daily activities (PAYER POLICY); state school-support criteria target broader functional disability rather than articulation alone; and the Australian-relevant alternative is the DEAP, which carries UK norms and was check-normed on 144 Queensland children aged 5:0 to 6:0, with Australian results reported separately in its manual. Edition and rights: the GFTA-3 (2015) is the current edition on Pearson's US, Canadian, and Australian sites as of September 2026, with no fourth edition announced; the GFTA-3 Spanish (2017) is an adaptation with its own norms based on Spanish speakers living in the US and Puerto Rico, not a translation, and the KLPA-3 companion is English only. The publisher classifies the GFTA-3 at qualification level B; record forms and stimulus books are protected consumables, transferring results into an electronic record and placing scoring on another platform are permission and license matters with Pearson, and a free public web scorer of GFTA-3 conversions has no authorization. A report may always contain the child's derived scores and your interpretation in your own words. GFTA-3 is a trademark of NCS Pearson, Inc. (Pearson's own materials mark GFTA and KLPA as trademarks of Pearson Education and its affiliates, and the 2015 test content is copyrighted by NCS Pearson, Inc.). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common GFTA-3 write-up errors reviewers flag The numbers behind these errors are specific. 34 CFR 300.304(b)(2) prohibits any single measure from deciding; Pearson's own FAQ states that dialectal variations are not counted as errors (and retains the correction of a webinar slide that said otherwise) and that Sounds-in-Words and Sounds-in-Sentences raw scores are not comparable; Crowe and McLeod's 2020 review of 18,907 US children places plosives, nasals, and glides at the 90 percent criterion by 3;11, affricates by 4;11, liquids by 5;11, and fricatives by 6;11; Morrison and Shriberg (1992) showed that articulation-test and conversational-sample results diverge in children with speech delay; Kirk and Vigeland (2014) found that norm-referenced phonological pattern tests failed many of the psychometric criteria expected of them; and Wisconsin's rule keys eligibility to intelligibility and a 30 percent stimulability criterion while California's articulation subsection names no number at all. The BastionGPT Clinical Advisory Board sees the same errors most often in GFTA-3 documentation reviews: The 85 cutoff written as the rule. "A standard score below 85 indicates an articulation disorder and qualifies the student." No federal rule says so, California's articulation criterion is intelligibility, developmental expectation, and adverse effect, Wisconsin's adds a stimulability threshold, and a publisher's classification cutpoint describes the test, not the law. Write the score, its interval, and the named rule, and let the team apply it. The articulation score standing in for intelligibility. "The GFTA-3 indicates 55 percent intelligibility." Sounds-in-Words counts consonant errors in single words, and the record-form rating is the examiner's judgment of imitated sentences. Intelligibility needs its own named method (the Intelligibility in Context Scale, a Percentage of Consonants Correct from a sized conversational sample, or a listener-based words-understood percentage), and an average single-word score with poor spontaneous intelligibility is a finding, not a dismissal. A phonological diagnosis with no pattern analysis. A low GFTA-3 score followed by "phonological disorder" with no KLPA-3 or equivalent process analysis, or the KLPA-3 written up as a second test. The GFTA-3 says how many consonants were wrong; the KLPA-3 re-analyzes the same responses to say whether the errors are organized into developmental or atypical patterns, which is the construct the diagnosis and the treatment approach depend on. Dialect and transfer features scored as errors. A community-dialect or Spanish-influenced production counted as a substitution, or English-edition standard scores reported as diagnostic for a sequential bilingual child. Pearson scores recognized dialectal variations as correct, the English norm sample's bilingual members were simultaneous learners who used English most, ASHA's position is that standard scores are not reported for a child the norm group does not represent, and the federal rules require nondiscriminatory assessment in the native language. The two raw scores compared, or the sentence task scored out of range. "He made 12 more errors in sentences than in words, showing a connected-speech effect." Pearson states the raw scores are not comparable because the tasks offer different production opportunities, and Sounds-in-Sentences yields no standard score below age 4. Compare the two standard scores' standing and the clinical pattern, and get connected-speech evidence from a spontaneous sample. The norm group left unstated. A percentile with no mention of the male or female table that produced it, or a nonbinary or transgender student scored silently against one table. The instrument offers only binary norms and the publisher has issued no guidance, so the report names the table, gives the reason, and rests the conclusion on criterion-referenced evidence rather than the choice. An age equivalent or an old chart doing decisional work. "Articulation age of 3 years 6 months" as the headline, or an error excused or flagged by a decades-old acquisition chart that places /r/ years later than the 2020 review's 5;11. Pearson's own guidance is that age equivalents should not be used for diagnostic or placement decisions; standard scores with intervals carry the peer comparison, and each error is compared with a named contemporary source. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on speech-language evaluation reports. Give it the facts (sections administered and platform, age and norm table, standard scores with intervals and percentiles, the error inventory by sound, KLPA-3 patterns with percent of occurrence, stimulability, intelligibility measures and the connected-speech sample, language history, the governing rule) and it drafts the results section: articulation and phonology in separate sections, intelligibility by a named method, each error compared with a named acquisition source, and the eligibility statement written against the rule, ready for your review. Cross-check a finished report for the gaps reviewers flag: a norm group left unstated, an intelligibility claim resting on the record-form rating, a phonological diagnosis with no pattern analysis, a dialect feature counted as an error, two raw scores compared, or an eligibility sentence built on a cutoff. Draft the companion paragraphs: the plain-language family summary, the bilingual or dialect validity statement, the nonbinary norm-group note, or the progress paragraph that repeats the same probes and conversational sample instead of re-administering the test. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do GFTA-3 scores mean? The Sounds-in-Words raw score is the number of consonant and cluster errors in single words, converted through age- and sex-specific norms into a standard score (mean 100, SD 15) with a confidence interval, a percentile rank, a test-age equivalent, and a Growth Scale Value; the comparison is with same-age peers in the male or female norm table used, and the report says which. Scores within 1 SD of the mean (85 to 115) are conventionally described as average and lower scores as below average to very low, in prose rather than a lookup table. Sounds-in-Sentences, given from age 4, has its own standard score on its own scale, and Pearson states the two raw scores are not comparable. Stimulability, the vowel and R analyses, and the sentence-task intelligibility rating are descriptive, not normed. Age equivalents show no rank among peers and, in Pearson's own guidance, should not be used for diagnostic or placement decisions. As of September 2026 the 2015 edition is current, with no fourth edition announced. What GFTA-3 score qualifies a child for speech services? No score does by itself, and three voices need separating. Federal law sets no cutoff, prohibits any single measure as the sole criterion (34 CFR 300.304(b)(2)), and requires adverse educational effect for the speech or language impairment category (LAW). State articulation rules are mostly functional: California requires reduced intelligibility or a speech-mechanism limitation that significantly interferes with communication, production below developmental expectation, and adverse effect, with no number; Wisconsin requires documented delay, intelligibility below the expected range not due to home language or dialect, and less than 30 percent stimulability for incorrect sounds; Texas association guidance builds on developmental norms, an intelligibility procedure, stimulability, and adverse effect (LAW and guidance, state by state). "Below 85" is a CONVENTION unless a rule adopts it, and where a publisher reports classification statistics at a cutpoint, that describes the test, not the law. So the report quotes the rule, compares each element with the evidence, and leaves the determination to the team. Does the GFTA-3 measure intelligibility? Not in the norm-referenced sense. The Sounds-in-Words score counts consonant errors in single words. The record form does carry an intelligibility rating, but it is the examiner's rating of each imitated sentence in the Sounds-in-Sentences task, summarized as a percentage: a clinician judgment of structured, repeated speech, not a normed score and not spontaneous conversation. A defensible intelligibility statement therefore names its method: the Intelligibility in Context Scale (a seven-item parent report across communication partners; McLeod, Harrison, and McCormack, 2012), a Percentage of Consonants Correct from a connected-speech sample of stated size (Shriberg and Kwiatkowski's 1982 convention reads above 85 percent as mild and below 50 percent as severe), or a listener-based percentage of words understood, as Texas guidance does with a 100-consecutive-word sample. An average single-word score with poor spontaneous intelligibility is a finding to investigate, not a reason to close the case. What is the difference between the GFTA-3 and the KLPA-3, and which one leads the write-up? They answer different questions from one set of responses. The GFTA-3 asks which consonants were produced accurately in the standardized single-word and sentence tasks; the KLPA-3 (Khan and Lewis, 2015) re-analyzes the same Sounds-in-Words responses to ask whether the errors are organized into phonological processes, scoring core developmental patterns and supplemental atypical ones by percent of occurrence, with its own standard score and percentile by age and sex. No second administration is involved, and the KLPA-3 is English only. The articulation section therefore reports sounds, positions, error types, consistency, and stimulability; the phonology section reports patterns and whether they are developmental for the child's age. Neither leads eligibility: the rule leads, and the report draws on whichever evidence answers it, with the treatment approach following the mechanism (motor-based work for an isolated distortion, pattern-based work for rule-governed processes). Which GFTA-3 norms do I use for a nonbinary or transgender student? Pearson provides separate male and female norm tables and, as of September 2026, no published guidance for a student who fits neither. The defensible write-up states that the instrument offers only binary norms, names the table used and why, reports the result as a comparison with that normative category rather than as a universal standing, notes that no eligibility conclusion rests on the choice, and leans on criterion-referenced evidence: the error inventory against a named acquisition source, KLPA-3 patterns, stimulability, and intelligibility in connected speech. If local policy permits scoring against both tables, label that as a sensitivity check rather than a publisher-approved procedure, and do not equate a student's gender identity with the test's norm variable. How do I write up the GFTA-3 for a bilingual or dialect-speaking child? Start with the language history: which languages, since when, which is dominant, which dialect, and whether the child is a simultaneous or sequential learner. Pearson's scoring counts recognized dialectal variations as correct (its FAQ retains the correction of a 2016 webinar slide that said the opposite), and the English norm sample's bilingual subgroup (13 percent) was limited to simultaneous learners who used English most, so the English standard score is not diagnostic for a sequential bilingual or English-learning child; ASHA's position is that standard scores cannot be reported for a child the norm group does not represent, and federal rules require nondiscriminatory assessment in the native language (34 CFR 300.304(c)(1)). Sample speech in both languages, treat cross-linguistic transfer as difference rather than error, consider the GFTA-3 Spanish (2017; its own norms from Spanish speakers in the US and Puerto Rico; not a translation), and keep whatever remains atypical in both languages as the clinical finding. Do I have to give Sounds-in-Sentences, and can I compare it with Sounds-in-Words? No universal rule requires it, and the report should state which sections were given rather than implying a complete kit. Sounds-in-Sentences is a structured sentence-imitation task (the examiner reads a short story, then the child repeats each sentence) normed on its own scale from age 4; for younger children it yields no standard score. It is useful when the referral concerns performance beyond single words, but it is not a substitute for a spontaneous connected-speech sample. Pearson states that the two raw scores are not comparable because the tasks offer different production opportunities; compare the two standard scores' standing and the clinical pattern, not the raw-error difference. If Sounds-in-Words is average and everyone says the child is hard to understand, the sentence task and a conversational sample are where the answer lives. Can I include GFTA-3 target words or the record form in my report? No, and the report does not need them. Record forms, stimulus books, and norms tables are the publisher's protected content at qualification level B, and a report that prints the word list leaks content while explaining nothing that a pattern description does not. Write the error inventory by sound, position, error type, and consistency (for example, velar stops fronted in all positions, clusters reduced, sibilants lateralized) and the KLPA-3 patterns by name and frequency. Pearson's permissions process covers transfer of results into an electronic record and placement of scoring on another platform, and unauthorized public scoring tools have no standing. Test security is also credibility: a team that sees stimuli in a report has reason to doubt the scores that follow. Can BastionGPT draft my GFTA-3 report? Yes. Give it the facts (sections administered and platform, age and norm table, standard scores with intervals and percentiles, the error inventory, KLPA-3 patterns with percent of occurrence, stimulability, intelligibility measures and the connected-speech sample, language history, and the governing rule) and it drafts the results section: articulation and phonology in their own sections, intelligibility by a named method, each error compared with a named acquisition source, and the eligibility statement written against the rule, ready for your review. It can also cross-check a finished report for a norm group left unstated, an intelligibility claim resting on the record-form rating, a phonological diagnosis with no pattern analysis, a dialect feature counted as an error, or an eligibility sentence built on a cutoff. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Publisher record, accessed September 2026: NCS Pearson, GFTA-3 product page (2015; ages 2:0 to 21:11; 5 to 15 minutes for Sounds-in-Words; qualification level B; separate male and female norms; FAQ on dialect scoring, raw-score comparability, and the bilingual norm subgroup), GFTA-3 and GFTA-3 Spanish brochure (sections, sentence-imitation task, intelligibility rating, stimulability, vowel and R analyses, picture sets, KLPA-3 core and supplemental processes), KLPA-3 product page (2015; uses GFTA-3 responses; English only), GFTA-3 Spanish product page (2017; not a translation; norms from Spanish speakers in the US and Puerto Rico), interpretation problems of age and grade equivalents , and Permissions and Licensing . Federal law: 34 CFR 300.8(c)(11) (speech or language impairment, including impaired articulation, that adversely affects educational performance) and 34 CFR 300.304 (variety of tools; no single measure as sole criterion; nondiscriminatory assessment in the native language; all areas of suspected disability), via Cornell LII. State rules and guidance: California, 5 CCR 3030(b)(11)(A) and (D) ; Wisconsin, PI 11.36(5)(b)1 (observation, criterion- or norm-referenced delay, intelligibility across environments not due to home language or dialect, less than 30 percent stimulable); Texas Speech-Language-Hearing Association, SI Disability Determination Guidelines for Articulation Disorders, revised 2020 (McLeod and Crowe acquisition table; 100-consecutive-word intelligibility procedure; 50 to 100 utterance sample; stimulability; skewed error distributions not used for eligibility; no regulatory authority). Acquisition evidence: Crowe K and McLeod S, 2020, American Journal of Speech-Language Pathology 29(4), 2155 to 2169, Children's English consonant acquisition in the United States: a review (15 studies; 18,907 children; 90 percent criterion by class); McLeod S and Crowe K, 2018, American Journal of Speech-Language Pathology 27(4), 1546 to 1571, Children's consonant acquisition in 27 languages: a cross-linguistic review (64 studies; 26,007 children; most consonants acquired by 5;0). Single-word versus connected speech and intelligibility: Morrison JA and Shriberg LD, 1992, Journal of Speech and Hearing Research 35(2), 259 to 273, Articulation testing versus conversational speech sampling ; Shriberg LD and Kwiatkowski J, 1982, Journal of Speech and Hearing Disorders 47(3), 256 to 270, Phonological disorders III: a procedure for assessing severity of involvement (Percentage of Consonants Correct); McLeod S, Harrison LJ, and McCormack J, 2012, Journal of Speech, Language, and Hearing Research 55(2), 648 to 656, The Intelligibility in Context Scale: validity and reliability of a subjective rating measure (seven items; five-point scale; alpha .93). Psychometric critique and practice survey: Kirk C and Vigeland L, 2014, Language, Speech, and Hearing Services in Schools 45(4), 365 to 377, A psychometric review of norm-referenced tests used to assess phonological error patterns , and 2015, 46(1), 14 to 29, Content coverage of single-word tests used to assess common phonological error patterns ; Skahan SM, Watson M, and Lof GL, 2007, American Journal of Speech-Language Pathology 16(3), 246 to 259, Speech-language pathologists' assessment practices for children with suspected speech sound disorders (333 respondents; commercial single-word tests, intelligibility estimates, and stimulability among the most used tasks); Buros Center for Testing, GFTA-3 listing (reviewed in the Twentieth Mental Measurements Yearbook). Professional guidance: ASHA Practice Portal, Speech Sound Disorders: Articulation and Phonology (single-word and connected-speech sampling; stimulability; intelligibility; orofacial examination; error-pattern analysis; dialect and language difference versus disorder; standard scores not reported for an unrepresented child). Canada and Australia: Ontario, Preschool Speech and Language Program (birth until school entry; self-referral); NDIS, supporting-evidence guidance for health professionals (functional impact across communication, social interaction, learning, and daily activities); Pearson Australia, DEAP product page (UK norms; check-normed on 144 Queensland children aged 5:0 to 6:0). ## Related templates CELF-5 Report Write-Up PLS-5 Report Write-Up Psychoeducational Evaluation Report IEP Psychoeducational Input Child & Adolescent Developmental Assessment ASQ-3 Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pls-5-report All note templates # PLS-5 Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The PLS-5 (Preschool Language Scales, Fifth Edition) is a norm-referenced language test for birth through 7 years 11 months that yields Auditory Comprehension, Expressive Communication, and Total Language standard scores. Speech-language pathologists and early-intervention evaluators use it in Part C and preschool eligibility evaluations. The write-up must label caregiver-reported evidence and apply the state's exact rule. This page covers how to write up PLS-5 results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Speech-language pathologists (SLPs) and early-intervention evaluators in Part C programs, preschools, clinics, and hospitals; publisher qualification level B Audience Part C eligibility teams and service coordinators, IFSP and preschool IEP teams, pediatricians and NICU follow-up programs, payers, NDIS and provincial program reviewers, families Typical length 400 to 800 words for the results section · administration 45 to 60 minutes, plus the caregiver questionnaire, a language sample, and hand scoring Format family Norm-referenced developmental language scale (Auditory Comprehension and Expressive Communication combining into Total Language; play-based elicitation, observation, and caregiver report; supplemental articulation screener, language sample checklist, and caregiver questionnaire) When it's used Part C early-intervention eligibility evaluations, preschool speech or language impairment and developmental-delay evaluations, late-talker and NICU follow-up referrals, bilingual Spanish-English evaluations with the Spanish edition, progress reevaluations Standards context Published by NCS Pearson (2011; Spanish and Australian and New Zealand editions 2012; no sixth edition announced as of September 2026); described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PLS-5? The PLS-5 (Preschool Language Scales, Fifth Edition; Zimmerman, Steiner, and Pond; NCS Pearson, 2011) is a norm-referenced developmental language assessment for children from birth through 7 years 11 months. Two scales, Auditory Comprehension and Expressive Communication, sample understanding and expression from preverbal interaction through emerging language and early literacy in play-based tasks with pictures, objects, and the examiner; each yields a standard score (mean 100, SD 15) with a confidence interval, a percentile rank, a Growth Scale Value for tracking change, and a language age equivalent, and the two combine into a Total Language score on the same metric. Three supplements travel with it: an Articulation Screener that takes under two minutes, a Language Sample Checklist, and the Home Communication Questionnaire, a caregiver form for children functioning in the birth-through-age-2 range whose answers, per the publisher's FAQ, allow many test items to be scored without administering them to the child. Administration runs about 45 to 60 minutes on paper with hand scoring (digital manuals and the digital questionnaire are delivered through Q-global, but scoring itself is manual), at publisher qualification level B, against a US normative sample of 1,400 children that excluded children born prematurely. The PLS-5 Spanish (2012) is a separately normed edition, standardized on 1,150 monolingual and bilingual Spanish-speaking children in the United States and Puerto Rico, that credits responses in either language into dual-language scores; the Australian and New Zealand Language Adapted Edition (2012) adapts wording but keeps the US normative sample; and a separate PLS-5 Screening Test exists for screening only. As of September 2026 the 2011 edition remains current on Pearson's US, Canadian, and Australian sites, with no sixth edition announced. Three facts carry the write-up. First, a PLS-5 score is not necessarily a summary of what the examiner saw: for infants and toddlers the standardized procedure lets caregiver answers on the Home Communication Questionnaire credit items, those credits carry full scoring weight, and no published study establishes that caregiver-credited items have the same diagnostic value as elicited ones, so a defensible report states which evidence came from report, from observation, and from elicitation. Second, the eligibility arithmetic belongs to the state, not the test: IDEA Part C requires each state to define developmental delay in its own terms (34 CFR 303.111), standard-deviation rules read straight off the standard score (85 sits 1 SD below the mean, 77.5 at 1.5 SD, 70 at 2 SD), percent-delay rules use the state's approved method, and federal law prohibits any single procedure from being the sole criterion (34 CFR 303.321(b)); the publisher's own FAQ adds that a low Auditory Comprehension or Expressive Communication score alone can qualify a child whose Total Language score does not, and says of rigid cutoffs applied without confidence bands and other data that they "do not align with IDEIA legislation, best testing practice, or appropriate PLS-5 test interpretation." Third, age: the norms exclude preterm children, the publisher permits correction for prematurity (36 weeks gestation or less) only through 24 months, and the report gives both ages and the authority for the one it used. The DAYC-2 page covers the broad developmental battery that often owns a program's domain-level eligibility arithmetic while the PLS-5 supplies language depth, the CELF-5 page covers school-age language from age 5, and the GFTA-3 page covers the full speech sound evaluation that the PLS-5's two-minute screener can only flag. ## Who uses the PLS-5 and when Speech-language pathologists on Part C early-intervention teams are the core users, giving the PLS-5 at home or in an arena evaluation as the communication component of the multidisciplinary process federal law requires, which must include an instrument, the child's history and a parent interview, functioning in all five developmental areas, other sources, and records, all within 45 days of referral and at no cost to the family. Preschool teams meet it again from age 3, inside Part B evaluations for the speech or language impairment category or a state's developmental-delay category, where the write-up feeds the eligibility team and the IEP input . Clinic and hospital SLPs, developmental pediatrics services, and NICU follow-up programs use it to characterize a late talker or a mixed receptive-expressive profile and to support medical necessity for treatment, billing the untimed evaluation code 92523. Bilingual evaluators use the PLS-5 Spanish for Spanish-English children, because the English edition was normed on children whose primary language is English. Outside the United States, Ontario and other provincial preschool speech and language programs use the US-normed edition inside access-based pathways with no score gate, and Australian clinicians use the Language Adapted Edition to supply standardized evidence for NDIS early childhood applications that are judged on functional impact. The readers are therefore a service coordinator applying a state rule, a receiving team re-deciding after a move, a physician or payer checking necessity, and a family entitled to a plain-language account. The broader evaluation frame lives on the developmental assessment page, global infant and toddler development on the Bayley-4 page, and the screening that precedes all of this on the ASQ-3 and M-CHAT-R pages. ## How to structure a PLS-5 results section No regulation prescribes a PLS-5 report format. What the federal evaluation rules, the state criteria, and the instrument's own design dictate is the content: the edition and the ages, the source of every credited behavior, uncertainty around every number compared with a threshold, the converging measures that make the instrument one procedure among several, and an eligibility statement written against the named rule. Each section below carries the pitfall that most often undermines it. Identification, edition, ages, and the governing rule. Open with the full test name and the edition actually given (PLS-5 English, PLS-5 Spanish with single-language or dual-language administration, or the Australian and New Zealand Language Adapted Edition), the materials (paper, or digital manuals and questionnaire through Q-global) and the fact that scoring is manual, the setting and mode (home, clinic, arena, or telepractice under the publisher's guidance), the evaluation dates, and the chronological age. If the child was born at 36 weeks or earlier, state the gestational age, the corrected age, the arithmetic, and the authority for using it (the publisher permits correction only through 24 months; the state's rule may say more). Name the state or program whose eligibility definition the report will apply. Pitfall: "PLS-5 administered in Spanish" for a child who was given the separately normed Spanish edition, or a corrected age with no gestational age, no arithmetic, and no authority. A reader cannot reproduce either. Evidence sources: reported, observed, elicited. State which credited behaviors came from the Home Communication Questionnaire and caregiver interview, which were observed spontaneously in play and routines, and which were elicited under the standardized procedure, without naming items. The procedure permits caregiver credit for early behaviors and gives it full scoring weight; the report's job is to preserve the provenance, because no published study establishes that reported and elicited credits carry the same diagnostic value. When caregiver report carried most of a scale, say so and treat the score as provisional pending direct observation. Pitfall: "N. demonstrated" when the credit came from the questionnaire. Reviewers flag it first, and a receiving team that later observes the child directly will find the gap. Behavioral and validity conditions. Document engagement, attention, and regulation, health on the day, hearing status (newborn screen, later screening, or audiology), vision, the languages spoken at home and the child's exposure to each, interpreter involvement, and any nonstandard conditions. State plainly whether the normative comparison fits this child: the English norms assume English as the primary language and exclude children born prematurely. Federal rules require assessment in the child's native language or the mode most likely to yield accurate information. Pitfall: English-edition standard scores reported as diagnostic for a bilingual toddler, or a session cut short by fatigue scored without comment. State the conditions and let them bound the interpretation. Scale scores with uncertainty. Report Auditory Comprehension, Expressive Communication, and Total Language as standard scores (mean 100, SD 15), each with its confidence interval and percentile rank and a descriptor in prose (scores within 1 SD of the mean are conventionally described as average), and say which age produced them. Lead with the receptive-expressive pattern rather than the total when the two scales diverge, because the total blends them and the publisher's guidance is that either scale alone can qualify a child. Record Growth Scale Values for later change measurement; report an age equivalent only where a form requires one, with its limitation stated. Pitfall: A bare 78 compared with a 77.5 line as if the integer were exact, or Total Language treated as the only number when a strong receptive score is hiding a qualifying expressive one. Converging evidence. Add the measures that make the PLS-5 one procedure among several: a language sample of stated length and context (vocabulary, word combinations, communicative functions, gesture use, intelligibility where relevant), an independent parent-report instrument such as the age-appropriate MacArthur-Bates CDI form, the Articulation Screener result if given (a screen, not a speech sound evaluation), developmental history, and hearing status. The publisher itself points evaluators to language sampling, observation, parent interview, and other measures when interpreting PLS-5 results. Pitfall: The PLS-5 standing alone. Federal law prohibits any single procedure as the sole criterion for eligibility, and a caregiver questionnaire embedded in the test is not an independent parent-report measure. Eligibility analysis against the named rule. Quote the jurisdiction's actual criterion and apply it in its own currency: a standard-deviation rule reads off the standard score, with the interval and the state's rounding convention deciding the boundary, and a percent-delay rule uses the state's approved method and instruments, labeled as the state's administrative method when it relies on age equivalents. Never convert a standard score into a percent delay. Then situate the score inside the multisource process: no single procedure decides, informed clinical opinion can establish eligibility that scores miss but never negate a qualifying result, and the determination belongs to the team. In a clinic, the same paragraph states the diagnosis and the functional limitations that establish medical necessity. Pitfall: "1.5 SD below the mean, which equals a 25 percent delay." It does not; the two are different currencies, and a state that accepts one route has not authorized the conversion. Family summary, recommendations, and next steps. Close with a plain-language paragraph the family can use: what is easier for the child, what is harder, how it shows up at home, what came from the family's own report, whether the program's criterion was met, and what happens next (team review, IFSP development within the 45-day window, transition planning near the third birthday with its 90-day conference requirement). Link each recommendation to a documented finding, name the progress measure (Growth Scale Values and a repeat parent-report inventory rather than re-comparing age-normed scores), and state the referrals the profile implies, hearing first. Pitfall: A report that ends at score classifications. Part C expects results explained to the family and their priorities to shape the plan; a template recommendation list with no thread back to the findings reads as an evaluation done to the family rather than with it. ### Blank template (copy and adapt) PLS-5 RESULTS SECTION SKELETON Child: [initials] Evaluation date(s): [ ] Evaluator: [name, credentials] Chronological age: [ ] Corrected age (if born at 36 weeks or earlier): [ ] (gestational age [ ]; correction permitted through 24 months; authority: [ ]) Referral question: [ ] Program + eligibility rule applied: [state/program; criterion quoted, e.g. 1.5 SD below the mean in one area / 25% delay] Instrument: Preschool Language Scales, Fifth Edition Edition: [English / Spanish (single or dual language) / Australian and NZ] Materials: [paper / Q-global manuals + questionnaire] Scoring: manual Setting + mode: [home / clinic / arena; in person / telepractice] Evidence sources: [behaviors credited from the Home Communication Questionnaire and caregiver interview; observed in play and routines; elicited] Behavioral + validity conditions: [engagement, health, hearing status, home languages and exposure, interpreter; normative comparison fits: yes / no] Scale scores (at [chronological / corrected] age, each with uncertainty): Auditory Comprehension: SS [ ] (95% CI [ ] to [ ]), percentile [ ], [descriptor] Expressive Communication: SS [ ], CI [ ], percentile [ ], [descriptor] Total Language: SS [ ], CI [ ], percentile [ ], [descriptor] Pattern statement: [receptive vs expressive; what the total blends] GSVs recorded: [ ] Age equivalent: [only if a form requires it] Converging evidence: [language sample: context, length, findings; parent-report measure (e.g. CDI form) result; Articulation Screener; history; hearing] Eligibility analysis: [criterion in its own currency (SD from standard scores; percent delay only by the state's method); result vs threshold; AC or EC alone if the rule allows; other sources; informed clinical opinion; team] Family summary (plain language): [strengths, needs, what the family reported, criterion met or not, next steps] Recommendations + next steps: [linked to findings; IFSP timeline; progress measures (GSVs, repeat parent report); referrals; transition if near three] Evaluator signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank results-section skeleton, ready to adapt. Neither reproduces test items, stimuli, record forms, the caregiver questionnaire, norms, or conversion tables. ## Sample PLS-5 write-up (fictional) Scenario: a 20-month-old former 35-week preterm infant referred by his pediatrician for few words, evaluated at home by a Part C speech-language pathologist, with caregiver-credited evidence labeled, both ages reported, an expressive-specific profile, and the eligibility statement written against a named state-style rule. All details are fictional. Patient: N.R., 20 months 12 days (19 months 7 days corrected) · Setting: Part C early intervention eligibility evaluation, communication component, family home · Clinician: L. Okonkwo, MS, CCC-SLP · Note date: 09/18/2026 Measures and conditions: Preschool Language Scales, Fifth Edition (PLS-5), English edition with US norms, paper administration with hand scoring, given at the family's home on 09/15/2026 as the communication component of a multidisciplinary Part C evaluation. The Home Communication Questionnaire was completed by N.R.'s mother on 09/12/2026 and reviewed with her by interview before testing. N.R. was born at 35 weeks gestation; chronological age at testing was 20 months 12 days and corrected age 19 months 7 days. Program policy permits correction for prematurity through 24 months, consistent with the publisher's guidance, so the norm-referenced scores below are reported at corrected age, with the chronological-age result given where it differs. Referral came from the pediatrician on 09/01/2026 for few words at the 18-month visit. The home is monolingual English, so the normative comparison is judged appropriate. Hearing: the newborn screen was passed, and a diagnostic audiology evaluation on 09/08/2026 was within normal limits bilaterally. N.R. engaged readily in floor play for about 40 minutes with one snack break, and standard administration was maintained. The Articulation Screener was not administered at this age; speech sounds were described from the play sample. Evidence sources: Behaviors that the standardized procedure permits to be credited from caregiver report were scored from the questionnaire and interview only where his mother described the behavior as current and consistent, and those credits are identified as caregiver-reported in the record and in this section rather than as examiner-observed. Comprehension behaviors were largely elicited or observed directly in play and routines. Expressive credits rested more heavily on caregiver report, and every reported expressive behavior was also heard or seen at least once during the session, so no credited behavior stands on report alone; the vocabulary his mother described matched what was observed and what the parent inventory recorded. Scale scores: At corrected age, the Auditory Comprehension standard score was 94 (95 percent confidence interval 87 to 101; 34th percentile), within the average range. Expressive Communication was 70 (64 to 76; 2nd percentile), well below the average range, with the entire interval below the 1.5 SD line of 77.5. Total Language was 80 (75 to 85; 9th percentile), below average, an intermediate value that blends the two scales and understates the expressive-specific pattern. Scored against chronological age instead, Expressive Communication would be 67 (1st percentile) and Auditory Comprehension 91; the pattern, and the conclusion below, are the same under either age. Descriptors follow the convention that standard scores within 1 SD of the mean are average. Growth Scale Values were recorded for reevaluation. Age equivalents are not reported: the program's criterion is stated in standard deviations, and the two metrics are not interchangeable. Converging evidence: A 30-minute play-based language sample at home on 09/15/2026 yielded nine different recognizable words used spontaneously, no two-word combinations, and frequent pointing, showing, giving, and vocalizing to request and comment; communicative intent, joint attention, and symbolic play were age-appropriate. On the MacArthur-Bates CDI Words and Sentences form, completed by his mother on 09/12/2026, productive vocabulary was 24 words, below the 10th percentile for age, with no word combinations, an independent parent-report result that converges with the expressive scale. The consonant inventory in the sample was limited to early-developing sounds (stops, nasals, and /h/) in simple syllable shapes, expected at this stage and to be reviewed as vocabulary grows. Eligibility analysis: Our state's Part C definition of developmental delay includes a score of at least 1.5 standard deviations below the mean in one developmental area on an appropriate standardized instrument, and the program applies that criterion to a receptive or an expressive scale score when the delay is confined to one; the definition also offers a percent-delay route that the program computes by its own approved method. In the communication area, the Expressive Communication score meets the standard-deviation criterion in its own currency, with the entire interval below 77.5; Total Language does not, and the publisher's own guidance is that a low Auditory Comprehension or Expressive Communication score can qualify a child whose Total Language score does not. No percent-delay figure is derived from the age equivalent or from the standard score. The finding converges with the caregiver questionnaire, the independent parent inventory, the language sample, the pediatrician's referral, and direct observation, and hearing has been ruled out as a cause. Under 34 CFR 303.321 this instrument is one procedure among several, the evaluator's informed clinical opinion supports the same conclusion, and eligibility is the multidisciplinary team's determination within the 45-day timeline that began with the 09/01/2026 referral. Family summary and recommendations: Reviewed with N.R.'s parents on 09/18/2026 in plain language: he understands much of what is said to him at home, follows familiar directions, and communicates clearly with gestures, sounds, and a small set of words; what is hard right now is saying words and putting them together, and his parents' questionnaire and observations matched what we saw. He met the program's definition of a communication delay in the expressive area. Recommended: the team meets with the family to confirm eligibility and develop the IFSP; parent-implemented language intervention coached in daily routines (following his lead, modeling single words and short phrases, responsive turn-taking), with the visit schedule set by the IFSP team; monitoring of speech sounds as vocabulary grows; reevaluation in six months using Growth Scale Values for change and a repeat parent inventory, with a hearing recheck if progress stalls. Transition planning toward preschool services will begin on the program's timeline as his third birthday approaches. This sample is fictional and for educational purposes. It does not describe a real child or record; the scores, dates, and details are invented to show write-up structure and are not clinical guidance. Scores are invented for illustration and correspond to no real child or record, and no norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The edition, the materials, both ages with the gestational basis and the correction authority, and the governing rule are stated before any score, so another evaluator (or a receiving state after a move) can reproduce the arithmetic. Every credited behavior has a source: caregiver-reported credits are labeled as such, the reliance of the expressive scale on report is acknowledged, and nothing from the questionnaire is written as examiner-observed performance. Each scale carries its confidence interval and percentile, the receptive-expressive pattern leads the total, the chronological-age result is shown, and the one qualifying score is the scale score the publisher's guidance says may be used, not a total that would have hidden it. The eligibility analysis stays in the rule's own currency, declines any percent-delay conversion, and rests the conclusion on converging measures (an independent parent inventory, a language sample, history, observation, hearing) inside the no-single-procedure process that 34 CFR 303.321 requires. The family summary does the federally expected work in plain language, and the recommendations trace to the findings with progress measured by Growth Scale Values and a repeat inventory rather than by re-comparing age-normed scores. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the write-up sits inside a legal machine, and the parts are citable. IDEA Part C requires each state to adopt its own definition of developmental delay (34 CFR 303.111) across the five developmental areas an evaluation must cover, communication among them (303.21); requires a multidisciplinary evaluation that includes an instrument, the child's history and a parent interview, functioning in each area, other sources, and records; prohibits any single procedure as the sole criterion; and requires informed clinical opinion, which may establish eligibility that instruments miss but may never negate a qualifying result (303.321). Evaluation, assessment, and the initial IFSP meeting must be complete within 45 days of referral (303.310), evaluation and assessment carry no fee to the family (303.521), and transition toward Part B preschool services requires a conference not fewer than 90 days before the third birthday (303.209) (LAW). From age 3, Part B eligibility runs through the speech or language impairment category, a communication disorder that adversely affects educational performance (300.8(c)(11)), or a state-optional developmental-delay category for ages 3 through 9 (300.8(b)), with the same no-single-measure rule (300.304(b)(2)) and the requirement to assess in the child's native language (300.304(c)(1)) (LAW). The state layer is where PLS-5 write-ups go wrong. A February 2026 federal technical-assistance summary counted 19 jurisdictions using 2 SD below the mean in one area, 19 using 1.5 SD in two or more areas, 18 using a 25 percent delay in one or more areas, 13 using a 30 to 40 percent delay in one area, and a tail of other rules, with most states offering more than one route. New York's regulation accepts a 12-month delay, a 33 percent delay in one domain or 25 percent in each of two, or 2.0 SD below the mean in one domain or 1.5 SD in each of two, and defines a delay confined to communication as 2.0 SD below the mean in that area; Illinois requires a delay of 30 percent or more measured with state-approved instruments, with informed clinical opinion where standardized testing is inappropriate; Texas requires a 25 percent delay in one or more areas, or 33 percent when expressive language is the only delay, established through its state-approved evaluation process (LAW and PROGRAM POLICY, state by state). Pearson's own guidance is a separate voice: use confidence intervals for eligibility decisions, remember that a low Auditory Comprehension or Expressive Communication score alone can qualify a child whose Total Language does not, and treat rigid cutoffs applied without other data as inconsistent with IDEA and good practice (publisher guidance); the familiar "1.5 SD everywhere" is a CONVENTION that no federal authority sets. In clinics, a PLS-5 supports but never establishes medical necessity: evaluations bill through the untimed code 92523 (or 92522 when only speech sound production is evaluated), and coverage, authorization, and frequency follow the plan's policy (PAYER POLICY). Canada and Australia change the frame from score-versus-threshold to function and access. Ontario's Preschool Speech and Language Program takes children from birth until they start school on a family's concern or self-referral, with no diagnosis and no score gate, and other provinces run analogous access-based pathways, so a Canadian report names the actual program framework and labels any local threshold a program CONVENTION unless a provincial policy source is cited; Pearson Canada sells the same US-normed English edition (PROGRAM POLICY, provincial). Australia's NDIS early childhood approach admits children under six on developmental delay without a diagnosis and judges evidence on functional impact, and the Pearson product sold there is a Language Adapted Edition (2012) standardized on the same 1,400 US children, so an Australian report never calls it Australian-normed (PAYER POLICY and CONVENTION). Edition, norms, and honesty about the evidence: the PLS-5 (2011) remains the current edition as of September 2026 on Pearson's US, Canadian, and Australian sites, with no sixth edition announced; the English norms were collected from children whose primary language is English and are not a bilingual norm set, which is what the separately normed PLS-5 Spanish (2012; 1,150 monolingual and bilingual Spanish-speaking children in the United States and Puerto Rico; dual-language scoring) exists for; the normative sample excluded children born prematurely, and the publisher permits correction for prematurity, defined as 36 weeks gestation or less, only through 24 months. Publisher-reported reliability is strong (split-half coefficients from .80 to .97; manual test-retest values of about .90 for the scales at preschool ages), and the publisher reports Total Language sensitivity of .83 and specificity of .80, figures the Columbia LEADERS Project's 2013 review rates as fair while questioning the reference standard and spectrum of the clinical samples. Independent peer-reviewed work has addressed structure and change rather than replicating those classification figures: a 2021 Mokken analysis of 169 Head Start preschoolers found high latent-class reliability for both scales but had to drop five Auditory Comprehension items to reach a moderate hierarchical scale, and a 2022 intervention study of 110 autistic toddlers found Growth Scale Values the most useful metric for change, with standard scores showing floor effects. No published study establishes that caregiver-credited items are diagnostically equivalent to elicited ones, which is why the source label is not optional. Rights follow the usual Pearson pattern: qualification level B, consumable record forms and the caregiver questionnaire that are not reproduced, transfer of results into an electronic record and any embedding of scoring logic as permission and license matters, and no authorization for the free web scorers and spreadsheet generators that circulate; a report may always contain the child's derived scores and your interpretation in your own words. PLS-5 is a trademark of NCS Pearson, Inc. (Pearson's product pages mark the name with the trademark symbol, and the test content is copyrighted by NCS Pearson, Inc.). BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common PLS-5 write-up errors reviewers flag The numbers behind these errors are specific. The ECTA Center's February 2026 summary counts 19 jurisdictions at 2 SD in one area, 19 at 1.5 SD in two or more, 18 at a 25 percent delay, and 13 at a 30 to 40 percent delay, most with more than one route; Pearson's own FAQ permits prematurity correction only through 24 months, defines prematurity as 36 weeks or less, and states that a low Auditory Comprehension or Expressive Communication score alone can qualify a child whose Total Language does not; the publisher reports Total Language sensitivity of .83 and specificity of .80 against split-half reliabilities of .80 to .97; Spaulding, Plante, and Farinella (2006) found sensitivity and specificity reported for only 9 of 43 child language tests, with acceptable accuracy for 5; Hsiao and colleagues (2021) had to drop five Auditory Comprehension items before the scale met a moderate hierarchy criterion in 169 Head Start children; Kwok and colleagues (2022) found standard scores subject to floor effects and Growth Scale Values most responsive to change in 110 autistic toddlers; and 34 CFR 303.321(b) prohibits any single procedure from deciding. The BastionGPT Clinical Advisory Board sees the same errors most often in PLS-5 documentation reviews: Eligibility written from the PLS-5 alone. "PLS-5 scores qualify the child for early intervention" with no state rule quoted and no other sources cited. The score can satisfy a quantitative element, but no single procedure may be the sole criterion, the evaluation must include history, parent interview, functioning in all areas, and records, informed clinical opinion can establish (never negate) eligibility, and the determination belongs to the team applying the named jurisdiction's definition. Percent delay manufactured from the test. "Expressive Communication at 1.5 SD below the mean, a 25 percent delay," or a percent delay computed from the language age equivalent in a state whose method never authorized it. Standard scores locate a child in the age distribution and do not convert to a percentage; percent delay comes only from the state's approved method and instruments, and where that method uses age equivalents the report labels it as the state's administrative rule, not a psychometric equivalence. Caregiver credit written as observed performance. "N. demonstrated" when the credit came from the Home Communication Questionnaire. The procedure legitimately accepts caregiver report for early behaviors and gives it full scoring weight, no study establishes its equivalence to elicited performance, and a questionnaire embedded in the test is not an independent parent-report measure; write the source (reported, observed, elicited) and pair the score with a separate inventory such as the CDI. Total Language treated as the eligibility number. A Total Language of 84 reported as the finding while Expressive Communication sits at 68. The total blends a strong receptive score with a weak expressive one, the publisher states that a low Auditory Comprehension or Expressive Communication score alone can qualify a child whose total does not, and a state rule that applies to the communication area may accept the scale score; lead with the pattern and say what the total hides. Corrected age silent, generic, or past the ceiling. "Corrected for prematurity" with no gestational age, no arithmetic, and no authority, or a correction applied at 30 months. The norms exclude preterm children, the publisher permits correction only through 24 months and defines prematurity as 36 weeks or less, and states differ on whether and how they adjust; report both ages, the policy relied on, which scores used which age, and whether the conclusion changes under the other. A bare score at the threshold. A 78 written against a 77.5 line, or a 70 against a 2.0 SD rule, as if the integer were exact. The publisher ties confidence intervals to eligibility and placement decisions; the report says where the interval falls relative to the criterion and, when it straddles the line, rests the conclusion on the converging measures rather than the point score. The edition, materials, or components misnamed. "PLS-5 administered in Spanish" for the separately normed PLS-5 Spanish (or an ad hoc translation of the English form presented as a standardized score), "Australian norms" for a language-adapted edition standardized in the United States, "scored on Q-global" for a manually scored test whose manuals are merely hosted there, or the two-minute Articulation Screener reported as an articulation evaluation. Name the edition, the administration mode, and the component, and send speech sound questions to a full evaluation. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on speech-language evaluation reports. Give it the facts (edition and materials, both ages with any correction arithmetic and its authority, which credits came from caregiver report, Auditory Comprehension, Expressive Communication, and Total Language scores with intervals and percentiles, the language sample and parent-inventory findings, observations, the state rule) and it drafts the results section: sources labeled, the receptive-expressive pattern ahead of the total, the eligibility analysis in the rule's own currency, and the family summary in plain language, ready for your review. Cross-check a finished report for the gaps reviewers flag: a caregiver credit written as observed, a percent delay derived from a standard score or an unauthorized age equivalent, a corrected age with no arithmetic or past 24 months, a bare score at the threshold, a total hiding a qualifying scale, or an eligibility sentence that names no rule. Draft the companion paragraphs: the plain-language family summary, the bilingual validity statement for a PLS-5 Spanish administration, the informed-clinical-opinion rationale when scores and functioning disagree, or the transition summary as the third birthday approaches. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do PLS-5 scores mean? The PLS-5 yields three standard scores on a mean-100, SD-15 scale: Auditory Comprehension, Expressive Communication, and Total Language, each with a confidence interval, a percentile rank, a Growth Scale Value, and a language age equivalent. Scores within 1 SD of the mean (85 to 115) are conventionally described as average and lower scores as below average to very low relative to age peers, in prose rather than a lookup table; 85 sits 1 SD below the mean, 77.5 at 1.5 SD, and 70 at 2 SD. The publisher's own advice is to use confidence intervals, not single scores, for classification, eligibility, or placement decisions. The two scales matter more than the total when they diverge, because the total blends them and the publisher states that a low score on either scale alone can qualify a child. Growth Scale Values answer a different question (how much the child's own ability changed between administrations) and are not peer rankings, and age equivalents have unequal intervals and no error band, so they never carry a decision. What PLS-5 score qualifies a child for early intervention? Whatever the child's state says, and nothing else. Federal law requires each state to define developmental delay and prohibits any single procedure from being the sole criterion (34 CFR 303.111 and 303.321) (LAW). The definitions differ: a February 2026 federal technical-assistance summary counted 19 jurisdictions using 2 SD in one area, 19 using 1.5 SD in two or more areas, 18 using a 25 percent delay, and 13 using a 30 to 40 percent delay, most with several routes. New York accepts a 12-month delay, 33 percent in one domain or 25 percent in each of two, or 2.0 SD in one domain or 1.5 SD in each of two, and defines a delay confined to communication as 2.0 SD below the mean; Illinois requires 30 percent or more measured with state-approved instruments; Texas requires 25 percent, or 33 percent when expressive language is the only delay (LAW and PROGRAM POLICY, state by state). So there is no universal PLS-5 cutoff. The defensible write-up quotes the current criterion, applies it in its own currency, reports the exact score with its confidence interval, uses the scale score where the rule allows it (a low Auditory Comprehension or Expressive Communication score alone can qualify a child whose Total Language does not, in the publisher's words), and leaves the determination to the multidisciplinary team. Can I compute percent delay from a PLS-5 age equivalent or standard score? Not from a standard score, ever, and from an age equivalent only where the state's own method says so. A standard score locates the child in the normative distribution for their exact age; percent delay divides a developmental-age shortfall by a reference age, treating developmental months as an equal-interval ruler, and the two answer different questions. States that offer both routes treat them as parallel legal alternatives, not translations: New York lists its percent-delay and standard-deviation criteria side by side, Texas establishes its official percent delay through its state-approved evaluation process, and Illinois ties its 30 percent criterion to approved instruments. A 2022 study of PLS-5 change scores found age equivalents, standard scores, raw scores, and Growth Scale Values behaving differently in the same children, which is the empirical reason not to treat an age-equivalent gap as interchangeable with a standard-deviation result. So the report applies a standard-deviation rule from the standard score, applies a percent-delay rule exactly as the state computes it (and labels an age-equivalent method as the state's administrative rule), and never announces one metric as the other. Can the PLS-5 be scored from the Home Communication Questionnaire when a toddler will not engage? Partly, and the report must say exactly where. For children functioning in the birth-through-age-2 range, the publisher's FAQ states that if the caregiver completes the questionnaire before testing, many items can be scored without administering them to the child, and the procedure gives those credits full weight. What no authority provides is evidence that caregiver-credited items carry the same diagnostic value as elicited ones, so the defensible write-up labels which behaviors were credited from report, which were observed, and which were elicited, corroborates reported behaviors through observation and routines where possible, treats a caregiver-heavy scale as provisional, and pairs the test with an independent parent-report inventory such as the age-appropriate MacArthur-Bates CDI form and a language sample. Under Part C, an incomplete standardized administration does not end the evaluation: the team uses appropriate procedures, informed clinical opinion, history, and observation, and no single procedure decides. What a report never does is present a partly administered test as a fully elicited score. How should corrected age be reported on a PLS-5? Completely. The publisher's FAQ states that the normative sample did not include children born prematurely, that clinicians may choose to adjust for prematurity through the age of 24 months, and that a premature birth means 36 weeks or less gestation; after 24 months no adjustment is appropriate. Report the gestational age, the chronological age, the corrected age, the arithmetic, the authority relied on (the publisher's guidance, and the state's rule, which may write prematurity adjustment into its own definition of delay or say nothing), which scores used which age, and whether the eligibility conclusion changes under the other age. Near a threshold the two ages can land on opposite sides of the line, and the honest report shows both rather than silently choosing the age that produces the preferred outcome. When the publisher's convention and the state's eligibility policy differ, the state's policy governs the eligibility analysis and the report says which it followed. Which edition do I use for a bilingual Spanish-English child, and how do I write it up? The PLS-5 Spanish (2012), not the English edition and not a translation of it. The English PLS-5 was normed on children whose primary language is English, and the publisher states it was not normed with bilingual children responding in another language, so an English-edition standard score is not diagnostic for a Spanish-English toddler. The Spanish edition was standardized on 1,150 monolingual and bilingual Spanish-speaking children in the United States and Puerto Rico and offers a dual-language administration that credits a response in either language into bilingual-conceptual Auditory Comprehension, Expressive Communication, and Total Language scores; a report names that edition and that mode, reports language exposure and dominance, the examiner's or interpreter's language competence, and converging evidence across both languages (a sample in each, an independent parent inventory in the home language, developmental history), because a disorder is inferred from a cross-linguistic pattern plus functional impact, not from limited experience in one language. Federal rules require assessment in the child's native language or the mode most likely to yield accurate information (34 CFR 303.321(a) for Part C and 300.304(c)(1) for Part B). PLS-5, CELF Preschool-3, DAYC-2, or REEL-4: which one does the referral need? By decision purpose and developmental level, not birthday. The PLS-5 spans birth through 7:11, works through play and permitted caregiver credit for children at very early communication levels, and gives the deepest receptive-expressive language picture for a toddler; it is often the right instrument for a minimally verbal three-year-old. The CELF Preschool-3 (2020; ages 3:0 to 6:11) begins at three and offers a more differentiated preschool-language profile for an engaged four- or five-year-old whose referral asks about content, structure, and classroom language, and the school-age CELF-5 takes over from age 5. The DAYC-2, covered on the DAYC-2 page , is a five-domain developmental battery (birth through 5:11) that a state program may designate as the official source of its domain-level percent delay while the SLP uses the PLS-5 for language depth, so the two are often paired rather than chosen between. The REEL-4 (birth through 36 months) obtains its receptive and expressive results by caregiver interview, and the Rossetti is criterion-referenced, useful for description but unable to supply a standard-deviation result a rule may require. Speech sound questions go to the GFTA-3 ; the PLS-5's screener only flags them. Can I include PLS-5 items, the record form, or the caregiver questionnaire in my report? No, and the report does not need them. Record forms, the Home Communication Questionnaire, picture and manipulative stimuli, scoring rules, and norms tables are the publisher's protected content at qualification level B; a report contains the child's derived scores, the edition and mode, your observations, the source of each credit at the level of report versus observation versus elicitation, and your interpretation in your own words, which is exactly what downstream readers use. Describing a behavior at the topic level (for example, following a familiar routine direction) is fine; reproducing an item, the questionnaire, or a conversion table is not. Pearson's permissions process covers transfer of results into an electronic record and any placement of scoring on another platform, and the free auto-scorers and spreadsheet generators that circulate online reproduce proprietary conversion logic without authorization. Test security is also credibility: a team that sees leaked stimuli in a report has reason to doubt the scores that follow. Can BastionGPT draft my PLS-5 report? Yes. Give it the facts (edition and materials, both ages with any correction arithmetic and its authority, which credits came from caregiver report, Auditory Comprehension, Expressive Communication, and Total Language scores with intervals and percentiles, the language sample and parent-inventory findings, observations, and the state rule) and it drafts the results section: sources labeled, the receptive-expressive pattern ahead of the total, the eligibility analysis in the rule's own currency, and a plain-language family summary, ready for your review. It can also cross-check a finished report for a caregiver credit written as observed, a percent delay derived from a standard score, a corrected age past 24 months or without its arithmetic, a bare score at a threshold, a total hiding a qualifying scale, or an eligibility sentence with no rule. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Publisher record, accessed September 2026: NCS Pearson, PLS-5 product page (2011; birth through 7:11; 45 to 60 minutes; qualification level B; manual scoring; score types; Home Communication Questionnaire for the birth-through-age-2 range; FAQ on prematurity, confidence intervals, rigid cutoffs, scale scores qualifying alone, English-primary norming), PLS-5 Spanish product page (2012; 1,150 monolingual and bilingual Spanish-speaking children in the United States and Puerto Rico; dual-language scores), and Permissions and Licensing ; Pearson Australia and New Zealand, PLS-5 Australian and New Zealand Language Adapted Edition (2012; 1,400 US children in more than 45 states; split-half .80 to .97; Total Language sensitivity .83 and specificity .80); Pearson Canada, PLS-5 . Federal law: eCFR, 34 CFR 303.111 (state definition of developmental delay), 303.21 (the five developmental areas), 303.321 (multisource evaluation; native language; no single procedure; informed clinical opinion), 303.310 (45-day timeline), and 303.209 (transition); Cornell LII, 34 CFR 300.8 (speech or language impairment; developmental delay ages 3 through 9) and 300.304 (no single measure; native language). State variation and examples: ECTA Center, Part C eligibility criteria summary (updated February 18, 2026; counts by criterion); New York, 10 NYCRR 69-4.23 via Cornell LII (12-month, percent-delay, and standard-deviation routes; communication-only rule); Illinois DHS, Early Intervention eligibility (30 percent or greater; approved instruments; informed clinical opinion); Texas HHS, ECI eligibility (25 percent; 33 percent for expressive language only; state-approved evaluation tool). Independent psychometric studies: Hsiao YY and colleagues, 2021, Journal of Speech, Language, and Hearing Research 64(10), 3983 to 3994, Hierarchy and reliability of the PLS-5: Mokken scale analysis (169 Head Start children; five Auditory Comprehension items excluded; latent-class reliability .94 and .96; manual test-retest .90, .91, .93); Kwok EYL and colleagues, 2022, Journal of Speech, Language, and Hearing Research, Measuring change during intervention using norm-referenced, standardized measures (110 autistic children aged 18 to 48 months; Growth Scale Values versus standard scores, raw scores, and age equivalents); Spaulding TJ, Plante E, and Farinella KA, 2006, Language, Speech, and Hearing Services in Schools 37(1), 61 to 72, Eligibility criteria for language impairment: is the low end of normal always appropriate? (sensitivity and specificity available for 9 of 43 tests; acceptable for 5). Independent reviews: Columbia University LEADERS Project, Test Review: PLS-5 English (2013; publisher sensitivity and specificity rated fair; reference-standard and spectrum concerns) and Test Review: PLS-5 Spanish (2013); Buros Center for Testing, tests reviewed in the Nineteenth Mental Measurements Yearbook (2014; PLS-5, PLS-5 Spanish, and both screening tests listed). Professional guidance and parent-report measures: ASHA Practice Portal, Late Language Emergence (multiple sources of information; parent report; language sampling; hearing screening); MacArthur-Bates Communicative Development Inventories (English and Spanish forms). Canada and Australia: Ontario, Preschool Speech and Language Program (birth until school entry; self-referral); NDIS, early intervention eligibility requirements and supporting-evidence guidance (functional impact; observed, assessed, or reported). ## Related templates CELF-5 Report Write-Up GFTA-3 Report Write-Up DAYC-2 Report Write-Up Child & Adolescent Developmental Assessment Bayley-4 Report Write-Up ASQ-3 Score Documentation M-CHAT-R/F Score Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/madrs All note templates # MADRS Score Documentation: Scoring Conventions, Response Criteria & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The MADRS (Montgomery-Åsberg Depression Rating Scale) is a clinician-rated, ten-item depression severity scale, published in 1979 and designed to be sensitive to change, with a total of 0 to 60. Psychiatrists, nurse practitioners, and research raters use it to track treatment-resistant depression, esketamine, TMS, and trial outcomes. It measures severity and change; it never diagnoses. This page covers how to document and interpret MADRS results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Trained clinician raters: psychiatrists, psychiatric nurse practitioners and nurses, psychologists, and research raters; no license or publisher qualification level applies, and reliability depends on training and a structured interview; the MADRS-S is completed by the patient Audience Interventional psychiatry, TMS, esketamine, and ketamine programs, prescribers managing treatment-resistant depression, payer utilization reviewers, clinical trial sponsors and monitors, referring clinicians, and quality abstractors, who will still need a PHQ-9 Typical length 4 to 8 chart lines (version and rater, interval, item profile, total with the convention named, baseline and change with the rule named, separate risk line, decision) · interview about 15 to 20 minutes with a structured guide (clinical team estimate) Format family Clinician-rated depression severity scale (10 items scored 0 to 6 on defined and intermediate steps, summed to 0 to 60, higher is worse; the MADRS-S is a separate nine-item self-report with its own total) When it's used Baseline and serial ratings in treatment-resistant depression, esketamine and TMS courses, ketamine and ECT programs, antidepressant trials, payer authorization and continuation packets, and measurement-based care where somatic symptoms confound a self-report Standards context Montgomery and Åsberg 1979 (British Journal of Psychiatry); copyright claimed by Stuart Montgomery and the Royal College of Psychiatrists; severity bands and remission cutoffs are later conventions; accepted by many payers, mandated by none, not accepted by the CMS and HEDIS depression-remission measures; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the MADRS? The Montgomery-Åsberg Depression Rating Scale (often written Montgomery-Asberg, and pronounced as the word MADRS) is a clinician-rated measure of depression severity published by Stuart A. Montgomery and Marie Åsberg in the British Journal of Psychiatry in April 1979 under the title "A new depression scale designed to be sensitive to change." The authors started from the 65-item Comprehensive Psychopathological Rating Scale, identified the 17 depressive symptoms most common in 54 English and 52 Swedish patients, and kept the ten items that changed most with antidepressant treatment and correlated best with overall change in 64 patients from four drug trials. Each item is rated 0 to 6, with descriptions at the even steps and intermediate points between them, and the ten ratings sum to a total of 0 to 60; there are no weights and no reverse-scored items. Described at the topic level, the items cover sadness as observed by the rater and as reported by the patient, inner tension, reduced sleep, reduced appetite, concentration difficulty, lassitude, reduced ability to feel, pessimistic thoughts, and suicidal thoughts. The original instructions let the rater choose the interval between ratings, weekly or otherwise, provided it is recorded, and the 1979 reliability work paired psychiatrists with general practitioners and with psychiatric nurses, so nothing in the instrument limits rating to physicians. Two derivatives are charted under their own names: the Structured Interview Guide for the MADRS, or SIGMA (Williams and Kobak, 2008), which supplies scripted probes, rates the past week, and reported a total-score intraclass correlation of 0.93 across 162 interviews by 81 rater pairs, and the MADRS-S (Svanborg and Åsberg, 1994), a separately published nine-item self-report with a recall window of about three days and its own scoring. Three facts carry the documentation. First, the MADRS measures severity and change in a depression that has already been diagnosed by interview; it is not a diagnostic instrument, and no total confirms or excludes major depressive disorder. Second, the numbers that get attached to it are conventions with dates and authors, not parts of the scale: the 1979 paper has no severity bands, the bands in general use (essentially absent up to 6, mild from 7 to 19, moderate from 20 to 34, severe from 35) are attributed to a short 1986 report by Snaith and colleagues, Müller and colleagues (2003) put the moderate-to-severe boundary at 31, response is conventionally a reduction of at least 50 percent from baseline, and published remission cutoffs run from 4 to 12, so a defensible entry names the version, the rater, the interval, the convention, the baseline, and the rule applied. Third, the suicidal thoughts item is a single screening item, so any rating above zero triggers a separate suicide risk assessment or C-SSRS , and an improving total never stands in for one. The PHQ-9 is the neighbor most often confused with it: a patient-reported screening and monitoring scale that United States quality measures count and the MADRS cannot replace, where the MADRS earns its place as the clinician-rated outcome in treatment-resistant depression, esketamine and TMS programs, and trials. ## Who uses MADRS documentation and when Interventional psychiatry is the MADRS's home: esketamine clinics, TMS programs, ketamine and ECT services, and treatment-resistant depression consults rate it at baseline and at intervals through a course, because it was the primary outcome of the landmark TMS trial (O'Reardon and colleagues, 2007, week-4 change in 301 patients) and of the esketamine development program, and because US payers list it among the scales they accept for authorization and continuation. Psychiatrists, psychiatric nurse practitioners, and trained nurses rate it in those programs; research raters rate it, usually with the SIGMA and after calibration, in antidepressant trials, where the total is the endpoint. General outpatient psychiatry reaches for it when a patient's pain, sleep disorder, or medical illness inflates the somatic items of a self-report, or when minimization is suspected and a clinician's judgment should carry the rating. In Canada and Australia it lives in rTMS services and research rather than in any payer rule. The neighbors are fixed. The PHQ-9 remains the screening and quality-measure instrument and is what many TMS clinics track with day to day; the BDI-II is the psychology self-report; a rating above zero on the suicidal thoughts item is documented in a suicide risk assessment or with the C-SSRS ; serial totals follow the conventions of the outcome measure note ; the score supports the decisions recorded in the medication management note , the psychiatric consultation note , and the treatment plan review ; and the baseline-and-current-score paragraph feeds the prior authorization request , which owns the letter mechanics while this page owns the scale-specific content those letters must carry. ## How to document MADRS results in the chart No law, payer, or professional body publishes a MADRS note format; what exists is a 1979 paper with rating instructions, a 2008 structured guide, trial protocols, and payer criteria that ask for a baseline and a recent score on an accepted scale. What survives review is an entry that names the instrument, the rater, and the interview method; states the interval rated; describes the item profile at the topic level; gives the total with its severity convention cited; reports the baseline, the change, and the response or remission rule applied; documents a separate risk assessment whenever the suicidal thoughts item is rated above zero; and records the treatment context and the decision. Each element below carries the pitfall that most often undermines it. Instrument, rater, and interview method. Write which instrument produced the number: the clinician-rated MADRS, a SIGMA-guided MADRS, or the MADRS-S self-report with its form and total range. Name the rater, their role, and their training or calibration, and the language or translation used. Clinician and self-report versions have different items, windows, and totals, and a structured interview changes reproducibility, so a reviewer, a payer, or a trial monitor needs this line before the total means anything. Pitfall: "MADRS 27" with no version, rater, or method, or a MADRS-S total submitted where the payer or protocol asked for the clinician scale. Interval rated. State the period the rating covers and the days since the previous rating. The 1979 instructions let the rater choose the interval, weekly or otherwise, provided it is recorded; the SIGMA and most trial protocols rate the past week; the MADRS-S asks about roughly the past three days. Serial totals are comparable only when the interval is the same, so say what it was rather than assuming the reader knows the convention. Pitfall: A bare total with no interval, or weekly and monthly ratings trended as if they covered the same period. Item profile at the topic level. Say which domains carry the total and which have resolved, in your own words: sadness as observed and as reported, inner tension, sleep, appetite, concentration, initiative and energy, emotional responsiveness, pessimism, and suicidal thoughts. The ten item scores live on the outcome flowsheet; the narrative describes the pattern, because a 24 carried by mood and pessimism and a 24 carried by sleep, appetite, and tension call for different decisions. Never paste anchor descriptions into the note. Pitfall: A total with no profile, or the scale's anchor text copied into the narrative as if it were the patient's account. Total and the named severity convention. Chart the total over 60 and, if a severity label is used, the convention and its source: the grade scores of Snaith and colleagues (1986) are the usual reference, Müller and colleagues (2003) place the moderate-to-severe boundary lower, and some programs adopt their own. Write the label as severity and change, never as a diagnosis, and keep the diagnosis on the line where the evaluation that made it lives. Pitfall: "MADRS 33, moderate depression" as a diagnosis, or a band presented as the scale's official interpretation. Baseline, change, and the rule applied. Give the baseline date and total on the same instrument, method, and rater type, the current total, the absolute change, the percent change, and the response or remission rule you are applying with its source (the plan's policy, the protocol, or CANMAT). State the result against that rule and keep the payer's continuation criterion separate from the remission convention; describe change from the previous visit rather than labeling it. Pitfall: "Responded" or "in remission" with no baseline, no percentage, and no rule, or a percentage computed from the wrong visit. The suicidal thoughts item and the separate assessment. Record the item rating and treat any rating above zero, any rise, or any clinical concern as the trigger for a separate, dated risk assessment (the suicide risk assessment structure or the C-SSRS), with the safety plan status and where the formulation is filed. A falling total is not evidence that risk has resolved; the item is a screen, not the assessment. Pitfall: Risk inferred from an improved total, or the single item charted as the risk assessment. Treatment context, decision, and companion records. Chart the treatment status at the time of rating (TMS session number, esketamine induction or maintenance week, medication changes since the last rating, adherence), the decision the score supports (continue, intensify, change, taper, request continuation), and what is filed separately: the REMS session record and Patient Monitoring Form for esketamine, the PHQ-9 kept for any quality measure that counts only the PHQ-9, and the authorization packet built from the baseline and current totals. Pitfall: A MADRS entered for a CMS or HEDIS remission measure that counts only the PHQ-9, or the REMS safety record folded into the outcome entry. ### Blank template (copy and adapt) MADRS DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Visit: [baseline / serial rating / continuation request / trial visit] Instrument: [clinician MADRS / SIGMA-guided MADRS / MADRS-S (form and range)] Rater: [name, role, training or calibration] Language: [ ] Interval rated: [past 7 days / other: state it] Days since last rating: [ ] Item profile (topic level): [domains carrying the total; domains resolved; item scores on the flowsheet, no anchor text] Total: [ ] of 60 Convention: [label and source, e.g. Snaith and colleagues 1986; severity and change, not a diagnosis] Baseline: [date] [ ] of 60, same method Change: [ ] points, [ ] percent Rule applied: [response or remission definition and source: payer policy, protocol, guideline] Result: [meets / does not meet] Suicidal thoughts item: [rating; any rating above zero: separate dated risk assessment, tool used, safety plan status, filed at ...] Treatment context: [TMS session / esketamine phase, REMS record separate / medication changes since last rating / adherence] Interpretation: [congruence with interview and collateral; minimizing or over-reporting noted; confounders] Decision: [continue / intensify / change / authorization packet: baseline, current, percent change, instrument, rater, interval] Other measures: [PHQ-9 for quality measures, charted separately] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale items or anchor descriptions. ## Sample MADRS documentation (fictional) Scenario: a US interventional psychiatry clinic's esketamine maintenance visit at week 12, where the psychiatric nurse practitioner rates a SIGMA-guided MADRS before the dose, compares it with the induction baseline under the plan's continuation criterion and the clinic's remission convention, documents the separate risk assessment the suicidal thoughts item triggered earlier in the course, and keeps the REMS session record and the PHQ-9 as their own documents. All details are fictional. Patient: J.M., 46 · Setting: Interventional psychiatry clinic, esketamine maintenance visit (week 12) · Clinician: L. Okonkwo, PMHNP-BC · Note date: 09/16/2026 Measure: Clinician-rated MADRS administered with the Structured Interview Guide (SIGMA) by me before today's dose, in English; I completed the clinic's rater training and calibration exercise before the induction phase. Rating covers the past seven days; 7 days since the 09/09/2026 rating. All ten items rated; item scores are on the outcome flowsheet. Total 14 of 60. Induction baseline 08/04/2026, same method and rater: 34 of 60. End of induction 08/28/2026: 19 of 60. Prior rating 09/09/2026: 17 of 60. Item profile: The total is carried by reduced emotional responsiveness, low initiative, and concentration difficulty. Observed and reported sadness, inner tension, pessimism, sleep, and appetite have improved to low ratings since induction. The suicidal thoughts item is rated 0 this week; it was rated above zero at induction and at the 09/09/2026 visit. Total and convention: 14 of 60. Under the grade scores of Snaith and colleagues (1986), the convention this clinic cites, 14 falls in the mild range; at induction, 34 fell at the top of the moderate range under the same convention and above the moderate-to-severe boundary that Müller and colleagues (2003) proposed, which is why the source is named. The MADRS measures severity and change and does not establish a diagnosis; the diagnosis of recurrent major depressive disorder, treatment-resistant, rests on the intake psychiatric evaluation and is unchanged. Change and rule applied: Absolute change from the induction baseline: 20 points; percent change: 58.8 percent reduction (34 to 14) on the same instrument, method, and rater. This meets the plan's continuation criterion, a reduction of at least 50 percent from baseline on a standardized depression rating scale with a score obtained within the past month. Remission under the convention this clinic has adopted from CANMAT, a MADRS total of 10 or less, has not been reached, so J.M. is charted as a responder and not in remission. Change from 09/09/2026: 3 points, described rather than labeled. Risk: The suicidal thoughts item is a single screening item, not a risk assessment. Because it was rated above zero at induction and on 09/09/2026, a separate risk assessment was completed at each of those visits and was repeated today: C-SSRS screener negative for the past month, no ideation reported since 09/09/2026, no suicidal behavior during the course, protective factors unchanged (spouse, work, treatment engagement), safety plan from 08/04/2026 reviewed with J.M. and unchanged. Formulation and plan are filed in the risk section of the record. Treatment context: Esketamine nasal spray, maintenance phase, week 12, dose and interval unchanged per the treatment plan; concurrent oral antidepressant unchanged since before induction; no other medication changes; adherence confirmed; alcohol and sleep reviewed and unchanged. Today's session safety documentation (observation start and end times, blood pressure at the intervals the program requires, sedation and dissociation checks, readiness for discharge) is on the REMS session record, and the Patient Monitoring Form will be submitted within the program's window; that record is separate from this outcome entry and does not depend on the MADRS. Decision and record: Continue maintenance esketamine at the current interval. Continuation authorization request submitted today with the instrument and version, the rater, the interval rated, the induction baseline (34 on 08/04/2026), the current total (14 on 09/16/2026), and the percent change. Next MADRS at the 09/30/2026 visit, same method. The clinic's PHQ-9 was also collected today for quality-measure reporting and is charted separately; it is not used to compute the MADRS change. J.M. reviewed the scores and agrees with the plan. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No scale items or anchor text are reproduced. ↑ Back to the template and downloads ### Why this sample works The instrument, the structured guide, the rater's training, and the interval are named, so the total can be reproduced and compared with the baseline on the same method. The profile says which domains carry the total and which have resolved, at the topic level and without copying anchor text, so the number is clinically readable. The severity label is tied to a named convention and written as severity, and the diagnosis is left with the evaluation that made it. Baseline, absolute and percent change, and the exact rule applied are all in the entry, so the payer's criterion and the remission convention are auditable and never conflated. The suicidal thoughts item triggers a separate, dated risk assessment, and the REMS session record and the PHQ-9 are kept as their own documents, so no reader mistakes one record for another. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the number is a convention; the processes around it are real. No federal or state LAW names the MADRS or prescribes a scoring note. The esketamine REMS is LAW in effect: under the FDA-required program, esketamine is dispensed and administered only in healthcare settings enrolled in the program, outpatients are enrolled, every patient is monitored for at least two hours after each dose with blood pressure reassessed at about 40 minutes, and the program's Patient Monitoring Form is completed after every outpatient session and submitted within seven days; the current prescribing information (2026) covers treatment-resistant depression as monotherapy or with an oral antidepressant, and none of it requires a MADRS, which is why the safety record and the outcome entry are separate documents. PAYER POLICY is where the MADRS appears, as one accepted option and with rules that differ by plan. UnitedHealthcare's commercial esketamine criteria ( effective July 1, 2026 ) require a documented baseline on at least one of the BDI, HAM-D, MADRS, PHQ-9, or QIDS and, for reauthorization, a baseline and a score from within the last month showing remission or a positive clinical response, with no MADRS number set in the policy; Carelon's clinical criteria for esketamine (published April 1, 2025) list seven standard rating scales (BDI, GDS, HAM-D, IDS-SR, MADRS, PHQ-9, QIDS), define the inadequate prior response as less than a 50 percent reduction on one of them, and require at least a 50 percent reduction from baseline for continuation; Aetna's Massachusetts TMS request form (version 12-25) accepts the GDS, PHQ-9, BDI, HAM-D, MADRS, QIDS, or IDS-SR before and after treatment, asks for greater than 50 percent improvement for retreatment, and offers a MADRS below 10 as a typical remission definition, a payer-form convention rather than a property of the scale. Medicare coverage of TMS runs through Medicare Administrative Contractor local coverage determinations (Novitas L34998, for example), each with its own documentation expectations for severity and response; check the current LCD and billing article for your contractor rather than assuming a MADRS is required or sufficient. QUALITY PROGRAMS are the trap: CMS Quality ID 370, Depression Remission at Twelve Months (2026 specification), is built on an initial PHQ-9 or PHQ-9M score greater than nine and remission as a score below five at twelve months, and the NCQA HEDIS measure Depression Remission or Response for Adolescents and Adults (DRR-E) defines response as a PHQ-9 reduction of 50 percent or more and remission as a PHQ-9 below five within four to eight months of the elevated score, so a clinic can satisfy an esketamine or TMS authorization with a MADRS and still owe a PHQ-9 for the measure. CONVENTION supplies the rest: trial statistical analysis plans define response as a reduction of at least 50 percent from baseline and remission at a protocol-specific total, and the same protocols specify rater training, the structured guide, and how missing items are handled, none of which binds a clinic unless it adopts them and says so in the chart. Canada and Australia set process, not a number, and the evidence explains why the chart must name its conventions. In Canada no federal or provincial LAW names the MADRS. Ontario Health's provincial rTMS program (PAYER POLICY) funds rTMS for adults 18 and older with treatment-resistant unipolar depression, takes referrals from a physician, nurse practitioner, or psychiatrist, determines eligibility by clinical assessment, and describes performance measures and data collection without naming a rating scale or a MADRS threshold; public drug plans largely do not cover esketamine, after the Canadian Drug Expert Committee recommended in December 2020 that it not be reimbursed for major depressive disorder and British Columbia listed it as a non-benefit effective August 3, 2021, so US-style continuation criteria do not transfer. CANMAT's 2023 guideline update (CONVENTION) recommends measurement-based care, lists the MADRS among clinician-rated scales and the PHQ-9 among patient-rated ones, defines early improvement as a reduction of at least 20 percent within two to four weeks of starting treatment and response as at least 50 percent from baseline, gives a MADRS total of 10 or less as its example of symptom remission (with 7 or less on the HAM-D and 4 or less on the PHQ-9), and treats a reduction under 20 percent at two to four weeks as a trigger to consider dose optimization or a switch. In Australia, rTMS has been on the Medicare Benefits Schedule since 1 November 2021 (LAW, through regulation; bare item numbers): 14216 and 14217 cover initial prescription and mapping and a course of up to 35 sessions for adults with major depressive disorder who have not improved on at least two classes of antidepressant and have had no previous rTMS, prescription and mapping are performed personally by a psychiatrist trained in rTMS, and the retreatment items 14219 and 14220 (up to 15 sessions) require a satisfactory response to the initial course, assessed with a validated major depressive disorder tool at least four months after the initial service, followed by relapse; no item names the MADRS. The RANZCP's Professional Practice Guideline 16 on rTMS administration and its position statement 79 (both November 2018; CONVENTION) call for regular monitoring of response and side effects without prescribing a scale. On the evidence: the 1979 paper reported inter-rater correlations of 0.89 to 0.97 between psychiatrists and general practitioners or psychiatric nurses and better separation of responders from non-responders than the comparison measure, and it contains no severity bands; the bands in general use are attributed to a three-page report by Snaith, Harrop, Newby, and Teale (1986); Müller and colleagues (2003) found that a total of 31 best separated moderate from severe depression against the Hamilton scale and the CGI in 85 patients (sensitivity 93.5 percent, specificity 83.3 percent); Riedel and colleagues (2010) found in 846 inpatients that a total of 7 or less best matched CGI-defined remission and a reduction of about 46 percent best matched CGI-defined response; Hawley and colleagues (2002) proposed remission below 9 or 10, Zimmerman and colleagues (2004) proposed 4 or less as a narrow and 9 or less as a broad definition, and the SUSTAIN-1 esketamine trial defined stable remission as 12 or less in at least three of the last four weeks of its optimization phase. Williams and Kobak (2008) reported a total-score intraclass correlation of 0.93 across 162 interviews by 81 rater pairs with the SIGMA; Carmody and colleagues (2006) found the scale unifactorial in 233 and 985 outpatients, with alphas of 0.90 and 0.92 and about twice the measurement precision of the 17-item Hamilton scale at average severity; Quilty and colleagues (2013) fit four symptom factors in 821 outpatients; Turkoz and colleagues (2021) linked a 6-point change to one CGI-S category and a 12-point change to two in 565 patients with treatment-resistant depression; and Leucht and colleagues (2017) linked a total below 8 to a CGI-S rating of borderline ill and a 48 to 57 percent reduction to a rating of much improved. The MADRS-S (Svanborg and Åsberg, 1994) has nine items, asks about the past three days, correlated 0.87 with the BDI in a 2001 comparison, and is published with 0 to 3 half-step item scoring and a 0 to 27 total (Fantino and Moore, 2009) while versions with 0 to 6 items and a 0 to 54 total also circulate, so the exact form belongs in the chart, and dozens of translations are catalogued by ePROVIDE, none of which proves equivalence, so the language and version are charted too. The suicidal thoughts item deserves its own rule. It is one screening item on a severity scale, rated on the same 0 to 6 steps as the others, and it is not a suicide risk assessment: a rating above zero, a rise from the previous rating, or any clinical concern prompts a separate, dated assessment of ideation, intent, plans, access to means, prior behavior, protective factors, and disposition, documented on the suicide risk assessment or C-SSRS structure and filed in the risk section, and a falling total never implies that risk has resolved. The esketamine prescribing information states that the drug's effectiveness in preventing suicide or in reducing suicidal ideation or behavior has not been demonstrated, one more reason to keep the risk record separate from the outcome entry. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). The MADRS is copyrighted by Stuart Montgomery and the Royal College of Psychiatrists, as recorded on ePROVIDE, and is made available for an individual clinician's or researcher's own use, copied in full with the full citation, with distribution to others or republication requiring the College's permission. The 1979 article is copyright Royal College of Psychiatrists and is published in the British Journal of Psychiatry by Cambridge University Press, the copy hosted by the American Psychological Association carries a 1978 Stuart Montgomery copyright line, and the SIGMA (Williams and Kobak, 2008) is copyright Royal College of Psychiatrists. The permission statement printed with the scale allows an individual clinician or researcher to copy it for their own use without asking, in full and acknowledging the source, and requires permission from the Royal College of Psychiatrists for copying and distribution to others or for republication in print, online, or by any other means; the College routes permission requests, including rating-scale requests, through the Request Permission link on the article's Cambridge Core page and may charge a fee. An EHR build that reproduces the item and anchor text, a printed pack handed to other clinicians, or a public web calculator that displays the scale is a distribution or republication question for the rights holders, and no blanket license for those uses had been published as of September 2026; recording independently written item scores, a total, and a narrative, as the template on this page does, is a different matter. No trademark is claimed for the name. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common MADRS documentation errors reviewers flag The numbers behind these errors are specific. The 1979 paper contains no severity bands; the bands in general use trace to a three-page 1986 report by Snaith and colleagues, and the moderate-to-severe boundary moved from 35 to 31 when Müller and colleagues tested it against the Hamilton scale and the CGI in 85 patients. In 846 inpatients the values that best matched clinician-judged remission and response were 7 or less and a reduction of about 46 percent (Riedel and colleagues, 2010), not 10 and 50; published remission cutoffs run from 4 (Zimmerman and colleagues, 2004) to 12 (the SUSTAIN-1 esketamine trial). With the structured guide the total-score intraclass correlation reached 0.93 across 81 rater pairs (Williams and Kobak, 2008); a 6-point change corresponded to one CGI-S category in 565 patients with treatment-resistant depression (Turkoz and colleagues, 2021); and the CMS and HEDIS remission measures count only the PHQ-9. No published audit counts how often the version, rater, interval, or baseline is missing from a MADRS chart entry; the payer criteria described under compliance considerations are what turn those omissions into a denial. The BastionGPT Clinical Advisory Board sees the same errors most often in MADRS documentation reviews: A total with no version, rater, or interval. "MADRS 27" cannot be reproduced: the clinician MADRS, a SIGMA-guided MADRS, and the MADRS-S are different instruments with different totals and recall windows, and the original instructions require the interval to be recorded. Write the instrument, the guide if one was used, the rater's name, role, and training, and the period rated, every time. Bands laundered into diagnoses or official cutoffs. "MADRS 33, moderate depression" charted as a diagnosis, or the 1986 grade scores presented as the scale's own interpretation. Name the convention and its source, write the label as severity, and let the diagnosis rest on the evaluation that made it; a total never confirms or excludes a disorder. Response or remission claimed without the arithmetic. "Responded" with no baseline, no percentage, and no rule, or a percentage computed from the wrong visit or a different instrument. Chart the baseline date and total on the same method, the current total, the absolute and percent change, and the rule with its source, and keep the payer's continuation criterion and the remission convention apart. The self-report substituted for the clinician scale. A MADRS-S total submitted where a payer, protocol, or referrer asked for the MADRS, or trended against clinician ratings. The MADRS-S has nine items, a window of about three days, and its own scoring, and none of the payer policies reviewed for this page names it. Label it MADRS-S, chart its form and total range, and keep the two series apart. The suicidal thoughts item treated as the risk assessment. A rating above zero left without a separate assessment, or "no safety concerns" inferred from an improved total. The item is a screen; document the dated risk assessment, the tool used, protective factors, the safety plan status, and where the formulation is filed, and repeat it whenever the rating rises. The wrong document for the job. A MADRS entered for a quality measure that counts only the PHQ-9, or the REMS session record and the outcome entry folded into one paragraph. Keep the PHQ-9 for CMS Quality ID 370 and HEDIS DRR-E, keep the two-hour monitoring record and the Patient Monitoring Form as the safety record, and let the MADRS entry carry the outcome and the authorization payload. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric progress notes and outcome-measure documentation. Give it the facts (instrument and guide, rater and training, interval, item scores or the profile, total, baseline date and total, the rule your payer or protocol applies, the suicidal-thoughts item rating with the separate assessment, treatment status, and the decision) and it drafts the documentation block: version and interval, the profile at the topic level, the total with the convention named, the change with the rule applied, the separate risk line, and the decision, ready for your review. Cross-check a finished note for the gaps reviewers flag: a total with no version or interval, a band charted as a diagnosis, a response claim with no baseline or rule, a MADRS-S mixed into a clinician series, a suicidal-thoughts rating with no separate assessment, or a MADRS offered where the measure counts only the PHQ-9. Turn the entry into the next document: the baseline-and-current-score paragraph for a prior authorization or continuation request, the outcome-measure entry for a serial record, or the plain-language summary a patient or referrer expects, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do MADRS scores mean, and what are the severity bands? A MADRS total is the sum of ten clinician ratings, each on a 0 to 6 scale with descriptions at the even steps and intermediate points between them, so the total runs from 0 to 60 and higher means more severe. The original 1979 paper contains no severity bands. The bands in general use, essentially absent or recovered up to 6, mild from 7 to 19, moderate from 20 to 34, and severe from 35 to 60, are attributed to a short 1986 report by Snaith and colleagues, and later work disagrees at the boundaries: Müller and colleagues (2003) found that a total of 31 best separated moderate from severe depression against the Hamilton scale and the CGI, and some tools splice several papers into five bands. So a chart entry reads "total 24 of 60; moderate under the Snaith and colleagues (1986) grade scores" rather than "MADRS 24 = moderate," names the source whenever a label is used, and never turns the label into a diagnosis: the MADRS measures severity and change in someone whose depression has already been diagnosed by interview. The item profile matters as much as the total, because the same 24 can be carried by mood and pessimism or by sleep, appetite, and tension. What counts as response and remission on the MADRS? Whatever rule you name, and you must name one, because neither is built into the scale. Response is conventionally a reduction of at least 50 percent from the baseline total; trial statistical analysis plans state it that way, CANMAT's 2023 update uses it, and several payers copy it. Remission has no consensus number: a total of 10 or less is the most common operational definition and CANMAT's example, Hawley and colleagues (2002) proposed below 9 or 10, Zimmerman and colleagues (2004) proposed 4 or less as a narrow and 9 or less as a broad definition, the SUSTAIN-1 esketamine trial required 12 or less in at least three of the last four weeks of its optimization phase, and Riedel and colleagues (2010) found in 846 inpatients that 7 or less best matched clinician-judged remission and that a reduction of about 46 percent best matched clinician-judged response. Chart the baseline date and total on the same instrument and method, the current total, the absolute and percent change, and the rule with its source, and keep the payer's criterion and your clinic's remission convention apart: a patient can meet a plan's continuation criterion and still not be in remission under the convention you cite. Change from the previous rating, as opposed to baseline, is described rather than labeled. Does insurance require a MADRS for esketamine or TMS, and is a 50 percent reduction the rule? No US payer reviewed for this page requires the MADRS by name; policies list it as one accepted instrument, and their rules differ (PAYER POLICY, not law). UnitedHealthcare's commercial esketamine criteria (effective July 1, 2026) require a documented baseline on at least one of the BDI, HAM-D, MADRS, PHQ-9, or QIDS and, for reauthorization, a baseline and a score from within the last month showing remission or a positive clinical response, with no numerical MADRS threshold in the policy. Carelon's esketamine criteria (published April 1, 2025) list seven scales, the MADRS among them, and require at least a 50 percent reduction from baseline for continuation. Aetna's Massachusetts TMS request form (version 12-25) accepts the GDS, PHQ-9, BDI, HAM-D, MADRS, QIDS, or IDS-SR before and after treatment and asks for greater than 50 percent improvement for retreatment, while offering a MADRS below 10 as a typical remission reference. Medicare TMS coverage runs through local coverage determinations with their own documentation expectations; check your contractor's current LCD. So the defensible sentence is "meets the plan's stated continuation criterion," with the baseline, the current total, and the arithmetic in the note, not "MADRS response is 50 percent." One trap: the CMS Depression Remission at Twelve Months measure (Quality ID 370) and the HEDIS DRR-E measure count only the PHQ-9 and PHQ-9M, so a MADRS that satisfies the authorization earns nothing there. The letter mechanics belong to the prior authorization request page; this page owns the scale-specific payload. MADRS, PHQ-9, or HAM-D: which one, when? Choose the MADRS when a clinician-rated severity and change measure is wanted: interventional psychiatry, treatment-resistant depression, trials, a payer packet that names it, or a patient whose pain, sleep disorder, or medical illness inflates the somatic items of a self-report, because the MADRS weights mood, anhedonia, and cognitive symptoms more heavily, and Carmody and colleagues (2006) found it unifactorial with about twice the measurement precision of the 17-item Hamilton scale at average severity. Choose the PHQ-9 for screening, primary care and collaborative care monitoring, and anything reported to CMS Quality ID 370 or HEDIS DRR-E, which count only the PHQ-9; it is faster, patient-reported, and mapped to DSM symptom criteria, and many TMS clinics track with it day to day. The HAM-D is the historical comparator with broader anxiety and somatic content, still named in many payer lists and older protocols. Whichever you choose, keep the same instrument and method across a course of treatment and, when a payer or trial names an instrument, use that one; a series that switches scales midway cannot show change. No validated conversion between the MADRS and the PHQ-9 is established for North American practice, so do not convert one into the other in the chart; the BDI-II page covers the other common self-report. Who can rate the MADRS, and do I need the SIGMA or rater training? Any clinician trained in the interview and the scale: one aim of the 1979 paper was an instrument usable by raters without specialist psychiatric training, its reliability work paired psychiatrists with general practitioners and with psychiatric nurses (inter-rater correlations of 0.89 to 0.97), and nothing in the instrument limits rating to physicians, so a psychiatric nurse practitioner or nurse in a TMS or esketamine program can rate it; whether a payer, employer, or protocol requires a particular credential is a separate question. The scale was published without standard interview questions, a known source of rater variability; the SIGMA (Williams and Kobak, 2008) supplies scripted probes and a past-week frame and reported a total-score intraclass correlation of 0.93 across 162 interviews by 81 rater pairs. Formal rater training, calibration, and independent raters are trial-protocol requirements, not properties of the scale; in a clinic the defensible minimum is a rater who knows the procedure, uses the same method every time, and records it. Chart the rater's name and role, whether the SIGMA was used, and any training or calibration, so a reviewer can tell a structured rating from an unstructured one and a clinician rating from a self-report. Is the MADRS-S the same as the MADRS, and can I submit it when a payer asks for a MADRS? No and no. The MADRS-S (Svanborg and Åsberg, 1994) is a separately published patient-completed scale with nine items, a recall window of about three days, and its own scoring: the form validated by Fantino and Moore (2009) scores each item 0 to 3 with half-steps for a 0 to 27 total, while versions scoring 0 to 6 per item for a 0 to 54 total also circulate, so the form and its total range belong in the chart. Self-ratings and clinician ratings agree well in group data, but the instruments differ in rater, items, window, and total, and none of the payer policies reviewed for this page names the MADRS-S; they name the MADRS and, separately, self-report scales such as the PHQ-9, BDI, or QIDS. Submit the instrument the payer or protocol named, label a self-report MADRS-S every time, never trend MADRS-S totals against clinician MADRS totals, and, when minimization is suspected, prefer the clinician scale, whose instructions allow the rater to draw on observed behavior and other clinical information when a definite answer cannot be obtained. The suicidal thoughts item is rated above zero. What do I document? A separate, dated risk assessment, every time. The item is one screening question on a severity scale and is not a risk assessment: document the item rating, then the assessment of ideation, intent, plans, access to means, prior behavior, protective factors, and disposition on the suicide risk assessment structure or with the C-SSRS , the safety plan status, and where the formulation is filed. Repeat it whenever the rating rises, and do not read a falling total as resolved risk; the esketamine prescribing information states that the drug's effectiveness in preventing suicide or in reducing suicidal ideation or behavior has not been demonstrated. Keep the risk record in the risk section rather than inside the outcome entry, so a payer packet built from the MADRS series is never the only place the assessment lives. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). What do the MADRS items ask, and can we put the scale in our EHR, on a form, or in a free web calculator? At the topic level the ten items cover sadness as observed by the rater and as reported by the patient, inner tension, reduced sleep, reduced appetite, concentration difficulty, lassitude, reduced ability to feel, pessimistic thoughts, and suicidal thoughts, each on a 0 to 6 scale; this page does not print or paraphrase the anchor descriptions, and the scale itself is the source for rating. Rights: the ePROVIDE record lists the copyright as claimed by Stuart Montgomery and the Royal College of Psychiatrists, the 1979 article is the College's copyright, and the permission statement printed with the scale allows an individual clinician or researcher to copy it in full, with the full citation, for their own use, while copying for distribution to others or republication in print, online, or by any other means requires the College's permission, which it handles through the Request Permission link on the article's Cambridge Core page and may charge for. An EHR template that reproduces the item and anchor text, a printed pack for other clinicians, and a public web calculator that displays the scale are distribution or republication questions for the rights holders; no blanket license for them had been published as of September 2026, and the free copies online are not evidence of one. Recording your own item scores, the total, and a narrative in the chart, as this page's template does, reproduces nothing. Can BastionGPT draft my MADRS documentation? Yes. Give it the facts (instrument and guide, rater and training, interval, item scores or the profile, total, baseline date and total, the rule your payer or protocol applies, the suicidal-thoughts item rating with the separate assessment, treatment status, and the decision) and it drafts the full entry: version and interval, the profile at the topic level, the total with the convention named, the change with the rule applied, the separate risk line, the decision, and the baseline-and-current-score paragraph for a continuation request, ready for your review. It can also check a finished note for a total with no version or interval, a band charted as a diagnosis, a response claim with no baseline or rule, a MADRS-S mixed into a clinician series, a suicidal-thoughts rating with no separate assessment, or a MADRS offered where the measure counts only the PHQ-9. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Montgomery SA, Åsberg M, 1979, British Journal of Psychiatry 134(4):382-389, a new depression scale designed to be sensitive to change (Cambridge University Press for the Royal College of Psychiatrists; copyright Royal College of Psychiatrists 1979); Williams JBW, Kobak KA, 2008, British Journal of Psychiatry 192(1):52-58, development and reliability of a structured interview guide for the MADRS (SIGMA) (162 interviews, 81 rater pairs, total-score intraclass correlation 0.93); Mapi Research Trust, ePROVIDE records for the MADRS and the SIGMA (copyright claimed by Stuart Montgomery and the Royal College of Psychiatrists; translations); Royal College of Psychiatrists, College copyright and permissions to use content (rating-scale permission requests through the Cambridge Core Request Permission route; fees may apply); American Psychological Association, depression assessment instruments (hosted copy of the scale with its rating instructions and a 1978 Stuart Montgomery copyright line; not reproduced here). Snaith RP, Harrop FM, Newby DA, Teale C, 1986, British Journal of Psychiatry 148:599-601, grade scores of the Montgomery-Åsberg Depression and the Clinical Anxiety Scales (the source of the bands in general use); Müller MJ, Himmerich H, Kienzle B, Szegedi A, 2003, Journal of Affective Disorders 77(3):255-260, differentiating moderate and severe depression using the MADRS (85 patients; boundary at 31); Riedel M and colleagues, 2010, Journal of Psychiatric Research 44(15):1063-1068, response and remission criteria in major depression: a validation of current practice (846 inpatients); Hawley CJ, Gale TM, Sivakumaran T, 2002, Journal of Affective Disorders 72(2):177-184, defining remission by cut off score on the MADRS ; Zimmerman M, Posternak MA, Chelminski I, 2004, Journal of Clinical Psychiatry 65(2):163-168, defining remission on the Montgomery-Åsberg Depression Rating Scale . Carmody TJ and colleagues, 2006, European Neuropsychopharmacology 16(8):601-611, the Montgomery-Åsberg and the Hamilton ratings of depression: a comparison of measures (unifactorial; about twice the precision of the HRSD-17); Quilty LC and colleagues, 2013, International Journal of Methods in Psychiatric Research 22(3):175-184, the structure of the MADRS over the course of treatment (821 outpatients; four factors); Leucht S and colleagues, 2017, Journal of Affective Disorders 210:287-293, what does the MADRS mean? Equipercentile linking with the CGI ; Turkoz I and colleagues, 2021, Acta Psychiatrica Scandinavica 143(3):253-263, clinically meaningful changes on depressive symptom measures in treatment-resistant depression (565 patients; 6 and 12 points). Svanborg P, Åsberg M, 1994, Acta Psychiatrica Scandinavica 89(1):21-28, a new self-rating scale for depression and anxiety states based on the CPRS , and 2001, Journal of Affective Disorders 64(2-3):203-216, a comparison between the BDI and the self-rating version of the MADRS (r = 0.869); Fantino B, Moore N, 2009, BMC Psychiatry 9:26, the self-reported MADRS is a useful evaluative tool in major depressive disorder (nine items; 0 to 3 with half-steps; past three days). O'Reardon JP and colleagues, 2007, Biological Psychiatry 62(11):1208-1216, efficacy and safety of TMS in the acute treatment of major depression (301 patients; MADRS change at week 4 as primary outcome); Daly EJ and colleagues, 2019, JAMA Psychiatry 76(9):893-903, esketamine nasal spray plus oral antidepressant for relapse prevention (SUSTAIN-1) , with the stable-remission and stable-response definitions in the ClinicalTrials.gov record NCT02493868 ; US Food and Drug Administration, Drugs@FDA, esketamine (NDA 211243) (current prescribing information: monotherapy or with an oral antidepressant; two-hour monitoring; blood pressure at about 40 minutes; REMS enrollment; suicide-prevention effectiveness not demonstrated); Spravato REMS program, program site and Patient Monitoring Form (completed after every outpatient session, submitted within seven days). UnitedHealthcare, Spravato prior authorization and medical necessity criteria (program 2026 P 2168-11, effective July 1, 2026; BDI, HAMD, MADRS, PHQ-9, QIDS; baseline and recent score within the last month); Carelon, Spravato medical drug clinical criteria CC-0086 (published April 1, 2025; at least 50 percent reduction from baseline for continuation); Aetna, Massachusetts repetitive TMS request form (GR-69757, 12-25; accepted scales; greater than 50 percent improvement for retreatment; remission reference values); CMS Medicare Coverage Database, LCD L34998, transcranial magnetic stimulation in adults with major depressive disorder ; CMS Quality Payment Program, 2026 Quality ID 370, Depression Remission at Twelve Months (PHQ-9 or PHQ-9M greater than nine; remission below five); NCQA, Depression Remission or Response for Adolescents and Adults (DRR-E) (PHQ-9; response 50 percent or more; remission below five). Lam RW, Kennedy SH, Adams C, and colleagues, 2024, Canadian Journal of Psychiatry 69(9):641-687, CANMAT 2023 update on clinical guidelines for management of major depressive disorder in adults , with the measurement-based-care definitions in CANMAT's question 5 resource (MADRS among clinician-rated scales; early improvement at least 20 percent; response at least 50 percent; MADRS 10 or less as the remission example); Ontario Health, neurostimulation procedures provincial program (publicly funded rTMS for adults with treatment-resistant depression; no scale named); CADTH Canadian Drug Expert Committee, esketamine (Spravato) recommendation (final recommendation December 2020: not to be reimbursed); British Columbia PharmaCare, esketamine drug coverage decision (non-benefit, August 3, 2021). MBS Online, item 14217 and item 14219 (rTMS items from 1 November 2021; explanatory note TN.1.28; retreatment response assessed with a validated major depressive disorder tool; bare item numbers only) and the rTMS factsheet ; RANZCP, Professional Practice Guideline 16: administration of rTMS (November 2018) and position statement 79 on rTMS ; Hamilton M, 1960, Journal of Neurology, Neurosurgery and Psychiatry 23:56-62, a rating scale for depression ; Kroenke K, Spitzer RL, Williams JB, 2001, the PHQ-9 (the quality-measure comparator). ## Related templates PHQ-9 Score Documentation Prior Authorization Request Outcome Measure Review Note Medication Management Follow-Up Note Suicide Risk Assessment C-SSRS Documentation Treatment Plan Review & Update Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/whodas All note templates # WHODAS 2.0 Documentation: Versions, Scoring & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The WHODAS 2.0 (World Health Organization Disability Assessment Schedule 2.0) is WHO's ICF-based measure of functioning and disability over the past 30 days, in 36-item and 12-item versions, self-, proxy-, or interviewer-administered. Psychiatrists, psychologists, and disability evaluators use it as the DSM-5 Section III functioning measure and as evidence in SSA, VA, 1915(i), and NDIS files. This page covers how to document WHODAS 2.0 results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Adult self-report, a proxy informant, or an interviewer-administered version; scored and interpreted by psychiatrists, psychologists, occupational therapists, social workers, and other clinicians; no publisher qualification level, WHO asks users to register, and some programs require administration training Audience Psychiatrists and psychologists, treatment teams reviewing plans, SSA and VA adjudicators, disability and workers' compensation evaluators, NDIS planners and delegates, state Medicaid 1915(i) reviewers, MIPS abstractors, payers and auditors Typical length 4 to 10 chart lines (version, mode, recall window, scoring method and coding, total, domain profile, percentile, linkage) · completion about 5 minutes for the 12-item and about 20 minutes for the 36-item version Format family Self-, proxy-, or interviewer-administered functioning and disability measure (36 or 12 items across six ICF domains, five-level difficulty scale, 30-day recall; simple sum or WHO complex scoring on a 0 to 100 metric) When it's used Psychiatric diagnostic evaluations, treatment plan reviews and outcome measurement, disability and workers' compensation evaluations, NDIS functional capacity and progress reports, 1915(i) needs-based eligibility, fitness-for-duty and capacity evaluations Standards context World Health Organization (manual 2010, Üstün and colleagues); WHO copyright, free clinical use with registration, licensing required for electronic or EHR use; DSM-5 Section III emerging measure; required by no federal law; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the WHODAS 2.0? The World Health Organization Disability Assessment Schedule 2.0 is WHO's generic measure of functioning and disability, built from the activity and participation components of the International Classification of Functioning, Disability and Health (ICF) through the WHO/NIH Joint Project, tested in 19 countries with more than 65,000 respondents, and published in the manual Measuring Health and Disability (Üstün, Kostanjsek, Chatterji, and Rehm, editors; WHO, 2010). It asks how much difficulty a person had, because of a health condition of any kind, over the past 30 days in six domains: cognition (understanding and communicating), mobility (moving and getting around), self-care (hygiene, dressing, eating, being alone), getting along (interacting with other people), life activities (household responsibilities and, where applicable, work or school), and participation (joining in community activities and the wider effects of the condition). Each item is rated on five levels from no difficulty to extreme difficulty or cannot do. The manual holds seven versions: a 36-item form that yields an overall score and six domain scores (about 20 minutes by interview) and a 12-item form for brief overall assessment (about 5 minutes; it explains 81 percent of the variance of the 36-item form), each in self-administered, interviewer-administered, and proxy-administered versions, plus a 12+24 interviewer hybrid that screens with the short set and adds the fuller items where difficulty is reported. WHO describes WHODAS 2.0 as an altogether different instrument from the earlier WHODAS II, which it supersedes. The load-bearing fact for documentation is that a WHODAS number has no identity of its own. WHO recognizes two scoring methods: simple scoring, an unweighted sum of the item responses, which the manual calls sufficient for busy clinical settings, and complex scoring, an item response theory method run through WHO's own template that weights items and levels and converts the result to a 0 to 100 metric, 0 meaning no disability and 100 full disability. The two are highly correlated (0.99 in a Swedish psychiatric sample) but produce different numbers. WHO's current templates also code the five responses 0 to 4 where the older manual text and the DSM-5 reproduction coded them 1 to 5, and WHO warns that the words did not change but the numbers did, so a raw sum is ambiguous until the coding is stated. And WHO publishes general-population norms and percentiles, not clinical cutoffs: none, mild, moderate, severe, and extreme are item-level response labels, not bands for the total, and CMS's 2026 specification for MIPS measure 502 states, "There is no recommended cutoff score." DSM-5 (2013) dropped the GAF and placed the 36-item self-administered WHODAS 2.0 in Section III as an emerging measure for further study, not as a mandated replacement, and DSM-5-TR (2022) kept it there; the retired scale it stands beside is covered on the GAF score page , and no conversion between the two exists. The diagnostic frame around the entry belongs to the psychiatric diagnostic evaluation ; the goals it informs belong to the treatment plan and its reviews . ## Who uses WHODAS 2.0 documentation and when Psychiatrists, psychiatric nurse practitioners, and psychologists chart it in a psychiatric diagnostic evaluation as the DSM-5 Section III functioning measure, then repeat it at treatment plan reviews as an outcome measure under the conventions of the outcome measure note ; US clinicians reporting MIPS measure 502 use the 12-item version for exactly that. Occupational therapists, psychologists, and social workers in Australia embed the 36-item version in an NDIS functional capacity assessment or an NDIS progress report , beside the LSP-16 and HoNOS. Disability, workers' compensation, and forensic evaluators cite it as corroborating self-report in a medical source statement , a workers' compensation report , or a fitness-for-duty evaluation ; state Medicaid 1915(i) programs such as North Dakota's and Michigan's administer the 36-item version for needs-based eligibility; VA clinicians and researchers use it in general mental health clinics. Neighbors win in three places: the GAF when a legacy form demands an Axis V number (entered beside, never converted from, the WHODAS), the HoNOS or LSP-16 when an Australian service needs its clinician-rated national measures, and a direct observational assessment of daily tasks when a decision turns on performance rather than reported difficulty. The mental status exam carries the observed symptom picture the WHODAS does not. ## How to document WHODAS 2.0 results in the chart No law, payer, or WHO rule prescribes a WHODAS 2.0 note format; WHO publishes the instrument, its templates, and population norms, and the systems that use the score set their own version, method, and threshold rules. What survives review is an entry that names the version and how the work items were handled, the respondent and mode, the recall window and completeness, the scoring method and response coding, the domain profile, an interpretation that uses percentiles rather than a WHO-attributed severity band and ties the pattern to diagnosis and plan, and the decision context and repeat plan. Each element below carries the pitfall that most often undermines it. Version, item count, and work items. Write the exact configuration: "WHODAS 2.0, 36-item, self-administered version" or "12-item, interviewer-administered," and the language and translation (WHO distributes partner translations without warranting their quality, so name the one used). State whether the work or school items applied: for a respondent not in paid work or school, the manual's pathway omits those items and scores the remaining 32, and the note should say "32-item scoring, respondent not working" rather than leaving the work items at no difficulty. If the 12+24 hybrid was used, name it; if a program mandates a configuration (North Dakota's 1915(i) benefit requires the 36-item interviewer version), cite the rule. Pitfall: "WHODAS 45" with no version, a 12-item total trended against a 36-item baseline, or an unemployed respondent's work items scored as no difficulty, which erases the vocational limitation the evaluation exists to describe. Respondent and administration mode. Record who completed it and how: self-administered on paper or a licensed electronic form, interviewer-administered in person or by telephone, or proxy-administered, with the proxy's relationship and the reason (impaired capacity, communication, or the program's request). A proxy result is a third party's view of the person's functioning, and agreement with self-report is better for observable domains such as mobility and self-care than for cognition and getting along, so chart it as a proxy report and keep it separate from any self-report on file. Note reading, language, sensory, or capacity limits that shaped the administration. Pitfall: A proxy score charted as the patient's own report, or "WHODAS completed" with no respondent named, so a later reader cannot tell whose account the number is. Recall window and completeness. Every version asks about the past 30 days. Write the window as dates ("the 30 days ending 09/15/2026") and say whether the month was typical: a result that covers an admission, a bereavement, a holiday, or a period off work is valid for that month and a poor baseline. State that all items were answered or which were missing, and whether the manual's limited imputation rule for a missing item or two was applied; a total computed with silently dropped items is not comparable with anything. Pitfall: A result that silently covers a hospital stay, or a total with missing items dropped without saying so. Scoring method and response coding. Name the method: simple scoring (an unweighted sum of the item responses, which the manual calls sufficient to describe the degree of functional limitation) or complex scoring (WHO's item response theory template, which weights items and levels and converts to the 0 to 100 metric). Then name the coding: WHO's current templates assign the five responses 0 to 4; the older manual text and the DSM-5 reproduction assigned 1 to 5, and WHO warns that the words did not change but the numbers did. Report the raw total with its maximum ("62 of 144 on 0 to 4 coding"), and the 0 to 100 summary only when WHO's template produced it; a raw sum expressed as a percentage is neither. If a program specifies a method (North Dakota's 1915(i) benefit requires the complex score), follow and cite it. Pitfall: A raw sum reported as a percentage or as a "complex score," or a 1 to 5 total from an old form compared with a 0 to 4 total from a new template as if 36 points of coding drift were improvement. Domain profile. For the 36-item version, report the six domain results beside the total and say which domains drive it and which sit near the floor: a psychiatric profile typically loads on cognition, getting along, life activities, and participation with mobility and self-care preserved, while a physical profile loads the reverse, and identical totals can hide opposite patterns. The 12-item version yields one overall score only, so do not read domain scores from it. Programs that use the profile for planning (1915(i) care plans, NDIS reports) need the domain results, not just the total. Pitfall: A total with no profile, so a mobility-driven and a cognition-driven 62 look identical, or domain scores reported from a 12-item administration. Interpretation: percentiles, not a WHO cutoff. Write that higher scores mean more reported difficulty, then locate the result against WHO's general-population norms in the manual if a percentile is useful, naming the norm source; do not attach a mild, moderate, or severe label to the total as if WHO defined it, because WHO defines none for the summary score and any band you use must be attributed to its actual source (a study, a program rule, or a local convention). Tie the pattern to the diagnosis as consistent with, not diagnostic of: the WHODAS is diagnosis-neutral and captures difficulty from any cause, so name other contributors (physical illness, environment) or their absence, and record congruence with the interview, the mental status exam , and collateral. Pitfall: "WHODAS 45, moderate disability" attributed to WHO, or the score offered as proof of a diagnosis or of disability. Decision context and repeat plan. Name the system the result will travel to (SSA, VA, workers' compensation, NDIS, a 1915(i) benefit, a payer, or a fitness-for-duty referral) and translate the findings into that system's own functional areas in prose, never as a crosswalk table, stating that self-report corroborates and does not establish eligibility, capacity, or causation; the sentence "WHODAS 58, therefore disabled" reserves a conclusion that belongs to the adjudicator. Link the profile to the treatment or support plan, then set the repeat: same version, mode, method, and coding, at the interval the purpose sets (MIPS measure 502 requires a follow-up 30 to 180 days after the index assessment), with change described in the profile rather than labelled clinically significant, because no single minimal important difference is established. Pitfall: "WHODAS 58, therefore disabled," or a follow-up on a different version, mode, or coding read as change. ### Blank template (copy and adapt) WHODAS 2.0 DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [diagnostic evaluation / treatment plan review / disability or NDIS report / outcome measurement] Version: [36-item / 12-item / 12+24 interviewer] Language + translation: [ ] Work or school items: [administered / not applicable: 32-item scoring used] Mode + respondent: [self / interviewer (in person, telephone) / proxy: relationship and reason] Reading, language, or capacity limits: [ ] Recall window: [the 30 days ending mm/dd/yyyy; typical or atypical month] Completeness: [all items answered / missing items and imputation applied or not] Scoring: [simple sum / WHO complex (IRT) template] Coding: [0 to 4 / 1 to 5] Total: [ ] of [maximum] 0 to 100 summary (if produced): [ ] Percentile (if used): [ ] Norm source: [WHO manual general population] Domain profile (36-item only): cognition [ ] mobility [ ] self-care [ ] getting along [ ] life activities: household [ ] work or school [ ] participation [ ] Driving domains: [ ] Near floor: [ ] Interpretation: [higher = more reported difficulty; no WHO cutoff or severity band; pattern consistent with, not diagnostic of, ...; other contributors: physical, environmental, or none; congruence with interview and collateral] Decision context: [SSA / VA / workers' compensation / NDIS / 1915(i) / payer / fitness for duty: findings translated into that system's own areas; self-report corroborates, does not establish eligibility or capacity] Prior results: [date, version, mode, method, coding; same-version change only; 12-item and 36-item not compared] Plan: [goals the profile supports / repeat: same version, mode, method, coding; interval (MIPS 502: 30 to 180 days) / where stored] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the instrument items, response scale, norms, or scoring templates. ## Sample WHODAS 2.0 documentation (fictional) Scenario: a US outpatient psychiatric diagnostic evaluation for an employed adult with recurrent major depressive disorder, where the psychiatrist documents a 36-item self-administered WHODAS 2.0 as the functioning measure, keeps a 12-item primary care result separate, ties the profile to the treatment plan, and fixes the repeat for the 90-day treatment plan review. All details are fictional. Patient: J.M., 37 · Setting: Outpatient psychiatry clinic, psychiatric diagnostic evaluation (intake) · Clinician: L. Ferreira, MD · Note date: 09/15/2026 Measure and administration: WHODAS 2.0, 36-item, self-administered version, English, completed on paper in the waiting room on 09/15/2026 before the diagnostic interview. J.M. works full time as a logistics coordinator, so the work items applied and all 36 items were scored (no 32-item pathway); all items answered, no imputation. Recall window: the 30 days ending 09/15/2026, which he describes as typical of the past three months, with no admission, leave, or bereavement. Scored by hand as a simple sum on the 0 to 4 response coding of WHO's current templates; WHO's complex (IRT) template was not run and no 0 to 100 summary is reported. Results: Total 62 of 144. Domain results, each the sum of that domain's item responses with the domain maximum in parentheses: cognition 14 (24), mobility 3 (20), self-care 2 (16), getting along 12 (20), life activities 18 (32), of which household 8 (16) and work 10 (16), participation 13 (32). Difficulty is concentrated in cognition, getting along, and the work portion of life activities, with participation elevated and mobility and self-care near the floor. Read against the general-population percentile table in the WHO manual, the total sits in the upper tail of the reference distribution, above the 90th percentile; no severity label is attached, because WHO defines none for the summary score. A 12-item WHODAS 2.0 completed at the primary care referral visit on 09/02/2026 is recorded separately in the referral documentation and is not compared with this result: different version, no domain scores. Interpretation: The pattern (cognitive, interpersonal, and work-role difficulty with preserved mobility and self-care) is consistent with, not diagnostic of, the recurrent major depressive episode documented in the diagnostic formulation above, and it matches his account of trouble concentrating and following conversations at work, withdrawal from colleagues and family, missed deadlines, and two absences this month. The WHODAS measures reported difficulty from any cause; J.M. has no physical condition that would load the mobility or self-care domains, and none is contributing here. Congruent with today's interview and mental status exam and with his partner's collateral account by telephone. The measure does not by itself establish diagnosis, work capacity, or disability. Linkage to plan: The profile is the functional baseline for the treatment plan dated 09/15/2026: goals target sustained concentration on work tasks, re-engagement with two social contacts each week, and restoration of the household routine, with the psychotherapy and medication interventions recorded in the plan itself. No disability, leave, or accommodation request is pending. If one arises, the WHODAS will be presented as corroborating self-report beside the interview, records, and collateral, translated into that system's own functional areas, and never as a conclusion about eligibility. Repeat and record: Repeat the 36-item self-administered version, same paper form, 0 to 4 coding, simple scoring, at the 90-day treatment plan review in early December; describe change in the domain profile rather than labelling it clinically significant, since no single minimal important difference is established for this population. Result entered in the outcome-measure record with version, mode, recall window, method, coding, total, and domain sums; the completed paper form is filed in the chart, and no GAF is entered because no form in this record requires one. J.M. was told the number describes his own report of difficulty over the past month and is not a diagnosis or a disability rating. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No instrument items, norms, or scoring materials are reproduced. ↑ Back to the template and downloads ### Why this sample works The configuration is complete: version, respondent, mode, language, recall dates, completeness, method, and coding are all named, so the total is reproducible and the December repeat can be compared with it. The domain profile sits beside the total with the driving and floor domains named, so a reader sees a cognition, interpersonal, and work-role pattern rather than an anonymous 62. The interpretation uses the manual's percentile position and no WHO-attributed severity band, states that the pattern is consistent with rather than diagnostic of the episode, and names the absence of physical contributors. The 12-item primary care result is kept separate and uncompared, and no GAF is manufactured, so no cross-version or cross-instrument conversion enters the record. The entry closes with linkage and limits: the profile feeds the treatment plan goals, the disability framing is set in advance, and the repeat is fixed to the same version, mode, method, and coding. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the authority map is layered, and none of it makes the WHODAS a decision rule. Under LAW, the Social Security Administration rates adult mental impairments on its own four functional areas (understanding, remembering, and applying information; interacting with others; concentrating, persisting, and maintaining pace; adapting and managing oneself) with its own limitation scale under Listing 12.00, and no SSA rule names the WHODAS or attaches a disability threshold to its score; the National Academies' 2015 report for SSA on psychological testing described self-report measures as corroborating symptoms rather than providing direct evidence of functional capacity, which is the frame a WHODAS should carry into a medical source statement (CONVENTION). Also under LAW, 38 CFR 4.126 directs VA rating agencies to base a mental-disorder evaluation on all the evidence of record bearing on occupational and social impairment rather than solely on an examiner's assessment at the moment of examination, so a WHODAS in a VA file is one strand of that record; peer-reviewed VA work uses it as an adjunct in general mental health clinics (CONVENTION), and the GAF the VA dropped from its examinations in 2014 does not convert to it. Under PAYER POLICY, CMS's 2026 MIPS quality measure 502 (Improvement or Maintenance of Functioning for Individuals with a Mental and/or Substance Use Disorder) accepts the 12-item WHODAS 2.0 or the Sheehan Disability Scale for patients 18 and older, requires a follow-up 30 to 180 days after the index assessment, defines improvement as any positive change in score and maintenance as no change, and states that there is no recommended cutoff score. State Medicaid 1915(i) benefits go further: North Dakota's needs-based eligibility policy (updated April 2026) requires the 36-item version with WHO's complex scoring, a complex score of 25 or higher (or a DLA-20 score of 5 or lower), trained administrators, and face-to-face administration even though the instrument itself does not require it, and Michigan uses the 36-item WHODAS 2.0 to support 1915(i) eligibility and Habilitation Supports Waiver screening in place of the Supports Intensity Scale (PAYER POLICY, state by state). Those program rules set the version, method, and threshold inside the program; outside them the number carries no threshold. State workers' compensation systems rate psychiatric impairment under their own schedules, none of which adopts the WHODAS as the rating instrument (LAW, state by state), and in fitness-for-duty and capacity evaluations it is corroborating self-report beside the interview, records, collateral, and any validity testing (CONVENTION). Canada and Australia accept it as evidence and require it nowhere. In Canada, the Ontario Disability Support Program determines disability from its own package, a mandatory Health Status Report and Activities of Daily Living Index completed by listed health professionals, to which functional assessments and specialist reports may be attached (PAYER POLICY, provincial); no provincial disability program reviewed for this page names the WHODAS, so a Canadian WHODAS is supplementary functional evidence, and the 12-item Canadian population study by Sjonnesen and colleagues (23,757 respondents; mean scores from 14.2 in the community to 23.1 in schizophrenia) is the reference most often cited beside it (CONVENTION). In Australia, the NDIS Act's access test asks for substantially reduced functional capacity in one or more of six areas (communication, social interaction, learning, mobility, self-care, self-management) (LAW); the NDIA's psychosocial disability guidance lists the LSP-16, HoNOS, and WHODAS as functional-assessment evidence, calls its Evidence of Psychosocial Disability form preferred but not compulsory, and states that those tools may not provide enough evidence on their own and are considered alongside a treating health professional's information and carer or support-worker statements (PAYER POLICY); RANZCP's guidance for psychiatrists (updated March 2025) lists the accepted tools in order as HoNOS, LSP-16, and WHODAS 2.0 for ages 17 and over, so the WHODAS sits inside the NDIS functional capacity assessment and the NDIS progress report rather than replacing either (CONVENTION). Reform is live: the Getting the NDIS Back on Track Act received assent on September 5, 2024, and on August 31, 2026 the Commonwealth opened a consultation, closing October 2, 2026, on how functional capacity should be assessed and what evidence should support access decisions, with a Technical Advisory Group advising, so an NDIS report written in late 2026 should note that the evidentiary framework is changing. On versions and norms: the 36-item and 12-item forms are not interchangeable (the short form explains 81 percent of the long form's variance and yields no domain profile), the 12+24 interviewer hybrid is a third configuration, the 2010 manual's general-population norms are the only WHO reference distribution, and WHO's own templates now code responses 0 to 4 where older materials, including the DSM-5 reproduction, used 1 to 5, so the coding is part of the record. The instrument carries no publisher qualification level: WHO asks users to register, the adult forms are for ages 18 and over (the WHODAS-Child is a separate research instrument, and a Canadian population study found the 12-item adult form valid for comparisons down to age 15 as a research finding, not a change of scope), and programs such as North Dakota's require completed administration training for the interviewer version. The WHODAS 2.0 is copyrighted by the World Health Organization and made available free of charge for clinical use on WHO's terms: the 2010 manual describes the standard instrument as in the public domain subject to completing WHO's online registration form and making no substantive changes without permission; the American Psychiatric Association's DSM-5 and DSM-5-TR reproductions, published with WHO's permission, state that clinicians may reproduce the material without permission for use with their own patients and that any other use, including electronic use, requires written permission from WHO; and WHO's WHODAS 2.0 page (accessed September 2026) directs anyone including the instrument in an electronic records or data capture system, or reproducing it in any way, to WHO's licensing process, under a permissions policy that requires permission for commercial uses. Paper use with your own patients is covered; an EHR build, an app, a public web scorer, or a paid product is not until WHO says so. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common WHODAS 2.0 documentation errors reviewers flag The numbers behind these errors are specific. In WHO's development work across 19 countries and more than 65,000 respondents, the 36-item total showed internal consistency of 0.86 and test-retest reliability of 0.98 (Üstün and colleagues, 2010); the 2026 systematic review of 143 studies in 43 countries found a mean alpha of 0.92 for the 36-item and 0.88 for the 12-item form, test-retest values of about 0.89 to 0.91, support for the six-domain structure in 74 percent of 36-item analyses but only 59 percent of 12-item analyses, and two unresolved scoring issues, inconsistent procedures between the manual's SPSS algorithm and WHO's Excel template and unclear handling of not-applicable responses (Federici and colleagues, 2026). Simple and complex person measures correlated 0.99 in 780 Swedish psychiatric outpatients, yet the same study found domain-level reliability problems and one misfitting item in the getting-along domain (Svanborg and colleagues, 2022). A 2023 systematic review found self and proxy ratings agree better in observable physical domains than in emotional ones (Hernández and colleagues). No single minimal important difference is established (Federici and colleagues, 2017); estimates run from about 3 points in depression and anxiety (Abdin and colleagues, 2025) to 4.87 points after low back pain rehabilitation (Ćwirlej-Sozańska and colleagues, 2020). No published audit counts how often a chart entry omits the version, mode, method, or coding; the program rules described under compliance considerations are what turn those omissions into an eligibility or weight problem. The BastionGPT Clinical Advisory Board sees the same errors most often in WHODAS 2.0 documentation reviews: A bare number. "WHODAS 45" could be a 12-item or 36-item total, a self or proxy report, a simple sum on 0 to 4 or 1 to 5 coding, or a complex score on the 0 to 100 metric, and nothing in the number says which. Write the version, respondent, recall window, method, and coding every time; a later reader, a payer, or an adjudicator cannot reconstruct them. A WHO severity band that WHO never published. "WHODAS 45, moderate disability" attributed to WHO, or a calculator's mild, moderate, and severe ranges pasted into the chart. WHO interprets the summary score against population norms and percentiles and defines no cutoff, and CMS's 2026 measure specification says the same; the none-to-extreme words are item response labels. Report the number and, if useful, the percentile with its norm source, and attribute any band to the study or program it came from. 12-item and 36-item results treated as interchangeable. A 12-item follow-up trended against a 36-item baseline, or domain scores reported from a 12-item administration. The short form explains 81 percent of the long form's variance, yields one overall score, and shows weaker support for the six-domain structure; describe the two results separately and compare only within a version. Work items scored as no difficulty, or the 32-item pathway unstated. An unemployed respondent's work and school items entered as no difficulty, which produces an artificially favorable profile and hides the vocational limitation the evaluation exists to describe. The manual's pathway omits those items and scores the remaining 32; say so, state whether the absence of work is itself clinically relevant, and keep the treatment constant at follow-up. Proxy and self-report charted as the same testimony. A carer's proxy form entered as the patient's own report, or a proxy baseline compared with a self-report follow-up as change. Proxy agreement is better for observable domains than for cognition, getting along, and emotional experience; name the proxy, the relationship, and the reason, and repeat with the same respondent. The score offered as the decision, or converted to a GAF. "WHODAS 58, therefore disabled," a WHODAS domain profile translated into an SSA rating or an NDIS determination as a crosswalk, or a total converted into a GAF for a legacy form. Every system applies its own criteria, self-report corroborates rather than establishes, and no validated conversion to the GAF exists; translate findings into the system's own areas in prose, and enter any required GAF separately with its own frame. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on psychiatric diagnostic evaluations and treatment plan reviews. Give it the facts (version and item count, who completed it and how, the recall window and dates, completeness, scoring method and coding, the total and domain results, the percentile if used, the diagnosis, the decision context, and any prior administration) and it drafts the documentation block: configuration named, domain profile described, interpretation kept to percentiles and consistent-with language, the system translation in prose, and the repeat plan, ready for your review. Cross-check a finished note for the gaps reviewers flag: a bare number with no version or method, a WHO-attributed severity band, a 12-item result trended against a 36-item baseline, work items scored as no difficulty for a non-working respondent, a proxy report charted as self-report, or a score presented as eligibility or converted to a GAF. Turn the entry into the next document: the functioning paragraph of a psychiatric diagnostic evaluation, the outcome section of a treatment plan review, or the corroborating self-report paragraph of a medical source statement or NDIS functional capacity assessment, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does a WHODAS 2.0 score mean, and is there a cutoff? Higher means more reported difficulty over the past 30 days; that is all a bare number says. A simple score is an unweighted sum of the item responses, so its maximum depends on the version and the coding (0 to 4 on WHO's current templates, 1 to 5 in older materials); a complex score is WHO's item response theory result on a 0 to 100 metric, where 0 is no disability and 100 full disability. WHO gives no clinical cutoff and no severity bands for the total: the manual interprets scores against general-population norms and percentiles, and the none-to-extreme words belong to the item response scale, not to the summary score. CMS's 2026 MIPS specification for measure 502 says the same: there is no recommended cutoff score. Where a threshold exists it is a program's rule (North Dakota's 1915(i) benefit uses a complex score of 25 or higher), and the note should cite that program, not WHO. The interpretable unit is the version, the method and coding, the total, the domain profile, and the percentile if one is used. 36-item or 12-item: which version do I document, and are they interchangeable? Choose by what the reader needs. The 36-item version yields an overall score and six domain scores, takes about 20 minutes by interview, and is the DSM-5 Section III form and the version state 1915(i) programs and most NDIS reports rely on, because the domain profile does the work. The 12-item version takes about 5 minutes, explains 81 percent of the variance of the long form, and yields one overall score; it suits brief outcome tracking and is the version MIPS measure 502 accepts. They are not interchangeable: a 12-item total cannot be trended against a 36-item baseline, the 2026 systematic review found weaker support for the six-domain structure in the short form, and no domain profile can be read from it. The 12+24 interviewer hybrid screens with the short set and adds the fuller items where difficulty is reported; name it if you use it. Whatever you choose, keep it constant for repeat administrations, and describe serial results under the conventions of the outcome measure note . Simple or complex scoring, and why does the 0 to 4 versus 1 to 5 coding matter? Both are WHO-recognized. Simple scoring adds the item responses without weighting; the manual describes it as practical for hand scoring and sufficient to describe the degree of functional limitation. Complex scoring runs the responses through WHO's item response theory template, which weights items and difficulty levels and converts the result to 0 to 100; it is the method for population comparison and the one some programs mandate. The two rank people almost identically (a 0.99 correlation in 780 Swedish psychiatric outpatients) but are different numbers, and the 2026 systematic review found the manual's SPSS algorithm and WHO's Excel template do not describe identical procedures, so name the method and, for complex scoring, the template used. Coding matters because WHO changed the numbers assigned to the unchanged response words from 1 to 5 to 0 to 4: on the 36-item form the same answers sum 36 points higher under the old coding, so an undocumented raw total from an old form and one from a new template cannot be compared, and a percentage computed from a raw sum is neither WHO score. Write the method and the coding every time. Did the WHODAS 2.0 replace the GAF, and can I convert between them? DSM-5 removed the GAF and the multiaxial system in 2013 and placed the 36-item self-administered WHODAS 2.0 in Section III, the chapter for emerging measures needing further study, as the suggested functioning measure; DSM-5-TR (2022) kept it there. That is a recommendation, not a mandate, and no national uptake figure exists, so write that DSM-5 introduced the WHODAS 2.0 as its standardized disability measure, not that it replaced the GAF in practice; some payer, state, and VA forms still carry a GAF field. The two are conceptually different: the GAF is a clinician's single 1 to 100 judgment blending symptom severity with functioning over a short current period for mental illness only, and the WHODAS is a respondent- or informant-based, diagnosis-neutral, domain-structured measure of difficulty from any cause over 30 days. No validated conversion exists, so never derive one from the other; when a form demands a GAF, enter it under its own frame as the GAF score page describes, record the WHODAS separately with its version and method, and state that no conversion was made. Do SSA, the VA, workers' compensation, or the NDIS require a WHODAS, and is a score enough to prove disability? No authority requires it, and no score proves anything by itself. SSA rates mental impairments on its own four functional areas with its own limitation scale (LAW) and names no instrument; the National Academies' 2015 report for SSA describes self-report measures as corroborating symptoms rather than as direct evidence of functional capacity, so present the WHODAS in a medical source statement as corroboration and translate the findings into SSA's areas in prose. The VA's rating rule, 38 CFR 4.126, requires an evaluation based on all the evidence of occupational and social impairment rather than an examiner's snapshot (LAW), which makes a WHODAS one strand of the record. Workers' compensation impairment is rated under state schedules, not the WHODAS (LAW, state by state). The NDIS access test asks for substantially reduced functional capacity in one or more of six statutory areas (LAW), and the NDIA lists the WHODAS with the LSP-16 and HoNOS as accepted functional evidence while stating those tools may not provide enough evidence on their own (PAYER POLICY); the NDIS functional capacity assessment page covers the report the WHODAS sits inside. The exceptions run the other way: programs such as North Dakota's 1915(i) benefit set a WHODAS threshold as one needs-based criterion, and CMS's MIPS measure 502 accepts the 12-item version for outcome reporting (PAYER POLICY). Everywhere else the number corroborates; the decision-maker decides. Can a family member or carer complete it, and how do I chart a proxy result? Yes, with the proxy-administered version, which WHO publishes for a family member, carer, or other observer to report on the person's functioning, and which the APA form points to when an adult cannot complete the self-report. Chart it as a proxy report: name the proxy's relationship, the reason proxy administration was used, and whether the person was present, and keep it separate from any self-report on file rather than blending the two. Agreement between self and proxy ratings is generally good but better for observable domains (mobility, self-care) than for cognition, getting along, and emotional experience, and in youth samples parents have rated less difficulty than the young people themselves, so a proxy profile can understate the domains that matter most in psychiatric evaluation. Some programs set the format: North Dakota's 1915(i) policy allows a face-to-face proxy assessment with the applicant's representative. Repeat administrations should use the same respondent; a proxy baseline and a self-report follow-up are not a change score. How do I handle the work and school items when my client is not working? Do not score them as no difficulty. The manual's pathway for a respondent who is not in paid work or school omits the work and school items and scores the remaining 32, and the manual describes the 32-item result as comparable with the 36-item score. Document it: "Work and school items not administered; respondent not working during the reference period; 32-item scoring used," and state whether the absence of work is itself clinically relevant, because leaving those items at zero produces an artificially favorable profile that hides the vocational limitation a disability or NDIS evaluation exists to describe. Keep the same treatment at follow-up so a return to work does not masquerade as a change in scoring; if the person has since started work, say that the 36-item and 32-item results are not directly comparable and describe the change in the other domains. The 2026 systematic review flagged handling of not-applicable responses as an unresolved point in WHO's instructions, which is one more reason to write down exactly what you did. Can we put the WHODAS 2.0 in our EHR, on our website, or in an app? Not without WHO's licensing. The layers are precise. WHO's 2010 manual describes the instrument as in the public domain in the sense of free access after completing WHO's online registration, with no substantive changes permitted without WHO's approval. The APA reproductions in DSM-5 and DSM-5-TR, published with WHO's permission, state that clinicians may reproduce the material without permission for use with their own patients and that any other use, including electronic use, requires written permission from WHO. And WHO's WHODAS 2.0 page, checked September 2026, directs anyone including the instrument in an electronic records or data capture system, or reproducing it in any way, to WHO's classifications licensing, under a permissions policy that requires permission for commercial uses. So a paper form used with your own patients is covered; an EHR build, a patient portal questionnaire, a public web scorer, or an app is an electronic use that needs WHO's written permission or license first, whether or not it is free, and vendor pages claiming that no permission is required are wrong as a blanket statement. This page and its templates stay on the right side of that line: structure in words, no items, no norms, no scoring template. Can BastionGPT draft my WHODAS 2.0 documentation? Yes. Give it the facts (version and item count, who completed it and how, the recall window and dates, completeness, scoring method and coding, the total and domain results, the percentile if used, the diagnosis, the decision context, and any prior administration) and it drafts the full entry: configuration named, domain profile described, interpretation kept to percentiles and consistent-with language, the system translation in prose, and the repeat plan, ready for your review. It can also check a finished note for a bare number with no version or method, a WHO-attributed severity band, a 12-item result trended against a 36-item baseline, work items scored as no difficulty for a non-working respondent, a proxy report charted as self-report, or a score presented as eligibility or converted to a GAF. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: World Health Organization, WHODAS 2.0 (36- and 12-item versions and administration times; 81 percent of variance; simple and complex scoring and the 0 to 100 metric; the 0 to 4 coding note; licensing for electronic records, data capture systems, and reproduction; translations disclaimer; accessed September 2026); Üstün TB, Kostanjsek N, Chatterji S, Rehm J, editors, 2010, Measuring Health and Disability: Manual for WHO Disability Assessment Schedule (WHODAS 2.0) (seven versions; general-population norms; registration and permissions); WHO, copyright, licensing, and permissions (permission required for commercial uses). Üstün TB, Chatterji S, Kostanjsek N, and colleagues, 2010, Bulletin of the World Health Organization 88(11):815-823, developing the WHODAS 2.0 (19 countries; more than 65,000 respondents; alpha 0.86; test-retest 0.98; 81 percent of variance); Federici S, Bracalenti M, Meloni F, Luciano JV, 2017, Disability and Rehabilitation, international systematic review (810 studies, 94 countries, 47 languages; no single minimal clinically important difference established); Federici S, Tosti A, Russo EA, Conigli L, 2026, Frontiers in Psychiatry, measurement characteristics of WHODAS 2.0 and WHODAS-Child (143 studies, 43 countries; reliability; factor structure; SPSS versus Excel scoring inconsistency; not-applicable handling). Svanborg C and colleagues, 2022, Journal of Patient-Reported Outcomes 6:45, Rasch analysis of the Swedish 36-item WHODAS 2.0 in mental disorders (780 outpatients; simple versus complex scoring correlation 0.99); Axelsson E and colleagues, 2017, JMIR Mental Health, the 12-item WHODAS 2.0 administered via the internet (alpha 0.83 to 0.92; test-retest 0.83); Theotokatos G and colleagues, 2023, Cureus, 12-item Greek version in welfare applicants (10,163 adults; no floor or ceiling effect); Sexton MB and colleagues, 2019, Psychiatry Research, factor analysis among veterans in a VA mental health clinic (464 veterans); Sjonnesen K and colleagues, 2016, Canadian Journal of Psychiatry, the 12-item WHODAS 2.0 in Canadians with mental disorders (23,757 respondents); Kimber M, Rehm J, Ferro MA, 2015, PLoS One, measurement invariance in youth (valid for comparisons from age 15). Hernández JD and colleagues, 2023, Journal of Rehabilitation Medicine, assessment by proxy of the SF-36 and WHODAS 2.0: a systematic review (better agreement in observable domains); Abdin E and colleagues, 2025, Frontiers in Rehabilitation Sciences, responsiveness and minimal important differences in depression and anxiety (about 3 points for the WHODAS); Ćwirlej-Sozańska A and colleagues, 2020, International Journal of Environmental Research and Public Health, the Polish 36-item WHODAS 2.0 in low back pain (MCID 4.87). American Psychiatric Association, DSM-5-TR online assessment measures (the 36-item self-, proxy-, and interviewer-administered WHODAS 2.0 in Section III; ages 18 and older; reproduction by clinicians for their own patients, written WHO permission for any other use including electronic use); Gold LH, 2014, Journal of the American Academy of Psychiatry and the Law 42(2):173-181, DSM-5 and the assessment of functioning (the GAF dropped; the WHODAS 2.0 offered for further study). United States: CMS Quality Payment Program, 2026 MIPS quality measure 502 (12-item WHODAS 2.0 or Sheehan Disability Scale; 30 to 180 days; no recommended cutoff score; bare code numbers only); Social Security Administration, Listing 12.00, mental disorders, adult (the four functional areas); eCFR, 38 CFR 4.126 ; National Academies of Sciences, Engineering, and Medicine, 2015, Psychological Testing in the Service of Disability Determination (self-report measures corroborate symptoms); North Dakota Health and Human Services, 1915(i) needs-based eligibility: WHODAS 2.0 and DLA-20 assessments (updated April 2026; complex score of 25 or higher; 36-item version and complex scoring required; face-to-face administration; trained administrators); Michigan Department of Health and Human Services, WHODAS 2.0 (36-item tool for 1915(i) eligibility and Habilitation Supports Waiver screening). Canada: Government of Ontario, Ontario Disability Support Program guide for health care professionals (updated March 10, 2026; Health Status Report and Activities of Daily Living Index; functional assessments may be attached; no instrument named). Australia: National Disability Insurance Agency, the general practitioner guide to the NDIS for people with psychosocial disability (LSP-16, HoNOS, and WHODAS may not provide enough evidence on their own; Evidence of Psychosocial Disability form preferred, not compulsory); RANZCP, NDIS: information for psychiatrists (updated March 31, 2025; HoNOS, LSP-16, WHODAS 2.0 for 17 and over); Australian Government Department of Health, Disability and Ageing, consultation opens on proposed NDIS access changes (August 31 to October 2, 2026; Technical Advisory Group); Federal Register of Legislation, National Disability Insurance Scheme Amendment (Getting the NDIS Back on Track No. 1) Act 2024 (assent September 5, 2024). ## Related templates GAF Score Documentation Psychiatric Diagnostic Evaluation Treatment Plan Review & Update Outcome Measure Review Note Disability Evaluation & Medical Source Statement NDIS Functional Capacity Assessment Report Fitness-for-Duty Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/sassi All note templates # SASSI Documentation: High or Low Probability, Limits & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The SASSI (Substance Abuse Subtle Screening Inventory) is a proprietary self-report screen published by The SASSI Institute in adult SASSI-4 and adolescent SASSI-A3 editions, whose publisher decision rules return a high or low probability of a substance use disorder. Addiction, DUI, probation, child-welfare, and EAP evaluators use it to decide who needs full assessment; it never diagnoses. This page covers how to document a SASSI result, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Addiction counselors, social workers, psychologists, and other human-service professionals who file the Institute's qualification form (Pearson lists the family at qualification level B); staff without professional training may administer and score under supervision; the qualified clinician interprets, gives feedback, and writes the report Audience Courts, probation and DUI programs, drug courts, child-welfare and custody evaluators, EAP and DOT Substance Abuse Professionals, treatment admission staff, referring clinicians, attorneys reviewing records, and payers Typical length 5 to 10 chart lines or one report paragraph (edition, mode, time frame, validity, category and limits, profile hypotheses, conclusion, recommendation, records) · administration and scoring about 15 minutes, plus interview and collateral review Format family Proprietary self-report screening inventory (face-valid scales plus subtle or indirect scales, validity indicators, and publisher decision rules that yield a high or low probability of a substance use disorder) When it's used Court-ordered DUI and probation evaluations, drug-court and diversion intake, child-welfare and custody assessments, EAP and post-violation DOT evaluations, addiction treatment admission and level-of-care workups, adolescent juvenile-justice and school referrals Standards context The SASSI Institute (Springville, Indiana): Adult SASSI-4 (2016) and Adolescent SASSI-A3 (2020) with proprietary items, scoring, and decision rules; no statute, payer policy, or federal rule names it; described here for documentation, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the SASSI? The Substance Abuse Subtle Screening Inventory is a family of brief, copyrighted self-report questionnaires published by The SASSI Institute of Springville, Indiana. Glenn A. Miller developed the original adult SASSI in the 1980s and an adolescent version followed in 1990; the current editions are the Adult SASSI-4 (user guide by Lazowski, Kimmell, and Baker; 2016) for ages 18 and up and the Adolescent SASSI-A3 (2020) for ages 13 to 18, both validated against DSM-5 substance use disorder diagnoses and both carrying a prescription drug scale. The publisher states a fourth- to fifth-grade reading level for the adult form, about fourth grade for the adolescent form, and about 15 minutes to administer and score. Each questionnaire has two parts: face-valid items on which the respondent acknowledges alcohol and other-drug experiences over a time frame the evaluator selects, and true-false items that feed subtle or indirect scales whose relationship to substance use is not obvious to the respondent, with validity indicators alongside. The publisher's decision rules, printed only in the edition's user guide and applied by its scoring key or the SASSI Online platform, combine the scales into a single categorical result: High Probability or Low Probability of a substance use disorder. A separate adult Spanish SASSI (publisher-reported accuracy of 84 percent, validated in its own sample) is not a translation of the SASSI-4, and Pearson lists the family at qualification level B. As of September 2026 the Institute has announced no SASSI-5 or SASSI-A4. Three facts carry the documentation. First, the result is a screening classification and the publisher says so: the Institute's guidance describes it as one piece of information for a diagnostic evaluation, states that the SASSI cannot determine whether a client is currently drinking, and warns that someone in long recovery will more than likely still screen as high probability because the subtle scales reflect traits that do not change quickly, so a high result never establishes current use or impairment and a low result never rules a disorder out. Second, everything that produces the result is proprietary: the items, scales, T scores, norms, profiles, score reports, and scoring keys are copyrighted, the copyright notice bars reproducing, paraphrasing, or translating the questions and copying the materials into computer storage, and the decision rules are neither published nor reproducible outside the licensed key or platform, which is why a defensible note reports the category in the publisher's words and keeps the scored profile in the testing file. Third, the subtle scales are contested: the Institute reports 92 percent overall accuracy for the SASSI-4 and 89 percent for the SASSI-A3 against clinician DSM-5 diagnoses, while the leading independent review (Feldstein and Miller, 2007) found sensitivity similar to public-domain screens, a high false-positive rate, and no empirical evidence for the claimed unique advantage of the indirect scales. That is the boundary against the CAGE , AUDIT-C , and DAST-10 , which are short, transparent screens with published scoring, and against the substance use assessment , where DSM-5 criteria and the diagnosis belong. ## Who uses SASSI documentation and when The SASSI is administered wherever a substance use question carries consequences and the respondent has a reason to minimize. In the publisher's SASSI-4 validation sample, 46 percent of respondents came from substance use treatment programs, 18 percent from criminal justice programs (community corrections, probation and parole, and drug courts), 16 percent from private practices, 9 percent from behavioral health centers, 6 percent from DWI and DOT screening and education programs, and 5 percent from government and community social service programs, which is a fair map of who charts it: addiction counselors at admission and level-of-care review beside the biopsychosocial assessment ; evaluators completing court-ordered DUI evaluations, presentence and probation assessments, and drug-court intake whose write-ups become a court report or a forensic evaluation report ; child-welfare and custody evaluators (the Institute's qualification form lists child protective services among user settings); employee assistance programs and DOT Substance Abuse Professionals in post-violation evaluations, where the fitness-for-duty evaluation page covers the occupational standard; and adolescent programs, schools, and juvenile courts using the SASSI-A3. The reader is rarely the author: a judge, probation officer, caseworker, employer, or opposing attorney will read the classification adversarially, which is why the entry has to state the edition, time frame, validity, category, limits, and the independent basis for the conclusion. Neighbors win in three places: the AUDIT-C and DAST-10 for routine primary care and SBIRT screening, where a transparent tool is expected and named in CMS guidance; the CAGE when four questions at the bedside are all the encounter allows; and a full substance use assessment whenever the question is diagnosis, severity, or placement, which no screen answers, with the treatment work that follows carried in a relapse prevention plan . ## How to document SASSI results in the chart No statute, payer, or accreditor prescribes a SASSI note format, and the publisher's contribution is a user guide, a set of appropriate-use and training statements, and a copyright notice that keeps the items, keys, and rules out of anything you write. What survives a court, a probation officer, or a records request is an entry that names the edition, form, and administration mode, states the time frame the respondent was given for the face-valid items, records the reading and comprehension conditions, reports the validity review in words, states the result category in the publisher's exact terms with its published limits, describes the profile qualitatively as hypotheses, and shows that the diagnostic conclusion and recommendation rest on interview, collateral, records, and toxicology rather than on the screen. Each element below carries the pitfall that most often undermines it. Edition, form, and administration mode. Write the exact instrument: Adult SASSI-4, Adolescent SASSI-A3, or Spanish SASSI, with the language, and how it was given: paper questionnaire scored with the publisher's key, SASSI Online (which returns the decision-rule outcome, a scale graph, and a narrative report), or the publisher's audio administration, plus who proctored it and where. Results from the retired SASSI-3 and SASSI-A2 belong to different instruments and are charted with their edition, never trended against a current result. Because the Spanish SASSI is a separate adult instrument rather than a translation, and the Institute recommends against using its scale scores to form clinical hypotheses, chart a Spanish administration by its classification alone. Pitfall: A note reading SASSI administered with no edition, or a SASSI-3 interpretation copied from an old handout and applied to a SASSI-4 result. Time frame and referral context. State the time frame the respondent was instructed to use for the face-valid section (lifetime, the past 12 months, the past six months, or since a specified date), because the publisher lets the evaluator choose it and the meaning of acknowledged or unacknowledged use changes with it. Record the referral source and question (court order, probation condition, child-welfare plan, EAP referral, treatment admission), the case number if the report will be filed, the date, and the fact that the respondent was told the result would be shared with the referring party. Pitfall: A classification reported with no time frame, so a court cannot tell whether the respondent denied any lifetime use or only use since the arrest. Reading level, comprehension, and accommodations. The publisher states a fourth- to fifth-grade reading demand for the adult form and about fourth grade for the adolescent form. Chart how you confirmed the respondent could read the questionnaire, whether the publisher's audio administration was used, whether the respondent asked what an item meant and how that was handled without rephrasing it, and any interpreter, vision, or fatigue issue. The copyright notice forbids paraphrasing or translating the questions, so an improvised explanation or translation is a nonstandard administration and is charted as one. Pitfall: A literacy problem or an item translated on the spot scored as a valid profile, or a helpful paraphrase of an item written into the note. Validity statement, in words. Report the publisher's validity check by name (the Random Answering Pattern indicator on the adult form, the Validity Check on the adolescent form), whether items were skipped or double-marked and how that was resolved, and your conclusion: the profile was interpretable for screening, interpretable with caution, or not interpretable and not used. Describe response style qualitatively: the Defensiveness scale indexes reluctance to endorse personal faults, and the publisher's guidance treats an elevated result as a reason to suspect a missed case, not as evidence of lying. Keep the numbers, the thresholds, and the profile sheet in the testing file. Pitfall: Valid profile with nothing behind it, T scores and cutoffs copied into the note, or defensiveness written up as deception. Result category, in the publisher's words. Write the classification exactly as the key or platform reports it, High Probability of a substance use disorder or Low Probability of a substance use disorder, with the edition and date, and follow it in the same sentence with its published limits: a screening classification, not a diagnosis; a low result does not rule out a disorder, especially with a defensive profile; a high result does not establish current use, intoxication, impairment, or relapse, and a person in long recovery may still screen high. Never restate the category as positive for addiction, passed, or failed. Pitfall: SASSI positive: alcohol dependence, or SASSI negative, no substance problem, either of which converts a probability screen into a finding it cannot support. Profile description, as hypotheses. Below the classification, describe what the scale pattern suggests in plain language and label it a hypothesis: limited acknowledgment of alcohol or other-drug consequences on the face-valid scales, a symptom or attribute pattern resembling that of people with a diagnosed disorder, a prescription drug scale flag that warrants specific follow-up, or, on the SASSI-A3, family and peer risk or attitudes toward use. The Correctional and Family scales describe resemblance to reference groups and support no inference about criminality or a family diagnosis. The Institute's own training guidance says the further an interpretation moves from the scores toward personality or behavioral conclusions, and the less collateral information exists, the more it must be treated as a hypothesis to investigate. Pitfall: The platform narrative pasted in as the formulation, or a single scale turned into a personality label, a prediction of offending, or a family diagnosis. Integration, conclusion, recommendation, and records. Name the evidence the conclusion actually rests on: the clinical interview with DSM-5 criteria reviewed, the substance use history, arrest, driving, and treatment records, collateral informants, toxicology or breath results, and any other instruments (the AUDIT-C or DAST-10 for a transparent comparison). State the diagnostic conclusion, or the absence of one, as reached independently of the screen, and write the recommendation (level of care under the ASAM Criteria, education, monitoring, or no treatment) so that it does not rest on the classification alone. Close with record handling: where the completed questionnaire, profile, and platform report are stored as test data, what goes into the report that leaves the office, the release or 42 CFR Part 2 consent on file, and any re-administration plan, remembering that the Institute recommends against routine repeats. Pitfall: A recommendation that cites only the classification, or a court report with the questionnaire and profile sheet stapled to it. ### Blank template (copy and adapt) SASSI DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Referral: [court / probation / child welfare / EAP or DOT SAP / treatment admission / other; case no.] Instrument: [Adult SASSI-4 / Adolescent SASSI-A3 / Spanish SASSI] Language: [ ] Administration: [paper, publisher's key / SASSI Online / audio] Proctored by: [ ] Time frame given for the face-valid items: [lifetime / past 12 months / past 6 months / since (date)] Reading and comprehension: [reading level confirmed how; audio used; questions asked and handled without rephrasing; interpreter; fatigue; none] Validity review: [items complete; random or inconsistent responding not flagged / flagged; response style described in words; profile interpretable / with caution / not interpretable, reason] Result category (publisher's words): [High Probability / Low Probability of a substance use disorder] Limits stated: [screen, not diagnosis; low does not rule out; high does not show current use or impairment] Profile hypotheses (no scores): [face-valid acknowledgment; symptom or attribute pattern; prescription drug flag; adolescent family, peer, or attitude risk; Spanish SASSI: classification only] Other instruments and toxicology: [AUDIT-C / DAST-10 / breath or urine result] Interview and collateral: [DSM-5 criteria reviewed; history; records; informants] Conclusion (independent of the screen): [diagnosis, or none, and its basis] Recommendation: [level of care; education; monitoring; not resting on the screen] Records and release: [questionnaire, profile, platform report stored as test data; contents of the outgoing report; release / Part 2 consent; repeat plan] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces items, scoring keys, decision rules, or profile layouts. ## Sample SASSI documentation (fictional) Scenario: a court-ordered DUI evaluation at a licensed outpatient addiction program, where an Adult SASSI-4 given through SASSI Online returns a low-probability classification with a guarded response style while the arrest record, interview, and collateral point the other way; the evaluator documents the edition, time frame, validity, and category in the publisher's terms, reaches the diagnosis independently, and handles the records for the court. All details are fictional. Patient: M.L., 34 · Setting: Outpatient addiction program, court-ordered DUI evaluation · Clinician: T. Alvarez, LCSW, CADC · Note date: 09/17/2026 Instrument and administration: Adult SASSI-4, English, administered on 09/17/2026 through SASSI Online on a program tablet in a private office, proctored by this evaluator, as one component of a substance use evaluation ordered by the county court after a 09/05/2026 arrest for driving under the influence (case number on file). The face-valid section was completed for the past 12 months, the time frame this program uses for first-offense evaluations. M.L. was told before starting that the result would be reported to the court. He completed high school and reads for his work; he read the on-screen instructions aloud without difficulty, asked no questions about item meaning, and used no audio or interpreter accommodation. All items were answered in about 14 minutes. Scoring and the classification were produced by the publisher's platform; the platform report and profile are stored in the program's testing file, and no item content appears in this record. Validity and response style: The platform's validity indicator did not flag random or inconsistent responding, and the profile was treated as interpretable for screening purposes. The response style was guarded: by the publisher's interpretive guidance the profile showed a defensive pattern, with little acknowledgment of alcohol-related consequences on the face-valid scales. Under the Institute's guidance that pattern raises the chance of a missed case and is not evidence of dishonesty; it is recorded here as a hypothesis weighed against the other information below, not as a finding about M.L.'s truthfulness. Result category: SASSI-4 classification as reported by SASSI Online: Low Probability of a substance use disorder. This is a screening classification, not a diagnosis, and it does not establish the absence of an alcohol use disorder; the publisher states that the instrument cannot determine current drinking and that a defensive profile can produce a low result in a person who has a disorder. The classification was weighed as one input and was not treated as dispositive. Interview, records, and collateral: Clinical interview of 55 minutes with DSM-5 alcohol use disorder criteria reviewed one by one. M.L. reported drinking on two or three evenings a week, usually four to five beers, and denied any consequence other than the arrest. The arrest report documents a breath alcohol concentration of 0.16 at 11:40 p.m. on a weeknight and a prior alcohol-related driving citation seven years earlier that M.L. did not mention until asked. With his written consent, his spouse described six or more beers on most Fridays and Saturdays over the past year, two arguments about his drinking, and one evening this summer when she drove him home from a friend's house because he could not. Program intake urine drug screen on 09/17/2026: negative on all panels; breath alcohol at intake 0.000. AUDIT-C completed on paper the same day for a transparent comparison: 7 of 12, positive under the program's convention for men. No prior treatment. Conclusion, reached independently of the screen: Alcohol use disorder, moderate, per DSM-5, on the basis of the interview and collateral: criteria met for hazardous use (driving after drinking on at least two occasions in the past year), larger amounts than intended, unsuccessful attempts to cut down (two, by his account), continued use despite recurrent interpersonal conflict, and tolerance. The low-probability SASSI-4 classification is inconsistent with this evidence; the inconsistency is explained by the guarded response style and by the instrument's stated inability to measure current drinking, and the diagnosis rests on the interview, the arrest record, the collateral account, and the AUDIT-C, not on the SASSI. M.L. was told the classification, what it does and does not mean, and the basis for the diagnosis; he disagrees with the diagnosis and his objection is recorded. Recommendation, records, and plan: Recommended to the court: enrollment in this program's outpatient level of care (ASAM Criteria Level 1) with the state DUI education series, weekly individual sessions, and monitored abstinence, based on the diagnosis, the six-dimension review, and M.L.'s stated willingness to attend; the recommendation does not rest on the SASSI classification. The court report will state the edition, administration mode, time frame, validity statement, classification, and limits exactly as written above and will attach no questionnaire, profile sheet, or platform report; those remain in the testing file as test data and will be released only under a signed release or a court order, in line with the program's test-security policy. Consent for disclosure to the court under 42 CFR Part 2 signed 09/17/2026, naming the court and the probation department as recipients. No repeat SASSI is planned, consistent with the Institute's guidance against routine re-administration; progress will be tracked with the AUDIT-C at monthly reviews, breath tests, and collateral check-ins. Report to the court due 09/30/2026; first treatment session 09/24/2026. This sample is fictional and for educational purposes. It does not describe a real patient or record; the dates, results, and details are invented to show documentation structure and are not clinical guidance. No SASSI items, scale scores, cutoffs, decision rules, or profile materials are reproduced. ↑ Back to the template and downloads ### Why this sample works The edition, language, administration platform, proctor, time frame, and reading conditions are all named, so the next reader knows exactly which instrument produced the category and under what conditions. The validity review and the guarded response style are written in words as hypotheses, with the publisher's own caution that defensiveness signals a possible missed case rather than dishonesty, and no scale score or threshold leaves the testing file. The classification appears in the publisher's exact terms with its published limits in the same breath, so nobody can read Low Probability as a clean bill or a high result as proof of current use. The diagnosis rests on named evidence (a criteria-based interview, the arrest record, the collateral account, toxicology, and a transparent second screen), and the note says plainly that the screen was inconsistent with it and why. The recommendation is tied to the diagnosis and the ASAM review, the court report's contents are specified, test materials stay out of the filing, the Part 2 consent is dated, and the repeat-testing decision cites the publisher's guidance. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: mandated evaluation almost never means a mandated SASSI. Under LAW, impaired-driving statutes and administrative codes require screening and assessment and prescribe who does it and how it is reported, not the brand: New York's Vehicle and Traffic Law 1198-a orders screening with a standardized written instrument developed by the state addiction services office and a formal assessment by a qualified professional, and names no instrument; the federal Substance Abuse Professional rule at 49 CFR 40.293 requires a face-to-face clinical evaluation, professional judgment, and an individualized recommendation after a DOT violation, and names no test. Where the SASSI is required, the requirement lives in a local program, contract, or court plan: a Utah county's 2024 to 2026 substance use disorder plan, for example, directs DUI-referred clients to complete the SASSI and a biopsychosocial assessment before treatment and refers those with a high-probability result for a complete substance use assessment (CONVENTION, local). Drug-court, probation, child-welfare, and custody practice standards call for validated screening and corroboration and leave the instrument to the evaluator, and the ASAM Criteria (fourth edition, 2023) place people by a six-dimension assessment in which a screen is at most a trigger (CONVENTION). Two publisher policies bind users by contract rather than by law: the Institute's terms of use bar using the questionnaires as a screening tool for job applicants, so a post-violation SAP evaluation or an EAP referral is a different act from pre-employment testing, and its qualification form and training statements decide who may buy and interpret the instrument. Under PAYER POLICY no code names the SASSI: a brief standardized instrument is usually reported under 96127 or inside the psychological testing family (96130, 96131, and 96136 to 96139) according to what was done and by whom, and the SBIRT codes (99408, 99409, G2011, G0396, G0397, and the state Medicaid codes H0049 and H0050) require a structured assessment and a brief intervention with time; the CMS SBIRT fact sheet (May 2025) names the AUDIT and the DAST as example tools and does not mention the SASSI, so bill SASSI as SBIRT is not a safe rule. When the evaluation is court-facing, the APA forensic guidelines and Ethics Code Standards 9.04 and 9.11 govern what leaves the office: the respondent's answers, scores, and your interpretation are test data, releasable with consent or legal compulsion, while the questionnaire, key, and profile layout are test materials that stay out of the filing; a record created by a federally assisted program is also a 42 CFR Part 2 record whose disclosure to a court needs the specific consent or order the Part 2 consent page describes. The edition and the evaluator's qualifications are foreseeable cross-examination targets, so date them. As of September 2026 the current editions are the Adult SASSI-4 (2016) and the Adolescent SASSI-A3 (2020); the SASSI-3 and SASSI-A2 they replaced still circulate in old handouts and study notes, and an interpretation drawn from them has no standing for a current result. The Institute requires a qualification form before a first purchase, accepts licensed or credentialed psychologists, physicians, psychiatric nurses, addiction counselors, counselors and therapists, social workers, and rehabilitation counselors, and lets staff without professional training administer and score under appropriate supervision; Pearson lists the family at qualification level B. Institute training is publisher policy, not law: the Institute states that training is not mandatory for practitioners whose credential or professional preparation includes assessment, offers Session I (administration and scoring) and Session II (clinical interpretation), and says of Session II that it is not required but highly recommended for anyone giving feedback to clients or writing reports (CONVENTION); a report author who has taken it should say so, and one who has not should expect the question. The same training guidance warns that the greater the inference and the thinner the collateral, the more any relationship between a score pattern and a clinical syndrome is a hypothesis to investigate, which is the sentence to keep in mind when a probation officer asks whether the SASSI showed lying. The evidence base argues for the same restraint: the publisher's SASSI-4 validation (Lazowski and Geary, 2016) reports 93 percent sensitivity and about 90 percent specificity against clinician DSM-5 diagnoses in 1,245 participants, with 79 percent sensitivity for the full scoring protocol against 47 percent for the face-valid scales alone when 120 adults in recovery were told to fake good; Feldstein and Miller's 36-study review found the indirect scales' internal consistency weak and no evidence for their claimed advantage; a court-mandated offender study (Hartigan and colleagues, 2021) found the SASSI-4 vulnerable to positive impression management, with the Defensiveness scale unable to separate partial denial of substance use from a generally favorable self-presentation; and a DSM-5 reanalysis of the retired SASSI-3 (Laux and colleagues, 2016) found 75 percent sensitivity, 77 percent specificity, and only fair agreement with counselor diagnoses. Canada and Australia change nothing about the note and everything about the expectation: the SASSI is sold and used in Canada, but provincial remedial driving programs such as British Columbia's Responsible Driver Program describe an assessment process and name no instrument, no provincial rule naming the SASSI was found in the research for this page, and Australian driver-fitness and public-health guidance points to the AUDIT and the ASSIST, so a SASSI outside the United States is charted as the evaluator's choice with its rationale. SASSI, SASSI-4, and SASSI-A3 are trademarks of The SASSI Institute (Springville, Indiana), which holds the copyright in the questionnaires, items, scales, T scores, norms, profiles, score reports, and scoring keys and whose copyright notice bars reproducing, paraphrasing, or translating the questions, creating derivative products, or copying the materials into computer storage without written permission, so a clinic may chart an authorized result but may not rebuild the questionnaire or its scoring in an EHR, a form, or a public web tool without a license from the Institute. Distributors such as Pearson sell the materials under the Institute's qualification rules. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common SASSI documentation errors reviewers flag The numbers behind these errors are specific. The publisher's SASSI-4 validation ( Lazowski and Geary, 2016 ) reports 92 percent overall accuracy against clinician DSM-5 diagnoses in 1,245 participants (sensitivity 93 percent, specificity 89 percent), 95 percent in criminal justice settings and 91 percent in DWI and DOT programs, and, in a fake-good study of 120 adults in recovery told to minimize, 79 percent sensitivity for the full protocol against 47 percent for the face-valid scales alone; the SASSI-A3 validation reports 89 percent overall accuracy (sensitivity 90 percent, specificity 85 percent) in 1,065 adolescents from 19 programs (The SASSI Institute, 2020). Against that, Feldstein and Miller (2007) reviewed 36 peer-reviewed reports with a combined N of 22,110 and concluded that sensitivity was similar to public-domain instruments, that the specificity findings pointed to a high false-positive rate, and that no empirical evidence supported the claimed unique advantage of the subtle scales; Laux and colleagues (2016) found the SASSI-3 at 75 percent sensitivity and 77 percent specificity against DSM-5 diagnoses in 241 participants, with a kappa of .42; and Hartigan and colleagues (2021) found the SASSI-4 vulnerable to positive impression management in court-mandated offenders. No published audit counts how often the edition, time frame, or validity statement is missing from SASSI entries; the local program rules, records-request law, and publisher terms described under compliance considerations are what turn those omissions into a problem. The BastionGPT Clinical Advisory Board sees the same errors most often in SASSI documentation reviews: The classification charted as a diagnosis. SASSI positive for alcohol dependence, or High Probability written into the diagnosis line. The publisher's own guidance calls the result a screening outcome to be used as one piece of a diagnostic evaluation, and a person in long recovery is expected to screen high. Write the category in the publisher's words, add the limits, and put the diagnosis, if any, on its own line with its criteria-based basis. Low Probability treated as exculpatory. A low result in a DUI, probation, or custody case written as no substance problem identified, sometimes with the platform's wording carried over. A low classification does not rule out a disorder, a defensive profile raises the chance of a missed case by the publisher's own account, and the instrument cannot measure current drinking. Chart the category, the response style in words, the conflicting evidence, and the conclusion the evidence supports. Defensiveness written up as lying, or the client beat the test. The Defensiveness scale indexes reluctance to endorse personal faults; the publisher's guidance reads an elevated result as a possible missed case, and a court-mandated offender study found the scale unable to separate partial denial of use from a generally favorable self-presentation. Describe the response style as a hypothesis, weigh it against collateral, and never write that the SASSI detected deception or that the respondent passed it. Edition, time frame, or administration mode missing. SASSI administered, low probability, with no edition, no language, no platform, and no time frame. A SASSI-3 handout applied to a SASSI-4 result, a Spanish SASSI charted as a SASSI-4, or a lifetime frame and a past-12-month frame trended against each other are all uninterpretable and, in a hearing, impeachable. Name the instrument, the mode, the proctor, and the frame every time. Scale scores, the profile sheet, or the platform narrative in the record that leaves the office. T scores and cutoffs copied into the progress note, a profile sheet attached to the court report, or the SASSI Online narrative pasted in as the formulation. These are the publisher's copyrighted materials and, under the APA distinction between test data and test materials, they belong in the testing file and are released only with consent or legal compulsion; the narrative also carries inferences the Institute itself says to treat as hypotheses. Report the category and a qualitative description, and keep the rest in the file. Requirement, billing, or employment use misattributed. Required by state law when the requirement is a county plan or a court contract, a claim filed as SBIRT on the strength of a SASSI alone, or an applicant screened at hiring with an instrument whose terms bar use as a screening tool for job applicants. Write required by this program or court, bill what was actually done under the codes that fit it, and confine occupational use to post-violation SAP evaluations and EAP referrals. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on substance use assessments and screening documentation. Give it the facts (edition, language, platform or paper, proctor, time frame, reading conditions, the validity review and response style in words, the category as reported, the interview findings with DSM-5 criteria, records, collateral, toxicology, other screens, and the recommendation) and it drafts the documentation block or the report paragraph: edition and time frame named, validity stated, the category in the publisher's words with its limits, the profile as hypotheses, and the conclusion tied to the evidence, ready for your review. Cross-check a finished evaluation for the gaps reviewers and opposing counsel flag: a category with no edition or time frame, a low result read as exculpatory, defensiveness written as deception, a recommendation resting on the screen alone, or scale scores and platform text that should have stayed in the testing file. Turn the entry into the next document: the substance use section of a court report or forensic evaluation, the referral to a level of care, or the plain-language explanation of the result for the client, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What does a SASSI High Probability or Low Probability result actually mean? It is a screening classification produced by the publisher's decision rules from the combination of face-valid scales (what the respondent acknowledged about alcohol and other-drug use over the time frame the evaluator chose) and subtle or indirect scales (true-false items whose relationship to substance use is not obvious), after a validity check. High Probability means the response pattern resembles that of people who were diagnosed with a substance use disorder in the validation studies and that a diagnostic assessment is warranted; the publisher's SASSI-4 validation reports 92 percent overall accuracy against clinician DSM-5 diagnoses, which is an estimate from that sample, not the chance that a particular respondent has a disorder. Low Probability means the pattern resembles those without a diagnosis, and it does not rule a disorder out, especially when the profile is defensive. Neither result establishes current use, intoxication, impairment, or relapse: the Institute states that the instrument cannot determine whether a client is currently drinking and that someone in long recovery will more than likely still screen high. Document the category in those exact words with the edition, the time frame, the validity statement, and the limits, and let the diagnosis come from the interview and collateral. Is the SASSI a diagnosis, and can a court, employer, or caseworker treat it as one? No. The publisher describes the result as a screening outcome that does not yield a clinical diagnosis and is one piece of information in a diagnostic evaluation, and its own validation used clinicians' independent DSM-5 diagnoses as the criterion the screen was tested against. A substance use disorder diagnosis comes from a criteria-based interview with history, collateral, records, and toxicology, documented in a substance use assessment ; placement comes from the ASAM Criteria's six dimensions; a fitness or return-to-duty opinion comes from functional evidence and the governing policy. When a referral source asks for the SASSI result alone, provide the category with its limits in the same sentence and the independent conclusion beside it, and say in the report that the recommendation does not rest on the classification. The independent literature is the reason for that restraint: sensitivity similar to public-domain screens and a high false-positive rate in the 2007 review, only fair agreement with counselor diagnoses for the SASSI-3 under DSM-5, and vulnerability to positive impression management in court-mandated offenders. Is the SASSI required for a DUI evaluation, probation, a DOT SAP evaluation, or a child-welfare case? Usually not by name, and the label matters in the report. LAW requires the evaluation, the evaluator's credential, and the report format: New York's Vehicle and Traffic Law 1198-a orders screening with a standardized instrument developed by the state addiction services office and a formal assessment by a professional, naming no brand; 49 CFR 40.293 requires the DOT Substance Abuse Professional to conduct a face-to-face clinical evaluation and use professional judgment, naming no test; state administrative codes for DUI evaluation prescribe uniform reports and risk classifications. Where the SASSI is required, the source is a county plan, a court contract, or a program protocol (a Utah county's 2024 to 2026 plan requires it with a biopsychosocial assessment before DUI treatment), which is CONVENTION or local rule, so write required by this program or court rather than required by state law unless the statute says so. Child-welfare and custody practice calls for validated, corroborated screening and names nothing, and the Institute's qualification form lists child protective services as a user setting. On billing, PAYER POLICY names no SASSI code either: report what was done under 96127 or the testing family, and use the SBIRT codes only when a structured assessment and a timed brief intervention were delivered, noting that the CMS SBIRT fact sheet names the AUDIT and the DAST as examples and not the SASSI. Can the SASSI detect lying, and what do I document when a client seems to have beaten it? It cannot, and the note should never say it did. The Defensiveness scale indexes reluctance to endorse personal faults or limitations; the publisher reads an elevated result as a reason to suspect a missed case and a possible reason to re-administer once rapport improves, not as proof of deception, and a court-mandated offender study found the scale unable to separate partial denial of substance use from a generally favorable self-presentation (Hartigan and colleagues, 2021). The publisher's own fake-good study cuts both ways: with 120 adults in recovery told to minimize, the full scoring protocol kept 79 percent sensitivity against 47 percent for the face-valid scales alone, which supports using the whole instrument and does not make any scale a lie detector. When a low result conflicts with the arrest record, collateral, or toxicology, write the category, describe the response style in words as a hypothesis, list the conflicting evidence, reach the conclusion the evidence supports, and explain the inconsistency; the phrase beat the test belongs nowhere in a record that a defense attorney, a judge, or a licensing board may read. SASSI-4, SASSI-A3, Spanish SASSI, SASSI-3: which edition is current and what changed? As of September 2026 the current adult edition is the SASSI-4 (2016; ages 18 and up; fourth- to fifth-grade reading level; about 15 minutes to administer and score; paper, SASSI Online, Scantron, and audio formats) and the current adolescent edition is the SASSI-A3 (2020; ages 13 to 18; about fourth-grade reading level; paper and SASSI Online). Both were validated against DSM-5 substance use disorder diagnoses and both carry a prescription drug scale; the SASSI-A3 also has adolescent scales for friends and family risk, attitudes, a validity check, and a secondary classification. The adult Spanish SASSI is a separate instrument validated in its own sample (publisher-reported accuracy 84 percent), not a translation of the SASSI-4, and the Institute recommends against using its scale scores to develop clinical hypotheses, so a Spanish administration is charted by its classification. The SASSI-3 and the SASSI-A2 (2001) are retired; their manuals, their cutoffs, and the study notes built on them do not apply to a current result, and the Institute has announced no SASSI-5 or SASSI-A4. State the edition and the date of administration in every entry. Who can administer and interpret the SASSI, and is Institute training required? The Institute requires a completed qualification form before a first purchase and accepts licensed or credentialed psychologists, physicians, psychiatric nurses, addiction counselors, counselors and therapists, social workers, rehabilitation counselors, and doctoral researchers; Pearson lists the family at qualification level B. Staff without professional training may administer and score under appropriate supervision, and interpretation, feedback, and report writing belong to the qualified clinician. Training is publisher policy, not law: the Institute states that training is not mandatory for practitioners whose credential or professional preparation includes assessment, offers Session I on administration and scoring and Session II on clinical interpretation (3.5 continuing-education hours each), and says Session II is not required but highly recommended for anyone giving feedback or writing reports. It also runs a free clinical helpline for customers on administration, scoring, and interpretation, by telephone only. In a court-facing report, state your license, your supervision arrangement if a technician administered the questionnaire, and your SASSI training, because opposing counsel will ask. What goes in the chart, and what goes to the court, attorney, or employer when records are requested? Separate the layers. The chart entry carries the edition, mode, time frame, reading conditions, validity statement, the category in the publisher's words with its limits, a qualitative profile description labeled as hypothesis, and the independent conclusion and recommendation. The completed questionnaire, the scored profile or platform report with its scale graph, and the narrative are stored in the testing file: under APA Ethics Code Standards 9.04 and 9.11 the respondent's answers, scores, and your interpretation are test data, releasable to the client or under a proper release or court order, while the questionnaire, scoring key, and profile layout are the publisher's test materials, protected by copyright and test-security obligations, and are not attached to a court report, sent to an employer, or placed in a portal. A DUI, probation, or custody report therefore restates the category and the basis for the conclusion in prose and attaches nothing from the publisher. When the program is federally assisted, the entry is also a 42 CFR Part 2 record, so disclosure to a court or probation office needs the specific consent or order described on the Part 2 consent page, and the Institute's terms of use require users to follow Part 2 and HIPAA. The records request and subpoena response pages cover the mechanics. How does the SASSI differ from the CAGE, AUDIT-C, and DAST-10, and when should I use those instead? The CAGE , AUDIT-C , and DAST-10 are short, transparent screens that are free for clinical use with attribution and have published scoring that anyone can read; the SASSI is a longer proprietary inventory whose items, scoring, and decision rules are copyrighted and whose selling point is the indirect scales meant to flag a disorder when the respondent minimizes. That difference decides the use. In primary care, emergency, and SBIRT workflows the public tools are expected and CMS names the AUDIT and the DAST as examples; in a court-ordered, probation, child-welfare, or EAP evaluation, where minimizing is likely, the SASSI is a common choice, the publisher's fake-good study is the argument for it, and the independent literature (sensitivity similar to public-domain instruments, a high false-positive rate, and vulnerability to impression management in offenders) is the argument for pairing it with a transparent screen, collateral, records, and toxicology rather than relying on it. Whichever you use, the note names the instrument, its time frame, and its convention, and the diagnosis comes from a substance use assessment , never from a screen. Can BastionGPT draft my SASSI documentation? Yes. Give it the facts (edition, language, platform or paper, proctor, time frame, reading conditions, the validity review and response style in words, the category as reported, the interview findings with DSM-5 criteria, records, collateral, toxicology, other screens, and the recommendation) and it drafts the documentation block or the report paragraph: edition and time frame named, validity stated, the category in the publisher's words with its limits, the profile as hypotheses, and the conclusion and recommendation tied to the evidence, ready for your review. It can also check a finished evaluation for a category with no edition or time frame, a low result read as exculpatory, defensiveness written as deception, a recommendation resting on the screen alone, or scale scores and platform text that belong in the testing file. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: The SASSI Institute, accessed September 2026: Adult SASSI-4 (ages, reading level, time, formats, accuracy claim) and products (SASSI-A3 and Spanish SASSI); qualification form ; training FAQ and SASSI Use and Training Requirements (July 7, 2022); clinical FAQs (screen, not diagnosis; current use; recovery; re-administration); clinical support helpline ; SASSI Online ; copyright notice and terms of use (no reproduction, paraphrase, translation, or computer storage; no use as a screening tool for job applicants; Part 2 and HIPAA). Pearson, Substance Abuse Subtle Screening Inventories (SASSI-4 2016, SASSI-A3 2020; qualification level B; published scale names). Publisher validation: Lazowski LE, Geary BB, Validation of the Adult Substance Abuse Subtle Screening Inventory-4 (SASSI-4) , European Journal of Psychological Assessment (advance online 2016; 2019;35(1):86-97) and the Institute's SASSI-4 reliability and criterion validity summary ; Tiburcio NJ, Baker SL, 2020, Estimates of the reliability and criterion validity of the Adolescent SASSI-A3 , The SASSI Institute; Adult SASSI-4 Guidelines (profile validity review; literacy and misunderstanding as sources of an unusable profile; collateral evidence). Independent reviews: Feldstein SW, Miller WR, 2007, Addiction, Does subtle screening for substance abuse work? A review of the SASSI (36 reports, N of 22,110); Lazowski LE, Miller GA, 2007, Addiction, SASSI: a reply to the critique of Feldstein and Miller ; Erford BT, Atalay Z, Bardhoshi G, 2020, Measurement and Evaluation in Counseling and Development, systematic review of psychometric characteristics of the SASSI-3 ; Laux JM and colleagues, 2016, The Professional Counselor, assessing the accuracy of the SASSI-3 using DSM-5 criteria (241 participants; sensitivity .75, specificity .77, kappa .42). Forensic use and test security: Hartigan SE, Rogers R, Williams MM, Donson JE, 2021, Journal of Psychopathology and Behavioral Assessment, Challenges for the SASSI-4 and InDUC-2R: positive impression management in offenders with substance use histories ; American Psychological Association, Specialty Guidelines for Forensic Psychology and Ethical Principles of Psychologists and Code of Conduct (Standards 9.04 and 9.11, test data and test materials). United States law and program rules: 49 CFR 40.293 (Substance Abuse Professional evaluation; no instrument named); New York State Senate, Vehicle and Traffic Law 1198-a (screening with a standardized instrument developed by the state addiction services office; no instrument named); Summit County, Utah, substance use disorder treatment plan 2024 to 2026 (SASSI and biopsychosocial assessment before DUI treatment); American Society of Addiction Medicine, The ASAM Criteria (fourth edition, 2023; six dimensions). Payer and Part 2: CMS Medicare Learning Network, May 2025, SBIRT services fact sheet (G2011, G0396, G0397; AUDIT and DAST as example tools; bare code numbers only); US Department of Health and Human Services, 2024, 42 CFR Part 2 final rule fact sheet ; eCFR, 42 CFR 2.11 definitions . Canada and Australia: Province of British Columbia, Responsible Driver Program (assessment process; no instrument named); Austroads, Assessing Fitness to Drive (substance misuse guidance; AUDIT named). ## Related templates CAGE Questionnaire Documentation AUDIT-C Score Documentation DAST-10 Score Documentation Substance Use Assessment Court Report & Treatment Progress Update Forensic Psychological Evaluation Report Biopsychosocial Assessment Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/dast-10 All note templates # DAST-10 Documentation: Scoring, Interpretation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The DAST-10 (Drug Abuse Screening Test) is a ten-item, yes-or-no self-report screen for problems related to drugs other than alcohol over the past 12 months, derived from Harvey Skinner's 1982 DAST and published by CAMH. Primary care, emergency, and behavioral health clinicians use it to decide who needs a substance use assessment; it never diagnoses. This page covers how to document and interpret DAST-10 results, with a fictional sample note. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Patient self-report on paper or a tablet, or asked by a clinician as an interview; scored and interpreted by the treating clinician under the qualified-professional oversight the CAMH guide asks for; no publisher qualification level applies Audience Primary care and emergency physicians, nurses, addiction and behavioral health clinicians, hospital and trauma-center SBIRT teams, corrections and workplace programs, payers and auditors reviewing screening and brief-intervention claims Typical length 3 to 6 chart lines (version and time frame, medication handling, total and topics endorsed, substances named, follow-up, intervention) · patient completion about 5 minutes Format family Brief case-finding screen (10 yes-or-no items about drugs other than alcohol, one point per answer in the problem direction, total 0 to 10, standard past-12-month window) When it's used Primary care and wellness visits after a positive single-question screen, emergency and trauma-center screening, addiction and behavioral health intakes, treatment follow-up on a documented shorter window, workplace, corrections, and research settings Standards context Skinner 1982 (Addictive Behaviors) with the CAMH Guide for Using the DAST (revised 2023); copyright Skinner and CAMH, free non-commercial reproduction with attribution and digital licensing through CAMH; USPSTF 2020 grade B lists the DAST among example tools and requires none; described here for documentation, no items reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the DAST-10? The Drug Abuse Screening Test was published by Harvey A. Skinner in Addictive Behaviors in 1982 as a 28-item self-report scale modeled on the Michigan Alcoholism Screening Test, and that paper also derived a 20-item form; the ten-item DAST-10 was built later from the items that discriminated best, and the author's guide reports it correlating .98 with the DAST-20. The instrument asks ten yes-or-no questions about involvement with drugs other than alcohol, and its standard instruction block sets the past 12 months as the window, defines drug use for scoring purposes as prescribed or over-the-counter medication taken beyond its directions plus any non-medical use of drugs, and lists cannabis, sedatives, stimulants, hallucinogens, solvents, and opioids as example classes. At the topic level the items cover non-medical use and using more than one drug at a time, control over use, memory and perceptual effects, guilt, complaints from family and neglect of responsibilities, illegal activity connected with obtaining drugs, withdrawal, and medical harm; one point is scored for each answer in the problem direction, one item is keyed in the opposite direction, and the total runs 0 to 10. It takes about five minutes, works as a self-report or a structured interview, is published and sold by the Centre for Addiction and Mental Health (CAMH) in Toronto in English and French, and carries LOINC panel 82666-9 with total score 82667-7. The DAST-20 remains available for fuller clinical assessment, the DAST-A (Martino, Grilo, and Fehon, 2000) is the adolescent adaptation, and a DAST-R with less stigmatizing wording was described as under development in the 2023 guide. Three facts carry the documentation. First, the familiar severity bands are not in the 1982 paper: they come from Skinner's CAMH-published Guide for Using the DAST, which introduces them as tentative guidelines, and the author's suggested case-finding threshold is a total of 3 or more, offered in the same tentative spirit pending research on alternative cut points; a synthesis of 346 DAST studies published online in 2024 (Johnson and colleagues, Measurement and Evaluation in Counseling and Development) reported pooled internal consistency around .81 to .84 and an aggregated diagnostic optimum of 7 for the DAST-10, which is the clearest evidence that a positive screen is a case-finding result and not a diagnosis, and that DSM-5 criteria belong in a substance use assessment . Second, the time frame decides the meaning: the standard instruction is the past 12 months, the guide says a screen used that way should not simply be repeated for another 12 months, and a follow-up administration reworded to six months (which the guide permits) is a modified instrument that must be charted as such. Third, the score is silent about the drugs: CAMH's own description says the DAST obtains no information about the types of drugs used, their frequency, or the duration of use, and the prescription-medication rule lives in the instruction block that many EHR builds and web reproductions drop, so the note has to add the substances and say how prescribed medication was handled. The CAGE and its CAGE-AID variant are the four-item consequence screens, the AUDIT-C covers alcohol, which the DAST excludes, and the WHO ASSIST and the TAPS tool are the substance-specific alternatives when the clinician needs to know which drug drives the risk. ## Who uses DAST-10 documentation and when The DAST-10 sits wherever a positive first question about drug use needs a structured second step. In US primary care the USPSTF's 2020 recommendation to ask adults about unhealthy drug use (grade B, wherever assessment and treatment can be offered or referred) has put a single question or the NIDA Quick Screen at the front of the visit and the DAST-10 behind it, so family physicians, internists, nurse practitioners, and embedded behavioral health consultants chart it at wellness visits and after a positive quick screen; CMS's SBIRT guidance names the AUDIT and the DAST as example structured assessment tools, which is why the same entry often supports a claim. Emergency departments and trauma centers use it when a fast drug-problem index fits the bedside, with the caution that it is not valid in a patient who is intoxicated or withdrawing. Addiction and behavioral health programs record it at the intake beside the substance use assessment and biopsychosocial assessment , and repeat it during treatment on a documented shorter window as a monitoring measure. Workplace, corrections, and court-linked programs use it too, where self-report validity is the documentation problem and instruments built for defensive responding, such as the SASSI , or collateral and toxicology data carry more weight. In Canada CAMH publishes the instrument and asks clinicians to screen every patient aged 10 and over for substance use and record the result in the cumulative patient profile; in Australia the RACGP frames routine or opportunistic screening inside the 5As and points general practice to the ASSIST-Lite rather than the DAST. Neighbors win in three places: the AUDIT-C for alcohol, which the DAST-10 excludes; the CAGE-AID when a four-item conjoint alcohol-and-drug probe is all the encounter allows; and the COWS or CIWA-Ar when the question is withdrawal severity, which no DAST answer can grade. ## How to document DAST-10 results in the chart No law, payer, or author prescribes a DAST-10 note format; what exists is an instruction block and a guide from the author, a copyright line that asks for attribution, and payer and accreditation rules that ask for a named structured tool, the result, the time spent, and an intervention. What survives review is an entry that names the version and the window actually used, says how prescribed medication and cannabis were handled, records the mode and the fitness to complete, charts the total with the topics endorsed, lists the substances in the patient's words, labels the result a screen with its band attributed and tentative, and documents the follow-up, the intervention with its minutes, and where the record lives. Each element below carries the pitfall that most often undermines it. Version and time frame. Write DAST-10, DAST-20, or DAST-A, and the window the patient was told to consider: the standard past 12 months, or the modified window actually used and why (the guide permits a shorter window, six months for example, for follow-up after treatment). A lifetime or 30-day adaptation is a different instrument with no demonstrated equivalence to the 12-month reading, so its band language does not transfer. When the score is a repeat, name the prior date and window so the two totals are compared on the same terms or not compared at all. Pitfall: "DAST 4" with no version or window, or a six-month follow-up total trended against a 12-month intake baseline as if the two were the same measure. Instruction block and prescribed medication. Record that the full instruction text was presented or read aloud, because it carries the definition that decides what counted: medication taken as prescribed, or as directed on the package, does not count; use beyond the directions, another person's prescription, a different route, and any non-medical use do. Cannabis counts whenever the use is non-medical, regardless of state or provincial law, and medically authorized cannabis is charted with how the patient interpreted it. When the instruction block was cut from the form or the EHR build, say so and say how the patient interpreted the items. Pitfall: Cannabis dropped because it is legal, or a prescribed opioid or stimulant counted (or excluded) with nothing in the note saying which and why. Mode, language, completeness, and fitness. Chart self-report (paper, tablet, portal) or clinician interview, the language or translation used (CAMH sells English and French; other adaptations vary in validation and may not carry the English cut points), that all ten items were answered, and who computed the total. Add that the patient was not intoxicated or in withdrawal at the time, because the guide says the DAST should not be given in either state; an emergency-department result taken from an impaired patient is charted with that limitation. Pitfall: A score taken from an intoxicated patient after an injury charted like a routine screen, or a translation charted as plain DAST-10 with the English threshold applied. Total, then the topics endorsed. Chart the total out of 10 and, in your own words, the topic areas the patient endorsed in the problem direction (a partner's complaints, control over use, withdrawal, medical harm, and so on), because the guide itself directs the reader to examine item responses for specific problem areas and the topics decide what the follow-up looks at. Score with the key on the licensed form: one item is keyed in the opposite direction, and reproductions circulate with reworded items and mismatched keys. Pitfall: A bare total, a key copied from an unlicensed PDF, or DAST-20 interpretation applied to a DAST-10 total. Substances named, in the patient's words. The instrument yields no drug type, frequency, or duration, so add them: each substance, route, typical amount and frequency, last use, and source (own prescription, someone else's, illicit supply), plus injection use, overdose history, and current medications. Screen alcohol separately (the AUDIT-C or the full AUDIT) and say so, since the DAST excludes it; the substance line is what the assessment, the intervention, and any later comparison rest on. Pitfall: A positive screen with no substance named, so the next reader cannot tell weekend cannabis from daily non-prescribed fentanyl. Interpretation: a screen, with a tentative band. Write the conclusion as a screening result (positive screen for further assessment; substance use disorder not established) and, when you use a band, attribute it: totals of 1 to 2 are described in the CAMH guide as a low level of problems, 3 to 5 as intermediate (moderate on older reprints), 6 to 8 as substantial, and 9 to 10 as severe, with 3 or more the author's suggested case-finding threshold, all labeled tentative by the author. Name the population caveat when it changes the reading: psychiatric populations have validated lower cut points, pregnancy studies show low sensitivity above the lowest cut point, adolescents have their own version and an insufficient-evidence rating from the USPSTF, and self-report can be shaped by legal, employment, or custody stakes. Pitfall: "DAST-10 = 7, substance use disorder" with no criteria assessed, or a zero in a court-referred patient charted as no problem. Follow-up, intervention, minutes, and record. Name what the result triggered and when: a full substance use assessment with DSM-5 criteria, a withdrawal assessment when current heavy use raises that risk, toxicology only if it changes management (screening means asking questions, not testing specimens). Document the brief intervention (feedback, the patient's goal, options offered), the referral made or declined, and the face-to-face minutes for any time-based code. State the rescreen plan and where the result lives: a primary-care DAST-10 is a HIPAA record; the same result created by a federally assisted Part 2 program carries Part 2 consent and redisclosure rules. A refusal is charted as a refusal, never as a zero. Pitfall: "Counseled on drug use" with no content, goal, or minutes, or a refused screen entered as 0 of 10. ### Blank template (copy and adapt) DAST-10 DOCUMENTATION BLOCK Date: [ ] Setting: [ ] Clinician: [ ] Encounter: [wellness / intake / ED or trauma / treatment follow-up / other] Version: [DAST-10 / DAST-20 / DAST-A] Prompted by: [single-question screen / routine / referral] Time frame: [standard past 12 months / modified: state the window and why] Instruction block: [full text presented / read aloud / omitted: state how items were interpreted] Prescribed medication as directed: [excluded; named] Cannabis: [non-medical, counted / medically authorized, how interpreted] Mode: [self-report: paper, tablet, portal / interview] Language: [ ] Completeness: [all 10 answered / missing: state] Total computed by: [ ] Fitness: [not intoxicated or in withdrawal at completion / limitation noted] Total: [ ] of 10 Topics endorsed (own words): [ ] Substances (patient's words): [substance, route, amount, frequency, last use, source; injection use; overdose history] Alcohol screened by: [AUDIT-C / AUDIT; result] Interpretation: [positive / negative screen for further assessment; band per the CAMH guide, tentative; disorder not established; population or self-report caveat, or none] Follow-up triggered: [substance use assessment with DSM-5 criteria / withdrawal assessment / toxicology if it changes management; date] Intervention: [feedback given; patient's goal; referral made, declined, or not indicated; face-to-face minutes] Plan: [rescreen window and date; treatment or counseling; next visit] Record: [code claimed (bare number); storage location; Part 2 status if created by a Part 2 program; consent on file] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the questionnaire items. ## Sample DAST-10 documentation (fictional) Scenario: a US family medicine visit in a clinic with an embedded behavioral health consultant, where a yes to the practice's single-question drug screen leads to a DAST-10 on the clinic tablet; the physician records the window and the medication rule, names the substances, keeps the result a screen, delivers a brief intervention, books the full assessment, and records the claim and the record status. All details are fictional. Patient: D.K., 31 · Setting: Family medicine, integrated behavioral health clinic · Clinician: P. Nakamura, DO · Note date: 09/15/2026 Screen: DAST-10, adult self-report version with the full instruction block (drugs other than alcohol; standard past-12-month frame; prescribed and over-the-counter medication counted only when taken beyond its directions or non-medically), completed in English on the clinic tablet on 09/15/2026 after D.K. answered yes to the practice's single-question drug screen at today's visit; all ten items answered; total computed by the EHR and confirmed by hand. Total 4 of 10. The topics endorsed in the problem direction (using more than one drug at a time, guilt, a partner's complaints, and control over use) were reviewed with D.K. and match the history below. No prior DAST on file. Alert and oriented, speech and gait normal; not intoxicated or in withdrawal at completion. Substances, in his words: Cannabis, smoked most evenings for about a year, one or two bowls, bought from a licensed dispensary and used to unwind, no medical authorization; alprazolam from a friend, about six times in the past three months, one tablet at a time to sleep after late shifts, last taken about ten days ago; no opioids, stimulants, or hallucinogens; never injected; no overdose history. Takes sertraline 50 mg daily as prescribed for two years, which was excluded from scoring under the instrument's definition when D.K. asked whether it counted. Alcohol screened separately today with the AUDIT-C: 2 of 12, negative under the practice convention. Tobacco: none. Interpretation: Positive screen for further assessment. On the standard past-12-month frame, a total of 4 sits in the range the CAMH guide's tentative interpretation guidelines label intermediate (moderate on older reprints) and meets the author's suggested case-finding threshold of 3 or more; the guide calls both tentative, and neither is a diagnosis. DSM-5 substance use disorder criteria were not formally assessed today. The instrument collects no drug type, frequency, or duration, which is why the substances above are recorded in D.K.'s words. The non-prescribed alprazolam is the clinical priority (unknown supply, sedative use alongside nightly cannabis); cannabis is counted because the use is non-medical, and its legal status does not change the screen. No population caveat applies (31-year-old man, English, original wording, full instructions). Follow-up assessment: Full substance use assessment booked with the clinic's behavioral health consultant for 09/22/2026 to review DSM-5 criteria for cannabis and sedative use, pattern and consequences, sleep, and D.K.'s goals. Withdrawal risk judged low today (intermittent benzodiazepine use, no daily use, no prior withdrawal); CIWA-Ar and COWS not indicated. Urine toxicology not ordered because it would not change today's management. Brief intervention: Delivered by the physician: feedback on what the screen does and does not show, the risk of counterfeit tablets and of combining a sedative with nightly cannabis, and a menu of options. D.K. chose a goal of no alprazolam from any source starting today and agreed to bring the sleep problem to the assessment visit; referral accepted. Face-to-face time for the structured screen and brief intervention: 20 minutes. Plan and record: Assessment visit 09/22/2026; physician follow-up in four weeks to review the goal, sleep, and the assessment findings, with the sertraline dose reviewed then. Rescreen with the DAST-10 on the standard frame no sooner than 12 months, or on a documented shorter window if the consultant uses it to monitor treatment. Screening and brief intervention claimed under 99408 (commercial plan), supported by the named tool, result, intervention content, and time above. Result recorded in the general medical record; this clinic is not a Part 2 program, so 42 CFR Part 2 does not attach to this entry and the internal referral needs no Part 2 consent. D.K. was told the screen is not a diagnosis and agreed with the plan. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. ↑ Back to the template and downloads ### Why this sample works The version, the full instruction block, and the standard 12-month window are named, so the total can be reproduced and a later repeat can be compared on the same terms or explicitly not compared. How prescribed sertraline, non-prescribed alprazolam, and legal-state cannabis were each handled is written down, which resolves the ambiguity the ten items alone leave open. The substances, route, frequency, source, and last use appear in the patient's words on their own line, supplying exactly what the instrument cannot collect. The result is labeled a screen, the band and the case-finding threshold are attributed to the CAMH guide as tentative, criteria are left to the booked assessment, and the withdrawal and toxicology decisions are recorded with their reasons. The brief intervention has content, a patient-chosen goal, a referral decision, and separate minutes, and the record status under Part 2 is stated, so the 99408 claim and the next visit are both supported. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the instrument is a convention; the rules around it are real. No federal or state LAW names the DAST-10 or prescribes a screening note format. The USPSTF recommendation of June 2020 (grade B; CONVENTION, with payer and accreditation teeth) covers screening adults 18 and older, including pregnant and postpartum persons, by asking questions about unhealthy drug use wherever accurate diagnosis, effective treatment, and appropriate care can be offered or referred; it gives adolescents 12 to 17 an insufficient-evidence statement, defines screening as asking questions rather than testing urine, saliva, or blood, states that screening tools are not meant to diagnose drug use disorders, and lists the DAST among several example tools (single questions, the NIDA Quick Screen, the ASSIST, the TAPS) without requiring any of them; a note on the USPSTF site says both recommendations are being updated, with no final statement published as of September 2026. Under PAYER POLICY the SBIRT codes are time-based: Medicare pays G2011 for 5 to 14 minutes, G0396 for 15 to 30 minutes, and G0397 for more than 30 minutes of structured assessment and brief intervention, commercial plans use 99408 and 99409, and state Medicaid programs may use H0049 and H0050 under their own documentation policies; the CMS SBIRT fact sheet (May 2025) names the AUDIT and the DAST as example structured assessment tools and states that the medical record must support the claim, so a bare "DAST-10 = 4" supports none of these codes and a documented tool, result, intervention, and time supports all of them. No MIPS or Joint Commission measure names the DAST-10; measure 431 and the hospital SUB measure set concern alcohol, and the American College of Surgeons trauma verification standards (2022) require alcohol screening with a documented brief intervention while treating drug screening as encouraged practice, so a hospital's drug-screening workflow is CONVENTION or local policy rather than an accreditation mandate. Under LAW, 42 CFR Part 2 as amended by the February 2024 final rule (effective April 16, 2024; compliance date February 16, 2026, now passed) attaches to records created or received by a federally assisted Part 2 program, not to the word drug: a DAST-10 charted by a primary care or emergency clinician outside a Part 2 program is a HIPAA record, the same result recorded at intake in a Part 2 program is a substance use disorder record with its own consent and redisclosure rules, and the question to ask is where the record originated and under what program and consent, not whether it mentions drugs; the Part 2 consent page covers the mechanics. Canada and Australia: no requirement, and the home institution's own documentation rule. In Canada, no federal or provincial LAW or payer names the DAST-10. CAMH, which publishes and sells the DAST-10 and DAST-20 in English and French, advises in its clinical guidance that every patient aged 10 and over be screened for substance use with the result recorded in the cumulative patient profile, that all use of cannabis and illegal drugs be followed by a brief assessment, and that daily and weekly use be documented for positive responses (CONVENTION); its instrument page states that the DAST collects no information on drug type, frequency, or duration. In Australia, the RACGP's alcohol and other drug guidance (updated June 2026; CONVENTION) frames routine or opportunistic screening inside the 5As, names the AUDIT-C for alcohol and the ASSIST-Lite and its online version for other drugs, and does not name the DAST-10; no MBS item names it either, and screening sits inside attendance and health-assessment items. In every jurisdiction the boundary is oversight rather than a qualification level: the CAMH guide says administration, scoring, and interpretation need oversight by a qualified professional and that the DAST is not designed for self-screening by the general public, so a patient-portal or EHR administration is fine when a clinician scores and interprets it, and a public self-scoring web tool is not. The validity limits are population-specific and belong in the note: a cut point of 2 was identified in 162 psychiatric outpatients with severe mental illness (Maisto and colleagues, 2000); among 1,349 psychiatric inpatients in India, 59 percent of those above the DAST-10 cut point had no substance use disorder at discharge (Carey, Carey, and Chandra, 2003); in 1,085 pregnant women sensitivity fell from 79 to 33 percent as the cut point rose from 1 to 3 (Lam and colleagues, 2015), which is why pregnancy-specific approaches are preferred in prenatal care; the adolescent DAST-A has its own validation and cut point (Martino, Grilo, and Fehon, 2000); and the guide's own limitations section notes that the item content is transparent enough for a respondent to fake, which is the reason a low score from a court-referred, employment-referred, or custody-involved patient is charted as a screen under those conditions, with collateral history, toxicology where clinically indicated, or an instrument built for defensive responding such as the SASSI considered rather than a clean bill written from the total. A translated DAST-10 is charted with its language, because the guide asks adapters to develop local cut points rather than import the English ones. The DAST-10 is copyrighted by Harvey A. Skinner and the Centre for Addiction and Mental Health, Toronto, and made available under a permission line that allows reproduction free of charge for non-commercial clinical, research, and training use with credit to the author; the 2023 guide asks users to contact the author or CAMH about permission, CAMH sells the printed forms and routes digital and EHR licensing through its publications office, and no current source grants blanket commercial or public-web rights, so an EHR vendor build, a paid app, or a public self-scoring tool should clear rights with CAMH first. The correct citation is Skinner HA, Addictive Behaviors, 1982, volume 7, pages 363 to 371, with credit to Dr. Harvey A. Skinner, York University, and CAMH. BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common DAST-10 documentation errors reviewers flag The numbers behind these errors are specific. Across the studies reviewed by Yudko, Lozhkina, and Fouts (2007) , DAST-10 sensitivity ranged from 41 to 95 percent and specificity from 68 to 99 percent depending on the cut point and the population, with the 41 percent coming from 1,349 psychiatric inpatients in India (Carey, Carey, and Chandra, 2003); a single question about past-year drug use was 100 percent sensitive and 73.5 percent specific for a current drug use disorder in 286 primary care patients and performed similarly to the DAST-10 (Smith and colleagues, 2010); a cut point of 2 fit 162 outpatients with severe mental illness (Maisto and colleagues, 2000); in 1,085 pregnant women sensitivity ran from 79 to 33 percent across cut points of 1 to 3 (Lam and colleagues, 2015); and a synthesis of 346 DAST articles reported pooled internal consistency of about .81 to .84 and an aggregated diagnostic optimum of 7 for the DAST-10 (Johnson and colleagues, 2024), against the author's tentative case-finding cut of 3. No published audit counts how often the version, the window, or the medication rule is missing from DAST-10 chart entries; the SBIRT and Part 2 rules described under compliance considerations are what turn those omissions into a claim or disclosure problem. The BastionGPT Clinical Advisory Board sees the same errors most often in DAST-10 documentation reviews: Bands charted as official, or a score charted as a diagnosis. "DAST-10 = 7, substance use disorder," or a chart macro that prints the severity bands as validated cut-offs. The bands are tentative guidelines from the CAMH guide, not the 1982 paper, the case-finding cut of 3 is the author's suggestion pending further research, and a pooled diagnostic optimum of 7 shows how far apart case finding and diagnosis sit. Write "positive screen; disorder not established," attribute the band, and assess DSM-5 criteria separately. Time frame missing or silently changed. A six-month follow-up total trended against a 12-month intake score, a lifetime adaptation read against the 12-month bands, or a repeat given three months later with no window stated. Name the window every time, label any modification, and compare totals only on the same window; the guide permits shorter windows for follow-up, not silent ones. Prescribed medication and cannabis handled without a word. A prescribed opioid or stimulant counted, or excluded, with nothing in the note saying so, or cannabis dropped from scoring because the state legalized it. The instruction block counts medication only when used beyond its directions or non-medically and counts cannabis whenever the use is non-medical; chart the handling whenever it affected an answer, and say whether the full instruction text was presented. A score with no substances. The DAST-10 obtains no information on drug type, frequency, or duration, so "DAST-10 = 5, referred" cannot tell the next reader what was used, how, or how often, and cannot support the assessment or the intervention. Record each substance, route, amount, frequency, source, and last use in the patient's words, and screen alcohol separately. Wrong key, wrong version, wrong population. DAST-20 interpretation applied to a ten-item total, a scoring key copied from an unofficial PDF whose wording and key do not match, an English cut point applied to a translation, or a low score in a pregnant, psychiatric, or court-referred patient read as no problem. Use the licensed form's key, name the version and language, and write the population or self-report caveat into the interpretation. Refusal, follow-up, intervention, or minutes undocumented. A refused screen entered as 0 of 10, a positive screen with no assessment named, or "counseled on drug use" with no content, goal, or minutes. Chart a refusal as a refusal with the history that was discussed, name the follow-up assessment and its date, and record the feedback, the patient's goal, the referral decision, and the face-to-face time; a Part 2 program result filed as an ordinary record is the same error on the disclosure side. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on substance use assessments and screening documentation. Give it the facts (version and window, whether the full instruction text was presented and how prescribed medication and cannabis were handled, mode and language, the total and topics endorsed, the substances in the patient's words, any assessment or withdrawal check and its date, the intervention and its minutes, and where the result is stored) and it drafts the documentation block: the screen result with the band attributed as tentative, the substance line, a screening-level interpretation with the population caveat, the follow-up, the intervention, the plan, and the record placement, ready for your review. Cross-check a finished note for the gaps reviewers flag: a missing version or window, medication or cannabis handling left unstated, a total with no substances or topics, a band charted as a diagnosis, a refusal entered as zero, or an intervention with no content or minutes. Turn the screen into the next document: the opening of a substance use assessment, the referral to an addiction program, or the SBIRT documentation a payer expects, ready to confirm against the record. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the DAST-10 scored, and what score is a positive screen? Each of the ten items is answered yes or no and one point is scored for each answer in the problem direction, with one item keyed in the opposite direction, so the total runs 0 to 10. There is no positive threshold inside the instrument. The CAMH-published Guide for Using the DAST suggests a total of 3 or more for case finding on the DAST-10 and describes its severity bands (0 none reported, 1 to 2 low, 3 to 5 intermediate or moderate, 6 to 8 substantial, 9 to 10 severe) as tentative guidelines pending further research on diagnostic validity; other cut points have been validated in particular populations (2 in outpatients with severe mental illness, 1 in a pregnancy study), and a 2024 synthesis of 346 DAST studies reported an aggregated diagnostic optimum of 7 for the DAST-10. Chart the total, the topics endorsed in your own words, the convention you applied with its source, and the conclusion as a screening result; never chart a band as a diagnosis, and never apply DAST-20 numbers to a DAST-10 total. DAST-10, DAST-20, DAST-A, or a different screen: which one, when? The DAST-10 is the brief case-finding form for primary care, emergency, workplace, and corrections settings; the DAST-20 keeps ten more items for fuller clinical assessment and research and is read on its own bands, which do not transfer to the ten-item total; the DAST-A (Martino, Grilo, and Fehon, 2000) is the adolescent adaptation with its own validation and cut point, and the USPSTF still rates the evidence for screening adolescents as insufficient. Reach past the DAST family when the encounter needs something else: the AUDIT-C or full AUDIT for alcohol, which every DAST excludes; the CAGE-AID when a four-item conjoint alcohol-and-drug probe is all the bedside allows; the WHO ASSIST or the TAPS tool when substance-specific risk (tobacco, alcohol, cannabis, stimulants, opioids) matters more than one aggregate problem score, which is also why Australian general practice guidance names the ASSIST-Lite; and the SASSI in court or employment contexts where defensive responding is expected. Whichever you use, the chart names it, its window, and its convention. Do prescribed opioids, stimulants, or benzodiazepines count on the DAST-10, and does cannabis count in a legal state? Prescribed medication taken as directed does not count; use beyond the directions, larger amounts, a different route, someone else's prescription, and any non-medical use do. That definition sits in the instrument's instruction block, not in the ten items, and the longer DAST's separate prescription-misuse item is absent from the DAST-10, which is why a form or EHR build that shows only the ten questions leaves the patient to guess whether a prescribed opioid or stimulant counts. Cannabis counts whenever the use is non-medical, regardless of state or provincial law, because it is a named example class in the instructions; medically authorized cannabis is the one case the legacy wording does not cleanly resolve, so chart how the patient interpreted it. Write the handling into the note whenever it affected an answer: "prescribed lisdexamfetamine as directed excluded; extra evening doses counted" is the shape of the sentence. Is a positive DAST-10 a diagnosis of substance use disorder? No. It is a positive case-finding screen that raises the probability of a drug use problem and says nothing about which drug, how often, or whether DSM-5 criteria are met; the USPSTF states that screening tools are not meant to diagnose drug use disorders, and the CAMH guide warns that a numerical score invites misinterpretation. The defensible chain in the chart is: the screen result under a named, attributed convention; the substances in the patient's words; a criteria-based substance use assessment ; a withdrawal assessment ( COWS or CIWA-Ar ) when current heavy use raises that risk; toxicology only when it changes management, because screening means asking questions rather than testing specimens; and then the diagnosis, if any, in its own line. Some state forms label high totals "probable substance use disorder"; chart the screen result and the assessment separately rather than importing that label. Can I bill SBIRT on a DAST-10? Not on the score alone. The SBIRT codes are PAYER POLICY and time-based: Medicare pays G2011 for 5 to 14 minutes, G0396 for 15 to 30 minutes, and G0397 for more than 30 minutes of structured assessment and brief intervention; commercial plans use 99408 (15 to 30 minutes) and 99409 (more than 30); state Medicaid programs may adopt H0049 and H0050 under their own documentation rules. The CMS SBIRT fact sheet (May 2025) names the AUDIT and the DAST as example structured assessment tools, so the DAST-10 qualifies as the tool; the record still has to show the result, the intervention with its content, and the face-to-face time, and the fact sheet states that the medical record must support the claim. Medicare's annual alcohol screen (G0442) and its counseling code (G0443) are alcohol benefits and do not apply to a drug screen. Minutes are the minutes actually spent face to face: a DAST-10 completed on a tablet in the waiting room supports the tool and the result, not the time. Does a DAST-10 result fall under 42 CFR Part 2? Only when the record is created or received by a federally assisted Part 2 program. Part 2 follows the source of the record, not the subject: a DAST-10 charted by a primary care or emergency clinician outside a Part 2 program is a HIPAA record and nothing more, while the same result recorded at intake in a Part 2 program is a substance use disorder record with Part 2 consent and redisclosure rules, and a copy received from the program carries the protection with it. The February 2024 final rule (effective April 16, 2024; compliance date February 16, 2026, now passed) aligned Part 2 with HIPAA on breach notification and enforcement, allowed a single consent for treatment, payment, and health care operations, and kept special protection against use of records in legal proceedings; it does not require a drug-screen result to be walled off in a separate part of the chart. Document where the result lives and, in a Part 2 program, which consent is on file; the Part 2 consent page covers the forms and the workflow. How reliable is a self-reported DAST-10 in legal, employment, custody, or pregnancy contexts? As reliable as the respondent's reason to answer honestly, which is the guide's own caveat: the item content is transparent, so results can be faked, and a numerical score can be given too much weight. In a court-referred, employment-referred, or custody-involved patient a total of 0 documents that no problem was reported under those conditions, not that none exists, so write the conditions into the interpretation and add collateral history, prior records, toxicology where clinically indicated, or an instrument designed for defensive responding such as the SASSI . In pregnancy the direction of error reverses: sensitivity fell from 79 to 33 percent as the cut point rose from 1 to 3 in 1,085 pregnant women (Lam and colleagues, 2015), and the USPSTF discusses pregnancy-specific instruments, so prenatal services usually document a pregnancy-oriented screen instead and say why. Psychiatric populations validate lower cut points, adolescents have their own version, and a translated DAST-10 does not inherit the English threshold; the note names the population and the convention in every one of these cases. How often should I rescreen, and can I trend the score? The standard instruction covers the past 12 months, and the CAMH guide says a screen used that way should wait at least another 12 months before it is repeated; the USPSTF sets no interval, so an annual DAST-10 is a workflow convention rather than an evidence-based interval. For treatment follow-up the guide permits altering the instructions to a shorter window, six months after discharge for example, and that is a modified administration: chart the window, record the current substances beside the total, and compare only totals taken on the same window, because a 12-month intake score and a six-month follow-up score are different measures. Trending is legitimate as a description ("7 of 10 at intake on the 12-month frame; 3 of 10 at six months on a six-month frame, not directly comparable"), never as a change score, since no validated change threshold exists. Serial results follow the conventions of the outcome measure note . Can BastionGPT draft my DAST-10 documentation? Yes. Give it the facts (version and window, whether the full instruction text was presented and how prescribed medication and cannabis were handled, mode and language, the total and topics endorsed, the substances in the patient's words, any assessment or withdrawal check and its date, the intervention and its minutes, and where the result is stored) and it drafts the full entry: the screen result with the band attributed as tentative, the substance line, a screening-level interpretation with the population caveat, the follow-up, the intervention, the plan, and the record placement, ready for your review. It can also check a finished note for a missing version or window, unstated medication or cannabis handling, a total with no substances, a band charted as a diagnosis, a refusal entered as zero, or an intervention with no content or minutes. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Skinner HA, 1982, Addictive Behaviors, The drug abuse screening test (the 28-item original and the 20-item derivation; the citation every reproduction asks for); Skinner HA, Guide for Using the Drug Abuse Screening Test (DAST), revised version 2023, Centre for Addiction and Mental Health, Toronto (copyright 1982 and 2023 by the author and CAMH; tentative interpretation guidelines; case-finding suggestion; 12-month instruction and follow-up windows; oversight by a qualified professional; not for public self-screening; DAST-R under development; obtainable from CAMH or the author, not linked here). Centre for Addiction and Mental Health, Drug Abuse Screening Test clinical tool page (self-report or structured interview; English and French; no information on drug type, frequency, or duration; digital licensing through [email protected] ) and CAMH Store product page (PZ074 and PZ096; published 1982/2019); NIH Common Data Elements Repository, DAST-10 form record (NIDA steward; copyright text permitting free non-commercial reproduction with credit to the author); LOINC, 82666-9 DAST-10 panel (total score 82667-7). Martino S, Grilo CM, Fehon DC, 2000, Addictive Behaviors, development of the DAST-A ; Gavin DR, Ross HE, Skinner HA, 1989, British Journal of Addiction, diagnostic validity of the DAST against DSM-III drug disorders (501 patients; 85 percent accuracy); Yudko E, Lozhkina O, Fouts A, 2007, Journal of Substance Abuse Treatment, comprehensive review of the psychometric properties of the DAST ; Johnson E and colleagues, published online September 2024 (volume 58, 2025), Measurement and Evaluation in Counseling and Development, psychometric synthesis of the DAST versions (346 articles). Maisto SA and colleagues, 2000, Psychological Assessment, the AUDIT and DAST-10 in adults with severe and persistent mental illness (162 outpatients; cut point of 2); Carey KB, Carey MP, Chandra PS, 2003, Journal of Clinical Psychiatry, the AUDIT and DAST-10 with psychiatric patients in India (1,349 inpatients); Smith PC, Schmidt SM, Allensworth-Davies D, Saitz R, 2010, Archives of Internal Medicine, a single-question screening test for drug use in primary care (286 patients; performance similar to the DAST-10); Lam LP and colleagues, 2015, Scientific Reports, validation of the DAST-10 among Chinese pregnant women (1,085 completed forms). US Preventive Services Task Force, June 2020, unhealthy drug use: screening (grade B for adults including pregnant and postpartum persons; I statement for adolescents; screening by asking questions; tools not meant to diagnose; example instruments including the DAST), and the companion statement noting that both screening recommendations are being updated . CMS Medicare Learning Network, May 2025, SBIRT services fact sheet (G2011, G0396, and G0397 minute ranges; AUDIT and DAST as example tools; the medical record must support the claim; H0049 and H0050; bare code numbers only), and April 2026, substance use screenings and treatment booklet (SBIRT components); SAMHSA, SBIRT . US Department of Health and Human Services, 2024, 42 CFR Part 2 final rule fact sheet (effective April 16, 2024; compliance February 16, 2026); eCFR, 42 CFR 2.11 definitions (records; Part 2 program); American College of Surgeons, VRC 2022 standards Q&A (alcohol screening and brief intervention standards). Centre for Addiction and Mental Health, Fundamentals of addiction: screening (screen all patients aged 10 and over; document in the cumulative patient profile; brief assessment after any cannabis or illegal drug use; document daily and weekly use). RACGP, Alcohol and other drugs: screening (updated June 3, 2026; routine or opportunistic screening within the 5As; AUDIT-C and ASSIST-Lite named). ## Related templates CAGE Questionnaire Documentation AUDIT-C Score Documentation Substance Use Assessment 42 CFR Part 2 Consent SASSI Documentation General Risk Screening Note COWS Protocol Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pcl-r-report All note templates # PCL-R Report Write-Up: Structure, Sample Language & Common Errors Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The PCL-R (Hare Psychopathy Checklist-Revised, Second Edition, 2003) is a 20-item clinician rating of psychopathic traits scored from a semi-structured interview plus collateral records. Forensic psychologists and psychiatrists use it in sentencing, civil commitment, parole, and dangerous-offender evaluations. A defensible write-up reports the total as a measurement band against a named comparison group and never as a diagnosis. This page covers how to document PCL-R results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Forensic psychologists and psychiatrists with PCL-R-specific training and forensic experience; publisher qualification level C Audience Judges and juries, prosecutors and defense counsel, parole boards and review tribunals, correctional treatment teams, opposing experts who will rescore the same file, the evaluee Typical length 400 to 900 words for the PCL-R section · administration 90 to 120 minutes of interview, about 60 minutes of collateral review, and 15 to 20 minutes of rating Format family Clinician-rated forensic personality construct measure (20 items rated 0 to 2 from a semi-structured interview plus collateral file review; total 0 to 40; two factors, four facets) When it's used Sexually violent predator and civil commitment evaluations, capital and other sentencing evaluations, parole and release reviews, Canadian dangerous-offender assessments, Australian high-risk offender proceedings, correctional treatment planning Standards context Published by Multi-Health Systems (2nd edition, 2003; no third edition announced as of September 2026); no statute names it; described here for write-up purposes, no test content reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the PCL-R? The PCL-R (Hare Psychopathy Checklist-Revised, 2nd Edition; Robert D. Hare; Multi-Health Systems, 2003, following the 1991 first edition) is a 20-item clinician rating scale for psychopathic traits in adults aged 18 and older. Each item is rated 0, 1, or 2 against lifetime functioning from a semi-structured interview combined with review of collateral records, for a total of 0 to 40; the publisher's stated timing is 90 to 120 minutes of interview, about 60 minutes of collateral review, and 15 to 20 minutes of rating, and the product is sold at qualification level C in handscored and software formats. The items group into two factors, an interpersonal and affective Factor 1 and a lifestyle and antisocial Factor 2, and four published facets (Interpersonal, Affective, Lifestyle, Antisocial). Its siblings are separate instruments with their own manuals and comparison samples: the PCL:SV (Screening Version; Hart, Cox, and Hare, 1995), an adult screen with a shorter interview, and the PCL:YV (Youth Version; Forth, Kosson, and Hare, 2003) for ages 12 to 18. As of September 2026 the second edition remains the current product on the publisher's United States and Canadian storefronts, with no third edition or norm update announced; MHS holds the copyright for the PCL-R family after acquiring Darkstone Research Group's assets in November 2019 and styles the mark Hare PCL-R with a trademark symbol. The load-bearing fact for the write-up is what the number is not. A PCL-R total expresses degree of correspondence with the construct the instrument measures; it is not a DSM-5-TR diagnosis (psychopathy is not one), not a probability of future violence, and not a legal finding. The 30-point threshold familiar from North American research is a classification convention for a dimensional construct, and a 2017 item-response analysis counted more than 8.5 million different rating combinations that sum to exactly 30. A second fact governs the courtroom: single-rater agreement in the manual's research samples is high, but opposing evaluators in real cases have agreed at intraclass correlations of .39 to .59, with differences usually favoring the retaining side, so the manual's standard error of measurement understates the uncertainty of scores offered in litigation. A defensible section therefore documents the frame before the number: the question and the role, the interview and collateral sources, the rater's qualification and any second rating, the total as a band against a named comparison group, the pattern at the level of the published labels, and the limits. The forensic evaluation report page covers the whole-report architecture this section sits inside, and the violence risk assessment page covers the instruments that carry the risk opinion. ## Who uses the PCL-R and when Forensic psychologists and psychiatrists are the users, and the settings are the highest-stakes ones in behavioral health: sexually violent predator and civil commitment evaluations (Texas's statute requires testing for psychopathy without naming an instrument), capital sentencing in the states that ask about future dangerousness, where a 2020 statement of concerned experts and a published rebuttal disagree about whether the PCL-R belongs there at all, parole and release reviews, pre-sentencing evaluations, Canadian dangerous-offender and long-term-offender assessments under Part XXIV of the Criminal Code and Correctional Service Canada psychological risk assessments, Australian high-risk offender proceedings in Queensland, New South Wales, Victoria, and Western Australia, and correctional treatment planning. Adolescents are assessed with the PCL:YV, with the field's own caution that psychopathy measures should not drive long-term placement or transfer decisions. The readers are judges and juries, prosecutors and defense counsel, parole boards, opposing experts who will rescore the same file, and the evaluee, whose liberty the label can affect. The section usually sits inside a forensic evaluation report or a court report beside personality inventories such as the MMPI-3 , PAI , or MCMI-IV , which answer different questions; it has no place in a capacity evaluation or a clinical psychological evaluation report unless the referral question genuinely turns on psychopathic traits. ## How to structure a PCL-R results section No statute, court rule, or publisher document prescribes a PCL-R report format. What the professional guidelines, the case law, and the instrument's own method dictate is the content: the question and the role, the two-part procedure with its sources, the rater's qualification, the total as a band against a named group, the pattern at the level of the published labels, the limits, and the hand-off to the risk instrument. Each section below carries the pitfall that most often undermines it in cross-examination. Referral question, legal standard, and role. Open with who ordered or retained the evaluation and on what date, the psycholegal question the PCL-R is offered to inform (treatment amenability, a statutory risk element, release planning), your role, and the notice the evaluee received about purpose, limits of confidentiality, and recipients. State what the section will and will not answer: a PCL-R total describes a trait pattern; it does not decide a statutory element or a sentence. Pitfall: A PCL-R section that opens with the score. Without the question and the role, the reader cannot tell what the number is for, and the label starts doing the legal work before the evidence is on the page. Instrument, procedure, and sources. Name the instrument in full with edition and publisher (Hare Psychopathy Checklist-Revised, 2nd Edition; Multi-Health Systems, 2003). Then document the two-part method: interview dates, durations, setting, and whether the evaluee participated fully, partially, or declined; every category of record reviewed with date ranges; collateral interviews; and records requested but not received. If no interview was possible, say that a file-only procedure was used and that the manual's file-review comparison samples apply; if any item could not be rated, say that the total was prorated per the manual, without reproducing the procedure. Pitfall: "PCL-R interview conducted" when the ratings rested on records, or a prorated total presented as if all 20 ratings were made. Both are impeachable on the first question about method. Rater qualification and second-rater procedure. State your licensure, PCL-R-specific training (the publisher's authorized program or documented equivalent), forensic experience, and how often you use the instrument. If a second trained rater scored the case, say whether the rating was independent and blind, what materials the second rater used, both totals, and whether the averaged score or a consensus is being reported. The author recommends averaging two independent ratings where feasible; report the disagreement rather than choosing the more convenient number. Pitfall: "Scored by a qualified examiner" with no training or second rating on the record. Publisher qualification level C is a purchasing standard, not evidence of competence with this instrument. Total score as a band with the named comparison group. Report the total with the manual's standard error of measurement as a band (about 3 points either side for a 68 percent band and roughly 6 for a 95 percent band, for a single rating in the manual's male offender sample as reported in the literature), name the comparison group and procedure that any percentile rests on (male justice-involved, male forensic psychiatric, or female justice-involved; standard or file-review sample), and give the percentile as a band in prose. If a research-convention threshold is mentioned, attribute it as a convention and say whether the band crosses it. Pitfall: A bare integer, or a percentile with no comparison group. The 2015 Canadian case-law review found single-rater agreement of .59 in court, so the manual's 2.90 understates the uncertainty of any score offered in litigation. Factor and facet pattern in prose. Describe the relative elevation across the published labels only (Factor 1, interpersonal and affective; Factor 2, lifestyle and antisocial; the four facets Interpersonal, Affective, Lifestyle, and Antisocial) and whether both raters saw the same shape. Pattern language tells the court something psychometrically real; item narrative reproduces protected content and reads as coaching. Pitfall: An item-by-item walk through the rating booklet, or the facet pattern rendered as a list of adjectives about the person. The same total can arise from millions of rating combinations, which is why the shape matters and the epithet does not. Limits and inferential boundaries. State that the result is not a diagnosis (psychopathy is not a DSM-5-TR category), not an individualized prediction of violence or recidivism, and not a legal finding; that thresholds are conventions that differ by research tradition; and that the field-reliability and adversarial-allegiance literature exists, with the safeguards you used named. Note any gap in the record base and its direction of effect on confidence. Pitfall: Limits omitted, or limits so total that they undercut the section. State what the score does support first (a structured description of documented traits), then what it does not. Integration with structured risk assessment and the formulation. Cross-reference the structured risk instrument that carries the risk opinion (HCR-20 Version 3, Static-99R, or another purpose-built tool, each in its own section under its own manual), the dynamic and protective factors, and the treatment and supervision conditions. Keep the PCL-R as one contributor to the formulation, and keep ultimate-issue language where your jurisdiction and role put it. Pitfall: "High psychopathy, therefore high risk." Comparative meta-analysis finds the PCL-R and eight other tools predict violence at similar moderate levels, with Factor 1 at chance among men; the score is a trait description, not the risk estimate. ### Blank template (copy and adapt) PCL-R SECTION SKELETON (forensic evaluation report) Evaluee: [initials] Age: [ ] Report date: [ ] Examiner: [name, credentials] Retained or appointed by: [court / prosecution / defense / agency] Order or retention date: [ ] Notice given to evaluee: [ ] Psycholegal question this section informs: [ ] Instrument: Hare Psychopathy Checklist-Revised, 2nd Edition (Multi-Health Systems, 2003) Procedure: [interview + collateral / file-only, reason] Interview: [dates, durations, setting; full, partial, or declined] Records reviewed: [category and date range for each] Collateral interviews: [ ] Requested, not received: [ ] Items not rated / prorating: [none / disclosed per the manual] Rater: [licensure, PCL-R training and year, forensic experience, frequency] Second rater: [independent and blind: yes / no; materials used; both totals; averaged or consensus] Total: [ ] reported as a band: [68% and 95% bands using the manual SEM] Comparison group and procedure: [male justice-involved / male forensic psychiatric / female justice-involved; standard or file-review sample] Percentile band (prose): [ ] Research-convention threshold, if named: [attributed as a convention; band crosses it: yes / no] Factor and facet pattern (published labels only): [relative elevation; both raters agree: yes / no] Limits: [not a diagnosis; not a prediction; conventions, not rules; field reliability acknowledged; safeguards used; record gaps and their effect] Integration: [structured risk instrument and its section; dynamic and protective factors; treatment and supervision conditions] Examiner signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank PCL-R section skeleton, ready to adapt. Neither reproduces items, rating criteria, the interview guide, norms, or conversion tables. ## Sample PCL-R write-up (fictional) Scenario: a court-ordered pre-sentencing evaluation of a 34-year-old man in which the court asked about psychopathic traits as they bear on treatment planning and supervision, written by a court-appointed psychologist with an independent second rater, a fictional mid-range total reported as a band, and the risk opinion kept in its own section. All details are fictional. Patient: D.K., 34 · Setting: Court-ordered pre-sentencing psychological evaluation, state superior court · Clinician: M. Hartley, PhD, licensed psychologist · Note date: 09/17/2026 Referral and legal question: By order dated 08/24/2026, the court appointed this examiner to conduct a pre-sentencing psychological evaluation and asked, among other questions, whether D.K. shows psychopathic traits relevant to treatment amenability and supervision planning. This section reports the Hare Psychopathy Checklist-Revised, 2nd Edition (PCL-R; Hare, 2003; Multi-Health Systems). It answers the descriptive question the court asked; it does not by itself answer any sentencing question, and the risk opinion appears in the violence risk section of this report. D.K. was informed of the purpose of the evaluation, the absence of confidentiality, and who would receive the report, and he agreed to proceed. Sources and administration conditions: The PCL-R was completed by the standard procedure: a semi-structured interview plus review of collateral records. The interview took place in two sessions at the county detention center on 09/09/2026 (2 hours) and 09/10/2026 (1 hour 30 minutes); D.K. participated fully in both. Records reviewed: the charging documents and the presentence investigation report (received 09/01/2026); jail classification and conduct records from the current detention (received 09/03/2026); probation records from 2019 to 2023; outpatient mental health and substance use treatment records from two agencies covering 2017 to 2024; school records through the tenth grade; and employment verification for three employers. A collateral telephone interview with D.K.'s older sister took place on 09/11/2026. Juvenile court records were requested on 09/02/2026 and had not been received by the date of this report; their absence is noted in the limits below. The index offense is described in the offense history section of this report and is not repeated here. All 20 items could be rated; no item was omitted and no prorating was needed. Rater qualification and second rating: This examiner is a licensed psychologist in forensic practice since 2015, completed the publisher-authorized PCL-R training program in 2021, and uses the instrument regularly in court-ordered evaluations. A second licensed psychologist trained in the instrument rated the case independently on 09/12/2026 from the recorded interview and the same record set, without seeing my ratings; no consensus meeting was held. Following the author's recommendation that two independent ratings be averaged where possible, both totals are reported below and the average is used. Total score with measurement error: My total was 23; the second rater's total was 21; the averaged total is 22. The two ratings differ by less than the standard error of measurement the manual reports for a single rating (about 3 points), which is the level of agreement expected under research conditions. Interpreted against the manual's standard-procedure comparison group of adult male justice-involved persons, an averaged total of 22 falls in the middle range of that group, in a percentile band around the median rather than in its upper range. Allowing roughly two standard errors either side, the 95 percent band spans approximately 16 to 28. That band does not reach the 30-point threshold used as a research convention in North American studies, and the band is the result: the single integer should not be read as an exact quantity. Factor and facet pattern: Described at the level of the published labels only: the profile was relatively more pronounced on the Lifestyle and Antisocial facets (Factor 2) than on the Interpersonal and Affective facets (Factor 1), which were comparatively modest. Both raters saw the same shape. In plain terms, the documented history weighs more heavily on the behavioral and lifestyle side of the construct than on the interpersonal and affective side. No item-level ratings are reproduced here; the rating booklet, item descriptions, and scoring criteria are the publisher's protected materials. Limits: This result is not a diagnosis: psychopathy is not a DSM-5-TR diagnostic category, and the PCL-R describes degree of correspondence with the construct the instrument measures. It is not a prediction of future behavior: comparative meta-analysis finds the PCL-R and other risk tools predict violence at similar moderate levels, with the interpersonal and affective factor adding little, so no individual likelihood is derived from this score. Thresholds such as 30 are classification conventions from research, not diagnostic or legal criteria, and this evaluation offers none as such. The published field-reliability literature, in which opposing evaluators have shown agreement well below research values, is acknowledged; the safeguards here were a court appointment rather than party retention, a predetermined record base, a recorded interview, and an independent second rating with both totals disclosed. The missing juvenile records could bear on the lifetime history the instrument rates, and the total is reported with that gap stated. Integration and formulation: The PCL-R contributes one structured description to the formulation. The violence risk section reports the HCR-20 Version 3 separately, including dynamic and protective factors and the supervision conditions under which risk would be manageable, and the treatment section addresses amenability using treatment history and current motivation rather than this score alone. Nothing in this section should be read as an opinion on the sentence; that question belongs to the court. This sample is fictional and for educational purposes. It does not describe a real person, case, or record; the scores, dates, and details are invented to show write-up structure and are not clinical or legal guidance. Scores are invented for illustration and correspond to no real person or record, and no item content, scoring criteria, or norm-table values are reproduced. ↑ Back to the template and downloads ### Why this sample works The edition, the two-part procedure, the interview dates and durations, the records by category and date, the collateral contact, and the records requested but not received are all stated before any score, which is what the AAPL guideline asks for and what cross-examination probes first. The rater's licensure, publisher-authorized training, and forensic experience are on the record, and the independent, blind second rating from the same record base, with both totals reported, converts a partisan-looking number into a documented one. The total is written as a band against a named comparison group and procedure, the percentile is a band in prose, and the research-convention threshold is named as a convention, not as a diagnosis or a decision rule. The factor and facet pattern stays at the level of the published labels with no item narrative, so the report shows the profile shape without reproducing the publisher's protected content or coaching the next evaluee. The limits paragraph does the defensibility work: not a diagnosis, not a prediction, field reliability acknowledged with the safeguards named, the record gap disclosed, and the risk opinion routed to the structured instrument's own section. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the write-up sits inside a legal machine, and the parts are citable. No statute names the PCL-R. Sexually violent predator laws require an assessment: Texas Health and Safety Code 841.023 requires a clinical assessment based on testing for psychopathy, a clinical interview, and other appropriate assessments and techniques, which is a requirement to test for the construct, not to use this product (LAW). Admissibility runs through Frye in some states and Daubert in the federal courts and others, is decided purpose by purpose, and in the published case-law reviews by DeMatteo and colleagues has seldom been the subject of a successful challenge; the capital context traces to Barefoot v. Estelle (1983), in which the Supreme Court allowed psychiatric future-dangerousness testimony at sentencing and left its reliability to cross-examination (case law, described here as CONVENTION and practice, not legal advice). The professional framework is the APA Specialty Guidelines for Forensic Psychology (2013) and the AAPL Practice Guideline for the Forensic Assessment (2015): identify the referral question and your role, list every source considered and every source requested but not received, state the limits of your opinions, and keep fee arrangements free of contingency (CONVENTION and ethics, not statute). The use dispute is live: the 2020 Statement of Concerned Experts in Psychology, Public Policy, and Law concluded that the PCL-R should not be used to predict serious institutional violence in capital sentencing, Olver and colleagues published a cautionary note in reply, and DeMatteo and Olver then co-authored a 2021 review of validity, reliability, and admissibility; a report that cites this exchange shows the court it knows the ground. Forensic evaluations are commissioned and paid by the court, a party, or an agency, not billed to health insurance under CPT, and the source of payment changes nothing about method or disclosure (PAYER POLICY). Canada and Australia require assessments, not this instrument. Criminal Code section 752.1 lets the court remand an offender for up to 60 days for assessment by a designated person before a dangerous-offender or long-term-offender application, with the report filed within 30 days, and section 753 sets the findings the court must make; Correctional Service Canada's Commissioner's Directive 712-1 makes a psychological risk assessment mandatory for conditional-release applications involving persistent or gratuitous violence, sexual offences, detention referrals, and indeterminate or life sentences, and defines it as an integrated evaluation using a variety of scientifically validated methods (LAW and agency POLICY); the PCL-R appears inside these assessments by professional custom (CONVENTION). Queensland's Dangerous Prisoners (Sexual Offenders) Act 2003 provides for examination by two psychiatrists who report independently on the assessed risk and its reasons, requires the report even if the prisoner does not cooperate, and sets an acceptable, cogent evidence standard to a high degree of probability; New South Wales's Crimes (High Risk Offenders) Act 2006, Victoria's Serious Offenders Act 2018 (in force as amended to October 2025), Western Australia's High Risk Serious Offenders Act 2020, and Commonwealth Division 105A run parallel regimes, and none names an instrument (LAW). Edition and norms: the second edition (2003) is current as of September 2026 with no successor announced, so date the edition statement in the report. The publisher's database of 10,896 justice-involved persons and patients is split by group and procedure (5,408 male justice-involved persons, 1,246 male forensic psychiatric patients, and 1,218 female justice-involved persons under the standard procedure; 2,622 male justice-involved persons and 402 male forensic psychiatric patients by file review), which is why a percentile means nothing until the group and procedure are named, and no female file-review or female forensic psychiatric sample is listed. Generalizability is an empirical question: a 21-study meta-analysis of 8,890 people found Black and White participants differing by less than one point on the total, and a 2011 comparison of North American and German file-based ratings found the Antisocial component not invariant across countries, so a report states a caution rather than adjusting a score. Qualification: MHS sells the PCL-R at level C (training or experience with tests plus an advanced degree in an appropriate profession, with state licensure possibly required), Hare's professional-use recommendations add registration with the body that regulates mental health assessment, forensic experience, adequate PCL-R training, use only with populations the instrument is validated for, and averaged independent ratings where possible, and MHS delivers and authorizes PCL-R training (a 16-hour continuing-education program combining an on-demand presentation, live question time with the author, and a two-day live webinar); no fixed number of supervised cases is an MHS rule, so the report states the training you actually have. This is a sensitive documentation type. The evaluee is usually in custody or facing loss of liberty, the label carries prejudice a court cannot easily set aside, and the report may be read by the evaluee, so write traits and their documented behavioral referents rather than the epithet, and keep the section free of detail that could identify victims or third parties beyond what the record already contains. If you or a client needs immediate support: call or text 988 (US), 9-8-8 (Canada), or Lifeline 13 11 14 (Australia). Before releasing raw scores, the rating booklet, or other test materials in response to a subpoena or a discovery demand, consult your attorney or board; state rules vary, APA Ethics Standards 9.04 and 9.11 distinguish test data from test materials, and the subpoena response page covers the procedure. Hare PCL-R is a trademark of Multi-Health Systems Inc., which holds the copyright for the PCL-R family of instruments and directs requests to reproduce excerpts, tables, or figures to [email protected] . BastionGPT is not affiliated with, or endorsed by, the publisher. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common PCL-R write-up errors reviewers flag The numbers behind these errors are specific. Opposing evaluators in 23 civil-commitment cases agreed on PCL-R totals at an ICC of .39 , with differences usually favoring the retaining party; 29 paired scores in US sexually violent predator case law agreed at .58 , only 41.4 percent of the differences fell within one standard error of measurement, and prosecution experts reported 30 or above in nearly 50 percent of cases against under 10 percent for defense experts; 102 Canadian criminal cases with two or more scores agreed at .59 , a level the authors note implies a real-world standard error far larger than the manual's 2.90; 108 forensic clinicians shifted their scores toward whichever side they believed had retained them; the nine-tool meta-analysis found the PCL-R no better than other instruments at predicting violence and Factor 1 at chance among men; and more than 8.5 million rating combinations sum to exactly 30. The BastionGPT Clinical Advisory Board sees the same errors most often in PCL-R documentation reviews: The threshold written as a diagnosis. "A total of 30 or above indicates psychopathy." Thirty is a North American research convention for a dimensional construct, lower figures were used in some UK and European research on contested grounds, millions of rating combinations sum to the same total, and psychopathy is not a DSM-5-TR diagnosis. Report the band, name the convention as a convention if it must be mentioned, and separate it from any diagnosis or legal finding. A bare integer with no measurement error or comparison group. "PCL-R total: 27." A single rating carries a manual standard error of about 3 points, scores offered in litigation have proven far less reliable than that, and a percentile without its group and procedure (standard or file-review; male justice-involved, male forensic psychiatric, or female justice-involved) hides an interpretive choice. Write the band, name the group, and give the percentile as a band in prose. The procedure misdescribed. A file-only or truncated assessment written up as a PCL-R interview, a refused interview left unmentioned, or a prorated total presented as if every rating were made. The publisher keeps separate file-review comparison samples for a reason: state the procedure, the reason, the missing sources, and the effect on confidence, and treat a refusal as an assessment condition rather than a trait. Item-level narrative or protected content. An item-by-item account of how the ratings were derived, paraphrased rating criteria, or a reproduced booklet page. These are the publisher's copyrighted materials sold at qualification level C, they add nothing a court can use, they read as coaching, and they invite a test-security challenge. Describe the pattern at the level of the published factor and facet labels. The score doing the risk opinion's work. "His PCL-R score places him at high risk of violence." The instrument was not built as a risk tool; comparative meta-analysis finds it predicts violence at the same moderate level as eight other instruments, with Factor 1 at chance among men. Route the risk opinion to the structured instrument in its own section and keep the PCL-R as a trait description feeding the formulation. Research reliability quoted as field reliability. "The PCL-R has excellent interrater reliability" with no mention of the adversarial field studies at .39 to .59 or the experimental allegiance effect. Cross-examination will supply the omission. Disclose the literature, then show the safeguards: retention disclosed, training on record, a predetermined record base, a recorded interview, and an independent second rating with both totals reported. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on forensic evaluation reports and violence risk assessments. Give it the facts (the court order or retention and the question, interview dates and participation, records reviewed and records not received, your training and the second rater's procedure, both totals, the comparison group, the facet pattern, the limits you want stated, and the risk instrument used) and it drafts the PCL-R section: sources listed, the total as a band against the named group, the pattern at the level of the published labels, the limits paragraph, and the cross-reference to the risk section, ready for your review. Cross-check a finished section for the gaps reviewers and cross-examiners flag: a bare integer, a percentile with no comparison group, a file-only procedure described as an interview, item-level narrative, a threshold written as a diagnosis, or a score doing the risk opinion's work. Draft the companion paragraphs: the refused-interview disclosure, the second-rater disagreement explanation, the field-reliability acknowledgment with your safeguards, or the plain-language summary of what the result does and does not mean for the court. See how clinicians use it day to day on the AI therapy notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do PCL-R scores mean? The PCL-R has 20 items, each rated 0, 1, or 2 by a trained clinician from a semi-structured interview plus collateral records, for a total of 0 to 40 that expresses how closely the documented lifetime pattern corresponds to the construct the instrument measures; the publisher describes the total as usable dimensionally or in a categorical framework. The items group into two factors (Factor 1, interpersonal and affective; Factor 2, lifestyle and antisocial) and four published facets (Interpersonal, Affective, Lifestyle, Antisocial). The manual reports percentiles and standardized scores for specific comparison samples, so a percentile only means something when the group and procedure are named. Two properties govern interpretation: the standard error of measurement for a single rating, about 3 points in the manual's male offender sample as reported in the published literature, which turns an integer into a band of roughly plus or minus 6 at 95 percent; and the fact that the same total can arise from millions of rating combinations, which is why the pattern matters more than the number. A total is never a diagnosis and never a prediction by itself. Is a PCL-R score of 30 a diagnosis of psychopathy? No. Thirty is the threshold used as a research convention in North American studies, the published literature (including the 2020 Statement of Concerned Experts) treats it as a convention for a dimensional construct rather than a diagnostic boundary, and psychopathy is not a DSM-5-TR diagnosis. Lower thresholds in the mid-20s have been used in UK and European research on the strength of cross-cultural equating work, and later analyses disputed whether a uniform downward adjustment is justified, so the same raw score does not carry the same meaning across settings. A 2017 item-response analysis counted more than 8.5 million rating combinations that sum to exactly 30, each describing a different profile. The write-up therefore never says that 30 equals a psychopath or that Europe uses 25; it reports the band, says which convention is under discussion and why it is relevant, states whether the band crosses it, and separates that statement from any psychiatric diagnosis or legal finding. When categorical language is unavoidable, call it an instrument-based classification convention and attribute it. Can the PCL-R be scored from records alone, and how do I write up a refused interview? Sometimes, and the report must say so prominently. The standard procedure is a semi-structured interview plus collateral review; the publisher maintains separate file-review-only comparison samples, which confirms that a records-only procedure is contemplated when an interview cannot be obtained and the records are extensive and reliable. The instrument cannot be completed from an interview alone, because collateral information is part of the method. Research comparing independent file-based ratings with full assessments cautions that the two are not automatically interchangeable, so the write-up states the date and circumstances of the offered interview, that the evaluee declined or participated partially, that ratings rested on the available records, which important sources were missing, that the file-review comparison sample was used for any percentile, and that confidence is correspondingly reduced. A refusal is an assessment condition to document, never a trait to score. The AAPL forensic assessment guideline asks for exactly this: list the sources considered, note information requested but not received, and state the limits that follow. A usable sentence: because the evaluee declined the interview, the PCL-R was completed from the collateral record where ratings could be supported, and this departure from the standard procedure limits the certainty of the interpretation. Which comparison group and norms should a PCL-R report cite? The one that matches the evaluee and the procedure, named in the report. The publisher's second-edition database contains 10,896 justice-involved persons and patients, split into standard-procedure samples of 5,408 male justice-involved persons, 1,246 male forensic psychiatric patients, and 1,218 female justice-involved persons, plus file-review-only samples of 2,622 male justice-involved persons and 402 male forensic psychiatric patients. Percentiles shift with the group, so a phrase like compared with PCL-R norms conceals a consequential choice. Write, for example, that the total was interpreted against the standard-procedure sample of adult male justice-involved persons, or against the file-review sample of male forensic psychiatric patients, and give the percentile as a band in prose rather than as a table. Note the gaps honestly: the product page lists no female file-review or female forensic psychiatric sample, and cross-cultural work has found the Antisocial component less invariant across countries than the other facets, so an evaluee outside the sampled populations gets a stated caution rather than a laundered percentile. How do I handle cross-examination about interrater reliability and adversarial allegiance? Do not deny the literature; distinguish population-level bias risk from evidence about your rating, then show the safeguards. The numbers are public: opposing evaluators in 23 Texas civil-commitment cases agreed at an ICC of .39, with differences usually favoring the retaining side (Murrie and colleagues, 2008); 29 paired scores in US sexually violent predator case law agreed at .58, with prosecution experts reporting scores of 30 or above in nearly half their cases against under 10 percent for defense experts (DeMatteo and colleagues, 2014); 102 Canadian criminal cases with two or more scores agreed at .59 (Edens and colleagues, 2015); and an experiment that led 108 forensic clinicians to believe one side had retained them shifted their scores in that side's direction (Murrie and colleagues, 2013). Independent raters working from the same files agreed with each other at .95 while agreeing with state and defense experts at .29 and .14, and even independent Swedish evaluation teams reached only .70 on retest. The defensible answer states who retained you, your PCL-R training, the predetermined record base, the recorded interview, the independent second rating with both totals disclosed, the band rather than the integer, and the acknowledgment that this literature exists. Choosing whichever score produces the more dramatic label is the error the literature documents. Is the PCL-R required by law in SVP, dangerous offender, or high-risk offender proceedings? No jurisdiction names it. What the law requires is an assessment or a report; the instrument is a professional choice (CONVENTION). Texas Health and Safety Code 841.023 requires a clinical assessment based on testing for psychopathy, a clinical interview, and other appropriate assessments and techniques, and names no instrument (LAW). Canada's Criminal Code section 752.1 lets the court remand an offender for up to 60 days for an assessment by a designated person before a dangerous-offender or long-term-offender application, with a report filed within 30 days, and names no instrument (LAW); Correctional Service Canada's Commissioner's Directive 712-1 makes a psychological risk assessment mandatory for certain conditional-release applications and defines it as an integrated evaluation using a variety of scientifically validated methods, again without naming a tool (agency POLICY). Queensland's Dangerous Prisoners (Sexual Offenders) Act 2003 provides for examination by two psychiatrists who report independently, requires the report even if the prisoner does not cooperate, and sets an acceptable, cogent evidence standard to a high degree of probability; New South Wales, Victoria, Western Australia, and the Commonwealth run parallel high-risk offender regimes, none of which names the PCL-R (LAW). Admissibility in the United States runs through Frye or Daubert according to the jurisdiction, is purpose-specific, and in the published case-law reviews has seldom been successfully challenged, which is a description of practice, not legal advice. Does the PCL-R predict violence, and how does it fit with the HCR-20 V3 or Static-99R? PCL-R scores correlate with violent and criminal outcomes at the group level, and psychopathic traits can be an important risk factor; that is different from an individualized probability. Yang, Wong, and Coid's 2010 meta-analysis compared nine risk instruments within the same studies and found all of them, and their subscales, predicting violence at about the same moderate level, with the exception of PCL-R Factor 1, which predicted violence only at chance level among men; the authors concluded that tools with this level of accuracy should not be used alone for decisions such as preventive detention. The write-up therefore keeps the PCL-R as a structured description of traits and routes the risk opinion to a purpose-built instrument in its own section: the HCR-20 Version 3 for structured professional judgment about violence (its authors recommend but no longer require a formal PCL-R rating for the relevant item), Static-99R and Stable-2007 for sexual recidivism, or another validated tool, each interpreted under its own manual with dynamic and protective factors and management conditions. Personality inventories such as the MMPI-3 , PAI , and MCMI-IV answer different questions and substitute for none of these. The violence risk assessment page covers the risk instruments themselves. Can I include PCL-R items, the interview guide, or scoring criteria in my report? No, and the report does not need them. The rating booklet, item descriptions, interview schedule, scoring criteria, and norms are copyrighted materials of Multi-Health Systems, which has held the copyright for the PCL-R family since acquiring Darkstone Research Group's assets in November 2019 and directs reproduction requests to [email protected] ; the instrument is sold at qualification level C. A report contains the evaluee's derived results and your interpretation: the total as a band, the comparison group, the factor and facet pattern at the level of the published labels, and your reasoning about limits. Under APA Ethics Standards 9.04 and 9.11, responses, scores, and interpretation are test data, while the booklet and criteria are test materials that stay out of the report, out of the public court file where the rules allow, and off any public page. An item-by-item narrative also reads as coaching: it tells the next evaluee what a low score looks like. Institutional or digital integration runs through the publisher's licensed platforms (MHS offers GEARS for agencies), a free web version of the checklist has no authorization and is not the instrument, and a subpoena for raw data is handled through counsel, as the subpoena response page describes. Can BastionGPT draft my PCL-R report? Yes. Give it the facts (the court order or retention and the question, interview dates and participation, records reviewed and records not received, your training and the second rater's procedure, both totals, the comparison group, the facet pattern, the limits you want stated, and the risk instrument used) and it drafts the PCL-R section: sources listed, the total as a band against the named group, the pattern at the level of the published labels, the limits paragraph, and the cross-reference to the risk section, ready for your review. It can also cross-check a finished section for the errors reviewers flag: a bare integer, a percentile with no comparison group, a file-only procedure described as an interview, item-level narrative, a threshold written as a diagnosis, or a score doing the risk opinion's work. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts, study figures, and legal claims on this page trace to these sources, last verified September 2026: Publisher record, accessed September 2026: Multi-Health Systems, Hare PCL-R: 2nd Edition product page (author; age 18 and older; interview plus collateral review; administration times; 20 items; qualification level C; handscored and software formats; normative sample composition; dimensional and categorical use; trademark styling), ordering and qualification levels , PCL-R training description (16 continuing-education hours), GEARS platform , and the Hare PCL:YV and Hare PCL:SV product pages; Robert D. Hare, PCL-R professional-use recommendations (degree, registration, forensic experience, training, averaged independent ratings) and MHS acquisition of Darkstone Research Group assets , November 18, 2019 (copyright; [email protected] ). Field reliability and adversarial allegiance: Murrie DC, Boccaccini MT, Johnson JT, Janke C, 2008, Law and Human Behavior, interrater (dis)agreement on Psychopathy Checklist scores in sexually violent predator trials (43 trials, 23 paired cases, ICC .39); DeMatteo D, Edens JF, Galloway M, Cox J, Smith ST, Formon D, 2014, Law and Human Behavior 38(3), 248 to 255, the PCL-R in US sexually violent predator evaluations (214 cases, 29 paired, ICC .58, 41.4 percent within one SEM, 30 or above in nearly 50 versus under 10 percent); Rufino KA, Boccaccini MT, Hawes SW, Murrie DC, 2012, Law and Human Behavior, independent raters versus opposing forensic experts (ICC .95 among independent raters; .29 and .14 with state and defense experts); Murrie DC, Boccaccini MT, Guarnera LA, Rufino KA, 2013, Psychological Science, are forensic experts biased by the side that retained them? (108 clinicians; effect sizes up to .85); Edens JF, Cox J, Smith ST, DeMatteo D, Sörman K, 2015, Psychological Assessment, PCL-R scores in Canadian criminal trials (102 cases, ICC .59; manual SEM 2.90); Sturup J and colleagues, 2014, Law and Human Behavior, field reliability among life-sentenced prisoners in Sweden (ICC .70 with independent evaluation teams). Prediction, thresholds, and the capital-sentencing dispute: Yang M, Wong SC, Coid J, 2010, Psychological Bulletin, meta-analytic comparison of nine risk assessment tools (similar moderate efficacy; Factor 1 at chance among men; not for preventive detention alone); Balsis S, Busch AJ, Wilfong KM, Newman JW, Edens JF, 2017, Journal of Personality Assessment, a statistical consideration regarding the PCL-R threshold (more than 8.5 million combinations summing to 30); DeMatteo D and colleagues, 2020, Psychology, Public Policy, and Law 26(2), 133 to 144, Statement of Concerned Experts on the use of the Hare PCL-R in capital sentencing ; Olver ME and colleagues, 2020, Psychology, Public Policy, and Law 26(4), 490 to 510, a cautionary note on DeMatteo et al. ; DeMatteo D, Olver ME, 2021, Journal of Personality Assessment, use of the PCL-R in legal contexts: validity, reliability, admissibility, and evidentiary issues ; DeMatteo D and colleagues, 2014, Psychology, Public Policy, and Law 20(1), 96 to 107, the PCL-R in United States case law . Construct debate and group generalizability: Skeem JL, Cooke DJ, 2010, Psychological Assessment, is criminal behavior a central component of psychopathy? and the reply to Hare and Neumann (the measure is not the construct); Skeem JL, Edens JF, Camp J, Colwell LH, 2004, Law and Human Behavior, ethnic differences in psychopathy: a meta-analysis (21 studies, 8,890 participants, less than one point); Mokros A and colleagues, 2011, International Journal of Law and Psychiatry, measurement invariance of file-based PCL-R ratings in North America and Germany ; Edens JF, Skeem JL, Cruise KR, Cauffman E, 2001, Behavioral Sciences and the Law, juvenile psychopathy and violence: a critical review (long-term placement decisions contraindicated). United States law and professional guidance: Texas Legislature, Health and Safety Code chapter 841 , section 841.023 (testing for psychopathy, a clinical interview, and other appropriate assessments; no instrument named); Barefoot v. Estelle, 463 U.S. 880 (1983), via Cornell LII ; American Psychological Association, Specialty Guidelines for Forensic Psychology (2013); American Academy of Psychiatry and the Law, Practice Guideline for the Forensic Assessment (2015; sources listed, information not received, limits stated). Canada: Criminal Code, section 752.1 (assessment remand of up to 60 days; report within 30 days) and section 753 (dangerous offender findings); Correctional Service Canada, Commissioner's Directive 712-1, Pre-Release Decision-Making (psychological risk assessment definition and mandatory referrals; no instrument named). Australia: Queensland, Dangerous Prisoners (Sexual Offenders) Act 2003 ; New South Wales, Crimes (High Risk Offenders) Act 2006 ; Victoria, Serious Offenders Act 2018 (in-force version as at 22 October 2025). Risk instruments named on this page: PAR, HCR-20 Version 3 (Douglas, Hart, Webster, and Belfrage; 20 violence risk factors), interpreted under its own manual and covered on the violence risk assessment page. ## Related templates Forensic Psychological Evaluation Report Violence & Homicide Risk Assessment Court Report & Treatment Progress Update MMPI-3 Report Write-Up PAI Report Write-Up Capacity & Competency Evaluation Psychological Evaluation Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/glasgow-coma-scale All note templates # GCS Documentation: Component Scores, Confounders & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Glasgow Coma Scale (GCS) is a bedside scale of impaired consciousness published by Teasdale and Jennett in 1974 that rates eye opening, verbal response, and motor response as three separate components summing to 3 to 15. Emergency, trauma, critical care, and neurosurgical nurses, paramedics, and physicians use it to describe a patient's level of response and to track change over time. This page covers how to document GCS findings, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Bedside nurses, paramedics and EMS clinicians, emergency and trauma physicians, intensivists, neurosurgical and stroke teams, and advanced practice providers; no license, training certificate, or publisher qualification level applies, and the rights holder asks only for acknowledgment Audience Receiving and relieving clinicians at handoff, trauma and neurosurgical consultants, rapid response and ICU teams, trauma registrars and coders, quality and peer-review staff, and later the attorneys and expert reviewers who reconstruct a neurological timeline from the chart Typical length 2 to 5 chart lines per assessment (time, E, V, M with the total when valid, stimulus, confounders or NT reasons, pupils, change and action) · assessment about 1 to 2 minutes at the bedside Format family Clinician-observed three-component scale (eye opening 1 to 4, verbal 1 to 5, motor 1 to 6; sum 3 to 15 only when all three are testable; NT recorded for a component that cannot be tested; GCS-Pupils extension 1 to 15) When it's used Prehospital and emergency assessment of head injury and altered consciousness, trauma resuscitation, serial neurological checks on wards and in critical care, sedation holds in ventilated patients, stroke and neurosurgical monitoring, and every handoff and transfer that carries a neurological status Standards context Teasdale and Jennett (Lancet, 1974) with the 2014 structured approach and NT rating (Teasdale and colleagues); free for clinical care and research with acknowledgment of the University of Glasgow and Sir Graham Teasdale; mandated by no law, embedded in trauma registries, triage rules, and ICD-10-CM coding; described here for documentation, no assessment aid reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Glasgow Coma Scale? The Glasgow Coma Scale is the bedside method of describing impaired consciousness that Graham Teasdale and Bryan Jennett published in the Lancet in July 1974, written so that nurses and doctors in different places and at different times could record the same patient the same way. It observes three kinds of behavior: whether and how readily the eyes open (four levels, from spontaneous opening down to none), the quality of the verbal response (five levels, from orientated conversation down to no sound), and the best motor response (six levels, from obeying commands through localizing, normal and abnormal flexion, and extension down to no movement); the six-level motor scale that separates abnormal flexion from extension followed in 1976 and is the version now recommended for all uses. The three ratings are the Scale; adding them to a single number from 3 to 15 produces the Score, which the authors introduced later as a summary index for groups and which the glasgowcomascale.org site describes as losing the detail the components carry. The 2014 structured approach (Teasdale and colleagues, Nursing Times and Lancet Neurology) did not change the instrument but standardized how it is done: check for factors that interfere with testing, observe spontaneous behavior, stimulate with spoken requests and then with physical pressure at the fingertip, the trapezius, or the supraorbital notch, and rate the best response; it renamed eye opening "to pain" as opening to pressure, discouraged rubbing the sternum, and introduced the not-testable rating, written NT, for a component that cannot be examined. The adult scale is used without modification from about age 5 (glasgowcomascale.org) and, in the Brain Trauma Foundation's prehospital guidance, from age 2, so pediatric adaptations of the verbal component (the Adelaide scale of Reilly, Simpson, and colleagues, 1988, and the pediatric GCS studied by the PECARN network) are a family of tools rather than one official chart, and a note should say which was used. The load-bearing fact for documentation is that the number is not the assessment. A total of 8 can be E2 V2 M4 or E1 V1 M6, two patients in very different states, and Reith and colleagues showed across 54,069 trauma patients that case fatality differed between component profiles sharing the same total for every sum from 4 to 14 except 6 and 7, and that the three components predicted outcome better than their sum. That is why the scale's own site, the ACS trauma program's TBI guideline, and NICE all ask for the components to be recorded and communicated with, not instead of, the total. Three conventions follow. A component that cannot be tested is recorded as NT with the reason (an endotracheal tube, periorbital swelling, paralysis) and no total is reported, because writing a 1 asserts a lowest response nobody observed and drags the trend down; the older habit of scoring an intubated patient's verbal response as 1 and adding a T (a "10T") is a legacy and registry convention, not the current structured approach. The motor score is the best arm response, with any weaker side written separately as a lateralizing sign rather than averaged in. And pupils are examined beside the scale, not inside it: the GCS-Pupils score (Brennan, Murray, and Teasdale, 2018) subtracts the number of unreactive pupils from the total to give a 1 to 15 prognostic index, and it needs the underlying components and pupil findings on the page. The GCS describes level of response and nothing else: sedation depth belongs to the RASS , delirium to the CAM-ICU , the consciousness item inside an early warning score to the MEWS and NEWS2 page (NEWS2 uses ACVPU, not the GCS), stroke deficits to the NIH Stroke Scale, and decision-making ability to a capacity evaluation ; a patient can score 15 and lack capacity, or score V4 from aphasia with normal arousal. ## Who uses GCS documentation and when Paramedics record the first GCS at the scene, and that field score, with its time and airway and drug context, follows the patient into the trauma registry and the coded record, so its components matter more than most prehospital guides admit. Emergency and trauma nurses and physicians repeat it on arrival, after resuscitation, and at every clinically meaningful transition (deterioration, intubation, leaving for and returning from CT), and trauma teams at ACS-verified centers abstract the eye, verbal, and motor values separately for the National Trauma Data Standard. ICU nurses chart it beside the RASS during sedation holds and mark the verbal component NT while the tube is in; neurosurgical, neurology, and stroke units run serial neurological checks on it at the interval the order or protocol sets; hospitalists and rapid response teams reach for it when an early warning score's AVPU or ACVPU item flags a change and a fuller description is needed. Pediatric emergency clinicians use a pediatric verbal adaptation for preverbal children and name it. The score also travels: a transfer-of-care summary and a discharge summary should carry the last components with their time and confounders, and an incident report after a fall should record the components at the moment of the event. Neighbors win in four places: the FOUR score when a patient is intubated and brainstem reflexes and breathing pattern matter, the RASS for sedation targets and depth, the CAM-ICU for delirium, and the NIH Stroke Scale for the focal deficits of stroke, none of which the GCS was built to measure. ## How to document GCS findings in the chart No law, payer, or the scale's authors prescribe a GCS note format; what exists is the 1974 scale, the 2014 structured approach with its NT rating, trauma-program and registry standards that keep the components separate, and coding rules that decide which score is reported. What survives review is an entry that time-stamps each assessment and places it in the course of care, records the three components before any total, names the stimulus and scores the best arm, writes NT with a reason instead of a 1, lists the confounders, records pupils beside the scale, and shows change in component terms with the action it prompted. Each element below carries the pitfall that most often undermines it. Time, phase of care, and version. Give every assessment a clock time and place it in the course of care: at the scene, on arrival, after airway and circulation were secured, before or after a sedative or paralytic, on the ward at a scheduled check, during a sedation hold. The Brain Trauma Foundation's prehospital guidance asks for the score after resuscitation and, where possible, before sedatives or paralytics or after they have worn off, and the ICD-10-CM coding rules key the code to when the score was taken (in the field, at ED arrival, at admission, 24 hours or more later), so an undated score is uncodable as well as uninterpretable. Name the assessor and the version: adult scale, or which pediatric adaptation. Pitfall: "GCS 14" with no time, so nobody can place it against the EMS score, the CT, or the propofol started twenty minutes earlier. Three components, then the total. Write eye, verbal, and motor in that order with their values (E3 V4 M6), then the total only when all three were testable. Where a number could be read two ways, add the observed behavior in a few words (opens eyes to spoken voice; converses but disorientated to time and place; obeys two-part commands with both arms). The components are the record; the total is shorthand for communication, triage, and group comparison, and the ACS TBI guideline says exactly that. A pediatric verbal scale is charted as such, never mapped onto the adult wording. Pitfall: A bare "GCS 11," which cannot tell the next reader whether the patient stopped talking or stopped obeying, the two very different events that produce the same number. Stimulus and best response. Record how the response was obtained: spontaneously, to a spoken request, or to physical pressure, and for pressure the site (fingertip, trapezius pinch, supraorbital notch), then use the same site at the next check so the trend compares like with like; the sternal rub is discouraged because it bruises and its responses are hard to interpret. Score the motor component from the arms, taking the best response, and describe a weaker or different response on the other side in its own sentence as a lateralizing finding. Leg movement can be a spinal reflex and is not scored. Pitfall: "M5" with no stimulus named, or a hemiparetic side silently averaged into the motor score and the lateralizing sign lost. Not testable (NT) instead of a 1. When a component cannot be examined, write NT and the reason: V NT (endotracheal tube), E NT (periorbital swelling), M NT (neuromuscular blockade, spinal injury, limb immobilization). Do not compute a total, because the sum would be falsely low, and record the components that remain as the tracked variables; the scale's site states both rules. The older notation of scoring an intubated patient as V1 with a T suffix survives in some EHR fields, trauma registries, and older teaching; where a system demands it, chart the bedside finding as NT with its reason as well, so the record shows that no lowest verbal response was ever observed. Pitfall: An alert, intubated patient charted "E4 V1 M6 = 11T," which asserts a verbal response of 1 nobody saw and understates the patient on every trend chart that follows. Confounders, named specifically. List what could alter the response and the timing of each: sedative, analgesic, or paralytic with drug, dose or infusion rate, and the time given or held; alcohol or other drugs with the level pending or known; aphasia or dysphasia; hearing loss; a language barrier and the interpreter used; hypotension or hypoxia at the time of testing; and any pre-injury baseline (dementia, developmental disability, chronic neurological disease) with the baseline itself recorded so change is measured against the right starting point, as NICE recommends. Record the response actually observed; no authority publishes an arithmetic correction for sedation, intoxication, aphasia, or dementia, and none should be applied in the chart. Pitfall: A low score read as brain injury while propofol runs, a low score written off to intoxication with no imaging, or "confused at baseline" with the baseline never recorded. Pupils and GCS-Pupils. Chart pupil size, symmetry, and reaction to light for each eye beside the scale, with the time, and repeat them with every check; pupils are not part of the GCS, and the site and the ACS guideline treat them as the second pillar of basic neuromonitoring. If your service uses the GCS-Pupils score, record the pupil reactivity score (0 for both pupils reacting, 1 for one unreactive, 2 for both unreactive), subtract it from the total, and label the result GCS-P, keeping the components and the pupil findings on the page; it is a prognostic index, not a replacement. When any component is NT there is no valid total, so a GCS-P cannot be computed without an imputation method that is named as such. Pitfall: "GCS-P 7" with no components or pupil findings, or pupil reactivity folded into the motor or eye score as if it were part of the scale. Change, escalation, and the next check. Compare with the previous assessment in component terms (motor fell from 6 to 5; eyes now open only to pressure) and state the time over which it happened, then record what was done: who was notified and when, the response, imaging or treatment ordered, and the new observation interval. No US, Canadian, or Australian authority sets a universal frequency; it comes from the order, the protocol, and the trajectory, and the most-cited external schedule and deterioration triggers are NICE's (half-hourly until 15, then a tapering schedule; urgent review for a sustained one-point drop, a two-point motor drop, or a three-point drop in eye or verbal). Deterioration should prompt urgent medical review, and the note should show it did. Pitfall: A trend of "13, 11, 9" with no times, no component detail, and no call, so the record shows a decline nobody appears to have acted on. ### Blank template (copy and adapt) GCS DOCUMENTATION BLOCK Date: [ ] Time: [ ] Setting: [field / ED / ward / ICU / transfer] Assessor: [ ] Phase of care: [scene / arrival / after resuscitation / after airway drugs / scheduled check / sedation hold] Scale version: [adult / pediatric: which] Eye opening: E [1 to 4 or NT] how obtained: [spontaneous / to sound / to pressure: fingertip] reason if NT: [ ] Verbal response: V [1 to 5 or NT] observed: [orientated / confused / words / sounds / none] reason if NT: [tube / aphasia / language / other] Motor response: M [1 to 6 or NT] best arm response to: [command / trapezius / supraorbital notch] weaker side, if any: [side and response] reason if NT: [ ] Total: [ ] of 15 (report only when E, V, and M are all testable) Pupils: right [size, reaction] left [size, reaction] PRS: [0 / 1 / 2] GCS-P (total minus PRS, only when a total exists): [ ] Confounders: [drug, dose or rate, time given or held / alcohol or drugs, level / aphasia / hearing / language and interpreter / hypotension or hypoxia / baseline before injury: state it] Change from prior: [prior E V M and time; which component moved; same stimulus site used] Action: [who notified, time; response; imaging or treatment; new check interval] Registry and coding notes: [field and arrival scores preserved; sedated or induced state flagged; components, not the total, documented] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the GCS Assessment Aid or any coma chart. ## Sample GCS documentation (fictional) Scenario: a US level II trauma center emergency department, where a trauma nurse documents serial GCS assessments for a 58-year-old man after a ladder fall with alcohol on his breath, records a deterioration in component terms with the escalation it prompted, switches to the NT convention after intubation, and hands the sequence to the ICU with the field and arrival scores preserved for the registry and the coders. All details are fictional. Patient: T.R., 58 · Setting: Emergency department, level II trauma center · Clinician: M. Okafor, RN (trauma nurse) · Note date: 09/12/2026 Arrival assessment and confounders (18:42): Adult GCS, 2014 structured approach, assessed by M. Okafor, RN, after airway, breathing, and circulation were confirmed on arrival at 18:42 on 09/12/2026. EMS scene score at 18:05 per the run sheet: E3 V4 M6, total 13, no airway drugs given. Arrival: E3 V4 M6, total 13; opens eyes to spoken voice, converses but is disorientated to time and place, obeys two-part commands with both arms; no pressure stimulus needed. Pupils 4 mm, equal, brisk to light bilaterally. Confounders: alcohol on breath, serum ethanol pending; no sedatives or analgesics given; no hearing, language, or speech barrier. His wife reports he was fully orientated and independent before the fall, with no dementia or prior neurological disease, so the pre-injury baseline is a full 15. Serial assessments and stimulus (19:15 and 19:45): 19:15: E3 V3 M6, total 12; still opens eyes to voice, now answers in single words and phrases that do not form a conversation, obeys commands with both arms; pupils unchanged. Trauma attending informed of the verbal change at 19:17; CT head, ordered on arrival, awaiting transport. 19:45: E2 V2 M5, total 9; eyes open only to fingertip pressure (right index finger); sounds without recognizable words; no longer obeys, localizes to trapezius pinch with the right arm and with the left arm, no asymmetry; the same stimulus sites will be used for every later check. Pupils: right 5 mm and unreactive to light, left 4 mm and brisk; the right pupil was reactive at 19:15. Pupil reactivity score 1; GCS-P 8, recorded here with the components and pupil findings it is built from. Deterioration and escalation (19:45 to 20:12): Change over 30 minutes: eye opening fell from 3 to 2, verbal from 3 to 2, motor from 6 to 5, total from 12 to 9, with a newly unreactive right pupil. Trauma attending at the bedside 19:47; neurosurgery paged 19:49 and at the bedside 20:02; CT head performed 20:05, result documented by the physicians; hypertonic saline given 20:12 per order and charted in the MAR. Ethanol result returned 19:50 and is filed with the labs; the decline was not attributed to alcohol, because the components changed over 30 minutes at the same stimulus sites and a pupil changed with them. Intubation and the NT convention (20:20 to 21:30): Intubated at 20:20 with etomidate and rocuronium (doses and times in the MAR); propofol infusion started 20:35. 20:40: E NT (neuromuscular blockade), V NT (endotracheal tube), M NT (neuromuscular blockade); no total reported; pupils right 5 mm unreactive, left 4 mm brisk, charted as the tracked variable until the blockade wears off. 21:30, blockade resolved per anesthesia: E1 V NT M4 (abnormal flexion to trapezius pinch, both arms), propofol running at the rate charted in the MAR; no total reported because the verbal component is not testable; sedation depth charted separately on the RASS. The verbal component was not scored as 1 and no T-suffixed total was written; the registry qualifier will record sedation and intubation. Handoff to ICU and record (22:05): Transferred to the neuro ICU at 22:05 with the sequence above in the transfer note as components with times: 18:05 field E3 V4 M6 (13); 18:42 arrival E3 V4 M6 (13); 19:15 E3 V3 M6 (12); 19:45 E2 V2 M5 (9), right pupil unreactive, GCS-P 8; 20:40 all three components NT under blockade; 21:30 E1 V NT M4 on propofol, no total. Neurological checks ordered hourly with pupils, using fingertip pressure and trapezius pinch as the standard stimuli. The field and arrival components are preserved for the trauma registry; the arrival components at 18:42 are the score the coding guidelines call for at presentation, and the scores after 20:20 are flagged as taken under sedation and not codable under the coma-scale rules. Wife updated at 22:10. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, times, dates, and details are invented to show documentation structure and are not clinical guidance. No assessment aid or scoring chart is reproduced. ↑ Back to the template and downloads ### Why this sample works Every assessment carries a clock time, the phase of care, the version, and the assessor, so the field score, the arrival score, and the post-intubation entries can be placed against the CT and the drugs. The components come first, with the observed behavior in words where a number could be read two ways, so the 12 and the 9 show which response changed rather than only that a number fell. The stimulus sites are named and reused, the motor score comes from the arms with symmetry stated, and the pupils are recorded beside the scale with the GCS-P labeled as an index built from them. After intubation the untestable components are NT with reasons, no total or T-suffixed score is written, the components that remain testable and the pupils are the tracked variables, and sedation is named as a confounder with the RASS kept separate. The deterioration is stated as component changes over a stated interval with the notification, response, and treatment times, and the handoff preserves the field and arrival scores for the registry and coders while flagging sedated scores as not codable. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the GCS, and the rules that do bind sit one step from the bedside. Under LAW, the Medicare Conditions of Participation require a nursing assessment for every inpatient (42 CFR 482.23) and a medical record whose entries are complete, dated, timed, and authenticated (42 CFR 482.24); neither prescribes the GCS, a component format, or a frequency, and the Joint Commission's assessment and reassessment standards do not either (CONVENTION), so a hospital that requires E, V, and M fields or hourly neurological checks is applying its own policy or a trauma-program standard, not a federal or accreditor mandate. The ICD-10-CM Official Guidelines for Coding and Reporting, whose use is required under HIPAA (LAW for coded claims), carry a coma-scale section that is the closest thing to a national documentation rule: the FY2026 guidelines (updated April 1, 2026) and the FY2027 guidelines posted by NCHS in June 2026 and effective October 1, 2026 state identically that the coma-scale codes R40.21- through R40.24- are used one from each subcategory with a matching seventh character for when the score was taken (in the field, at ED arrival, at hospital admission, 24 hours or more after admission, or at an unspecified time); that at a minimum the initial score documented on presentation is reported and it may be an EMT's or an emergency department score; that R40.24- is assigned only when a total alone is documented; that when several scores fall within the first 24 hours after admission only the admission score is coded; that no coma-scale codes are reported for a patient in a medically induced coma or a sedated patient; and that the codes are secondary and cannot accompany R40.2A. Section I.B.14 lets coders take the score from clinicians other than the patient's provider (the guidelines name the emergency medical technician) while the associated diagnosis must come from the provider. Under CONVENTION with registry teeth, the ACS National Trauma Data Standard (2026 and 2027 admission-year dictionaries) collects the eye, verbal, and motor values as separate fields with qualifiers for sedation, intubation, and eye obstruction, which is why verified trauma centers abstract components and never a bare total; the ACS TQP Best Practices Guidelines for the Management of Traumatic Brain Injury (2024) direct clinicians to assess and document each component, to document only the reactions of the best arm, to record the type of stimulus and keep it as the standard for later checks, and to treat severity classification from the sum alone as an oversimplification; and the 2021 National Guideline for the Field Triage of Injured Patients replaced the old total of 13 or less with an inability to follow commands, defined as a motor score below 6, as its mental-status criterion. The Brain Trauma Foundation defines severe TBI for its recommendations as a GCS of 3 to 8 after resuscitation and asks prehospital clinicians to score after resuscitation and, where possible, before sedatives or paralytics or after they have worn off; the teaching that a GCS of 8 or less means intubation is dogma rather than rule, and Hatchimonji and colleagues (2021) found intubation within an hour of arrival associated with slightly higher mortality and longer stays among 6,676 trauma patients with GCS 6 to 8. Under PAYER POLICY, no national payer requires a GCS format; GCS thresholds live inside proprietary level-of-care criteria, and because coma and coma-scale codes influence severity-of-illness and MS-DRG assignment, clinical documentation integrity teams query for the components, the timing, and the sedation status the guidelines turn on. Canada, Australia, and the UK-origin guidance everyone cites. In Canada, no federal or provincial LAW and no Accreditation Canada standard names the GCS or sets a component or interval rule (CONVENTION at most); the Canadian Triage and Acuity Scale uses level of consciousness among its modifiers, provincial trauma registries collect the components with their own intubation and sedation handling, and the Canadian CT Head Rule (Stiell and colleagues, Lancet 2001; 3,121 adults with GCS 13 to 15) makes failure to reach 15 within two hours of injury a high-risk criterion, with the high-risk factors 100 percent sensitive for neurological intervention in the derivation, which is the clearest argument that the timed trajectory, not one number, is the clinical variable. In Australia, the NSQHS Recognising and Responding to Acute Deterioration Standard (second edition; CONVENTION binding on accredited services) requires observation charts, track-and-trigger systems, and escalation pathways but names no consciousness instrument, so the general observation chart carries AVPU or ACVPU and the full GCS is applied when a fuller neurological assessment is needed; the Australasian Triage Scale descriptors published by ACEM reference the GCS, state ambulance services publish their own GCS procedures (some still in the older "to pain" wording, one more reason to name the stimulus), the Victorian State Trauma Registry dictionary carries a qualifier for whether the first GCS was a legitimate score or one affected by sedation, intubation, or paralysis, and ACSQHC has a third edition of the standards in development in 2026, so no new requirement should be asserted until it is final. The UK-origin rules bind no one in the three countries but are the most-cited external schedule: NICE NG232 (2023) asks that monitoring and every communication and record describe the three components alongside the total, sets an observation schedule after head injury of half-hourly until a GCS of 15 and then half-hourly for two hours, hourly for four hours, and two-hourly thereafter, defines urgent-review triggers (a one-point drop sustained for 30 minutes, a two-point motor drop, a three-point eye or verbal drop), and requires a pre-injury baseline for people with dementia, chronic neurological disorders, or learning disabilities; NEWS2 (Royal College of Physicians, 2017) uses ACVPU, not the GCS. Two boundaries hold everywhere. Version: the 2014 terminology (pressure, sound, abnormal flexion) and the NT rating are the current standard, older wording and the T suffix survive in some charts and registries, and pediatric adaptations differ by service, so the note names the version and the stimulus, and a change of version mid-trend is charted as one. Reliability: the systematic review by Reith and colleagues (2016) rated only 7 of 52 studies good quality, found the components more reliable than the sum, and identified training, level of consciousness, and stimulus type as the factors that move agreement, which is the evidence behind writing the stimulus and the behavior and not only the numbers. The Glasgow Coma Scale was published by Graham Teasdale and Bryan Jennett in the Lancet in 1974 and has been used worldwide since without a license; the glasgowcomascale.org permissions page, maintained by the Royal College of Physicians and Surgeons of Glasgow, states that the GCS, the GCS Aid, the GCS-P score, the GCS-P Age CT prognostic charts, and the verbal-score imputation tool may all be used for clinical care and clinical research at no cost with no license required, and asks that the copyright of the University of Glasgow in the scale, and of Sir Graham Teasdale in the Aid, the GCS-P, the charts, and the imputation tool, be acknowledged; the 2014 Nursing Times paper describes the structured-approach materials as copyright-free for clinical care, education, and academic work. That permission covers the bedside and the classroom; it is not a general grant to republish the structured assessment aid or its pictorial materials, which this page describes and does not reproduce. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common GCS documentation errors reviewers flag The numbers behind these errors are specific. The systematic review by Reith and colleagues (2016) found 52 reliability studies , rated 7 good, 18 fair, and 27 poor, found agreement higher for the components than for the sum, and identified education and training, the level of consciousness, and the type of stimulus as the factors that move it; Teasdale's 2014 update cites survey work showing at least seven different physical stimulus techniques in use. Across 54,069 trauma patients pooled from TARN, the Victorian State Trauma Registry, and CRASH, case fatality differed significantly between component profiles sharing the same total for every sum from 4 to 14 except 6 and 7, and the three components together predicted outcome better than the sum in each dataset (Reith and colleagues, Injury 2017). Among 6,676 patients with GCS 6 to 8 in the 2016 National Trauma Data Bank, intubation within an hour of arrival was associated with higher mortality (odds ratio 1.05, 95 percent confidence interval 1.03 to 1.06) and with ICU and hospital stays 14 and 27 percent longer (Hatchimonji and colleagues, 2021). In the derivation of the Canadian CT Head Rule among 3,121 adults with GCS 13 to 15 , failure to reach 15 within two hours was one of five high-risk factors that together were 100 percent sensitive for neurological intervention (Stiell and colleagues, 2001). No published audit counts how often a chart carries a total without components, a 1 where NT belongs, or an unnamed stimulus; the coding, registry, and trauma-program rules described under compliance considerations are what turn those omissions into a data-quality and defensibility problem. The BastionGPT Clinical Advisory Board sees the same errors most often in GCS documentation reviews: A total without its components. "GCS 11" cannot tell the next reader whether the patient stopped talking or stopped obeying, and the same total carries different mortality depending on which components produced it. Write E, V, and M in order with the total after them, add the observed behavior where a number is ambiguous, and trend the components, not the sum. A 1 where NT belongs. An intubated patient charted V1 or "10T," or a patient under propofol and rocuronium charted E1 V1 M1 as if neurologically comatose. A component that cannot be tested is NT with its reason, no total is reported, and the components that remain testable and the pupils become the tracked variables; sedation depth goes on the RASS, and coders may not assign coma-scale codes for a sedated or medically induced state in any case. Stimulus and side unrecorded. "M5" with no site named, a sternal rub used and left out, or a hemiparetic arm averaged into the motor score. Name the stimulus (spoken request, fingertip pressure, trapezius pinch, supraorbital notch), keep the same site for serial checks, score the best arm, and write the weaker side in its own sentence as a lateralizing sign. Confounders missing, or the score corrected. A low score read as brain injury with sedation running, written off to alcohol without imaging, or silently adjusted upward for dementia or aphasia. Record the response actually observed and, beside it, the drug with dose and time, the intoxicant and level, the language deficit or barrier, and the pre-injury baseline; no authority publishes a correction factor, and the chart should not invent one. Pupils folded in, or a GCS-P with no parts. Pupil reactivity counted into the eye or motor score, or "GCS-P 7" standing alone. Pupils are a separate examination recorded beside the scale with size, symmetry, and reactivity for each eye; the GCS-Pupils score subtracts the number of unreactive pupils from a valid total and is labeled as that index, with the components and the pupil findings on the same page. A trend without times or action, and database arithmetic mistaken for the exam. A sequence of totals with no clock times, no component detail, and no record of who was called, or a numeric total typed into an EHR or registry field from an NT component because the software demanded a number. Write each change as a component change over a stated interval with the notification and response, and let registry qualifiers and any labeled imputation carry the database need; the bedside record preserves the inability to test. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on emergency, trauma, and critical care nursing documentation. Give it the facts (the time and phase of care, the scale version and assessor, the three component values or NT reasons, how each response was obtained and the stimulus site, the pupil findings, the confounders with drug, dose, and time, the previous assessment, and the notification and response) and it drafts the entry: components in order with the total only when valid, the observed behavior in words, NT with reasons, confounders and baseline, pupils and any labeled GCS-P, the change in component terms, and the action taken, ready for your review. Cross-check a finished note or flowsheet narrative for the gaps reviewers flag: a total with no components, a 1 or a T suffix where NT belongs, an unnamed stimulus, a lost lateralizing sign, an unrecorded sedative or baseline, pupils folded into the scale, or a decline with no time and no call. Turn the sequence into the next document: the neurological status lines of a transfer-of-care or discharge summary with the last components, time, and confounders, the timeline paragraph of an incident report after a fall, or a plain-language update for a family, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How do I write a GCS in the chart: the components, the total, or both? Both, in that order. Write the time, then eye, verbal, and motor with their values (E3 V4 M6), then the total only when all three components were testable, and add the observed behavior in a few words where a number could be read two ways (opens eyes to voice; confused conversation; obeys commands with both arms). The components are the assessment; the total is a summary the scale's authors introduced for communication and group comparison, and the same total can come from very different patients: an 8 can be E2 V2 M4 or E1 V1 M6. The scale's own site, the ACS trauma program's TBI guideline (2024), and NICE NG232 (2023) all ask for the components to accompany the total in every record and every handoff, and a large pooled analysis found that mortality differed between component profiles with identical totals for nearly every sum. The blank template and the sample on this page carry the format; the transfer-of-care summary page shows the last components traveling with the patient. How do I document the GCS for an intubated or sedated patient, and is it 1T or NT? NT. When a component cannot be tested, the current structured approach (Teasdale and colleagues, 2014, and the glasgowcomascale.org FAQ) records it as not testable with the reason, and no total is reported, because a sum with a 1 standing in for an untestable component is falsely low and implies the patient is worse than observed. So an alert intubated patient is E4 V NT (endotracheal tube) M6, not E4 V1 M6 = 11T, and a patient under a neuromuscular blocker is E NT, V NT, M NT with the drug and time named, not a GCS of 3. Sedation without paralysis is different: the eye and motor responses that can be observed are recorded with the sedative, its rate, and the time given or held as a confounder, and the trend is read across sedation holds; sedation depth itself is charted on the RASS . The older notation of scoring the verbal component as 1 and adding a T persists in some EHR fields, registries, and teaching; where your system demands it, chart the bedside finding as NT with its reason as well. For coding, the ICD-10-CM guidelines direct coders not to assign coma-scale codes for a sedated patient or a medically induced coma at all. Which stimulus do I use and record, and what if the two sides respond differently? Use the sequence the 2014 structured approach sets out and write down what you used. Observe first, then speak or shout a request, then apply physical pressure: at the fingertip, by pinching the trapezius, or at the supraorbital notch, escalating only as needed. Rubbing the sternum is strongly discouraged because it bruises and produces responses that are hard to interpret, and pressure behind the jaw is not recommended for routine use. Record the site ("localizes to trapezius pinch") and use the same site at the next check so the trend compares like with like; the ACS TBI guideline (2024) asks for exactly that, and stimulus type is one of the factors the reliability literature shows changes agreement between raters. The motor score is the best response from the arms: document only the arm reactions, not the legs, which can move by spinal reflex. When the two arms differ, score the better one and describe the other in its own sentence ("left arm localizes; right arm flexes abnormally") as a lateralizing sign, which is clinical information the GCS number was never designed to hold. How often should the GCS be repeated, and what change should trigger escalation? There is no universal rule in the United States, Canada, or Australia: the interval comes from the order, the unit protocol, the diagnosis, and the trajectory, and national standards (CMS, the Joint Commission, Accreditation Canada, the NSQHS deterioration standard) require adequate monitoring and escalation without naming a GCS frequency. The most-cited external schedule is NICE NG232 (2023) for head injury: half-hourly until the GCS reaches 15, then half-hourly for two hours, hourly for four hours, and two-hourly after that, with urgent review for a one-point drop sustained 30 minutes (more weight on a motor drop), a two-point motor drop, or a three-point drop in eye or verbal response. The Brain Trauma Foundation asks prehospital clinicians to repeat the score to detect change, and the ACS TBI guideline asks for a baseline as soon as threats to life are managed and repeat assessments at clinically meaningful transitions. Whatever your interval, chart an unscheduled reassessment whenever something changes, write the change in component terms over the interval it took, and record who was notified, when, and what followed, because a documented decline with no documented call is the pattern reviewers and attorneys look for. The MEWS and NEWS2 page covers the consciousness item inside early warning scores, which is AVPU or ACVPU rather than the GCS. What do mild, moderate, and severe mean, and does a GCS of 8 mean the patient must be intubated? The severity labels are a trauma convention, not part of the scale. Totals of 13 to 15 are conventionally called mild, 9 to 12 moderate, and 3 to 8 severe; the bands came from trauma outcome studies in the late 1970s and early 1980s and spread through ATLS and later guidelines, the scale's authors did not define them in 1974, and the ACS TBI guideline (2024) cautions that classifying injury from the sum alone is an oversimplification. Two refinements matter for the chart. The Brain Trauma Foundation defines the severe TBI population for its recommendations as a GCS of 3 to 8 after resuscitation, so a score taken during hypotension or before sedatives have worn off should not be labeled as if it were the resuscitated score. And a total of 13 behaves more like moderate than mild injury in several studies, which is one reason the 2021 US field triage guideline moved from a total of 13 or less to an inability to follow commands (a motor score below 6) as its mental-status criterion. On intubation: no guideline states that a GCS of 8 mandates it; the decision rests on airway protection, oxygenation and ventilation, mechanism, and trajectory, and among 6,676 trauma patients with GCS 6 to 8 in the 2016 National Trauma Data Bank, intubation within an hour of arrival was associated with slightly higher mortality and longer stays (Hatchimonji and colleagues, 2021). Document the components, the timing relative to resuscitation and drugs, and the clinical reasons for the airway decision, not a threshold. What is the GCS-Pupils score, and how do I document it? The GCS-Pupils score (GCS-P) is a prognostic index published by Brennan, Murray, and Teasdale in the Journal of Neurosurgery in 2018 from pooled IMPACT and CRASH data on about 15,900 patients with traumatic brain injury. It takes the conventional total and subtracts a pupil reactivity score: 0 when both pupils react to light, 1 when one does not, 2 when neither does, so the index runs from 1 to 15 and extends the scale's floor below 3 for the most severely injured. The scale's site states that it is not intended to replace separate assessment and reporting of each component; it is shorthand for severity and prognosis. Documentation follows from that: record E, V, and M with the total, record pupil size, symmetry, and light reaction for each eye with the time, then the pupil reactivity score and the GCS-P labeled as such (E2 V2 M5, total 9; right pupil unreactive, PRS 1; GCS-P 8). Pupil size does not enter the arithmetic, only reactivity. If any component is NT there is no valid total, so a GCS-P cannot be computed unless a named imputation method is used and labeled; do not fill in a verbal value to make the subtraction work. How is the GCS coded in ICD-10-CM, and can a nurse's or paramedic's score be used? Through the R40.2- coma-scale codes, under rules that reward component documentation. The ICD-10-CM Official Guidelines (FY2026, updated April 1, 2026, and the FY2027 edition posted in June 2026 and effective October 1, 2026, which carries the section forward unchanged) state that the coma-scale codes R40.21- through R40.24- are used with traumatic brain injury codes and in any setting that collects the information, one code from each subcategory, with a seventh character that shows when the score was taken (in the field, at ED arrival, at hospital admission, 24 hours or more after admission, or unspecified) and must match across the three codes. At a minimum the initial score documented on presentation is reported, and it may be the EMT's or the emergency department score; R40.24- is used only when a total alone is documented and no components; when several scores fall in the first 24 hours after admission only the admission score is coded; the codes are secondary, cannot be used with R40.2A, and are never assigned for a sedated patient or a medically induced coma. Section I.B.14 lists the coma scale among the items coders may take from clinicians other than the patient's provider, naming the emergency medical technician as the typical documenter, while the associated diagnosis must come from the provider. So a paramedic's or nurse's E, V, and M with a time is codable data; a bare total costs the record specificity, and an undated score cannot be placed in the seventh-character scheme at all. How do I document the GCS in a young child, or in a patient with dementia, aphasia, or a language barrier? Name the version and the confounder, and never adjust the number. Children: the adult scale is used without modification from about age 5 according to the scale's authors, and from age 2 in the Brain Trauma Foundation's prehospital guidance, so the age boundary is a service decision; below it, a pediatric verbal adaptation (the Adelaide scale of Reilly and colleagues, 1988, or the pediatric GCS the PECARN network studied in more than 42,000 children, where it was somewhat less accurate than the standard scale for injury on CT but equivalent for clinically important injury) replaces orientation questions with age-appropriate vocal and social behavior. Write which one you used, describe the behavior that earned the verbal rating, mark NT any component a preverbal child cannot perform on command, and keep the same version across the trend. Dementia, developmental disability, or chronic neurological disease: record the pre-injury baseline from a carer or the record and compare against it, as NICE NG232 asks; a chronically disorientated patient can have a verbal score below 5 without any change in arousal, and no validated correction exists. Aphasia or dysphasia: the verbal component cannot separate a language deficit from reduced consciousness, so record the observed response and the language deficit in its own sentence. Language barrier or deafness: adapt the method (an interpreter, a culturally appropriate examiner, written communication) and note it; where the verbal component still cannot be validly tested, record NT with the reason rather than a low number. Can BastionGPT draft my GCS documentation? Yes. Give it the facts (the time and phase of care, the scale version and assessor, the three component values or the NT reasons, how each response was obtained and the stimulus site, the pupil findings, the confounders with drug, dose, and time, the previous assessment, and who was notified and what followed) and it drafts the full entry: components in order with the total only when valid, the observed behavior in words, NT with reasons, confounders and baseline, pupils and any labeled GCS-P, the change in component terms, and the action taken, ready for your review. It can also check a finished note for a total with no components, a 1 or a T suffix where NT belongs, an unnamed stimulus, a lost lateralizing sign, an unrecorded sedative or baseline, pupils folded into the scale, or a decline with no time and no call. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Teasdale G, Jennett B, 1974, Lancet 2(7872):81-84, assessment of coma and impaired consciousness: a practical scale ; Teasdale G, Allan D, Brennan P, McElhinney E, Mackinnon L, 2014, Nursing Times 110(42):12-16, forty years on: updating the Glasgow Coma Scale (structured approach; pressure; NT; stimulus sites; sternal rub discouraged; adult scale from age 5; materials copyright-free for clinical care, education, and academic work); Teasdale G, Maas A, Lecky F, Manley G, Stocchetti N, Murray G, 2014, Lancet Neurology 13(8):844-854, the Glasgow Coma Scale at 40 years . Glasgow Coma Scale official site (Royal College of Physicians and Surgeons of Glasgow), accessed September 2026: permissions (free for clinical care and research, no license, acknowledgment of the University of Glasgow and Sir Graham Teasdale), FAQ (NT rather than 1; no total when a component is NT; components always described; children over 5), recording the GCS (E2V2M4 and E1V1M6 both total 8; serial charting; deterioration prompts urgent review), and GCS-Pupils ; Brennan PM, Murray GD, Teasdale GM, 2018, Journal of Neurosurgery 128(6):1612-1620, the GCS-Pupils score . Reith FCM, Van den Brande R, Synnot A, Gruen R, Maas AIR, 2016, Intensive Care Medicine 42(1):3-15, the reliability of the Glasgow Coma Scale: a systematic review (52 studies; 7 good, 18 fair, 27 poor; components more reliable than the sum); Reith FCM, Synnot A, van den Brande R, Gruen RL, Maas AIR, 2017, Neurosurgery 80(6):829-839, factors influencing the reliability of the Glasgow Coma Scale ; Reith FCM, Lingsma HF, Gabbe BJ, Lecky FE, Roberts I, Maas AIR, 2017, Injury 48(9):1932-1943, differential effects of the Glasgow Coma Scale Score and its components: 54,069 patients . Centers for Disease Control and Prevention, National Center for Health Statistics, ICD-10-CM Official Guidelines for Coding and Reporting, FY2026, updated April 1, 2026 , and FY2027, effective October 1, 2026 (Sections I.B.14 and I.C.18.e; bare code numbers only on this page); CMS, ICD-10 code files ; eCFR, 42 CFR 482.24 and 42 CFR 482.23 . American College of Surgeons, 2024, TQP Best Practices Guidelines: the Management of Traumatic Brain Injury (assess and document each component; best arm; stimulus type documented and kept standard; sum for group-level use; pediatric GCS for preverbal children); ACS, National Trauma Data Standard data dictionary (2026 and 2027 admission years); ACS and the National Expert Panel on Field Triage, 2021, National Guideline for the Field Triage of Injured Patients , and Newgard CD and colleagues, 2022, Journal of Trauma and Acute Care Surgery 93(2):e49-e60, recommendations of the National Expert Panel on Field Triage ; Brain Trauma Foundation, guidelines for prehospital management of TBI, second edition (GCS after resuscitation and before sedatives or paralytics; adult protocol from age 2) and severe TBI guidelines ; Hatchimonji JS and colleagues, 2021, European Journal of Trauma and Emergency Surgery 47(6):2073-2079, questioning dogma: does a GCS of 8 require intubation? (6,676 patients; OR 1.05). Pediatric adaptations: Reilly PL, Simpson DA, Sprod R, Thomas L, 1988, Child's Nervous System 4(1):30-33, assessing the conscious level in infants and young children: a paediatric version of the Glasgow Coma Scale ; Borgialli DA and colleagues (PECARN), 2016, Academic Emergency Medicine 23(8):878-884, performance of the pediatric Glasgow Coma Scale score in children with blunt head trauma (42,041 children, 10,499 under 2). Canada and Australia: Stiell IG and colleagues, 2001, Lancet 357(9266):1391-1396, the Canadian CT Head Rule for patients with minor head injury (3,121 adults; failure to reach GCS 15 within 2 hours; 100 percent sensitivity); Accreditation Canada, standards ; Australian Commission on Safety and Quality in Health Care, Recognising and Responding to Acute Deterioration Standard ; Australasian College for Emergency Medicine, triage and the Australasian Triage Scale ; Victorian State Trauma Registry, 2024, data dictionary version 6.4 (GCS qualifier). UK-origin references: National Institute for Health and Care Excellence, 2023, NG232, head injury: assessment and early management (three separate responses in every record; observation schedule; deterioration triggers; pre-injury baseline); Royal College of Physicians, 2017, National Early Warning Score (NEWS) 2 (ACVPU, not the GCS). ## Related templates RASS Score Documentation CAM-ICU Documentation MEWS & NEWS2 Early Warning Score Documentation Transfer-of-Care Summary Discharge Summary Incident & Adverse Event Report Capacity & Competency Evaluation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/braden-scale All note templates # Braden Scale Documentation: Risk Score to Interventions & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Braden Scale for Predicting Pressure Sore Risk is a copyrighted nursing instrument (Bergstrom, Braden, Laguzza, and Holman, 1987) that rates six subscales for a total of 6 to 23, lower meaning higher pressure injury risk. Bedside, long-term-care, and home health nurses use it to decide who needs prevention measures and which ones. This page covers how to document the score and the interventions it triggers, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Registered nurses at the bedside, in long-term care, and in home health, with licensed practical or vocational nurses where facility policy and state scope allow; no publisher qualification level applies, the licensor's training materials come with a Braden Scale II license, and the nurse who observed the patient enters the ratings Audience Wound, ostomy, and continence nurses and skin integrity teams, nurse managers and unit educators, attending physicians and advanced practice providers, dietitians and therapists, coders determining present-on-admission status, quality and patient safety staff, CMS and state surveyors, accreditation reviewers, and expert reviewers in pressure injury claims Typical length 4 to 10 chart lines (trigger and version, six subscale ratings and total, facility risk category, separate skin inspection, intervention for each low subscale, reassessment date) · 2 to 5 minutes by hand after the patient has been observed Format family Observer-rated pressure injury risk assessment (six subscales, five scored 1 to 4 and friction and shear 1 to 3, total 6 to 23, lower total meaning higher risk; risk categories set by facility protocol from the authors' recommendations) When it's used Admission and transfer assessments in acute and critical care, scheduled reassessment under unit policy, change of condition, long-term care admission and periodic reassessment feeding MDS Section M, home health start of care and recertification, perioperative and rehabilitation admissions; the Braden Q and Braden QD in pediatrics Standards context Bergstrom, Braden, Laguzza, and Holman (Nursing Research, 1987); copyrighted by Barbara Braden and Nancy Bergstrom, licensed since April 2021 through Health Sense Ai, an HD Nursing subsidiary; named by CMS and accreditation bodies only as an example of a validated tool and mandated by none; described here for documentation, no descriptors reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Braden Scale? The Braden Scale for Predicting Pressure Sore Risk is a nurse-rated instrument that estimates a patient's risk of developing a pressure injury. Barbara Braden and Nancy Bergstrom published the conceptual model in Rehabilitation Nursing in 1987 and, with Ann Laguzza and Victoria Holman, the instrument itself in Nursing Research the same year (volume 36, pages 205 to 210). It rates six subscales, named here by their published labels with the intent of each in original words: sensory perception (whether the patient can notice pressure discomfort and act on it or call for help), moisture (how wet the skin stays through the day), activity (how much the patient is up and about), mobility (how well the patient repositions without help), nutrition (how well the patient is actually eating), and friction and shear (how much the skin drags against bed and chair surfaces during movement and transfers). Five subscales are scored 1 to 4 and friction and shear 1 to 3, so the total runs from 6 to 23 and a lower total means higher risk. The descriptors that define each rating are the copyrighted text of the instrument and are not reproduced or paraphrased on this page; the licensed form is the source of record. The authors founded Prevention Plus to license the scale; on April 20, 2021 Health Sense Ai, a subsidiary of HD Nursing, completed its acquisition of Prevention Plus and announced that it would be the sole licensor going forward, and in 2022 it released the Braden Scale II, developed by Amy Hester in consultation with Braden and Bergstrom, with updated terminology and descriptors, available with a toolkit under a facility license. Two pediatric siblings have their own rights holder: the Braden Q (Curley and colleagues, Nursing Research, 2003), validated in children from about three weeks to eight years of age and adding a tissue perfusion and oxygenation subscale, and the Braden QD (Curley and colleagues, Journal of Pediatrics, 2018), validated from preterm infants through age 21 and built to capture medical-device-related as well as immobility-related risk. The load-bearing fact for documentation is that the total is a risk signal and not a care plan. The authors, the Hartford Institute for Geriatric Nursing, and the international guideline bodies all say that a low rating on any single subscale should drive its own preventive measure regardless of the total, because two patients with the same total can reach it by different routes: one through immobility, another through moisture or poor intake. The scale also does not examine the skin. It predicts risk; a separate visual and tactile inspection establishes whether an injury already exists, and the pressure injury staging page covers how that finding is described. The risk categories attached to the total are conventions rather than part of the 1987 instrument: the original validation set its cutoff at 16, the 1998 multisite study by Bergstrom, Braden, Kemp, Champagne, and Ruby moved the best overall cutoff to 18 and found that reassessment 48 to 72 hours after admission predicted better than the admission score, a 2024 systematic review of 34 intensive care studies found most empirically derived ICU cutoffs in the range of 12 to 14, and the Cochrane review by Moore and Patton (2019) found no reliable evidence that a structured risk tool reduces pressure ulcer incidence compared with clinical judgment. None of that makes the scale useless; it means the value lies in what the ratings prompt a nurse to do and to write down. A home health agency may adopt the Braden under its own policy, but the current OASIS item set carries no named risk-tool item (the OASIS assessment page covers the items that do exist), and the Morse Fall Scale is the structural peer for the same lesson on the falls side: a risk total that must not be mistaken for a plan. ## Who uses Braden Scale documentation and when Bedside nurses on medical, surgical, orthopedic, and rehabilitation units chart it on admission, on transfer, at the interval unit policy sets, and whenever the patient's condition changes; critical care nurses chart it beside sedation and delirium scores such as the RASS and the CAM-ICU , where sedation, ventilation, and vasopressors cluster patients at high risk and make the subscale detail more useful than the total. Long-term care nurses use it at admission, at the reassessment interval the facility protocol sets, and with any change of condition, and the result informs MDS 3.0 item M0100 and the care plan that CMS surveyors read under F686. Home health nurses record it at the start of care and recertification under agency policy, together with the caregiver teaching it prompts. Wound, ostomy, and continence nurses and skin integrity teams audit the entries; nurse managers and educators calibrate raters; dietitians and therapists receive the referrals that low nutrition, mobility, and activity ratings generate; physicians and advanced practice providers document the diagnosis and present-on-admission status that a coder needs, which depend on the skin inspection rather than the score. The Braden page owns one job: the risk assessment, the numeric ratings, the facility's risk category, and the link from each low subscale to a preventive action and a reassessment date. Neighbors take over from there: an existing wound is described on the pressure injury staging page, the risk status and prevention regimen travel in the transfer of care summary and the discharge summary , a facility-acquired injury is reported through the incident report when policy treats it as an event, and the interdisciplinary plan the ratings feed belongs in the care coordination note . ## How to document a Braden Scale assessment in the chart No regulator prescribes a Braden note format, and no authority in the United States, Canada, or Australia requires the Braden by name; what exists is a copyrighted instrument, facility protocols that set the reassessment interval and the risk categories, and survey and coding rules that judge whether an identified risk led to action. What survives review is an entry that states the trigger and the version, records all six ratings with the total, names the facility category as policy, documents a separate skin and device-site inspection, ties every low subscale to a specific intervention that was actually delivered, records the patient's response, and sets the next reassessment date and the triggers that bring it forward. Each element below carries the pitfall that most often undermines it. Trigger, timing, and version. Open with why the assessment is being done and which instrument was used: admission, transfer, the scheduled interval under a named policy, or a change in condition (new sedation or immobility, incontinence, falling intake, fever, surgery), then the version (original Braden Scale or Braden Scale II under the facility license), the date and time, and the rater. The 1998 multisite validation used admission and 48 to 72 hour reassessment, and CMS sets no interval; an every-shift or weekly schedule is the facility's policy and should be cited as such. When reassessment is triggered by a change, name the change, because that sentence is what turns a repeated number into evidence of monitoring. Pitfall: A time-stamped duplicate total with no trigger, or a note that presents the unit's every-shift schedule as a CMS requirement. Six subscale ratings and the total. Enter all six ratings as numbers in the published order with the total, and keep them visible in the note or flowsheet, not just the total. The total conceals the mechanism, and the ratings are less reliable than they look: Kottner and Dassen found totals for the same nursing home residents differing by as much as nine points between trained raters, with sensory perception and nutrition the least reliable subscales, and Magnan and Maklebust found that 1,391 web-trained nurses classified risk correctly 82.6 percent of the time, with the mid-range hardest. Rate the patient you observed today; do not carry yesterday's ratings forward as a new assessment, and say in one clause what the low ratings rest on in ordinary clinical words (intake percentages, how the patient moved during care, continence episodes), never in the scale's descriptor language. Pitfall: "Braden 14" with no subscale ratings, or ratings auto-populated from the last shift and signed as a fresh assessment. Facility risk category, named as policy. Write the category the facility protocol assigns to the total and cite the protocol, because the categories are conventions layered on the instrument rather than the instrument's own rule: the 1987 validation used a cutoff of 16, the 1998 and 2002 multisite studies supported 18, most empirically derived intensive care cutoffs fall around 12 to 14, and facilities adopt the authors' later graded recommendations in their own policies. A category tells the reader which protocol tier applies; it does not by itself prescribe a mattress or a turning interval, and a note that reads "Braden 14 therefore every two hours" should be written only when that linkage is literally the organization's approved protocol. Pitfall: A category presented as the scale's rule, or a single universal cutoff applied to an ICU patient whose unit has validated a different one. Skin and device-site inspection, charted separately. Record a head-to-toe visual and tactile inspection of the pressure points and every device contact site as its own act, with the findings: intact, or the location, laterality, and description of any change. For darker skin tones, compare with the person's usual tone and palpate for warmth, firmness, and tenderness rather than relying on visible redness. The inspection, not the score, establishes whether an injury was present on admission, and an existing injury is staged and described under the staging conventions by the nurse and diagnosed by the provider; a coder may take the stage from nursing documentation but the diagnosis from the provider's note, so the two must agree. Pitfall: The Braden score charted as the skin assessment, or a sacral injury found on day four with no admission inspection to show it was not there. An intervention for each low subscale. For every subscale rated low, name the measure actually started or continued and how it is delivered: for mobility and activity, the pressure redistribution surface by type, the repositioning schedule with the real interval and assistance level, heel offloading, and the mobilization plan; for moisture, the cause being managed, the cleansing and barrier regimen, and the continence or microclimate measures; for nutrition, intake monitoring, weights, and the dietitian referral, without equating a low rating with malnutrition; for friction and shear, the transfer technique and equipment and head-of-bed limits; for sensory perception, the surveillance and teaching added because the patient may not report discomfort. Do this even when the total sits above the facility threshold, which is the position of the authors, HIGN, and the international guideline, and write the schedule you actually use rather than one the total is assumed to dictate. Pitfall: "Prevention protocol initiated" or "skin bundle in place" with no intervention that answers a specific low rating. Patient response, refusal, and preferences. Chart what was delivered and how the patient responded, and when repositioning or a device is declined, chart the offer, the patient's stated reason where appropriate, the education given, the alternatives offered or accepted, who was notified under policy, and the revised plan. The international guideline builds patient preference and comfort into repositioning decisions, and survey and serious-event frameworks recognize refusal and clinical instability as circumstances that require documentation when preventability is judged. The rating itself does not change because an intervention was refused; the feasibility of the plan does, and the note should say so. Pitfall: "Refused" with nothing after it, or a repositioning log showing care that the patient declined. Reassessment date, triggers, and plan revision. Close with the next scheduled reassessment date under the policy and the changes that bring it forward: transfer, surgery, new sedation or immobility, new incontinence, falling intake or weight, fever or hemodynamic decline, and any new skin finding. When a total changes, say which subscale moved and why in clinical terms, and revise the plan to match; a score should improve only because the patient's condition improved, never because a better mattress was placed, since the scale rates the patient and not the interventions. Carry the risk status, the surface, and the regimen into the handoff and the transfer of care summary , because transitions are reassessment points in every guideline. Pitfall: "Continue to monitor" with no date, a total that "improved" after a mattress change, or a care plan never revised after the ratings moved. ### Blank template (copy and adapt) BRADEN SCALE DOCUMENTATION BLOCK (risk assessment and linked plan) Date / time: [ ] Setting: [ ] Rater: [ ] Trigger: [admission / transfer / scheduled per policy (name it) / change of condition (state the change)] Version: [original Braden Scale / Braden Scale II under facility license] Subscale ratings (numbers only): sensory perception [ ] moisture [ ] activity [ ] mobility [ ] nutrition [ ] friction and shear [ ] Total: [ ] of 23 Prior total and date: [ ] Subscale that changed: [ ] Basis for the low ratings (plain clinical facts): [intake percentages, weights, observed movement during care, continence episodes, sedation] Facility risk category: [ ] per [protocol name]; categories are policy, not the scale's rule Skin and device-site inspection (separate act): [pressure points and each device site; intact or location, laterality, description; present on admission determination; staging documented separately if an injury exists] Interventions by low subscale (what is actually in place): mobility / activity: [surface type; repositioning interval and assist level; heel offloading; mobilization plan] moisture: [cause managed; cleansing and barrier regimen; continence plan] nutrition: [intake monitoring; weights; dietitian referral and date] friction and shear: [transfer technique and equipment; head-of-bed limit] sensory perception: [surveillance; patient and caregiver teaching] Patient response: [accepted / declined: offer, reason, education, alternatives, who was notified, plan revised] Reassessment: [next date per policy; earlier if ...] Care plan revised: [ ] Communication: [provider, dietitian, therapy, skin team, handoff, MDS or OASIS] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's subscale descriptors or rating anchors. ## Sample Braden Scale documentation (fictional) Scenario: a US skilled nursing facility's scheduled weekly Braden reassessment, three weeks after an 84-year-old resident's admission following hip fracture repair; the nurse records all six ratings, finds that only the nutrition rating has fallen, names the facility's risk category as policy, charts a separate skin inspection, ties the changed rating to a dated dietitian referral, and resets the reassessment. All details are fictional. Patient: E.W., 84 · Setting: Skilled nursing facility, long-term care unit; scheduled weekly reassessment · Clinician: M. Okafor, RN · Note date: 09/17/2026 Trigger, version, and rater: Scheduled weekly reassessment, week three of the four weekly reassessments the facility's pressure injury prevention policy (NSG-114) requires after admission; the interval is facility policy, not a CMS requirement. Original Braden Scale as built in the facility's electronic record, rated by M. Okafor, RN, at 09:40 on 09/17/2026 after morning care and after review of the meal intake and weight records for 09/10/2026 through 09/17/2026. Prior totals on the same instrument: 15 on admission 08/27/2026 (admitted after right hip fracture repair in late August), 15 on 09/03/2026, 15 on 09/10/2026. Ratings and total: Sensory perception 3, moisture 3, activity 2, mobility 2, nutrition 2, friction and shear 2; total 14 of 23 (15 on 09/10/2026). The only change is nutrition, from 3 to 2. Basis in the record: meal intake documented at 25 to 50 percent for 11 of the 21 meals since 09/10/2026 and refused at 3 of them, against 75 percent or more at most meals the week before; weight 61.2 kg today against 62.6 kg on 08/27/2026, a 2.2 percent loss; E.W. reports poor appetite and intermittent nausea since a new medication was started on 09/08/2026. The unchanged ratings rest on the same observations as last week: she transfers bed to wheelchair with two-person assist and a slide sheet, shifts her weight in the chair when prompted, spends about six hours a day in the wheelchair and walks 15 meters in therapy with a walker and standby assist, is continent of bowel and of bladder apart from occasional nighttime urgency, and reports discomfort reliably. Facility risk category: Moderate risk under NSG-114, which sets its categories by total from the authors' published recommendations; last week's total placed her in the policy's at-risk tier. The category selects the protocol tier that applies. The interventions below answer the individual ratings, and this note applies the protocol without restating it. Skin and device-site inspection (separate act): Head-to-toe visual and tactile inspection completed with morning care at 08:30: sacrum, coccyx, both ischial tuberosities, both trochanters, both heels, elbows, scapulae, and occiput intact, with no discoloration, warmth, or induration on palpation; light skin tone; right hip incision healed with no erythema; hearing aid contact sites and eyeglass bridge intact. No pressure injury. Unchanged from the admission inspection of 08/27/2026, which documented intact skin at every site. Staging not applicable. Interventions by low subscale: Nutrition (2, changed): dietitian referral entered 09/17/2026 at 10:15 for assessment within 72 hours per policy; meal intake to be recorded at every meal and reviewed daily; weekly weights continued, next on 09/24/2026; mid-morning and evening snacks offered daily in line with her stated preference for smaller, more frequent meals; attending notified by secure message at 10:40 of the appetite change following the 09/08/2026 medication start, order for the dietitian consult received at 13:15, medication review requested; daughter (health care representative) informed by phone at 14:05. Mobility and activity (2, unchanged): pressure redistribution foam mattress in place since admission; assisted repositioning every three hours overnight on the individualized schedule agreed with E.W. on 08/28/2026 (the policy default of every two hours was relaxed to three at her request, with intact skin at every weekly inspection since); weight shift at least hourly in the wheelchair on a pressure redistribution cushion; heels floated on a pillow under the calves in bed; therapy five days a week continuing. Friction and shear (2, unchanged): two-person assist with a slide sheet for bed repositioning, no dragging, head of bed at 30 degrees or lower except at meals. Moisture (3) and sensory perception (3): scheduled toileting before bed and at 02:00 continued; no barrier product indicated; she reports discomfort reliably, so no added surveillance. Response, reassessment, and communication: E.W. accepts repositioning and toileting. She declined the evening snack on 09/12/2026 and 09/15/2026 (offered, reason given as feeling full, yogurt accepted on 09/15/2026 instead), documented on the intake record. Next scheduled Braden 09/24/2026 (week four), then quarterly with the MDS and with any change of condition; reassess sooner for acute illness, new incontinence, reduced mobility, further weight loss, intake below 50 percent for three consecutive days, or any new skin finding. Care plan revised today: nutrition problem added with the goal of intake at 50 percent or more at two of three meals by 09/30/2026 and weight stable at the 09/24/2026 weigh-in; pressure injury prevention interventions continued as above. Reported at shift handoff and to the skin integrity team meeting on 09/18/2026; MDS coordinator notified for the quarterly assessment. This sample is fictional and for educational purposes. It does not describe a real patient or record; the ratings, dates, and details are invented to show documentation structure and are not clinical guidance. No subscale descriptors or rating anchors are reproduced. ↑ Back to the template and downloads ### Why this sample works The trigger names the facility policy and its week count, the version and rater are stated, and the prior totals sit beside today's, so a surveyor reads monitoring rather than a repeated number. All six ratings appear with the total, the one that moved is identified, and its basis is written in intake percentages, weights, and observed movement rather than in the scale's descriptor language. The risk category is attributed to the facility protocol, so nobody reads it as the scale's rule, and the interventions answer the individual ratings instead of the category. The skin and device-site inspection is a separate act with locations, palpation findings, and the admission baseline, which is what a present-on-admission determination and any later staging would rest on. The changed rating has a dated referral, a notification chain, and a resident-preference detail behind it; the declined snacks are documented with the alternative; and the reassessment date, the earlier triggers, and the care plan revision close the loop surveyors check under F686. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the Braden, and the rules that do exist judge what the score led to. In long-term care, 42 CFR 483.25(b)(1) (LAW, surveyed as F686) requires a facility to ensure that a resident who enters without pressure ulcers does not develop them unless the resident's clinical condition demonstrates they were unavoidable, and that a resident with pressure ulcers receives the care needed to promote healing, prevent infection, and prevent new ones; the interpretive guidance in Appendix PP of the State Operations Manual frames avoidability around four steps (evaluate the resident's condition and risk factors, implement interventions consistent with the resident's needs and professional standards, monitor and evaluate their impact, and revise them), and the CMS pressure ulcer critical element pathway (CMS-20078) sends surveyors to the comprehensive assessment, the pressure relief devices, the repositioning schedule, the scheduled skin inspection, and the care plan, asking whether risk was comprehensively assessed, whether a care plan with interventions and measurable goals followed, whether it was implemented, and whether effectiveness was reassessed and the plan revised (F656 and F657 carry the care plan and revision requirements). The RAI Manual for MDS 3.0 item M0100 (LAW as to the assessment, PAYER POLICY as to its use) offers a formal tool such as the Braden or Norton as one response option beside clinical assessment and states that "the requirements do not mandate the use of any specific assessment tool" other than the RAI itself. In hospitals, Section 5001(c) of the Deficit Reduction Act of 2005 put stage 3 and 4 pressure ulcers among the 14 hospital-acquired condition categories for which Medicare pays no additional amount when the condition was not present on admission, for discharges on or after October 1, 2008 (PAYER POLICY); the present-on-admission indicator has been reported since October 1, 2007; the ICD-10-CM Official Guidelines let a coder take the pressure ulcer stage from documentation by clinicians who are not the patient's provider while the diagnosis must come from the provider; and the determination rests on the admission skin inspection, never on the Braden total, with no coding deadline by which a provider must document present-on-admission status, so early baseline inspection is good practice rather than a coding rule. The separate Hospital-Acquired Condition Reduction Program scores a patient safety composite that includes the pressure ulcer rate (PSI 03). In home health, OASIS collection is required (LAW and PAYER POLICY), but OASIS-D retired the M1300 and M1302 risk-tool items in January 2019 and the current item set records pressure ulcer and injury status, so an agency uses the Braden under its own policy and should not describe it as an OASIS requirement (the OASIS assessment page covers the items that exist). Accreditation and benchmarking are CONVENTION: Joint Commission standards ask accredited organizations to assess and periodically reassess pressure injury risk and act on it, NDNQI and Magnet track hospital-acquired pressure injury as a nurse-sensitive indicator, and none names the Braden or sets its interval. Two habits deserve the same label. "Braden every shift" is a unit policy, not a federal rule; the 1998 validation reassessed at 48 to 72 hours. "Turn every two hours" is a protocol default, not a guideline mandate: the TURN trial (Bergstrom and colleagues, 2013; 942 residents in 27 US and Canadian nursing homes on high-density foam) found pressure ulcer incidence of 2.5, 0.6, and 3.1 percent with 2-, 3-, and 4-hour repositioning (P = .68), TEAM-UP (Yap and colleagues, 2022; 992 residents in nine nursing homes) reported no incident injuries across 2-, 3-, and 4-hour intervals against a 5.24 percent baseline, and the 2026 International Guideline individualizes the interval to the person, the surface, and the skin response while advising against routinely extending it for everyone. In litigation and survey defense (CONVENTION), repeated risk scores with no documented response to the identified risk leave a chart open to the retrospective question of whether the assessment changed care; no published dataset quantifies how often plaintiffs' experts target that pattern, and this page does not claim one. Canada and Australia require a validated process and name no tool, and the evidence explains why the ratings matter more than the total. In Canada, no federal or provincial statute names the Braden; Accreditation Canada's Required Organizational Practices have included pressure injury prevention with risk assessment on admission and at intervals for the programs they cover, and the current wording should be confirmed in the standards set your organization is surveyed against (CONVENTION with accreditation force); the Registered Nurses' Association of Ontario best practice guideline on the assessment and management of pressure injuries for the interprofessional team (third edition, 2016) and Wounds Canada's best practice recommendations carry the clinical standard, and provincial long-term care and health authority toolkits reproduce the scale under license with the 1988 copyright line of Braden and Bergstrom. In Australia, the NSQHS Comprehensive Care Standard (second edition) requires health services to have pressure injury prevention systems consistent with best-practice guidelines, comprehensive skin inspections within best-practice time frames, and patient information and equipment for prevention (actions 5.21 to 5.23; CONVENTION with accreditation force); pressure injury sits on the ACSQHC hospital-acquired complications list, the National Aged Care Mandatory Quality Indicator Program counts pressure injuries in residential aged care, and the Waterlow score is used alongside the Braden, so the note names which instrument produced the total. On the evidence: the 1987 paper reported inter-rater correlations of .83 to .94 for nurses' aides and licensed practical nurses and .99 for registered nurses, with 100 percent sensitivity at a cutoff of 16 in its two samples; the 1998 multisite study of 843 adults in tertiary hospitals, VA medical centers, and skilled nursing facilities found 18 the best overall cutoff and reassessment at 48 to 72 hours more predictive than the admission score; Bergstrom and Braden (2002) found that 18 performed similarly in Black and White participants (sensitivity and specificity of about 75 and 76 percent against 70 and 77 percent), so the dark-skin problem is one of detecting early injury during the skin inspection, not of adjusting the arithmetic; Huang and colleagues (2021) pooled 60 studies and 49,326 participants to a sensitivity of 0.78, a specificity of 0.72, and an area under the curve of 0.82 with 18 as the optimal cutoff, calling the predictive validity moderate; Mehicic, Burston, and Fulbrook (2024) found inconsistent psychometrics across 34 intensive care studies, total-score intraclass correlations of 0.66 to 0.96, and most ICU cutoffs in the range of 12 to 14; Pancorbo-Hidalgo and colleagues (2006) found the Braden the best-validated scale across 33 studies (sensitivity 57.1 percent, specificity 67.5 percent, odds ratio 4.08); and Moore and Patton's Cochrane review (2019) found two trials with 1,487 participants and evidence too uncertain to show that structured risk tools reduce pressure ulcer incidence compared with clinical judgment. In pediatrics, the Braden Q was validated in 322 children on bed rest from about three weeks to eight years of age, the Braden QD in 625 patients from preterm infants through age 21 at eight centers with an area under the curve of 0.78, and the Braden QD's total runs in the opposite direction (higher meaning more risk), so a pediatric note names the instrument and version before the number. On versions: the licensor describes the Braden Scale II (2022) as keeping the six subscales with updated terminology and descriptors, and an institution's rights under an older implementation are not automatically the same as a current Braden Scale II license, so the record names the version, and an electronic build carries the licensor's authorization for the provider rather than an assumption that the vendor cleared it. The Braden Scale for Predicting Pressure Sore Risk is a copyrighted instrument: it was published by Bergstrom, Braden, Laguzza, and Holman in Nursing Research in 1987, and authorized copies carry the copyright line of Barbara Braden and Nancy Bergstrom, 1988, with the instrument's full name and no changes to wording or scoring. Prevention Plus, the company the authors founded to license it, was acquired on April 20, 2021 by Health Sense Ai, a subsidiary of HD Nursing, which announced that it would be the sole licensor of the Braden Scale going forward and now describes itself as the owner of the Braden Scale and Braden Scale II copyrights; permission to use the Braden Scale II (2022) is granted only through a Health Sense Ai license agreement sold by care setting, the public pages publish no blanket free-use grant for websites, calculators, forms, or electronic builds of the adult scale, and a 2023 Advances in Skin and Wound Care piece (Delmore and Ayello) reported the licensor's position that facilities already using the original scale could continue for the time being and that licensing runs to providers rather than to record vendors. No enforcement action was found in either research pass, which is a null finding and not a permission. The Braden Q and Braden QD are copyrighted by Martha A. Q. Curley (Braden QD, 2018), who permits unmodified reproduction for research and clinical practice, including hospital forms and hospital-based electronic records, and requires written permission for commercial use, external republication, and translation. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Braden Scale documentation errors reviewers flag The numbers behind these errors are specific. In two nursing homes, trained raters' totals for the same residents differed by up to nine points , with intraclass correlations of 0.73 to 0.95 for the total and far lower for sensory perception and nutrition, and years of experience made no difference (Kottner and Dassen, 2008); after web-based training, 1,391 registered nurses at three medical centers classified risk level correctly 82.6 percent of the time and struggled most in the middle of the range (Magnan and Maklebust, 2008); a 2024 review of 34 intensive care studies found total-score reliability of 0.66 to 0.96 and most empirically derived ICU cutoffs at 12 to 14 rather than 18 (Mehicic, Burston, and Fulbrook); the 1998 multisite validation found reassessment at 48 to 72 hours more predictive than the admission score (Bergstrom and colleagues); the TURN trial found no difference in pressure ulcer incidence among 2-, 3-, and 4-hour repositioning of at-risk residents on high-density foam ( 2.5, 0.6, and 3.1 percent , P = .68; Bergstrom and colleagues, 2013); and the Cochrane review found only two trials, 1,487 participants , and no reliable evidence that a structured tool lowers incidence compared with clinical judgment (Moore and Patton, 2019). No published audit counts how often a chart carries a Braden total with no linked intervention; the F686 survey pathway and the hospital-acquired condition and present-on-admission rules described under compliance considerations are what turn that omission into a deficiency or a payment problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Braden Scale documentation reviews: A total with no subscale ratings. "Braden 14" alone tells the next nurse, the surveyor, or the expert reviewer nothing about whether the risk is immobility, moisture, or intake, and it cannot be checked against the interventions. Chart all six ratings in the published order with the total, and put the basis for each low rating in ordinary clinical words (intake percentages, weights, how the patient moved during care, continence episodes), never in the scale's descriptor language. Ratings copied forward or scored from the chart. Yesterday's ratings auto-populated and signed as a fresh assessment, or a score assigned from the flowsheet without observing the patient. The scale is an assessment, and raters disagree by up to nine points on the same resident, so a copied total can hide real change while looking like continuity. If policy calls for a reassessment, observe, rate, and name the trigger; if nothing has changed, say that the ratings were re-observed and are unchanged. A cutoff or category written as the scale's rule. "Braden 18 or below requires a specialty mattress," a universal cutoff applied to an ICU patient, or "at risk" with no protocol cited. The cutoffs and categories are conventions: 16 in the 1987 validation, 18 in the 1998 and 2002 multisite studies, 12 to 14 in most ICU studies, and graded categories in the authors' later recommendations that facilities adopt as policy. Write the facility category with the protocol it comes from, and let the individual ratings, not the tier, choose the interventions. The score treated as the skin assessment. A Braden total charted where the skin inspection should be, or an injury discovered on day four with no admission inspection to show it was absent. The scale predicts risk and does not examine the skin. Document a separate visual and tactile inspection of the pressure points and device sites with findings, use palpation and the person's usual tone for darker skin, and let the inspection, not the score, carry the present-on-admission determination and any staging. A bundle with no subscale behind it, or a schedule attributed to the total. "Prevention protocol initiated," "skin bundle in place," or "Braden 14 therefore every two hours." A low rating on any subscale calls for its own measure regardless of the total, and the repositioning interval, the surface, the moisture regimen, and the dietitian referral each answer a specific rating. Name the measure, how it is delivered, and the rating it answers; write the interval you actually use, and cite the protocol if a total does map to an interval in your organization. No reassessment date, a plan never revised, or a score credited to the mattress. "Continue to monitor" with no date, a total that "improved" after a new surface was placed, a care plan unchanged after the ratings moved, or "refused" with nothing after it. The scale rates the patient, not the interventions, so a score changes only when the patient's condition does; every entry needs the next date and the triggers that bring it forward; a changed rating needs a revised plan; and a refusal needs the offer, the reason, the education, the alternatives, and who was told. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on nursing assessments and skin integrity documentation. Give it the facts (the trigger and version, the six ratings and total with the prior totals, the basis for the low ratings, the facility category and protocol, the skin and device-site findings, the measures in place or started for each low rating, the patient's response, the notifications, and the next reassessment) and it drafts the documentation block: trigger and version, ratings and total, category attributed to policy, separate inspection, intervention for each low subscale, response, and the reassessment and communication lines, ready for your review. Cross-check a finished entry for the gaps reviewers flag: a total with no ratings, ratings copied forward, a category or cutoff written as the scale's rule, the score standing in for the skin inspection, a bundle with no subscale behind it, a refusal with no context, or a missing reassessment date. Turn the entry into the next document: the pressure injury risk problem and goal for the care plan, the dietitian or therapy referral, the caregiver teaching for a home health visit, or the risk status and regimen line for the handoff, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do the six Braden subscales measure, and how is the total scored? The Braden Scale rates six subscales, named here by their published labels with the intent of each in original words: sensory perception (whether the patient can notice pressure discomfort and act on it or call for help), moisture (how wet the skin stays through the day), activity (how much the patient is up and about), mobility (how well the patient repositions without help), nutrition (how well the patient is actually eating), and friction and shear (how much the skin drags against surfaces during movement and transfers). Five subscales are scored 1 to 4 and friction and shear 1 to 3, the total runs from 6 to 23, and a lower total means higher risk. The descriptors that define each rating are the copyrighted text of the instrument, so this page does not reproduce or paraphrase them; rate from the licensed form your organization uses and write the basis for a low rating in ordinary clinical words. The total is a risk signal: a low rating on any one subscale calls for its own preventive measure, which is why the six ratings, not just the total, belong in the record. Which Braden score means a patient is at risk: 18, 16, or something else? It depends on the population and the protocol, which is why the note cites the protocol. The original 1987 validation set its cutoff at 16, where sensitivity was 100 percent in both study samples. The 1998 multisite study of 843 adults in tertiary hospitals, VA medical centers, and skilled nursing facilities found 18 the best overall cutoff, and Bergstrom and Braden confirmed 18 in Black and White participants in 2002; Huang and colleagues' 2021 meta-analysis of 60 studies also landed on 18 with only moderate predictive validity (sensitivity 0.78, specificity 0.72). In intensive care, a 2024 systematic review found most empirically derived cutoffs at 12 to 14, because sedation, ventilation, and vasopressors cluster nearly every ICU patient below 18. The graded categories most facilities use come from the authors' later recommendations and live in facility protocols, not in the 1987 instrument. Document the total, the facility category, and the protocol that defines it, and never present a cutoff as a physiologic law or as a rule that by itself prescribes a surface or a turning interval. How often does the Braden Scale have to be repeated? Does CMS require it every shift? No regulator sets a Braden interval, and CMS does not require the Braden at all: 42 CFR 483.25(b) requires assessment, prevention, monitoring, and revision, the RAI Manual states that the requirements do not mandate any specific assessment tool, and F686 guidance names validated tools as examples. "Every shift" in acute care and "weekly for four weeks, then quarterly and with change of condition" in long-term care are facility conventions, and the 1998 multisite validation itself reassessed at 48 to 72 hours after admission, where prediction improved. What the evidence and the guidelines do support is reassessment on admission, on transfer, when the clinical picture changes (new sedation or immobility, surgery, incontinence, falling intake, fever, hemodynamic decline), at transitions of care, and before discharge. Follow your policy, cite it in the note, and name the trigger for every repeat, because a dated trigger is what distinguishes monitoring from a repeated number; and never carry ratings forward without re-observing the patient. Does a Braden score replace the skin assessment, and how does it relate to staging and present-on-admission status? No. The Braden predicts the risk of a future injury; it does not examine the skin, and a low total says nothing about whether an injury already exists. Chart a separate visual and tactile inspection of the pressure points and every device contact site, with findings by location and laterality, and for darker skin tones compare with the person's usual tone and palpate for warmth, firmness, and tenderness rather than relying on redness. If an injury is found, it is described and staged under the conventions on the pressure injury staging page, and the provider documents the diagnosis; the ICD-10-CM guidelines let a coder take the stage from nursing documentation but the diagnosis only from the provider, so the two must agree. Present-on-admission status for the hospital-acquired condition rules rests on that admission inspection, never on the Braden total, and CMS sets no coding deadline for documenting it, so the baseline inspection is good practice and evidence rather than a payment rule. The total is above our threshold but one subscale is low. What do I document? Treat the low subscale as the finding. The authors, the Hartford Institute for Geriatric Nursing, and the international guideline bodies all say that low subscale ratings should drive prevention regardless of the total, because the total is an aggregate that can mask a single actionable problem. A patient whose only low rating is moisture needs the cause managed, a cleansing and barrier regimen, and a continence or microclimate plan today, not when unrelated subscales deteriorate enough to pull the total down; a patient whose only low rating is nutrition needs intake monitoring, weights, and a dietitian referral, without the rating being equated with malnutrition. Write the six ratings, name the low one and its basis in clinical words, record the measure started and how it is delivered, and set the reassessment. An entry that reads, in effect, total above the facility threshold, moisture rating identifies actionable risk, moisture management started, reassess on the stated date, is exactly what a surveyor or reviewer wants to find. What do I chart when a patient refuses repositioning, or when the score improves after a mattress change? For a refusal, chart the offer, the patient's stated reason where appropriate, the education given about the risk, the alternatives offered or accepted (a smaller shift, a different position, a change of surface or cushion, a later time), who was notified under policy, and the revised plan. Do not record care that was not delivered, and do not leave the word refused standing alone. The 2026 International Guideline builds comfort and preference into repositioning decisions, including at the end of life, and survey and serious-event frameworks recognize refusal and clinical instability as circumstances that require documentation when preventability is judged. The rating does not change because a measure was declined; the feasibility of the plan does, so the plan is what you revise. For the mattress: a new surface is an intervention, not one of the six subscales, so the total should improve only if the patient's own condition improved (better intake, more independent movement, resolved incontinence). Write which rating changed and why in clinical terms, keep the surface in place, continue the inspections that show whether the surface is working, and never discontinue a prevention measure because the number rose. Braden Q or Braden QD: which version applies to children and adolescents? Use the instrument your organization has adopted for the population in which it was validated, and name it in the note. The Braden Q (Curley, Razmus, Roberts, and Wypij, Nursing Research, 2003) adapted the adult framework for children, added a tissue perfusion and oxygenation subscale, and was validated in 322 children on bed rest from about three weeks to eight years of age, so it is not evidence for a 15-year-old. The Braden QD (Curley and colleagues, Journal of Pediatrics, 2018) was validated in 625 patients from preterm infants through age 21 at eight centers, was built to predict both immobility-related and medical-device-related injury, and scores in the opposite direction from the adult scale and the Braden Q: a higher Braden QD total means more risk. Three consequences for the chart: the instrument and version come before the number, a Braden QD total cannot be trended against a Braden Q or adult total, and device-site inspection is part of the pediatric entry. The Braden Q and QD are copyrighted by Martha A. Q. Curley, whose published permission allows unmodified clinical and research use, including hospital forms and hospital-based electronic records, with written permission required for commercial use, external republication, and translation. Can we put the Braden Scale in our EHR, on a paper form, or on our website? Only with the rights holder's permission. The Braden Scale is copyrighted (Barbara Braden and Nancy Bergstrom, 1988), and since April 20, 2021 its licensing has been administered by Health Sense Ai, a subsidiary of HD Nursing, which acquired Prevention Plus, announced itself as the sole licensor, and describes itself as the owner of the Braden Scale and Braden Scale II copyrights. Its public pages say that permission to use the Braden Scale II is granted only through a Health Sense Ai license agreement, sold by care setting, and they publish no blanket grant for websites, calculators, custom forms, or electronic builds of the adult scale; a 2023 Advances in Skin and Wound Care piece reported the licensor's position that facilities already using the original scale could continue for the time being and that licensing runs to the provider rather than to the record vendor, so an EHR build should rest on the facility's own authorization rather than an assumption that the vendor cleared it. The fact that the anchored form appears on educational and calculator sites is not permission, and neither is being free or non-commercial. What needs no license is the concept: numeric fields for six ratings and a total, your own documentation prompts in original words, and the structure on this page. The Braden QD has a different and more permissive published permission from its own rights holder. Can BastionGPT draft my Braden Scale documentation? Yes. Give it the facts (the trigger and version, the six ratings and total with the prior totals, the basis for the low ratings, the facility category and protocol, the skin and device-site findings, the measures in place or started for each low rating, the patient's response, the notifications, and the next reassessment) and it drafts the full entry: trigger and version, ratings and total, category attributed to policy, separate inspection, an intervention for each low subscale, response, and the reassessment and communication lines, ready for your review. It can also check a finished entry for a total with no ratings, ratings copied forward, a category written as the scale's rule, the score standing in for the skin inspection, a bundle with no subscale behind it, a refusal with no context, or a missing reassessment date. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Bergstrom N, Braden BJ, Laguzza A, Holman V, 1987, Nursing Research 36(4):205-210, The Braden Scale for Predicting Pressure Sore Risk (six subscales; inter-rater reliability; cutoff of 16); Braden BJ, Bergstrom N, 1987, Rehabilitation Nursing 12(1):8-12, a conceptual schema for the study of the etiology of pressure sores; Bergstrom N, Braden B, Kemp M, Champagne M, Ruby E, 1998, Nursing Research 47(5):261-269, multisite study of the predictive validity of the Braden Scale (843 subjects; cutoff of 18; reassessment at 48 to 72 hours); Bergstrom N, Braden B, 2002, Nursing Research, predictive validity among Black and White subjects ; Ayello EA, Braden B, 2002, Advances in Skin and Wound Care 15(3):125-131, how and why to do pressure ulcer risk assessment . Rights and licensing: Health Sense Ai, bradenscale.com (owner of the Braden Scale and Braden Scale II copyrights; permission for the Braden Scale II only through a license agreement; accessed September 2026); PR Newswire, April 20, 2021, Health Sense Ai acquires Prevention Plus (HD Nursing as sole licensor going forward); Delmore BA, Ayello EA, 2023, Advances in Skin and Wound Care 36(6):332-335, Braden scales for pressure injury risk assessment (licensing position; Braden Scale II); Regenstrief Institute, LOINC panel 38228-3 (Braden scale panel with its copyright and permission notice). Pediatric versions: Curley MA, Razmus IS, Roberts KE, Wypij D, 2003, Nursing Research, predicting pressure ulcer risk in pediatric patients: the Braden Q Scale (322 children); Curley MAQ and colleagues, 2018, Journal of Pediatrics 192:189-195, predicting pressure injury risk in pediatric patients: the Braden QD Scale (625 patients, eight centers, preterm through 21 years); Curley MAQ, Braden QD Scale permissions (copyright 2018; unmodified clinical and research use permitted; commercial use, republication, and translation by permission). CMS long-term care: eCFR, 42 CFR 483.25 quality of care (paragraph (b)(1), pressure ulcers; F686); CMS, pressure ulcer critical element pathway, form CMS-20078 (assessment, care plan, implementation, reassessment and revision); CMS, MDS 3.0 Resident Assessment Instrument manual (item M0100; no specific tool mandated); LeadingAge, coding risk for pressure ulcers: M0100 and M0150 (response options; Braden and Norton as examples). CMS hospital and home health: CMS, hospital-acquired conditions (Deficit Reduction Act section 5001(c); 14 categories including stage III and IV pressure ulcers; discharges on or after October 1, 2008); CMS, ICD-10-CM Official Guidelines for Coding and Reporting, FY 2026 (pressure ulcer stage from non-provider clinician documentation; present-on-admission guidelines); CMS, Hospital-Acquired Condition Reduction Program ; CMS, OASIS data sets (current item set; no named risk-tool item since OASIS-D retired M1300 and M1302 in 2019). Guidelines and toolkits: NPIAP, EPUAP, and PPPIA, International Pressure Injury Guideline, 2026 edition (published in chapters, including repositioning and skin and tissue assessment ; risk assessment chapter pending as of September 2026); Hartford Institute for Geriatric Nursing, Try This: predicting pressure injury risk (do not rely only on the total score; prevention based on low subscale scores, attributed to Dr. Braden, NPUAP, EPUAP, and PPPIA); AHRQ, Preventing Pressure Ulcers in Hospitals toolkit ; Wound, Ostomy, and Continence Nurses Society, wocn.org (pressure injury prevention and serious reportable event resources). Psychometrics and prevention evidence: Pancorbo-Hidalgo PL and colleagues, 2006, Journal of Advanced Nursing 54(1):94-110, risk assessment scales for pressure ulcer prevention: a systematic review (33 studies); Huang C and colleagues, 2021, Nursing Open, predictive validity of the Braden Scale in adults: meta-analysis (60 studies, 49,326 participants); Mehicic A, Burston A, Fulbrook P, 2024, Intensive and Critical Care Nursing, psychometric properties of the Braden scale in intensive care: a systematic review (34 studies); Kottner J, Dassen T, 2008, International Journal of Nursing Studies, an interrater reliability study of the Braden scale in two nursing homes ; Magnan MA, Maklebust J, 2008, Advances in Skin and Wound Care, multisite web-based training in using the Braden Scale (1,391 nurses); Moore ZEH, Patton D, 2019, Cochrane Database of Systematic Reviews CD006471, risk assessment tools for the prevention of pressure ulcers (two trials, 1,487 participants); Bergstrom N and colleagues, 2013, Journal of the American Geriatrics Society 61(10):1705-1713, Turning for Ulcer ReductioN (TURN) (942 residents, 27 nursing homes); Yap TL and colleagues, 2022, Advances in Skin and Wound Care 35(6):315-325, TEAM-UP trial results (992 residents, nine nursing homes). Canada and Australia: Registered Nurses' Association of Ontario, long-term care toolkit: Braden risk assessment scale (used under the Braden and Bergstrom 1988 copyright line) and assessment and management of pressure injuries for the interprofessional team, third edition (2016); Wounds Canada, best practice recommendations ; Accreditation Canada, Required Organizational Practices (confirm the current pressure injury prevention wording in the applicable standards set); Australian Commission on Safety and Quality in Health Care, NSQHS Comprehensive Care Standard (actions 5.21 to 5.23, preventing and managing pressure injuries) and hospital-acquired complications list . ## Related templates Pressure Injury Staging Documentation Morse Fall Scale Documentation OASIS Assessment Documentation Transfer-of-Care Summary Discharge Summary Incident & Adverse Event Report Care Coordination Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/apgar-score All note templates # Apgar Score Documentation: Components, Timing & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Apgar score is a five-component description of a newborn's condition at 1 and 5 minutes after birth (heart rate, respiratory effort, muscle tone, reflex irritability, and color), devised by Virginia Apgar in 1953. Labor and delivery nurses, midwives, obstetricians, and neonatal teams use it to record the newborn's transition and response to resuscitation. This page covers how to document Apgar scores with the support in progress, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The clinician responsible for the newborn at the time, usually a labor and delivery or newborn care nurse, a midwife, a neonatal nurse practitioner, or a pediatric or neonatal physician; NRP provider status is the usual local expectation, no license or publisher qualification level applies, and most policies keep the score away from the clinician delivering the infant Audience Neonatal and pediatric teams receiving the infant, NICU and nursery nurses, obstetricians and midwives closing the delivery record, birth-registry and vital-records staff, perinatal quality abstractors, coders, and expert reviewers in birth-injury claims Typical length 4 to 12 chart lines (birth time and context, five components and total at 1 and 5 minutes, extended scores to 20 minutes when indicated, support in progress at each time, scorer, cord gas, registry items) · 1 to 3 minutes to record after each scoring time Format family Observer-rated newborn condition score (five components each 0 to 2, total 0 to 10, assigned at fixed minutes after birth; the AAP and ACOG expanded reporting form pairs each score with the resuscitation in progress) When it's used Every live birth at 1 and 5 minutes in hospital, birth center, and home settings; extended to 10, 15, and 20 minutes when the 5-minute total is below 7; carried into the resuscitation record, the newborn admission note, the birth-certificate facility worksheet, and perinatal registries Standards context Virginia Apgar (1953), a public domain method; the AAP and ACOG joint statement (2015, reaffirmed 2021) sets timing, extended scoring, and expanded reporting; the NRP 9th edition (2025) governs the resuscitation itself; required by law only as a registry item; described here for documentation, no form or scoring table reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Apgar score? The Apgar score is a fixed-time description of a newborn's condition. Virginia Apgar, an anesthesiologist, proposed it in 1953 in Current Researches in Anesthesia and Analgesia (volume 32, pages 260 to 267) as a way to compare infants and obstetric practices with something better than impression, and she and colleagues reported a much larger series in JAMA in 1958. Five signs are observed at fixed minutes after birth and each is rated 0, 1, or 2 according to whether it is absent, present but reduced, or fully present: heart rate, respiratory effort, muscle tone, reflex irritability (the response to stimulation such as suctioning), and color; the five ratings sum to a total of 0 to 10. The familiar mnemonic (Appearance, Pulse, Grimace, Activity, Respiration) was a later teaching aid by L. Joseph Butterfield, recorded in 1961 correspondence and published as a JAMA letter in 1962, not part of the original method. The modern conventions come from the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists, whose joint statement (Pediatrics 2015, volume 136, pages 819 to 822, and ACOG Committee Opinion 644, reaffirmed 2021, replacing statements of 1996 and 2006) sets the score at 1 and 5 minutes for every infant and every 5 minutes through 20 minutes when the 5-minute total is below 7, and encourages an expanded Apgar score reporting form that records the resuscitation in progress beside each score. In Australia and New Zealand, ANZCOR extends scoring every 5 minutes until heart rate and breathing are normal. The score is not a resuscitation guide: NRP begins resuscitation before the 1-minute score exists, so the number records the infant's condition and response and never decides whether or how to intervene. The load-bearing fact for documentation is that the total is a summary and the record needs what it summarizes. A 5-minute score of 5 assigned to an infant receiving positive-pressure ventilation is not the same event as a 5 in an infant breathing on its own, and the AAP and ACOG say so directly: there is no accepted standard for scoring every component in an infant under resuscitation, so the defensible entry carries the five components, the support in progress at that minute, and the name of the person who observed them. Three limits follow from the same statement and belong in the note whenever the score is low. The score alone is not evidence of asphyxia; the 2014 ACOG and AAP report Neonatal Encephalopathy and Neurologic Outcome treats a score below 5 at both 5 and 10 minutes as one neonatal sign among several, none diagnostic by itself, beside umbilical artery acid-base status, neuroimaging, and organ injury. It does not predict an individual infant's mortality or neurologic outcome, even though very low scores track both at population scale (Casey and colleagues, 2001; Moster and colleagues, 2001). And it triggers nothing except its own follow-through: an umbilical arterial cord gas when the 5-minute total is 5 or less, extended scoring when it is below 7, and, in the United States, a 10-minute registry item when it is below 6. The score also has neighbors it must not absorb: the NRP resuscitation record is the timed event log, the newborn admission assessment is the full examination, the Ballard examination estimates maturity, Sarnat staging grades encephalopathy, and the Bishop score is the obstetric sibling that describes the cervix before labor rather than the infant after birth. Downstream, the transfer of care summary and the discharge summary carry the scores forward with their context rather than restating the method. ## Who uses Apgar score documentation and when Labor and delivery nurses and newborn care nurses assign and chart most Apgar scores in hospital births, because the clinician responsible for the newborn, not the one delivering, is usually the scorer under unit policy; nurse-midwives and community midwives do the same in birth centers and at home under their jurisdictional standards. Neonatal nurse practitioners, neonatologists, pediatric hospitalists, and pediatric residents assign or confirm the scores when a resuscitation team attends, and they are the readers who need the components and the support timeline in the NICU admission note. Obstetricians and maternal-fetal medicine physicians read the scores when they close the delivery record and when a category II or III tracing, an operative delivery, or a shoulder dystocia makes the newborn's condition part of the obstetric narrative. Birth-registry and health information staff transcribe the 5-minute and, when required, the 10-minute total to the facility worksheet; perinatal quality abstractors and coders read the scores beside the provider's diagnoses; and expert reviewers in birth-injury litigation read them for context, which is why the context has to be there. The Apgar page owns one job: the components, the timing, the support in progress, and the scorer. Neighbors take over from there: the resuscitation itself is documented in the NRP record, the newborn's examination in the admission assessment, a hospital transfer in the transfer of care summary , the discharge in the discharge summary , an unexpected outcome that the organization treats as a safety event in the incident report , a pediatric or neurology opinion in the consultation report , and the multidisciplinary plan for a NICU family in the care coordination note . ## How to document Apgar scores in the delivery record No law, payer, or professional body prescribes an Apgar note format. What exists is Virginia Apgar's method, the 2015 AAP and ACOG joint statement on timing, extended scoring, and expanded reporting, NRP guidance that resuscitation runs on its own timeline, and registry specifications that collect one or two totals. What survives review is an entry that fixes the time of birth and the clock, records the five components and the total at each scoring time, continues to 20 minutes when the 5-minute total is below 7, names the support in progress beside every score, identifies who assigned it, attributes any interpretation and states its limits, and closes with the cord gas, the registry items, and the handoff. Each element below carries the pitfall that most often undermines it. Time of birth, the clock, and context. Open with the time of complete birth (delivery of the whole body), the clock it came from, and, where cord clamping was delayed or resuscitation began on the cord, the clamping time and the time resuscitation started; the 2015 AAP and ACOG statement suggests recording exactly those times, and every Apgar minute runs from birth, not from clamping or arrival at the warmer. Add the context that changes what a component means without implying hypoxia: gestational age in completed weeks, delivery mode and any instrument, maternal magnesium sulfate, opioids within a few hours of birth, or general anesthesia, the fluid, and any known congenital condition. A synchronized labor-room clock or a resuscitation timer started at birth is good practice and a local policy matter, not a published Apgar rule, so name the source of your times rather than asserting a standard. Pitfall: Scores timed from cord clamping or from the warmer, or a preterm infant's low tone and reflex scores entered with no gestational age beside them. The five components at 1 and 5 minutes. Record each component as its own number at each time, then the total: heart rate, respiratory effort, muscle tone, reflex irritability, and color, each 0, 1, or 2, for a total of 0 to 10, with 0 a legitimate value. The components describe what was observed at that minute in the conventional categories, and the total summarizes them, not the other way around; two infants with a total of 5 can differ completely in heart rate and respiratory effort, which is exactly what the receiving team and any later reviewer need to know. Assign at the minute mark. Scores written later from memory run higher than contemporaneous observation (O'Donnell and colleagues, 2006, found delivery-room scores averaging 2.4 points above the same infants scored from video), so an entry made after the fact is labeled as such. Pitfall: "Apgars 8 and 9" with no components, or a 1-minute score written at 30 minutes of life from what the team recalls. Extended scores at 10, 15, and 20 minutes. When the 5-minute total is below 7, continue scoring every 5 minutes through 20 minutes, with the components and the support in place at each time, which is the AAP and ACOG convention; ANZCOR's guidance for Australia and New Zealand extends the same idea to scoring every 5 minutes until heart rate and breathing are normal. Do not stop because the 10-minute total reached 7, and do not skip a time point because the team was busy: a missing 15-minute score in a record that runs to 20 minutes reads as a gap. Keep the clinical rule distinct from the registry rule: the US Standard Certificate asks for the 10-minute total only when the 5-minute total is below 6, so an infant scored 6 at 5 minutes needs the clinical extended scores but has no 10-minute registry item. Pitfall: Scoring stopped at 5 minutes because the infant was breathing, or the 10-minute score omitted from the birth-registry worksheet when the 5-minute total was 4 or 5. Support in progress beside each score. For every scoring time, write what was being done to the infant at that moment: supplemental oxygen and its concentration, positive-pressure ventilation, CPAP, an advanced airway, chest compressions, epinephrine, and when each began and ended. This is the expanded Apgar score reporting concept the AAP and ACOG encourage (the form itself is theirs; record the same information in your own fields), and it exists because a score assigned during resuscitation is not equivalent to the same number in a spontaneously breathing infant and because there is no accepted standard for scoring some components under support: respiratory effort in an intubated infant is the classic problem, and Lopriore and colleagues (2004) showed clinicians score such infants inconsistently. Record the component you observed and the support beside it; do not score an imagined untreated infant. The second-by-second event log belongs in the NRP resuscitation record; the Apgar entry is the fixed-time snapshot that points to it. Pitfall: A 5 assigned under PPV charted like a 5 in a breathing infant, or interventions listed in a block with no link to the minute each applied to. Who assigned the score, and how. Name the person who assigned each score and their role, and whether the score was assigned at the minute mark or entered later. No national body designates the scorer: the AAP and ACOG ask perinatal professionals to be consistent, Apgar's own 1966 advice favored an observer other than the person delivering the infant, and most hospital policies give the score to the clinician responsible for the newborn rather than to the obstetrician or midwife delivering, so cite your policy rather than a rule. When two clinicians observed different things, record both observations with names; there is no published hierarchy under which a physician's number replaces a nurse's contemporaneous score, and the component detail resolves most disputes better than a contested total. A score added after the fact is a late entry with its actual date, time, author, and reason, never an overwrite. Pitfall: No scorer named, or a physician's later total silently replacing the newborn nurse's contemporaneous components. Interpretation and its limits. Write what the scores describe and attribute any band you use: the AAP and ACOG, drawing on the 2014 Neonatal Encephalopathy and Neurologic Outcome report, describe a 5-minute score of 7 to 10 in a term or late-preterm infant as reassuring, 4 to 6 as moderately abnormal, and 0 to 3 as low, and those words belong to them, not to the instrument. Then state the limits the same statement sets: the score alone is not evidence of asphyxia, does not predict an individual infant's mortality or neurologic outcome, and was not used to decide whether or how to resuscitate. Name the context that lowers components without hypoxia (prematurity, maternal magnesium or opioids, general anesthesia, congenital neuromuscular or cardiac disease) and, where a cord gas exists, read the two together as independent measures. For color, say what you assessed (trunk, tongue, and mucous membranes) and record oximetry beside it; the component is the least reliable of the five and carries documented bias against darker-skinned infants. Pitfall: "Apgar 3, birth asphyxia" or "Apgar 9, no hypoxic event" as a conclusion from the numbers alone, or a preterm infant's expected low tone read as depression. Cord gas, registry items, and handoff. Close with what the scores triggered. When the 5-minute total is 5 or less, the AAP and ACOG recommend an umbilical arterial blood gas from a clamped cord segment, and placental pathology may be valuable; record the result or that the sample could not be obtained. Enter the registry items your jurisdiction actually collects: in the United States the facility worksheet for the 2003 Standard Certificate takes the 5-minute total and, when it is below 6, the 10-minute total, never the 1-minute score; Canadian provincial registries and the Australian perinatal collections have their own field sets. Hand off to the newborn or NICU team with the components at each time, the support timeline, the cord gas, and the times, and make sure the delivery note, the resuscitation record, the newborn admission note, and the discharge summary carry the same numbers. Tell the parents what was documented and what it does and does not mean, and record that conversation. Pitfall: Cord gas skipped at a 5-minute score of 5, the 1-minute score sent to the registry, or a NICU admission note whose scores disagree with the delivery record. ### Blank template (copy and adapt) APGAR SCORE DOCUMENTATION BLOCK (components, timing, and support) Date: [ ] Setting: [hospital / birth center / home] Scorer (name, role): [ ] Complete birth (time, clock used): [ ] Cord clamped: [time] Resuscitation started: [time / not required] Gestational age: [ ] weeks [ ] days Delivery: [spontaneous vaginal / assisted / cesarean] Maternal medications: [magnesium / opioid / general anesthesia / none] 1 minute [time]: heart rate [ ] respiratory effort [ ] tone [ ] reflex [ ] color [ ] Total [ ] of 10 Support in progress: [none / oxygen (percent) / PPV / advanced airway / compressions / epinephrine] 5 minutes [time]: heart rate [ ] respiratory effort [ ] tone [ ] reflex [ ] color [ ] Total [ ] of 10 Support in progress: [ ] Extended scores (required when the 5-minute total is below 7): 10 minutes [time]: components [ ] Total [ ] Support: [ ] 15 minutes [time]: components [ ] Total [ ] Support: [ ] 20 minutes [time]: components [ ] Total [ ] Support: [ ] Support timeline: [PPV start and stop / CPAP / airway / compressions / medication, each with times; NRP resuscitation record filed separately] Scoring notes: [assigned at the minute mark by the scorer above / late entry with actual date, time, and reason / differing observations, both observers named] Interpretation: [description at each time; a score under support is not equivalent to a spontaneous score; context: prematurity, medications, congenital condition; no inference of asphyxia from the score alone] Cord gas: [umbilical arterial gas when the 5-minute total is 5 or less: pH, pCO2, base deficit / not indicated / not obtainable] Placenta to pathology: [yes / no] Registry items: [5-minute total; 10-minute total when the 5-minute is below 6 (US); local field set] Disposition and handoff: [ ] Parents informed: [who, when, what was explained] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the AAP and ACOG expanded reporting form. ## Sample Apgar score documentation (fictional) Scenario: a US community hospital labor and delivery unit, where a term infant born by vacuum-assisted vaginal delivery after a category II tracing needs positive-pressure ventilation; the newborn care nurse on the resuscitation team assigns and charts the components at 1, 5, 10, 15, and 20 minutes with the support in progress at each time, the cord gas the 5-minute score triggered, the registry items, and the NICU handoff. All details are fictional. Patient: Infant M., female, 39 weeks 2 days · Setting: Community hospital labor and delivery unit; vacuum-assisted vaginal birth, resuscitation team present · Clinician: K. Alvarez, RN (newborn care nurse, NRP provider) · Note date: 09/14/2026 Birth, clock, and context: Complete birth (delivery of the whole body) at 03:12:40 on 09/14/2026 by the labor room clock, which is synchronized to the electronic record; the resuscitation timer was started at that moment and every time below runs from it. Vacuum-assisted vaginal birth after a category II tracing with recurrent late decelerations in the second stage; 39 weeks 2 days by early ultrasound; estimated weight about 3,400 g. Maternal epidural analgesia (bupivacaine with fentanyl) since 21:40 on 09/13/2026; no magnesium sulfate, no systemic opioid in the 4 hours before birth, no general anesthesia. Fluid clear. The infant was apneic and limp at birth, so the cord was clamped at 03:13:05 rather than after the planned 60 seconds, and the infant was moved to the warmer with initial steps (warm, dry, stimulate, position) begun at 03:13:10. 1-minute score, assigned during PPV: At 1 minute (03:13:40): heart rate 1 (about 80 per minute by auscultation), respiratory effort 0 (apneic), muscle tone 1 (some flexion), reflex irritability 1 (grimace to suctioning), color 0 (pale, central cyanosis); total 3 of 10. Positive-pressure ventilation with a T-piece resuscitator in room air had begun at 03:13:20, 40 seconds after birth, when apnea persisted after the initial steps and the heart rate was below 100; the 1-minute score was assigned during PPV and is not equivalent to a 3 in a spontaneously breathing infant. Pulse oximeter placed on the right hand at 03:13:50. 5-minute score and support: At 5 minutes (03:17:40): heart rate 2 (132 per minute on the oximeter), respiratory effort 1 (irregular gasps between delivered breaths), muscle tone 1, reflex irritability 1, color 0 (central cyanosis; SpO2 68 percent against the NRP target of about 80 to 85 percent at 5 minutes); total 5 of 10. PPV continuing, oxygen increased to 30 percent at 03:16:00 per the oximetry targets; no chest compressions, no advanced airway, no epinephrine. Because the 5-minute total is below 7, scoring continued at 10, 15, and 20 minutes; because it is 5 or less, an umbilical arterial blood gas was drawn from the clamped cord segment at 03:19 (per the AAP and ACOG recommendation) and the placenta was sent for pathology. Extended scores with the support in progress: At 10 minutes (03:22:40): heart rate 2, respiratory effort 1 (spontaneous but irregular; PPV stopped at 03:20:30 when sustained spontaneous breathing began, and CPAP 5 cm H2O with 30 percent oxygen was started), muscle tone 1, reflex irritability 2, color 1 (acrocyanosis, SpO2 88 percent); total 7. At 15 minutes (03:27:40): heart rate 2, respiratory effort 2 (regular, mild subcostal retractions), muscle tone 1, reflex irritability 2, color 1 (SpO2 93 percent, oxygen weaned to 25 percent); total 8; CPAP continuing. At 20 minutes (03:32:40): heart rate 2, respiratory effort 2, muscle tone 2 (active movement), reflex irritability 2, color 1 (acrocyanosis, SpO2 95 percent in 21 percent oxygen); total 9; CPAP continuing for the retractions. Each score was assigned at the minute mark with the support then in place; the second-by-second event log is in the NRP resuscitation record filed with this note. Who assigned the scores: All scores were assigned contemporaneously by K. Alvarez, RN, the newborn care nurse on the resuscitation team, who was not delivering the ventilation. The delivering obstetrician (S. Brennan, MD) did not assign scores. The pediatric hospitalist (L. Nair, MD) arrived at 03:16 and led the resuscitation from that point; at 5 minutes Dr. Nair's independent observation matched the five components above, and no discrepancy needed recording. Times are from the resuscitation timer started at complete birth. This entry was completed at 04:05 from the timer and the resuscitation record; it contains no late entries, and any later addition will be made as a dated late entry with its reason rather than as a change to these values. Interpretation and limits: The 5-minute total of 5 falls in the range the AAP and ACOG describe as moderately abnormal for a term infant; the word describes condition at that minute and is not a diagnosis. The 1- and 5-minute scores were assigned during PPV and are not equivalent to scores in a spontaneously breathing infant. The trajectory (3, 5, 7, 8, 9) documents the response to ventilation. Umbilical arterial gas at 03:19: pH 7.14, pCO2 60 mmHg, base deficit 8 mmol/L, which does not meet the acid-base thresholds (pH below 7.0 or base deficit of 12 mmol/L or more) that the 2014 ACOG and AAP neonatal encephalopathy report lists among the signs consistent with an acute intrapartum event, and the score was above 5 at 10 minutes; no inference of intrapartum asphyxia is drawn from the scores, and none could be drawn from them alone. Color was assessed on the trunk, tongue, and mucous membranes with oximetry recorded beside it. Maternal epidural analgesia is not expected to depress tone or respiration; no magnesium, systemic opioid, or general anesthesia exposure; no congenital anomaly noted on the initial examination. Cord gas, registry, handoff, and parents: Transferred to the NICU at 03:40 on CPAP 5 cm H2O in 21 percent oxygen for respiratory monitoring and a blood glucose check; handoff to the NICU nurse and Dr. Nair included the components at each time, the support timeline (PPV 03:13:20 to 03:20:30, CPAP from 03:20:30, oxygen 21 to 30 to 21 percent), the cord gas, and the times, and the NICU admission note carries the same values. Birth-registry facility worksheet: 5-minute score 5; because the 5-minute score is below 6, 10-minute score 7; the 1-minute score is not a registry item; abnormal conditions of the newborn: assisted ventilation immediately following delivery and NICU admission. Placenta sent to pathology at 03:45. Dr. Nair spoke with both parents at 03:50 and explained that the scores describe how the infant looked at each minute and how she responded to breathing support, and that they do not predict her development; documented in the parent communication note. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, times, dates, and details are invented to show documentation structure and are not clinical guidance. No reporting form or scoring table is reproduced. ↑ Back to the template and downloads ### Why this sample works Complete birth time, the clock, cord clamping, and the start of resuscitation are fixed first, so every minute in the record can be reconstructed and no reader has to guess which clock the scores ran on. Each score appears as five components with the total and the support in place at that minute, so a 5 under PPV cannot be read as a 5 in a breathing infant, and the trajectory shows the response to ventilation. Scoring continued through 20 minutes because the 5-minute total was below 7, the cord gas was drawn because it was 5 or less, and the 10-minute registry item was completed because it was below 6, each rule kept distinct. The scorer is named with her role, the delivering physician is recorded as not scoring, the hospitalist's concurring observation is documented, and the late-entry policy is stated, which is what a reviewer asks first. The interpretation attributes its one descriptive word to the AAP and ACOG, reads the cord gas as an independent measure against the 2014 encephalopathy thresholds, draws no asphyxia inference, and records what was assessed for color with oximetry beside it. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the rules around the score are real, narrow, and mostly about registries and coding. Under LAW, state vital-records statutes and regulations require facilities to report birth-certificate items, and the 2003 revision of the US Standard Certificate of Live Birth (item 51) collects the 5-minute total for every birth and the 10-minute total only when the 5-minute total is below 6; it does not collect the 1-minute score or any component, and states vary in their own confidential medical data (Virginia's regulation, for example, lists the Apgar score at one minute and five minutes), so check your state's facility worksheet specification rather than assuming one national field set. Also under LAW as to coding, ICD-10-CM contains no code for an Apgar score and no rule that assigns a diagnosis from one: birth asphyxia (P84) is coded only from the provider's documented diagnosis, never from a low score, oxygen, suctioning, or meconium, and the FY 2027 Official Guidelines, effective October 1, 2026, change nothing on that point. The Food and Drug Administration classifies the Apgar timer as a Class I device at 21 CFR 880.2930, a fact about equipment rather than documentation. Under PAYER POLICY, no Medicare, Medicaid, or commercial policy conditions payment on Apgar components or on the expanded form; the delivery attendance and newborn resuscitation codes (99464 and 99465) are supported by the documented interventions and times in the resuscitation record, not by the score. The rest is CONVENTION with professional force. The AAP and ACOG joint statement (2015, reaffirmed 2021) sets 1 and 5 minutes for every infant, extended scores every 5 minutes through 20 minutes when the 5-minute total is below 7, an umbilical arterial cord gas when the 5-minute total is 5 or less, and the expanded reporting form, and states that the score is not used to determine the need for resuscitation, is not evidence of asphyxia, and does not predict individual outcome. The Neonatal Resuscitation Program's 9th edition, released by the AAP and the American Heart Association on October 22, 2025 with instructors required to teach it by June 1, 2026, governs the resuscitation itself, and the 2025 AHA and AAP guideline places the discussion of redirecting care at about 20 minutes of no detectable heart rate despite high-quality resuscitation, with its evidence review cautioning that an Apgar of 0 at 10 minutes, taken alone, does not reliably predict death or major impairment. The 2014 ACOG and AAP report Neonatal Encephalopathy and Neurologic Outcome (second edition) treats a score below 5 at both 5 and 10 minutes as one neonatal sign, beside an umbilical artery pH below 7.0 or a base deficit of 12 mmol/L or more, neuroimaging, and multisystem injury, none sufficient alone, a smaller role than the 2003 first edition gave a score of 0 to 3 beyond 5 minutes. The Joint Commission removed PC-01 (elective delivery) from ORYX reporting effective January 1, 2026 while remaining its steward for certification, retained PC-06 (unexpected complications in term newborns), whose specification runs on coded diagnoses and outcomes and contains no Apgar element, and has no measure that requires component-level Apgar documentation or the scorer's identity; a unit that says the Joint Commission requires those is quoting its own policy. In litigation (CONVENTION), Apgar values appear in every birth-injury chronology beside the fetal heart rate record, the resuscitation course, the cord gases, imaging, and the examinations, and plaintiff-side material sometimes presents a low score as proof of causation; the AAP and ACOG position that the score alone cannot establish asphyxia is the clinical standard, and a contemporaneous, component-level record with the support in progress is the document that carries it. No published dataset counts how often missing context decides a claim, and this page does not claim one. Canada and Australia collect the score through registries, run resuscitation on the same NRP and ANZCOR frameworks, and require no note format, and the evidence explains why the components and the support matter more than the total. In Canada, no federal or provincial LAW prescribes Apgar documentation; the Canadian Perinatal Surveillance System draws its Apgar data from hospital and provincial sources rather than from the national live-birth registration, and provincial perinatal registries such as BORN Ontario (a prescribed registry under Ontario's health privacy legislation, capturing hospital, home, and birth-centre births) and the BC Perinatal Data Registry specify their own Apgar fields, so a Canadian unit follows its registry dictionary (LAW for the reporting duty, CONVENTION for the content). The Canadian Paediatric Society administers NRP in Canada and launched the 9th edition on June 1, 2026, with instructors to transition by September 1, 2026 and all providers and organizations fully transitioned by January 2027; there is no freestanding SOGC Apgar statement, and none should be invented. In Australia, state and territory perinatal data collections are notified under public health LAW and feed the Perinatal National Minimum Data Set, which carries Apgar scores at 1 and 5 minutes; the Australian Institute of Health and Welfare reports the 5-minute score as a National Core Maternity Indicator (1.5 percent of liveborn babies born at or after term had a 5-minute score below 7 in 2023), NSW Health describes scoring at 1 and 5 minutes as routine and reports the same term indicator, and ANZCOR Guideline 13.9 (2021) sets the clinical convention that scoring continues every 5 minutes until heart rate and breathing are normal, states that "Interventions for depressed newborns should not await Apgar scoring," and requires full documentation of observations, interventions, and times for clinical and medicolegal reasons (CONVENTION). On the evidence: O'Donnell and colleagues (2006) had 42 observers score video of 30 newborns at 5 minutes and found wide variability in every observable component, with the scores assigned in the delivery room averaging 2.4 points higher than the video scores; Bashambu and colleagues (2012) showed 335 neonatologists video of infants from 24 to 40 weeks and found near-perfect agreement for a term infant (kappa 0.89 to 0.94) but poor agreement for preterm infants on support, with kappa for respiratory effort as low as 0.07; Lopriore and colleagues (2004) found clinicians scoring intubated and resuscitated infants inconsistently; Hegyi and colleagues (1998) found in 1,105 preterm infants under 2,000 g that scores tracked gestational age and birth weight and that color correlated least with the other components; Edwards and colleagues (2023) found in 977 neonates that providers assigned Black infants lower scores at 1 minute (odds ratio 0.63) and 5 minutes (0.64) after adjusting for cord gases, gestational age, and complications, driven by the color component (odds ratio 0.52 at 1 minute); and Fair and colleagues (Pediatric Research, 2024) reviewed ten studies covering more than 39 million infants and found that a score of 3 or less predicted mortality across ethnic groups while visual cyanosis assessment was unreliable in every group and tongue color and pulse oximetry indicated oxygen need regardless of ethnicity. At population scale the score keeps its value: Casey and colleagues (2001) studied 151,891 liveborn singletons and found neonatal mortality of 244 per 1,000 term infants with a 5-minute score of 0 to 3 against 0.2 per 1,000 with a score of 7 to 10, and the low score more predictive of death than an umbilical artery pH of 7.0 or below; Moster and colleagues (2001) followed 235,165 Norwegian term children and found a score of 0 to 3 in 0.1 percent, with neonatal death 386 times and cerebral palsy 81 times more frequent than after a score of 7 to 10; and Ehrhardt and colleagues (2026) found among 7,900 very preterm infants in Europe that the share scored below 7 at 5 minutes ranged from about 14 to 40 percent between countries, warning against using the score in prognostic models without local validation. The Specified and Combined Apgar of Rüdiger and colleagues (the TEST-Apgar study, 2015) formalizes the separation of condition from intervention as a research instrument; it is not the AAP and ACOG form and should not be entered as one. The Apgar score is in the public domain as a method: Virginia Apgar published it in 1953 in Current Researches in Anesthesia and Analgesia, a scoring method is not copyrightable under 17 U.S.C. 102(b), no trademark attaches to the eponym, and any hospital record, EHR, or web tool may implement the five components and their 0 to 2 values in its own words and layout, while the National Library of Medicine, which hosts the 1953 article, notes that the article text itself may remain under copyright. The 2015 joint statement (Pediatrics 136:819 to 822; Obstetrics and Gynecology 126:e52 to e55) and the expanded Apgar score reporting form it carries are copyrighted by the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists, all rights reserved, with reuse handled through the AAP permissions process, so this page describes the form and links to it rather than reproducing it; the NRP algorithm and materials belong to the AAP and the American Heart Association on the same footing. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Apgar score documentation errors reviewers flag The numbers behind these errors are specific. Delivery-room scores averaged 2.4 points higher than scores assigned from video of the same 30 infants by 42 observers, with wide variability in every component (O'Donnell and colleagues, 2006); 335 neonatologists agreed almost perfectly on a term infant (kappa 0.89 to 0.94) and poorly on preterm infants receiving support, with kappa for respiratory effort as low as 0.07 (Bashambu and colleagues, 2012); providers assigned Black neonates lower scores at 1 minute (odds ratio 0.63 ) and 5 minutes (0.64) with cord gases held constant, driven by the color component (Edwards and colleagues, 2023); the share of very preterm infants scored below 7 at 5 minutes ranged from about 14 to 40 percent across European countries in a cohort of 7,900 (Ehrhardt and colleagues, 2026); and among 151,891 infants a 5-minute score of 0 to 3 carried a term neonatal mortality of 244 per 1,000 against 0.2 per 1,000 for 7 to 10 (Casey and colleagues, 2001), which is population risk, not an individual prognosis. The AAP and ACOG state that there is no accepted standard for scoring an infant under resuscitation and that the predictive reliability of such a score has not been studied. No published audit counts how often a delivery record carries totals without components or scores without the support in progress; the registry, coding, and litigation realities described under compliance considerations are what turn those omissions into a problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Apgar documentation reviews: A total with no components. "Apgars 8 and 9" or "5-minute Apgar 5" alone tells the NICU nurse, the coder, and the expert reviewer nothing about whether the infant was bradycardic, apneic, or merely dusky, and it cannot be checked against the interventions. Chart the five component values at every scoring time, in the conventional order, with the total after them. Scoring stopped early, or the registry rule confused with the clinical rule. Scoring ended at 5 minutes because the infant was breathing, or at 10 minutes because the total reached 7, when the 5-minute total was below 7 and the AAP and ACOG convention runs through 20 minutes; or a 10-minute score omitted from the US facility worksheet when the 5-minute total was 4 or 5. Keep the two thresholds distinct (below 7 for extended scoring, below 6 for the registry item) and score every 5 minutes through 20. A score under resuscitation charted as if the infant were breathing alone. A 5-minute 5 assigned during PPV recorded with no mention of the ventilation, or interventions listed in a block with no link to the minutes they applied to. The support in progress belongs beside each score with its start and stop times, because the same number means different things with and without it; describe it in your own fields if your record has no expanded reporting section. No scorer, a reconstructed score, or a disagreement erased. No name on the score, a 1-minute value entered at 30 minutes of life from memory, or a physician's later total replacing the nurse's contemporaneous components. Name the scorer and role, assign at the minute mark, enter anything later as a dated late entry with its reason, and record differing observations with both names rather than one official number. Asphyxia, prognosis, or a diagnosis code inferred from the number. "Apgar 3 at 1 minute, birth asphyxia," P84 coded from a low score without a provider diagnosis, "Apgar 9, no hypoxic event," or a parent told that a 10-minute score predicts development. The AAP and ACOG position is that the score alone is neither evidence of asphyxia nor an individual predictor; the 2014 encephalopathy framework needs acid-base, neurologic, imaging, and organ findings together; and coders take P84 from the provider's statement only. Write what the scores describe and what else was measured. Context and color left unstated. A 29-week infant's tone and reflex scores of 1 with no gestational age beside them, a low 1-minute score after general anesthesia with no mention of the anesthesia, or color scored pink-or-blue in a darker-skinned infant with no note of what was assessed and no oximetry. Record gestational age, maternal medications, and congenital conditions as context, and for color record the sites assessed (trunk, tongue, mucous membranes) with the saturation beside it, because the color component is the least reliable of the five and carries documented bias. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on labor and delivery and newborn care documentation. Give it the facts (complete birth time and the clock, gestational age and delivery mode, maternal medications, the five component values and total at each minute, the support in progress at each time with start and stop times, who assigned each score, the cord gas result, the registry items, and the disposition) and it drafts the documentation block: birth and context, components and totals at 1 and 5 minutes and the extended times, the support beside every score, the scorer and late-entry status, an interpretation with attributed language and stated limits, and the cord gas, registry, and handoff lines, ready for your review. Cross-check a finished delivery record for the gaps reviewers flag: totals without components, scoring stopped before 20 minutes when the 5-minute total was below 7, a score under PPV with no support named, no scorer, a cord gas missing at a 5-minute total of 5 or less, the 1-minute score sent to the registry, or an asphyxia inference drawn from the numbers alone. Turn the entry into the next document: the newborn course paragraph of a NICU admission or transfer of care summary, the birth-registry worksheet entries in the order the form asks for them, or a plain-language explanation for parents of what the scores describe and what they do not predict, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What are the five Apgar components, and how is the score assigned and totaled? Five signs are observed at fixed minutes after birth and each is rated 0, 1, or 2 according to whether it is absent, present but reduced, or fully present: heart rate, respiratory effort, muscle tone, reflex irritability (the response to stimulation), and color. The five ratings sum to 0 to 10, and 0 is a legitimate total; consumer pages that give the range as 1 to 10 are wrong. Scores are assigned at 1 and 5 minutes after complete birth for every liveborn infant, timed from birth rather than from cord clamping or arrival at the warmer, by the clinician responsible for the newborn. The number describes condition at that minute in the presence of whatever support is being given; it is not a test the infant passes or fails, and it does not decide whether to resuscitate, because resuscitation begins before the first score exists. This page describes the components conceptually and reproduces no scoring table; rate from your organization's record and put the five numbers, not just the total, in the chart. When do I keep scoring past 5 minutes, and how long do I continue? When the 5-minute total is below 7, the AAP and ACOG convention is to assign a score every 5 minutes through 20 minutes: at 10, 15, and 20 minutes, each with its five components and the support in progress. Scoring does not stop because the 10-minute total reached 7; the sequence runs to 20 minutes. ANZCOR's guidance for Australia and New Zealand phrases the same idea as scoring every 5 minutes until heart rate and breathing are normal. Keep the clinical rule apart from the registry rule: the 2003 US Standard Certificate asks for a 10-minute total only when the 5-minute total is below 6, so an infant scored 6 at 5 minutes gets extended clinical scores and no registry 10-minute item, while an infant scored 5 gets both. An extended score assigned during continued ventilation is annotated as such, and a 10-minute score of 0 is documented like any other with the resuscitation in progress; the decision to continue or redirect care belongs to the NRP algorithm and the team, not to the score. How do I document an Apgar score assigned while the infant is being resuscitated? Record the components you actually observed at the minute mark and, beside them, the support in place at that moment: supplemental oxygen and its concentration, positive-pressure ventilation, CPAP, an advanced airway, chest compressions, epinephrine, with the time each began and ended. That pairing is the idea behind the expanded Apgar score reporting form the AAP and ACOG encourage; the form's text and layout are theirs, but the information belongs in every record, in your own fields if necessary. Two points from the 2015 statement justify the effort: a score assigned during resuscitation is not equivalent to a score in a spontaneously breathing infant, and there is no accepted standard for scoring every component in an infant under support. Respiratory effort in an intubated infant is the classic ambiguity, and Lopriore and colleagues (2004) found clinicians scoring such infants inconsistently. Do not score an imagined untreated infant, do not leave the total unqualified, and point to the NRP resuscitation record for the timed event log; the Apgar entry is the fixed-time snapshot that sits beside it. Who should assign the Apgar score when the whole team is resuscitating? No national body designates the scorer, so the honest answer is your policy plus a name. The AAP and ACOG ask perinatal professionals to be consistent in how they score during resuscitation; Apgar's own 1966 advice favored an observer other than the person delivering the infant; and most hospital policies give the score to the clinician responsible for the newborn, typically the labor and delivery or newborn care nurse, a midwife, or a member of the neonatal team, rather than to the obstetrician who is managing the mother. Whoever it is, the record names them and their role, states that the score was assigned at the minute mark, and, if two clinicians observed different things, carries both observations with both names. There is no published hierarchy under which a physician's number overrides a nurse's contemporaneous score or the reverse, and component-level detail resolves most disagreements better than a contested total. A score entered later is a dated late entry with its reason, never a silent replacement, and retrospective scores run higher than contemporaneous ones, which is one more reason to write the name and the time. Does a low Apgar score mean birth asphyxia, and does it predict cerebral palsy? No on both counts, and the sources are explicit. The 2015 AAP and ACOG statement says the score alone cannot be considered evidence of or a consequence of asphyxia and does not predict an individual infant's mortality or neurologic outcome. The 2014 ACOG and AAP report Neonatal Encephalopathy and Neurologic Outcome lists a score below 5 at both 5 and 10 minutes as one neonatal sign consistent with an acute intrapartum event, alongside an umbilical artery pH below 7.0 or a base deficit of 12 mmol/L or more, neuroimaging findings, and multisystem organ injury, with none sufficient by itself. What is true is population risk: Casey and colleagues (2001) found term neonatal mortality of 244 per 1,000 with a 5-minute score of 0 to 3 against 0.2 per 1,000 with 7 to 10, and Moster and colleagues (2001) found neonatal death 386 times and cerebral palsy 81 times more frequent after a score of 0 to 3 than after 7 to 10 in 235,165 Norwegian term children, yet most infants with low scores do not develop cerebral palsy. For the chart, write what the scores describe, record the cord gas and the examination as independent measures, draw no causal conclusion from the numbers, and remember that coders assign P84 (birth asphyxia) only from the provider's documented diagnosis, never from the score. What goes on the birth certificate, and what do registries collect in the US, Canada, and Australia? In the United States the 2003 Standard Certificate of Live Birth (item 51) collects the 5-minute total for every birth and the 10-minute total only when the 5-minute total is below 6; the 1-minute score and the components are never registry items, and state confidential medical data can differ (Virginia's regulation lists Apgar scores at one and five minutes), so the facility worksheet specification for your state governs. In Canada the national live-birth registration generally does not carry the Apgar; provincial perinatal registries do, BORN Ontario and the BC Perinatal Data Registry among them, each with its own data dictionary, and the Canadian Perinatal Surveillance System draws on those sources for national reporting. In Australia, state and territory perinatal data collections feed the Perinatal National Minimum Data Set, which carries scores at 1 and 5 minutes, and the Australian Institute of Health and Welfare reports the share of term liveborn babies with a 5-minute score below 7 as a National Core Maternity Indicator (1.5 percent in 2023). In every jurisdiction the registry takes totals; the clinical record still needs the components and the support, because the registry item is a summary of a summary. How should I score and document color in a newborn with darker skin? Score the conventional component as the instrument defines it, say what you assessed, and put objective oxygenation beside it. Color was the least reliable component from Apgar's own 1953 report onward, and the modern evidence is specific: Edwards and colleagues (2023) found providers assigned Black neonates lower scores at 1 and 5 minutes with cord gases held constant, driven by the color component, and Fair and colleagues (2024) reviewed ten studies covering more than 39 million infants and found visual cyanosis assessment unreliable in every group, with tongue color and pulse oximetry indicating oxygen need regardless of ethnicity. No authority has published a race-adjusted rubric, and inventing one in the chart would make the score incomparable, so the documentation answer is context rather than modification: record the sites you assessed (trunk, tongue, mucous membranes), the preductal saturation from the oximeter at each scoring time against the NRP minute-by-minute targets (about 80 to 85 percent at 5 minutes and 85 to 95 percent at 10 minutes), and the oxygen concentration in use. Healthy term infants commonly lose a color point at 1 minute and often at 5 minutes for acrocyanosis, so a 9 is not a deficit that needs explaining. Can we put the Apgar score or the AAP and ACOG expanded form in our EHR, on a paper record, or on a website? The method, yes; the form, not without permission. A scoring method is not copyrightable under US law (17 U.S.C. 102(b) excludes procedures, processes, and systems), so a hospital record, an EHR build, or a free web calculator may implement the five components and their 0 to 2 values in its own words and layout, and LOINC codes exist for the 1-, 5-, and 10-minute scores (9272-6, 9274-2, and 9271-8) for structured capture. The National Library of Medicine, which hosts Apgar's 1953 article, notes that the article text itself may still be under copyright, and the 2015 joint statement and the expanded Apgar score reporting form are copyrighted by the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists, with reuse handled through the AAP permissions process; so describe the form's content in your own fields (each score paired with the support in progress and the times) rather than scanning or redrawing it, and link to the publishers for the original. The NRP algorithm and materials belong to the AAP and the American Heart Association under the same logic. Nothing on this page or in its downloads reproduces the form. Can BastionGPT draft my Apgar score documentation? Yes. Give it the facts (complete birth time and the clock, gestational age and delivery mode, maternal medications, the five component values and total at each minute, the support in progress at each time with start and stop times, who assigned each score, the cord gas result, the registry items, and the disposition) and it drafts the full entry: birth and context, components and totals at 1 and 5 minutes and the extended times, the support beside every score, the scorer and late-entry status, an interpretation with attributed language and stated limits, and the cord gas, registry, and handoff lines, ready for your review. It can also check a finished delivery record for totals without components, scoring stopped before 20 minutes when the 5-minute total was below 7, a score under PPV with no support named, a missing scorer, a cord gas missing at a 5-minute total of 5 or less, or an asphyxia inference drawn from the numbers alone. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Apgar V, 1953, Current Researches in Anesthesia and Analgesia 32:260 to 267, A Proposal for a New Method of Evaluation of the Newborn Infant (National Library of Medicine copy; the item may remain under copyright); Apgar V, 1966, Pediatric Clinics of North America 13(3):645 to 650, the newborn (Apgar) scoring system: reflections and advice ; Butterfield J, Covey MJ, 1962, JAMA 181(4):353, Practical Epigram of the Apgar Score (the mnemonic). American Academy of Pediatrics Committee on Fetus and Newborn and ACOG Committee on Obstetric Practice, 2015, Pediatrics 136(4):819 to 822, The Apgar Score , and ACOG, Committee Opinion No. 644 (reaffirmed 2021; timing, extended scoring, expanded reporting form, cord gas at 5 or less; copyright ACOG and AAP); ACOG and AAP, 2014, Pediatrics 133(5):e1482, Neonatal Encephalopathy and Neurologic Outcome, second edition, executive summary ; AAP, licensing and permissions . Resuscitation: American Heart Association and AAP, 2025, Pediatrics 157(1):e2025074352, Part 5: Neonatal Resuscitation (redirection of care discussion at about 20 minutes); AAP, Neonatal Resuscitation Program (9th edition released October 22, 2025; instructors to teach it by June 1, 2026); Canadian Paediatric Society, NRP in Canada (9th edition launch June 1, 2026; instructors by September 1, 2026; full transition by January 2027); ANZCOR, 2021, Guideline 13.9: After the Resuscitation of a Newborn ; AHA, AAP, and ILCOR, 2023, Pediatrics 151(2):e2022059631, Recommended Guideline for Uniform Reporting of Neonatal Resuscitation: the Neonatal Utstein Style . US registry, coding, and measures: CDC National Center for Health Statistics, 2003, U.S. Standard Certificate of Live Birth (item 51: score at 5 minutes; score at 10 minutes if the 5-minute score is less than 6); Virginia Administrative Code, 12VAC5-550-100, birth certificate items (Apgar at one and five minutes in confidential data); CMS and NCHS, ICD-10-CM Official Guidelines for Coding and Reporting, FY 2027 (effective October 1, 2026; no Apgar code); Regenstrief Institute, LOINC 9274-2, 5 minute Apgar score ; Legal Information Institute, 21 CFR 880.2930, Apgar timer (Class I); The Joint Commission, Specifications Manual for Joint Commission National Quality Measures, Perinatal Care, v2026A and PC-06 measure information form (no Apgar element). Canada and Australia: Public Health Agency of Canada, Canadian Perinatal Surveillance System ; BORN Ontario, data holdings ; Australian Institute of Health and Welfare, National Core Maternity Indicators: Apgar score of less than 7 at 5 minutes for birth at or after term (1.5 percent in 2023) and National Perinatal Data Collection: baby measurements (Apgar at 1 and 5 minutes); NSW Health, Mothers and Babies 2024: Apgar score . Reliability and bias: O'Donnell CPF, Kamlin COF, Davis PG, Carlin JB, Morley CJ, 2006, Journal of Pediatrics 149:486 to 489, interobserver variability of the 5-minute Apgar score (42 observers, 30 infants; delivery-room scores 2.4 points higher than video); Bashambu MT and colleagues, 2012, Pediatrics 130:e982 to e987, interobserver agreement of Apgar scoring in preterm infants (335 neonatologists); Lopriore E and colleagues, 2004, BMJ 329:143 to 144, correct use of the Apgar score for resuscitated and intubated newborn babies ; Hegyi T and colleagues, 1998, Pediatrics 101:77 to 81, the Apgar score and its components in the preterm infant (1,105 infants); Edwards SE, Wheatley C, Sutherland M, Class QA, 2023, American Journal of Obstetrics and Gynecology 228:229.e1 to e9, associations between provider-assigned Apgar score and neonatal race (977 neonates); Fair FJ, Furness A, Higginbottom G, Oddie SJ, Soltani H, 2024, Pediatric Research 97:939 to 952, systematic review of Apgar scores and cyanosis in Black, Asian, and ethnic minority infants (ten studies, more than 39 million infants). Outcomes and research variants: Casey BM, McIntire DD, Leveno KJ, 2001, New England Journal of Medicine 344:467 to 471, the continuing value of the Apgar score for the assessment of newborn infants (151,891 infants); Moster D, Lie RT, Irgens LM, Bjerkedal T, Markestad T, 2001, Journal of Pediatrics 138:798 to 803, the association of Apgar score with subsequent death and cerebral palsy in term infants (235,165 children); Ehrhardt H and colleagues, 2026, BJOG 133:649 to 659, five-minute Apgar scores and prognostic value in very preterm infants: a multinational cohort study (7,900 infants; 14 to 40 percent across countries); Rüdiger M and colleagues, 2015, BMC Pediatrics 15:18, TEST-Apgar: the Specified and Combined Apgar . ## Related templates Bishop Score Documentation Transfer-of-Care Summary Discharge Summary Incident & Adverse Event Report Consultation Response & Report Care Coordination Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/rass-score All note templates # RASS Documentation: Target Versus Actual, Timing & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Richmond Agitation-Sedation Scale (RASS) is a ten-level observational scale, published by Sessler and colleagues in 2002, that places a patient on one continuum from deep sedation through calm alertness to severe agitation. ICU and step-down nurses, intensivists, respiratory therapists, and pharmacists use it to set and check sedation targets and to gate delirium screening. This page covers how to document RASS assessments, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Bedside ICU and step-down nurses at each assessment, with intensivists, advanced practice providers, respiratory therapists, and pharmacists reading and acting on it; no license, training certificate, or publisher qualification level applies, and the rights holder permits free clinical use Audience Relieving nurses at handoff, intensivists and the rounding team who set the sedation target, pharmacists reviewing sedative and analgesic orders, respiratory therapists timing awakening and breathing trials, delirium screeners, quality staff counting ICU Liberation bundle assessments, and later the reviewers who reconstruct a sedation course from the chart Typical length 1 to 3 chart lines per assessment (time, assessed value, ordered target, stimulus where relevant, sedation on board, action if off target) · assessment takes under a minute at the bedside Format family Clinician-observed ten-level ordinal scale on one continuum (negative values for increasing sedation depth, zero for calm and alert, positive values for increasing agitation; the stimulus needed to obtain a response is built into the negative values) When it's used Continuous sedation of mechanically ventilated adults, titration of sedative and analgesic infusions, spontaneous awakening trials, the arousal check before every CAM-ICU, agitation episodes in the ICU, emergency department, and step-down unit, palliative sedation with the RASS-PAL, and every handoff and transfer that carries a sedation status Standards context Sessler and colleagues (AJRCCM, 2002) with longitudinal validation by Ely and colleagues (JAMA, 2003); copyright registered by Virginia Commonwealth University in 2012 with free use for health systems, clinicians, and researchers; named by the SCCM PADIS guideline and the ICU Liberation bundle, mandated by no law; described here for documentation, no level descriptors reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the RASS? The Richmond Agitation-Sedation Scale is a bedside observational scale of sedation and agitation developed at Virginia Commonwealth University in Richmond, Virginia, by a team that was multidisciplinary by design: the eight authors of the November 2002 paper in the American Journal of Respiratory and Critical Care Medicine (Sessler, Gosnell, Grap, Brophy, O'Neal, Keane, Tesoro, and Elswick) included physicians, nurses, pharmacists, and a biostatistician, and the first reliability panel was two physicians, two nurses, and one pharmacist. The scale has ten levels on a single continuum: four positive values describe increasing agitation, from restlessness to behavior that endangers the patient or staff; zero describes a patient who is awake and calm without any prompting; and five negative values describe increasing depth of sedation, separated by how much stimulation it takes to get a response and by the quality of that response. The assessment runs as a sequence rather than a menu. The clinician first watches the undisturbed patient, and a patient who is alert, or who is restless or agitated, is rated from observation alone. If the patient is not spontaneously alert, the clinician speaks, saying the name and asking for eye opening and eye contact, and rates by whether eye contact is made and how long it is held. Only when there is no response to voice does the clinician use physical stimulation, and the two deepest values are separated by whether that produces any response at all. The value therefore carries the stimulus inside it: a negative number already tells the next reader whether voice or touch was needed. The wording that defines each level is the scale's copyrighted text and is not reproduced here; the published scale is the source of record. Ely and colleagues validated the scale over time at Vanderbilt (JAMA, June 2003): in 38 patients, two nurses agreed with a weighted kappa of 0.91 across 290 paired observations, against 0.64 for the Glasgow Coma Scale; in 275 mechanically ventilated patients, RASS values correlated with the GCS (r 0.91), with an attention screen (r 0.78), and with bispectral EEG (r 0.63), and tracked change across consecutive ICU days. That Vanderbilt group, now the Critical Illness, Brain Dysfunction, and Survivorship Center, made the RASS the arousal step that precedes the CAM-ICU , which is why the two are usually taught together. The load-bearing fact for documentation is that a RASS value describes an observed state, and the record needs two more things before it describes sedation management: the target the team ordered and the sedation the patient had on board when the value was taken. The scale does not need a target to be valid; Ely and colleagues reported that their ICU had no sedation protocol and no routinely identified objective sedation targets during the validation, and the scale performed anyway. What the target and the drug context add is intent: whether an assessed value is what the team wanted, an overshoot after a bolus, an unintended drift, or a deliberate exception such as deep sedation under neuromuscular blockade. Three boundaries follow. The RASS is not a consciousness scale for brain injury: the Glasgow Coma Scale describes eye, verbal, and motor responses after injury and does not measure sedation depth, so in a sedated neurological patient the two are charted side by side rather than one standing in for the other. The RASS is not a delirium tool: it decides whether the patient is arousable enough for the CAM-ICU (at the two deepest values the screen is recorded as unable to assess), and the delirium features are a separate assessment, so a calm and alert patient can still be delirious. And the RASS is not a pain scale: PADIS asks for a behavioral pain instrument such as the CPOT or BPS in patients who cannot self-report, and pain is assessed and treated before sedation is escalated. Alternatives and variants have their own names and are charted under them: the Riker Sedation-Agitation Scale (1999) is the other sedation scale the SCCM guidelines treat as valid and reliable; the older Ramsay scale runs in the opposite direction and separates agitation poorly; the RASS-PAL (Bush and colleagues, 2014) is a palliative care modification with pilot evidence only; the modified RASS (Chester and colleagues, 2012) adds an attention item for ward delirium screening; and the adult scale has been validated in critically ill children (Kerson and colleagues, 2016; Tapia and colleagues, 2021) without adult targets automatically transferring. ## Who uses RASS documentation and when Critical care nurses do most of the charting: on admission, at the interval the unit protocol sets, with each titration of a sedative or analgesic infusion, before and after a bolus, before every CAM-ICU , and before, during, and after a spontaneous awakening trial. Intensivists and advanced practice providers write the target as a value or a range in the sedation order and revisit it on rounds; pharmacists read the assessed values against the target when they reconcile infusion rates, cumulative doses, and benzodiazepine exposure; respiratory therapists time the spontaneous breathing trial to the awakening trial and need the RASS trajectory to do it; and quality staff count documented arousal assessments for the ICU Liberation bundle. Beyond the ICU, emergency and step-down nurses use it during agitation episodes and the continuous sedation of ventilated patients waiting for a bed, procedural teams use it beside their required monitoring, and palliative care teams use the RASS-PAL during palliative sedation and agitated delirium, naming the variant. The value travels with the patient: a transfer-of-care summary should carry the current target, the last assessed value with its time, active infusions, the last CAM-ICU result, and the awakening trial status, and an incident report after a self-extubation or a restraint-related injury should record the RASS at the moment of the event with its precipitant. Neighbors win in four places: the GCS when the question is consciousness after brain injury, the CAM-ICU for delirium, the CPOT or BPS for pain, and, on general wards where the RASS was never validated, a ward opioid sedation score or the ACVPU consciousness item inside an early warning score such as MEWS or NEWS2 . ## How to document RASS assessments in the chart No law, payer, or guideline prescribes a RASS note format; what exists is a copyrighted scale, a validation literature that ties it to goal-directed sedation, guideline and bundle conventions that expect a documented target and regular arousal assessments, and unit protocols that set the interval. What survives review is an entry that carries a clock time and its place in the course of care, the ordered target as written, the assessed value with the stimulus where it is not self-evident, the sedation and analgesia on board at that moment, what was done when actual and target differed, the awakening trial trajectory and the CAM-ICU it leads to, and an honest entry rather than a number under neuromuscular blockade. Each element below carries the pitfall that most often undermines it. Time, assessor, and the moment in the course of care. Give every assessment a clock time and place it: scheduled per protocol, after a titration, minutes after a bolus, during a procedure, before or during an awakening trial, on arrival from the emergency department. The interval is protocol, not regulation: Critical Care BC asks for depth of sedation every four hours or more often, Vanderbilt's trauma protocol monitors each shift, one UK cardiothoracic unit requires hourly entries, and the SCCM ICU Liberation data manual counts documented level-of-arousal assessments per 24 hours with a goal of at least six. Name the assessor, and when a value was taken within about half an hour of a bolus or a rate change, say so in the same line, because a post-bolus value describes a transient peak rather than the infusion. Pitfall: A column of values with no times, so nobody can tell which one followed the 13:10 propofol bolus and which one describes the running infusion. The ordered target, as written. Chart the target exactly as the order states it, a single value or a range, beside every assessed value or in the header of the flowsheet the values sit in, and name the order it comes from and the date it was last reviewed. Do not translate the order into a phrase such as light sedation, and do not silently narrow a range to one number. If the ordered target is deeper than the guideline preference, record the indication (neuromuscular blockade, status epilepticus, refractory intracranial hypertension, severe ventilator dyssynchrony); if no target has been ordered, obtain one and chart the request, because a value with no target cannot be read as on or off course. Vanderbilt's trauma protocol states that "goal and actual RASS should be specified on ALL patients," and Critical Care BC asks for a documented daily RASS goal with sedatives titrated to it. Pitfall: Assessed values with no target anywhere in the record, or an order for -1 to 0 charted as a target of -2 because that is what the unit usually uses. The assessed value, with the stimulus where it matters. Follow the sequence (observe, then voice, then physical stimulation) and record the value it produces, not the value the infusion rate would predict. A negative value already implies the stimulus that was needed, so a stimulus note is not required beside every number, but write it when the picture is ambiguous: eyelid swelling, hearing loss or a language barrier, a neurological injury that alters eye opening, a patient who looks calm with eyes open but does not attend to voice, or a value that disagrees with the last one. A patient asleep who wakes fully to voice is scored by that response, not as deeply sedated. Where the observed behavior would let two raters pick different values, add three or four words of behavior; a 2025 study found an ICU nurse and two trained raters chose the identical value in only 35 percent of encounters and flagged ambiguity in 16 percent. Pitfall: A zero charted because the eyes were open, or a deep value charted because they were closed, with nothing to show what the patient did when spoken to. Sedation and analgesia on board. Record what the patient had at the time of the assessment: each sedative and analgesic infusion with its rate, any bolus with the clock time, the last rate change and when, and the non-drug measures in use. In an electronic record this is often a matter of aligning the RASS row with the medication administration record rather than retyping doses, but the note that explains an off-target value should name the drug and the time in words. When agitation is the finding, chart the pain assessment made with a behavioral tool before any sedative response, because PADIS puts routine pain assessment and treatment ahead of sedative escalation and the ICU Liberation bundle keeps pain, sedation choice, and delirium as separate elements. Pitfall: A value taken twenty minutes after a procedural bolus read as the effect of the infusion, or an agitated patient given more sedative with no pain score anywhere near the entry. Off target: recognition, cause, action, recheck. When the assessed value differs from the target, the entry has four parts. Recognition: state the direction, deeper or lighter than target, rather than leaving the reader to compare two columns. Cause: for a deeper value, a recent bolus, drug accumulation, a new neurological change, or a deliberate exception; for a lighter or agitated value, pain, hypoxemia or dyssynchrony, a full bladder or other discomfort, withdrawal, delirium, a procedure, or an environmental trigger. Action: the titration made under the medication order or unit protocol, with the new rate and time, or the reason the team chose not to titrate, and the notification when the pattern or the indication needs a prescriber. Recheck: the follow-up value at the interval the order or protocol sets. The scale itself orders nothing; the titration authority is the order. Pitfall: A value two levels deeper than target followed by "continue current infusion," or a score-to-sedative reflex with no cause considered. Awakening trial trajectory and the CAM-ICU link. For a spontaneous awakening trial, chart the pre-trial value with its time, the safety screen result, the time each sedative was stopped and which infusions continued, the values during the trial with their times, the criteria watched for failure (sustained agitation, respiratory distress or a high respiratory rate, desaturation, an arrhythmia, or other instability under your protocol) and whether any occurred, the coordination with the respiratory therapist for the breathing trial, and, if sedation was restarted, the reason, the time, and the rate. Once the patient is above the arousal gate and a delirium screen is due, record the CAM-ICU result after the trajectory, not before it; at the two deepest RASS values the screen is recorded as unable to assess, any lighter value allows it, and a calm and alert value does not by itself say anything about delirium. Pitfall: "Propofol held per SAT" with no values, times, or restart rate, or a CAM-ICU charted as negative on a patient whose RASS that hour made it unassessable. Neuromuscular blockade, restraint, and handoff. Under effective neuromuscular blockade, do not manufacture a value: the scale depends on visible responses that paralysis abolishes, so chart that the RASS cannot be assessed because of pharmacologic paralysis, carry forward the last assessment before the blocker with its time, record the sedative and analgesic regimen and the blocker's start time, and name the unit's monitoring strategy for sedation depth, processed EEG if used; the 2016 SCCM guideline calls for sedation and analgesia before and during blockade, and the 2026 SCCM guideline for ARDS treats scale-based versus non-scale evaluation before blockade, and monitoring versus no monitoring during it, as open questions with very low certainty evidence. When agitation ends in restraint, the RASS describes the behavior and does not authorize the restraint: document the behavior and precipitant, the less restrictive measures tried, the order, the times applied and removed, and the response, as 42 CFR 482.13(e) and your policy require. At handoff and transfer, carry the target, the last value with its time, the infusions, the last CAM-ICU, and the awakening trial status. Pitfall: RASS -5 charted every hour on a paralyzed patient as if it were a sedation assessment, or "RASS +3, restraints applied" with nothing else. ### Blank template (copy and adapt) RASS DOCUMENTATION BLOCK (assessment with target and context) Date: [ ] Time: [ ] Setting: [ICU / step-down / ED / palliative unit] Assessor: [ ] Scale version: [RASS / RASS-PAL / pediatric use: state it] Ordered target: [single value or range, exactly as written] Order date and prescriber: [ ] Indication if deeper than the unit default: [neuromuscular blockade / seizures / ICP / severe dyssynchrony / other] Assessed RASS: [ ] Stimulus if not self-evident: [observation / voice / physical] Behavior in a few words (when ambiguous): [ ] Confounders: [hearing, language, eyelid swelling, neurological injury, sleep] Moment in care: [scheduled / after titration / [ ] minutes after bolus / procedure / awakening trial / arrival] Sedation and analgesia on board: [drug, rate, last change and time / bolus and time / non-drug measures] Relation to target: [at target / deeper than target / lighter or agitated] Likely cause: [ ] Action under the order or protocol: [new rate, time / none, reason] Prescriber notified: [who, time, response] Recheck due: [time] Recheck value: [ ] Awakening trial (if today): pre-trial RASS and time [ ] safety screen [pass / fail: why] sedative stopped [drug, time] values during [ ] failure criteria observed [none / which] breathing trial [with RT, time, result] restarted [drug, rate, time, reason] CAM-ICU link: [performed at RASS [ ], time, result / unable to assess at -4 or -5 / due at ] Neuromuscular blockade: [RASS not assessable, paralyzed since [time] / last pre-blockade RASS and time / sedation and analgesia running / monitoring strategy] Agitation event: [behavior, precipitant, pain score and tool, measures tried, medication and time, restraint order and times if used, response] Handoff: [target, last RASS and time, infusions, last CAM-ICU, awakening trial status] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's level descriptors or scoring card. ## Sample RASS documentation (fictional) Scenario: a US academic medical ICU, day three of mechanical ventilation for pneumonia with acute hypoxemic respiratory failure in a 64-year-old woman; the day nurse charts the ordered target beside each assessed value, a spontaneous awakening trial that leads into a CAM-ICU, an expected overshoot after a procedural bolus, an agitation episode with its cause and a brief restraint, and the handoff. All details are fictional. Patient: L.M., 64 · Setting: Medical ICU, academic hospital; day 3 of mechanical ventilation · Clinician: J. Ortega, RN, CCRN · Note date: 09/14/2026 Target, overnight context, and pre-trial assessment (06:50): RASS as built in the unit flowsheet, assessed by J. Ortega, RN, at 06:50 on 09/14/2026. Ordered target: RASS -1 to 0, written by the intensivist in the sedation order of 09/12/2026 and reaffirmed on rounds 09/13/2026; the unit protocol (MICU-SED-04) sets assessment every two hours, with every titration, and before every CAM-ICU, which is policy rather than a regulatory interval. On board: propofol 30 mcg/kg/min, unchanged since 22:00 on 09/13/2026; fentanyl 75 mcg/h. Assessed 06:50: RASS -2; she opened her eyes to her name but did not hold eye contact; no physical stimulation needed. Deeper than target by one level with no bolus in the preceding six hours, consistent with overnight accumulation. CPOT 1. Night values of -1 at 02:00 and -2 at 04:00 were charted against the same target. Spontaneous awakening trial safety screen at 07:00, per protocol: no active seizures or alcohol withdrawal, no escalating sedative doses for agitation, no neuromuscular blockade, no active myocardial ischemia, no raised intracranial pressure concern; screen passed and the trial planned with the respiratory therapist (R. Chen, RRT). Awakening trial, breathing trial, and the CAM-ICU (07:05 to 08:45): Propofol stopped 07:05; fentanyl continued at 75 mcg/h for analgesia. 07:35: RASS -1, opens eyes to her name and holds eye contact, squeezes hands on request; RR 20, SpO2 97 percent on FiO2 0.50 and PEEP 8, synchronous with the ventilator. 08:05: RASS 0, awake and calm without prompting, nods appropriately, no reaching for lines; RR 22, SpO2 96 percent, heart rate 92, no arrhythmia. No failure criteria (no sustained agitation, RR under 35, SpO2 above 88 percent, no distress or arrhythmia). Spontaneous breathing trial started 08:10 by R. Chen, RRT, on pressure support; stopped 08:32 for RR 38 with accessory muscle use and SpO2 90 percent, returned to prior settings, result and settings in the respiratory flowsheet. Sedation not restarted: at RASS 0 and comfortable, the target is met without an infusion; the propofol order was left active for use if needed. CAM-ICU performed 08:30 at RASS 0, above the arousal gate: negative, recorded in the delirium flowsheet; next screen due 20:00 under the once-per-shift policy. Intensivist informed at 08:45 of both trial results; plan from rounds: target stays -1 to 0 with no continuous sedative unless required, fentanyl decreased to 50 mcg/h at 09:00 for CPOT 0 to 1. Procedural bolus and an expected off-target value (13:10 to 14:30): 13:10: right radial arterial line replaced by the intensivist; propofol 30 mg intravenous bolus and fentanyl 50 mcg given under the procedural order, both charted in the medication administration record with times. 13:30: RASS -3, deeper than the target of -1 to 0; she moved and briefly opened her eyes when her name was called but made no eye contact; no physical stimulation needed. Cause recorded: expected transient effect of the 13:10 bolus, no infusion running, no new neurological finding, pupils equal and reactive. No titration to make (no sedative infusion); recheck set for 14:00 per protocol. 14:00: RASS -1, eye contact to voice held; 14:30: RASS 0, target met. No prescriber notification required under the protocol for a bolus-related value that returned to target within the recheck interval. Agitation episode: cause, intervention, response (16:20 to 17:05): 16:20: RASS +3; after repositioning she became restless, then repeatedly reached for the endotracheal tube and pulled at the new arterial line; SpO2 89 percent with high peak pressures and ventilator asynchrony. Cause search: suctioned for a thick secretion plug at 16:21, SpO2 95 percent by 16:24; CPOT 5 with grimacing and muscle rigidity, fentanyl 50 mcg bolus given 16:25 under the pain order; bladder catheter patent and draining; no new medication started today, no alcohol history; CAM-ICU negative this morning and repeated at 16:50 (below). Soft bilateral wrist restraints applied 16:23 under the intensivist's order for the immediate physical safety of the airway after verbal reassurance, hand-holding, and reorientation failed to stop her reaching for the tube; the order, the less restrictive measures tried, her behavior, and her response are documented in the restraint record as hospital policy and 42 CFR 482.13(e) require. No additional sedative given; the intensivist agreed at 16:30 to treat pain and the airway first and to reassess before any propofol. 16:50: RASS 0, calm, follows commands, ventilator synchronous; CAM-ICU repeated at RASS 0 because of the acute behavior change: negative. Restraints removed 17:05 at the earliest point her behavior allowed, RASS 0 sustained for 15 minutes, removal time charted. Daughter present from 17:15 and updated. Handoff and record (18:45): Handoff to the night nurse at 18:45 with the sedation state in one line: target RASS -1 to 0 per the 09/12/2026 order, reaffirmed today; last assessed RASS 0 at 18:30 by observation alone; no sedative infusion running since 07:05, propofol order active; fentanyl 50 mcg/h, last bolus 16:25; awakening trial passed today and breathing trial stopped at 08:32 for tachypnea; CAM-ICU negative at 08:30 and 16:50, next due 20:00; restraint episode 16:23 to 17:05 documented in the restraint record; recheck RASS every two hours and with any titration per MICU-SED-04. Values entered in the sedation flowsheet beside the target and the medication administration record; this narrative filed as the nursing note for 09/14/2026. Every RASS value in this record is an observation at its clock time; none was estimated from the infusion rate. This sample is fictional and for educational purposes. It does not describe a real patient or record; the values, times, doses, dates, and details are invented to show documentation structure and are not clinical guidance. No scale descriptors or scoring card are reproduced. ↑ Back to the template and downloads ### Why this sample works The ordered target sits beside every assessed value with the order date, so each number reads as on target, deeper, or lighter without a second lookup, and the unit interval is named as policy rather than regulation. Every value carries a clock time and its place in the course of care, so the -2 after an unchanged overnight infusion, the -3 twenty minutes after a bolus, and the +3 during asynchrony read as three different clinical stories rather than three numbers. The awakening trial is a trajectory (safety screen, stop time, values, criteria watched, breathing trial, restart decision), and the CAM-ICU is recorded after the patient cleared the arousal gate, with the once-per-shift schedule visible. Each off-target value has a cause, an action or a documented reason for no action, and a recheck, and the agitation episode shows secretions, hypoxemia, and pain addressed before any sedative, with the restraint documented under its own rules. The handoff carries target, last value with time, infusions and last bolus, trial results, CAM-ICU results, and the next due screen, which is what the receiving nurse, the pharmacist, and a later reviewer need to reconstruct the day. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the RASS, and the rules that bind sit one step from the bedside. Under LAW, the Medicare Conditions of Participation require a medical record that contains practitioner orders, nursing notes, medication records, vital signs, and the other information needed to monitor the patient's condition, with entries complete, dated, timed, and authenticated (42 CFR 482.24), and a nursing assessment and care plan for every inpatient (42 CFR 482.23); neither names a sedation scale, an interval, or a target field, so a RASS entry is defensible under them when it lets a reader reconstruct what was ordered, observed, given, and done. The restraint rules are LAW with specific documentation content: 42 CFR 482.13(e) allows restraint only to ensure the immediate physical safety of the patient, staff, or others, requires the least restrictive effective intervention and an order from a physician or other licensed practitioner, requires discontinuation at the earliest possible time regardless of the length of the order, and requires the record to show the patient's behavior and the intervention used, the alternatives or less restrictive interventions attempted, the condition or symptoms that warranted it, and the patient's response; a RASS value is none of those things, and it is not an order. CMS's anesthesia services condition (42 CFR 482.52) and its interpretive guidelines (S&C-10-09 of December 2009, revised by S&C-11-10 of January 14, 2011) place anesthesia, deep sedation, and monitored anesthesia care for procedures under pre- and post-anesthesia evaluation and rescue-capacity requirements along the ASA continuum; they name no sedation scale, and neither memorandum mentions the continuous sedation of mechanically ventilated ICU patients, so the ICU record's obligations come from the general record and nursing conditions and from professional standards rather than from the anesthesia condition. Under CONVENTION, the SCCM PADIS guideline (Devlin and colleagues, Critical Care Medicine, 2018) suggests light rather than deep sedation in mechanically ventilated adults (a conditional recommendation on low-quality evidence), achieved by daily sedative interruption or nurse-protocolized targeted sedation, and suggests propofol or dexmedetomidine over benzodiazepines; the 2025 focused update (Lewis and colleagues) adds a conditional suggestion for dexmedetomidine over propofol where light sedation or delirium reduction is the highest priority and otherwise leaves the 2018 framework in place. Neither document sets a documentation interval, a single universal target number, or a rule that the target, the drug time, and the CAM-ICU must share a row, and the definitions of light sedation used across the trials and guidelines differ, which is one more reason the patient's own ordered target is the reference point. The SCCM ICU Liberation bundle turns the convention into a metric: its data collection manual defines the C element by the number of documented level-of-arousal assessments per 24 hours using a standard scale such as the RASS or SAS, with a goal of at least six, and treats a RASS of -4 or -5 as exempt from the D element delirium assessment; in the Collaborative's 15,226 adults across 68 ICUs, complete bundle performance was associated with lower odds of death within seven days (adjusted hazard ratio 0.32), next-day mechanical ventilation (adjusted odds ratio 0.28), coma (0.35), and delirium (0.60) (Pun and colleagues, 2019), and a documented target sedation level was independently associated with higher odds of an awakening trial (adjusted odds ratio 1.68) and a breathing trial (1.46) being performed (Balas and colleagues, Chest, 2022). The Critical Care Societies Collaborative's Choosing Wisely list (2014) advises against deep sedation of ventilated patients without a specific indication and without daily attempts to lighten it. Joint Commission provision-of-care standards on sedation and anesthesia address pre-sedation assessment, monitoring, and recovery for procedures and name no scale, and its restraint standards parallel the CMS rules (CONVENTION with accreditation force). The CDC NHSN ventilator-associated event algorithm (January 2026 protocol) is built on ventilator settings, temperature, white cell count, antimicrobials, and microbiology, contains no sedation element, and states that it is for surveillance and is not a clinical definition algorithm, so a sedation practice or a missing RASS never triggers a VAE. Under PAYER POLICY, no CMS national coverage rule, NCCI edit, or Hospital IQR or MIPS measure keys payment or a score to a RASS value; ICU Liberation metrics are a voluntary quality program, and a hospital's own RASS documentation metric is local policy. Under neuromuscular blockade, the 2016 SCCM guideline (Murray and colleagues) states as good practice that analgesic and sedative drugs should be used before and during blockade, the 2026 SCCM guideline for adults with ARDS suggests either a scale-based or a non-scale evaluation of sedation depth before blockade and either monitoring or no monitoring during it, both on very low certainty evidence, and a 2022 meta-analysis of 24 studies and 1,235 patients found a correlation of 0.66 between the RASS and the bispectral index (Heavner and colleagues), so processed EEG is related to but not interchangeable with the scale; awareness with paralysis was reported in 2.6 percent of 383 ventilated emergency department patients (the ED-AWARENESS study, 2021) and in 3.4 percent of patients in the rigorously conducted studies pooled by the same group, which is the reason the record shows sedation and analgesia running rather than a manufactured value. Three habits deserve the same label: an hourly RASS by regulation, a target of -2 for every patient, and mandatory BIS under paralysis are unit conventions or assumptions, not rules any of these authorities state. Canada and Australia name no scale and set no interval, and the reliability record explains why the stimulus and the drug time belong in the entry. In Canada, no federal or provincial LAW names the RASS. The clearest guidance is provincial: Critical Care BC's ICU Liberation guide (Provincial Health Services Authority, July 22, 2025; CONVENTION) states that depth of sedation should be assessed every four hours or more using a validated tool, naming the RASS, that each patient should have a daily RASS goal documented with sedatives titrated to meet it, that the goal and the effectiveness of analgesic and sedative choices are discussed on daily interdisciplinary rounds, and that RASS and pain scores are documented every four hours or more to support handover. Health Standards Organization's HSO 11001:2026 Critical Care Services, the standard relevant to Accreditation Canada programs, was in public review as a draft dated March 2026 as of September 2026; the draft requires that organizational leaders provide the team with validated assessment tools for clients receiving critical care and describes reassessment with changes in clinical status, with input from clients or care partners, or at intervals defined by the organization, and it names no sedation scale, so no Accreditation Canada interval or RASS requirement should be asserted and the draft should be cited as a draft until adoption is published. The Choosing Wisely deep-sedation recommendation also appears on the Canadian critical care list. In Australia, the ANZICS-led SPICE III trial (Shehabi and colleagues, New England Journal of Medicine, 2019; 3,904 patients) targeted a RASS of -2 to +1 in both arms and found 90-day mortality of 29.1 percent in each, which is why that range appears in Australasian order sets as a trial target rather than a mandate. The NSQHS Recognising and Responding to Acute Deterioration Standard (second edition; CONVENTION with accreditation force) requires observation charts, track-and-trigger systems, and escalation without naming a consciousness or sedation instrument; general ward charts record consciousness as AVPU or ACVPU and opioid-induced sedation on a short ward sedation score, neither of which is a RASS, so a RASS on an Australian ward is a local choice and the ward sedation score should never be charted as one. The ACSQHC Delirium Clinical Care Standard (2021) asks for a validated tool rather than naming one, and its supporting material lists the modified RASS among tools used to monitor for incident delirium; the joint college guideline for patients sedated for acute behavioural disturbance (CICM IC-22 and ANZCA PS63) describes response to voice or stimulation as the way to monitor depth of sedation, treats tools such as the RASS as potentially helpful, and asks that drug, dose, time, and response be recorded, within its own scope of acute behavioural disturbance rather than routine ICU sedation. Two boundaries hold everywhere. Version: the RASS-PAL (Bush and colleagues, 2014) rests on a pilot of 13 clinicians rating 10 patients at five time points with inter-rater intraclass correlations of 0.84 to 0.98, whose authors asked for further validation, particularly for delirium; the modified RASS (Chester and colleagues, 2012) adds an attention item and, in 95 medical inpatients, detected delirium with 64 percent sensitivity and 93 percent specificity as a single assessment and 74 and 92 percent when used serially; and the adult scale was validated in critically ill children across 100 encounters (weighted kappa 0.825 between nurse and researcher; Kerson and colleagues, 2016) and 139 episodes in 14 pediatric ICUs (weighted kappa 0.946; Tapia and colleagues, 2021), so the note names the version and does not import adult targets. Reliability: Sessler and colleagues (2002) reported agreement of r 0.956 and weighted kappa 0.73 among five investigators in 192 encounters and r 0.964 and kappa 0.80 between a nurse educator and 27 trained bedside nurses in 101 encounters, and Ely and colleagues (2003) a weighted kappa of 0.91 between nurses; in routine practice, Fuchita and colleagues (2025) found that an ICU nurse and two trained investigators chose the identical value in only 35 percent of 79 encounters, that nurses recorded a lighter value than the investigators in 52 percent, and that at least one assessor noted uncertainty in 16 percent, and a 2026 prospective audit of 76 nursing assessments in a UK cardiothoracic unit found 75 percent accurate against an expert rater but only 33 percent documented hourly as that unit's policy required, with poorer agreement at deeper sedation (Szentgyorgyi and colleagues). Completion and accuracy are different problems, and both are the reason to write the behavior and the drug time when the number is not self-explanatory. The Richmond Agitation-Sedation Scale was published by Sessler and colleagues in the American Journal of Respiratory and Critical Care Medicine in November 2002; the article is copyright 2002 the American Thoracic Society, and Virginia Commonwealth University holds a registered copyright in the scale itself (2012; VCU Tech 12-063, inventors Sessler, Grap, Brophy, and Elswick), which the university describes as allowing free use by health systems, clinicians, and researchers while pharmaceutical companies must license it for clinical trials comparing sedative medications, and which its technology transfer office lists as available for licensing to industry. The Vanderbilt ICU Delirium and Cognitive Impairment Study Group, now the Critical Illness, Brain Dysfunction, and Survivorship Center, validated the scale and built it into the CAM-ICU workflow; icudelirium.org states that the CAM-ICU and its educational materials, which carry the RASS on the pocket card, are copyright 2002 E. Wesley Ely and Vanderbilt University, that the group deliberately made them unrestricted for implementation and clinical use with the copyright line requested on pocket cards and educational materials, and that other uses go to [email protected] . That permission is Vanderbilt's, for Vanderbilt's materials; it is not a general license to republish the scale's level descriptors, which this page describes and does not reproduce. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common RASS documentation errors reviewers flag The numbers behind these errors are specific. In the original validation, five investigators (two physicians, two nurses, and one pharmacist) agreed at r 0.956 and weighted kappa 0.73 across 192 encounters, and a nurse educator and 27 trained bedside nurses at r 0.964 and kappa 0.80 (Sessler and colleagues, 2002); in contemporary practice an ICU nurse and two trained investigators chose the identical value in only 35 percent of 79 encounters , nurses recorded a lighter value than the investigators in 52 percent, and an assessor flagged uncertainty in 16 percent (Fuchita and colleagues, 2025), and a 2026 audit of 76 nursing assessments found 75 percent accurate but only 33 percent documented hourly as the unit required, with most discrepancies one level (Szentgyorgyi and colleagues). On the target side, a goal-directed sedation guideline in one medical ICU raised the share of orders carrying a patient-specific RASS goal from 21.3 to 85.4 percent and RASS assessments per 24 hours from 4.7 to 11.4, while the share of assessments at goal rose only from 31.4 to 44.1 percent (DeGrado and colleagues, 2011); across the ICU Liberation Collaborative, a documented target sedation level was associated with higher odds of an awakening trial ( adjusted odds ratio 1.68 ) and a breathing trial (1.46) being performed (Balas and colleagues, 2022), and complete bundle performance with lower odds of coma (0.35) and delirium (0.60) in 15,226 adults (Pun and colleagues, 2019). No published audit counts how often a RASS entry lacks its target, its clock time, or the sedation on board; the bundle metrics and the compliance reviews described under compliance considerations are what turn those omissions into a quality finding. The BastionGPT Clinical Advisory Board sees the same errors most often in RASS documentation reviews: A column of values with no target. Values every two hours with no ordered target anywhere in the flowsheet, so no reader can say whether a -3 was intended. Chart the target as the order states it, with its date, beside the values or in the header they sit under, and when none has been ordered, obtain one and chart the request; the scale has descriptive validity on its own, but sedation management does not. The target rewritten, or a house target assumed. "Light sedation" typed in place of an order for -1 to 0, a range narrowed to one number, or -2 charted as the target because that is what the unit usually does. Preserve the order as written, record the indication when a deeper target is intentional, and never present a guideline preference or a calculator default as the patient's order. A value with no clock or drug context. A -3 charted twenty minutes after a procedural bolus and read as the infusion, or an agitated value with no note of what was running. Give every value a time and its place in the course of care, name the drug, rate, or bolus time that explains an off-target value, and label a post-bolus value as transient. Score to sedative, cause skipped. An agitated value answered with more propofol before pain, hypoxemia, dyssynchrony, a full bladder, withdrawal, or delirium were considered, or restraints applied because the number reached +3. Chart the pain score and tool, the reversible causes checked, the measures tried, the medication given under its order with the time, and the recheck; restraint documentation follows 42 CFR 482.13(e), not the RASS. A manufactured value under paralysis, or a delirium screen at the wrong level. RASS -5 charted hourly on a patient receiving cisatracurium as if it measured sedation depth, a CAM-ICU charted as negative when the RASS that hour made it unassessable, or a RASS of 0 read as absence of delirium. Under blockade write that the scale cannot be assessed, carry the last pre-blockade value forward, and record sedation, analgesia, and the monitoring plan; record the CAM-ICU only at -3 or lighter and after the arousal check. An awakening trial with no trajectory, or a handoff without the sedation state. "Propofol held per SAT" with no pre-trial value, stop time, values during, failure criteria, breathing trial coordination, or restart rate; a transfer note that says "sedated, stable" with no target, value, time, infusion, or CAM-ICU. Chart the trial as a sequence and hand off the five items: target, last value with time, infusions and last bolus, last CAM-ICU, and trial status. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on critical care nursing and sedation documentation. Give it the facts (the ordered target and its date, each assessed value with its clock time and the moment in care, the stimulus or behavior where it was ambiguous, the infusions and boluses with times, the pain score and tool, any awakening trial sequence and CAM-ICU result, any blockade or restraint event, and the recheck) and it drafts the entry: target beside actual, the relation to target with its cause, the action taken under the order, the trial trajectory and the delirium screen in order, and the handoff line, ready for your review. Cross-check a finished flowsheet narrative or note for the gaps reviewers flag: values with no target, a target rewritten, a value with no time or drug context, a sedative given before a pain assessment, a manufactured value under paralysis, a CAM-ICU at an unassessable level, or an awakening trial charted as a single word. Turn the day into the next document: the sedation status lines of a transfer-of-care summary, the timeline paragraph of an incident report after a self-extubation or restraint event, or a plain-language update for a family, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the RASS scored, and what do the positive, zero, and negative values mean? The RASS has ten levels on one continuum. Positive values (+1 to +4) describe increasing agitation, from restlessness to behavior that endangers the patient or staff; zero is a patient who is awake and calm without prompting; negative values (-1 to -5) describe increasing depth of sedation, separated by how much stimulation it takes to get a response and by the quality of that response. The assessment is a sequence: observe the undisturbed patient first, and rate an alert or agitated patient from observation alone; if the patient is not alert, speak, say the name and ask for eye opening and eye contact, and rate by whether contact is made and how long it is held; only if voice produces nothing, use physical stimulation, and the two deepest values differ by whether anything happens at all. A negative value therefore already tells the next reader which stimulus was needed. The wording that defines each level is the scale's copyrighted text (Virginia Commonwealth University, 2012), so score from the published card your unit uses and describe the behavior in your own words when a value could be read two ways. It takes under a minute at the bedside, and it is neither a pain scale nor a delirium screen. Do I chart the target or the actual RASS, and what if the order is a range or no target was ordered? Both, every time the flowsheet allows it. The target belongs in the record exactly as the order states it, a single value or a range (-1 to 0, or -2 to +1), with the order date, never translated into a phrase such as light sedation and never narrowed to one number. The assessed value sits beside it with its time, so each entry reads as at target, deeper, or lighter; Vanderbilt's trauma protocol states that "goal and actual RASS should be specified on ALL patients," and Critical Care BC asks for a documented daily RASS goal with sedatives titrated to it. If the order is a range, chart the range and note which end the team is aiming for when rounds decided that. If no target has been ordered, obtain one and chart the request, because a value with no target cannot be read as on or off course and a goal-directed protocol has nothing to titrate to; in one medical ICU, a goal-directed guideline raised orders with a patient-specific RASS goal from 21.3 to 85.4 percent. If the ordered target is deeper than the guideline preference for light sedation, preserve it, record the indication (neuromuscular blockade, status epilepticus, refractory intracranial hypertension, severe dyssynchrony), and clarify an unexplained deep target with the team rather than substituting your own; a target of -2 for everyone is a habit, not a guideline. How often does the RASS have to be documented? Is there a regulatory interval? No law or accreditor in the United States, Canada, or Australia sets a RASS interval, and PADIS does not either; the interval comes from the medication order and the unit protocol, plus the clinical moments that call for an unscheduled value (a titration, a bolus, a procedure, an awakening trial, a change in behavior, and the arousal check before every CAM-ICU). Real protocols differ: Critical Care BC (2025) asks for depth of sedation every four hours or more, Vanderbilt's trauma protocol monitors each shift, one UK cardiothoracic unit requires hourly entries, some units require a value with every titration of a continuous sedative, and the SCCM ICU Liberation data manual counts documented level-of-arousal assessments per 24 hours with a goal of at least six. Follow your protocol, cite it in the note as policy, and remember that frequency and accuracy are separate problems: in the 2026 UK audit, 75 percent of nursing values matched an expert rater but only 33 percent were documented hourly as required. A row of hourly numbers is not evidence of good sedation management if the target, the drug context, and the response to off-target values are missing, and a correctly assessed patient can still fail a local frequency metric. Which RASS values allow a CAM-ICU, and does a RASS of 0 mean no delirium? The RASS is the arousal check that comes first. In the Vanderbilt workflow, a RASS of -4 or -5 means the patient cannot be assessed for delirium and the CAM-ICU is recorded as unable to assess for that attempt; any value from -3 through +4 clears the gate and the delirium features are assessed, so -3 is not automatically too sedated. The SCCM ICU Liberation data manual treats -4 and -5 the same way, as exempt from the delirium element. Chart the RASS with its time first and the CAM-ICU result after it, and repeat the screen when the patient becomes assessable again if one is due; the RASS and the CAM-ICU are linked but not synchronous, since sedation is assessed many times a day while delirium is usually screened once a shift, so the absence of a CAM-ICU beside every RASS is not a defect. A RASS of 0 says the patient is calm and alert; it says nothing about delirium, which is answered only by the CAM-ICU features, and a patient can be delirious at 0, at -2, or at +2. Delirium itself is a medical diagnosis documented by the provider; the screen is nursing documentation. What RASS do I chart for a patient on neuromuscular blockade, and is BIS required? Do not manufacture one. The scale depends on visible responses, and a paralyzed patient will look unresponsive whatever their awareness, so a charted -5 under effective blockade is not a sedation assessment and can hide inadequate sedation. Chart that the RASS cannot be assessed because of pharmacologic paralysis, with the blocker and its start time; carry forward the last assessment taken before the blocker with its time; record the sedative and analgesic regimen with rates; and name the unit's strategy for judging sedation depth during blockade, whether processed EEG or clinical signs. The 2016 SCCM guideline on sustained blockade states as good practice that analgesic and sedative drugs should be used before and during blockade, and the 2026 SCCM guideline for adults with ARDS suggests either a scale-based or a non-scale evaluation before blockade and either a monitoring-based strategy or no monitoring during it, both on very low certainty evidence, so BIS is not mandatory; a 2022 meta-analysis found a correlation of only 0.66 between the RASS and the bispectral index, so the two are related and not interchangeable. The reason the entry matters is awareness with paralysis: 2.6 percent of 383 ventilated emergency department patients in the ED-AWARENESS study and 3.4 percent in the rigorously conducted studies pooled by the same group. When the blocker wears off, resume scoring and write the first assessable value with its time. How do I document a spontaneous awakening trial with the RASS? As a trajectory, not an event. Record the pre-trial value with its time and the target; the safety screen result under your protocol (the original ABC trial excluded patients on a sedative infusion for active seizures or alcohol withdrawal, on escalating sedation for agitation, on neuromuscular blockade, or with active myocardial ischemia or raised intracranial pressure); the time each sedative was stopped and which infusions, usually the analgesic, continued; the values during the trial with their times and the behavior that produced them; the failure criteria you watched for (sustained agitation, respiratory distress or a high respiratory rate, desaturation, arrhythmia, hemodynamic instability) and whether any occurred; the coordination with the respiratory therapist for the breathing trial and its result; and, if sedation was restarted, the reason, the time, and the rate (the ABC protocol restarted at half the prior dose, a protocol choice rather than a rule). Once the patient has cleared the arousal gate and a delirium screen is due, chart the CAM-ICU after the trajectory. Daily interruption traces to Kress and colleagues (2000) and the paired trial to Girard and colleagues (2008); PADIS names daily interruption and nurse-protocolized targeted sedation as the ways to achieve light sedation, and the sample on this page shows one full day. RASS or GCS, SAS, Ramsay, and pain scales: which one do I chart, and when? Chart the one that answers the question, and name it. The RASS answers how sedated or agitated the patient is and whether a delirium screen can proceed. The Glasgow Coma Scale answers level of consciousness after brain injury through eye, verbal, and motor components; the two correlated at r 0.91 in the Vanderbilt validation but measure different things, and in a sedated neurological patient both are charted, with the sedative named as a confounder of the GCS. The Riker Sedation-Agitation Scale (1999, seven levels, weighted kappa 0.92 in its validation) is the other tool the SCCM guidelines treat as valid and reliable; a unit uses one or the other, and no conversion between them belongs in a chart. The Ramsay scale is older, runs in the opposite direction, and separates agitation poorly; if a legacy order uses it, chart under that name and ask for the order to be rewritten. Pain is a separate construct: PADIS asks for the CPOT or BPS in patients who cannot self-report, and pain is assessed and treated before sedation is escalated, because an agitated RASS is often a pain finding. On general wards and in Australian observation charts, consciousness is recorded as AVPU or ACVPU inside an early warning score (see the MEWS and NEWS2 page) and opioid-induced sedation on a short ward sedation score; neither is a RASS, and the RASS was validated in critical care populations rather than in every hospitalized adult. Can we put the RASS in our EHR, on a printed flowsheet, or on our website? For a health system's own clinical use, yes, on the rights holder's stated terms. Virginia Commonwealth University registered a copyright in the scale in 2012 (VCU Tech 12-063) and states that the protection allows free use by health systems, clinicians, and researchers while pharmaceutical companies must license it for clinical trials comparing sedative medications; its technology transfer office lists the scale as available for licensing to industry. So a hospital printing the scale on its own flowsheet or building it into its own EHR sits inside that permission, with attribution to Sessler and colleagues (2002) and Ely and colleagues (2003) as the convention. A software vendor redistributing the descriptor wording in a commercial product, a paid app, or a public website that republishes the complete scale is a different case: VCU's statement names health systems, clinicians, and researchers, not public republication, and the wide appearance of the table on calculator sites is not a license. The Vanderbilt CIBS Center's permission covers its own CAM-ICU materials, which carry the RASS on the pocket card; it asks for its copyright line and refers other uses to [email protected] , and it is not a substitute for VCU's position on the scale itself. Free, non-commercial, or attributed does not by itself create a license, which is why this page describes the scale in original words and reproduces no level descriptors. Can BastionGPT draft my RASS documentation? Yes. Give it the facts (the ordered target and its date, each assessed value with its clock time and the moment in care, the stimulus or behavior where it was ambiguous, the infusions and boluses with times, the pain score and tool, any awakening trial sequence and CAM-ICU result, any blockade or restraint event, and the recheck) and it drafts the full entry: target beside actual, the relation to target with its cause, the action taken under the order, the trial trajectory and the delirium screen in order, and the handoff line, ready for your review. It can also check a finished note for values with no target, a target rewritten, a value with no time or drug context, a sedative given before a pain assessment, a manufactured value under paralysis, a CAM-ICU at an unassessable level, or an awakening trial charted as a single word. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Sessler CN, Gosnell MS, Grap MJ, Brophy GM, O'Neal PV, Keane KA, Tesoro EP, Elswick RK, 2002, American Journal of Respiratory and Critical Care Medicine 166(10):1338-1344, the Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients (five investigators, two physicians, two nurses, and one pharmacist; 192 encounters; r 0.956, kappa 0.73; nurse educator and 27 nurses, 101 encounters, r 0.964, kappa 0.80; copyright 2002 the American Thoracic Society); Virginia Commonwealth University, November 6, 2023, inventor of world standard for determining ICU patient comfort named VCU innovator (copyright 2012; free use by health systems, clinicians, and researchers; pharmaceutical companies license for trials); VCU TechTransfer and Ventures, Richmond Agitation-Sedation Scale listing (VCU Tech 12-063; registered copyright 2012; available for licensing to industry). Ely EW, Truman B, Shintani A, and colleagues, 2003, JAMA 289(22):2983-2991, monitoring sedation status over time in ICU patients (38 and 275 patients; weighted kappa 0.91 against 0.64 for the GCS; r 0.91 with the GCS, 0.78 with attention screening, 0.63 with bispectral EEG; no sedation protocol or routine targets in the study ICU); Vanderbilt Critical Illness, Brain Dysfunction, and Survivorship Center, monitoring delirium in the ICU (copyright 2002 E. Wesley Ely and Vanderbilt University; unrestricted implementation and clinical use with the copyright line; other uses to [email protected] ; delirium assessment once per shift; accessed September 2026); Vanderbilt University Medical Center Trauma and Surgical Critical Care, June 2026, delirium practice management guideline (goal and actual RASS specified on all patients; RASS and CAM-ICU each shift; CAM-ICU unable to assess at RASS -4 or -5). Guidelines: Devlin JW and colleagues, 2018, Critical Care Medicine 46(9):e825-e873, clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption (PADIS) , with the SCCM guideline page (light over deep sedation, conditional; propofol or dexmedetomidine over benzodiazepines); Lewis K and colleagues, 2025, Critical Care Medicine 53(3):e711-e727, focused update to the PADIS guideline (dexmedetomidine over propofol where light sedation or delirium reduction is the highest priority); Barr J and colleagues, 2013, Critical Care Medicine 41(1):263-306, the 2013 PAD guideline ; Murray MJ and colleagues, 2016, Critical Care Medicine 44(11):2079-2103, clinical practice guidelines for sustained neuromuscular blockade ; Society of Critical Care Medicine, 2026, guidelines for the administration of neuromuscular blockade in adults with ARDS (scale-based or non-scale evaluation before blockade; monitoring or no monitoring during; very low certainty); Heavner MS and colleagues, 2022, Pharmacotherapy 42(8):667-676, correlation between bispectral index and clinical sedation scales (24 studies, 1,235 patients; RASS 0.66). ICU Liberation: Society of Critical Care Medicine, the ABCDEF bundle and ICU Liberation Bundle data collection manual (documented level-of-arousal assessments per 24 hours, goal of at least six, RASS or SAS; RASS -4 or -5 exempt from delirium assessment); Pun BT and colleagues, 2019, Critical Care Medicine 47(1):3-14, results of the ICU Liberation Collaborative in over 15,000 adults ; Balas MC and colleagues, 2022, Chest 162(3):588-602, factors associated with spontaneous awakening trial and spontaneous breathing trial performance (documented target sedation level, adjusted odds ratios 1.68 and 1.46); DeGrado JR and colleagues, 2011, Journal of Pain Research 4:127-134, evaluation of a local ICU sedation guideline on goal-directed administration of sedatives and analgesics (RASS goal in orders 21.3 to 85.4 percent; assessments per 24 hours 4.7 to 11.4); Kress JP and colleagues, 2000, New England Journal of Medicine 342(20):1471-1477, daily interruption of sedative infusions ; Girard TD and colleagues, 2008, Lancet 371(9607):126-134, the Awakening and Breathing Controlled trial ; Critical Care Societies Collaborative, 2014, Choosing Wisely critical care recommendations . Reliability in practice and the alternative scale: Fuchita M and colleagues, 2025, Critical Care Explorations 7(9):e1302, inter-rater agreement of RASS assessments in adult patients receiving mechanical ventilation (79 encounters, 58 nurses; identical value in 35 percent; nurses lighter in 52 percent; uncertainty noted in 16 percent); Szentgyorgyi L and colleagues, 2026, Nursing in Critical Care 31(4):e70531, accuracy and documentation of RASS assessments in a cardiothoracic critical care unit (76 assessments; 75 percent accurate; 33 percent documented hourly); Riker RR, Picard JT, Fraser GL, 1999, Critical Care Medicine 27(7):1325-1329, prospective evaluation of the Sedation-Agitation Scale (45 patients, 69 paired assessments, weighted kappa 0.92). Variants and special populations: Bush SH and colleagues, 2014, BMC Palliative Care 13:17, the RASS modified for palliative care inpatients (RASS-PAL) (13 clinicians, 10 patients, five time points; ICC 0.84 to 0.98); Kerson AG and colleagues, 2016, Journal of Intensive Care 4:65, validity of the RASS in critically ill children (100 encounters, 50 patients, 2 months to 21 years; weighted kappa 0.825); Tapia R and colleagues, 2021, Frontiers in Pediatrics 9:795487, validity and reliability of the RASS in pediatric intensive care patients: a multicenter study (28 observers, 14 PICUs, 139 episodes; weighted kappa 0.946); Chester JG, Beth Harrington M, Rudolph JL, 2012, Journal of Hospital Medicine 7(5):450-453, serial administration of a modified RASS for delirium screening (95 veterans); Shehabi Y and colleagues (SPICE III), 2019, New England Journal of Medicine 380(26):2506-2517, early sedation with dexmedetomidine in critically ill patients (target RASS -2 to +1; 3,904 patients); Pappal RD and colleagues, 2021, Critical Care Medicine 49(3):e304-e314, awareness with paralysis in mechanically ventilated patients: systematic review and meta-analysis (3.4 percent in high-quality studies), and 2021, Annals of Emergency Medicine 77(5):532-544, the ED-AWARENESS study (2.6 percent of 383 patients). United States law and surveillance: eCFR, 42 CFR 482.24 medical record services , 42 CFR 482.23 nursing services , 42 CFR 482.13 patient's rights (paragraph (e), restraint or seclusion, including the (e)(16) documentation list), and 42 CFR 482.52 anesthesia services ; CMS, January 14, 2011, S&C-11-10, revised hospital anesthesia services interpretive guidelines (ASA continuum; no sedation scale named); CDC National Healthcare Safety Network, January 2026, Patient Safety Component Manual chapter 10, ventilator-associated event (surveillance definition, not a clinical definition algorithm; no sedation element). Canada and Australia: Critical Care BC, Provincial Health Services Authority, July 22, 2025, ICU Liberation element C: choice of analgesic and sedation (RASS every four hours or more; daily documented RASS goal; titration to goal); Health Standards Organization, March 2026, HSO 11001:2026 Critical Care Services, public review draft (validated assessment tools; intervals defined by the organization; no scale named); Australian Commission on Safety and Quality in Health Care, Delirium Clinical Care Standard (2021) and Recognising and Responding to Acute Deterioration Standard ; College of Intensive Care Medicine of Australia and New Zealand, IC-22, guideline for safe care for patients sedated in health care facilities for acute behavioural disturbance (joint with ANZCA PS63, ACEM, and RANZCP). ## Related templates CAM-ICU Documentation Glasgow Coma Scale (GCS) Documentation MEWS & NEWS2 Early Warning Score Documentation Transfer-of-Care Summary Incident & Adverse Event Report CIWA-Ar Protocol Documentation Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/morse-fall-scale All note templates # Morse Fall Scale Documentation: Risk Score to Interventions & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Morse Fall Scale is a nurse-rated screen for a patient's risk of falling, published by Janice Morse and colleagues in 1989, that scores six items for a total of 0 to 125, higher meaning higher risk. Hospital, rehabilitation, long-term care, and home health nurses use it to decide who needs which fall prevention measures. This page covers how to document the score and the interventions it triggers, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Registered nurses on admission and at reassessment in acute care, rehabilitation, long-term care, and home health, with licensed practical or vocational nurses where facility policy and state scope allow; no publisher qualification level or training certificate applies, and the nurse who watched the patient move records the result Audience Relieving and receiving nurses at handoff, charge nurses and nurse managers, attending physicians and advanced practice providers, pharmacists and physical therapists who receive the referrals, patient safety and quality staff, coders determining present-on-admission status after an injury, CMS and state surveyors, accreditation reviewers, and the post-fall reviewers and attorneys who reconstruct the sequence Typical length 4 to 10 chart lines (trigger and version, items that scored with the total, facility category, a measure for each scoring item, teaching and response, reassessment triggers) · 2 to 4 minutes by hand after the patient has been observed walking Format family Nurse-rated fall risk screen (six weighted items, some with graded responses, total 0 to 125, higher total meaning higher risk; risk categories set by facility policy from published examples, not by the instrument) When it's used Admission and transfer assessments in acute care and rehabilitation, scheduled reassessment under unit policy, change of condition, after any fall, long-term care admission and periodic reassessment beside the MDS, home health start of care under agency policy; pediatric units use tools validated in children Standards context Morse, Morse, and Tylko (Canadian Journal on Aging, 1989); copyrighted, free to use in clinical care with the author's permission per the Fall TIPS FAQ; carried as an example tool by AHRQ and the VA and mandated by no regulator or accreditor in the United States, Canada, or Australia; described here for documentation, no items or weights reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Morse Fall Scale? The Morse Fall Scale is a nurse-rated screen for the likelihood that a hospitalized adult will fall. Janice M. Morse developed it with Robert M. Morse and Suzanne J. Tylko and published it in the Canadian Journal on Aging in 1989 (volume 8, pages 366 to 377), from a case-control study whose discriminant analysis correctly classified 80.5 percent of patients; her book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008) carries the scoring guidance and the level-of-risk chapter that facilities still cite. The scale weights six items, some with graded responses, for a total of 0 to 125, and a higher total means higher risk. Named here as topics in original words, the items ask whether the patient has fallen recently or during this admission; whether more than one active medical diagnosis is present; what the patient relies on to walk, because dependence on an aid, or on whatever is within reach, marks unsteadiness; whether an intravenous line or access device is in place, because it tethers and encumbers; how the patient actually walks when observed; and whether the patient's judgment of what they can safely do matches their real ability, the item that flowsheets usually label mental status. The item wording, the response anchors, and the point weights are the copyrighted text of the instrument and are not reproduced or paraphrased on this page; your facility's licensed form or approved electronic build is the source of record. The load-bearing fact for documentation is that the total is a screening number whose meaning is set locally. The familiar bands, with 25 and 45 as the usual edges, appear as an example in AHRQ's Preventing Falls in Hospitals toolkit (Tool 3H) and in the VA's falls toolkit, whose policy template leaves the high-risk score blank for each unit to determine and points to Morse's book, which advises calibrating the cutoff to the unit's own patients; validation studies have found the best cutoff at 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), 51 in a Korean electronic record (Baek and colleagues, 2014), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and no regulator in the United States, Canada, or Australia names the scale or a threshold. The original validation reported a positive predictive value of 10.3 percent, so most patients the scale calls high risk never fall, and a 2026 VA study of 5,004 inpatients found the mean total above 45 in fallers and non-fallers alike while individual items still carried signal (Oppegaard and colleagues). Current guidelines have moved further: NICE (2025) advises against using falls risk prediction tools to predict who will fall, and the ACSQHC's 2025 hospital guideline treats every older inpatient as high risk and directs effort to the person's own risk factors. The lesson for the chart is the same in every setting: the items that scored, not the total, choose the interventions, and the record has to show that link. Locally altered forms (a medication item added, a rule that any in-facility fall makes the patient high risk, a 51-point band) are local instruments rather than the published scale, which Morse herself addressed in a 2006 commentary, so the note names the version or build it used. The Braden Scale is the structural peer on the skin side: a risk total that must never be mistaken for a plan. ## Who uses Morse Fall Scale documentation and when Medical, surgical, orthopedic, and rehabilitation unit nurses score it on admission, on transfer, at the interval unit policy sets, when the patient's condition changes (a new sedating, opioid, or blood pressure medication, new confusion, a new line, a change in mobility), and after any fall; critical care nurses record it beside the RASS and the CAM-ICU , where sedation and delirium alter the self-appraisal and mobility items at once, and where a patient who does not walk can score low while transfers and toileting remain the moments of danger. Long-term care nurses use it, or a facility variant, at admission, with the MDS assessment cycle, with any change of condition, and after any fall, feeding the Section J fall items and the care plan that surveyors read under F689. Home health nurses record a fall risk assessment at the start of care under agency policy, sometimes the Morse and often a tool built for community-dwelling patients, and the OASIS assessment page covers what the item set itself asks. Charge nurses and nurse managers audit the entries; pharmacists and physical therapists receive the medication review and mobility referrals that scoring items generate; physicians and advanced practice providers document injuries and present-on-admission status after a fall, which rest on the clinical assessment and not on the score; patient safety and quality staff read the sequence after every event. The Morse page owns one job: the assessment, the items that scored, the facility's category named as policy, and the link from each scoring item to a specific measure and a reassessment trigger. Neighbors take over from there: a fall is reported through the incident report while the clinical reassessment stays in the chart, the risk status and measures travel in the transfer of care summary and the discharge summary , the medication review belongs in the medication management note , and the interdisciplinary plan the items feed belongs in the care coordination note . Pediatric units use a tool validated in children, such as the Humpty Dumpty Falls Scale, because the Morse was developed and validated in adults. ## How to document a Morse Fall Scale assessment in the chart No regulator prescribes a Morse note format, and no authority in the United States, Canada, or Australia requires the Morse by name; what exists is a copyrighted instrument, facility policies that set the reassessment interval and the risk categories, and survey, accreditation, and coding rules that judge whether an identified risk led to action. What survives review is an entry that states the trigger and the version, names the items that scored in plain words with the total, attributes the category to policy, ties every scoring item to a specific measure that was actually delivered, records teaching and the patient's response, sets the next reassessment and the changes that bring it forward, and, after a fall, documents a new clinical assessment rather than a copied score. Each element below carries the pitfall that most often undermines it. Trigger, timing, version, and rater. Open with why the assessment is being done: admission, transfer in, the scheduled interval under a named policy, a change in condition (a new opioid, sedative, or antihypertensive, new confusion, a new line, a change in how the patient moves), or a fall. Then the version (the published scale as built in the electronic record, or a named facility variant), the date and time, and the rater who watched the patient move. Every-shift scoring is a unit policy: the VA's falls toolkit names admission, transfer, change in condition, and after a fall as the triggers and leaves routine frequency to the facility, and neither CMS nor the Joint Commission sets an interval, so cite the policy rather than presenting the habit as a rule. A trigger named in one clause is what turns a repeated total into evidence of monitoring. Pitfall: A total time-stamped every twelve hours with no trigger, or a note that presents the unit's every-shift habit as a CMS or Joint Commission requirement. The items that scored, in plain words, then the total. Write which of the six items scored and what each rests on in ordinary clinical facts: the date and place of the last fall, the active diagnoses, the aid the patient uses, the line that is in, what you saw when the patient walked (step length, steadiness, use of the hands, hesitation), and what the patient said about getting up alone against what you observed. Then the total against its maximum. Rate the patient you observed today: the gait and self-appraisal items are assessments of the patient, not of the chart, which AHRQ's toolkit says in as many words, and a total copied from the last shift is not a reassessment. The item topics can be named in your own words; the instrument's wording and weights belong to the licensed form. Pitfall: "Morse 85" with no items named, or a gait item rated from the previous entry while the patient lay in bed the whole shift. Facility risk category, named as policy, with any override. Write the category the facility policy assigns to the total and cite the policy, because the categories are conventions layered on the instrument: the 25 and 45 edges most facilities use appear as an example in AHRQ's toolkit and in VA practice, Morse's own guidance is to calibrate the cutoff to the unit, and published optima run from 40 to 55 by population. Then record any clinical judgment override and its reason: a patient who does not walk and scores low on the mobility items but climbs out of bed, or a patient in the high band whose only scoring item has resolved. The category tells a reader which protocol tier applies; the items, not the tier, decide what is done, and a sentence such as high risk therefore alarm should be written only where that linkage is literally the organization's approved protocol. Pitfall: A category presented as the scale's rule, or a bedbound, agitated patient labeled low risk with no override recorded. An intervention for each scoring item. For every item that scored, name the measure in place or started and how it is delivered: for a prior fall, the rounding interval, the call light within reach, and the teach-back; for multiple diagnoses, the medication review requested and from whom; for a walking aid, the aid within reach and the level of assistance for every walk; for a line, the pole on the side of the aid, tubing managed, disconnection when policy allows; for an unsteady gait, the gait belt, the footwear, the bed height, the therapy referral; for a mismatch between the patient's confidence and ability, the supervision plan, the toileting schedule, and who stays with the patient. Add the universal environmental checks with the time. Fall TIPS, the toolkit built on the scale, exists to make exactly this link, and in 37,231 patients on 14 units it was associated with 15 percent fewer falls and 34 percent fewer injurious falls (Dykes and colleagues, 2020). A bed or chair alarm is one measure among others, not the plan: CMS's long-term care guidance cautions against relying on position-change alarms as the sole intervention, and no authority reads a high total as a reason for restraint. Pitfall: "Fall precautions maintained" or "high fall risk bundle in place" with no measure that answers a specific scoring item. Teaching, preferences, and refusals. Record the teaching given to the patient and family, in their words where you can, with a teach-back of the one behavior that matters most (call before getting up), and the patient's stated preferences. When a measure is declined, chart the offer, the reason given, the education, the alternatives offered or accepted, who was notified under policy, and the plan as revised; CMS's long-term care guidance expects a resident's choice to be honored while the team works to mitigate the risk, and a signed waiver does not end the duty. A family member at the bedside is part of the plan, not a scored item, so it changes the measures and the supervision, never the number. Pitfall: "Refused bed alarm" with nothing after it, or a daughter staying overnight charted as if the total had dropped. Reassessment triggers and a changed total. Close with the next scheduled reassessment under the policy and the changes that bring it forward: transfer, a new sedating, opioid, or blood pressure medication, new confusion, a change in mobility, a line placed or removed, and any fall. When the total changes, say which item changed and why in clinical terms, because two items can move in opposite directions and hide each other: a line removed lowers the total on the same day that new confusion raises the risk, and a note that reports only the lower number tells the next reader the opposite of the truth. De-escalate measures item by item when an item resolves, and carry the scoring items, the measures in place, and the next reassessment into the handoff and the transfer of care summary . Pitfall: A total that fell because a line came out read as lower risk, while the patient's new confusion went unrecorded. After a fall: the clinical entry and the incident report. A fall triggers a new assessment, not an edit. Chart the injury check and vital signs, the neurological checks where the head may have struck, the provider notification with time and response, the imaging or treatment ordered, a fresh Morse in which the fall now scores, the contributing factors you can see (toileting urgency, footwear, lighting, a line, a medication given), the revised measures, the patient's and family's understanding, and the next reassessment. Report the event through the organization's safety reporting system separately; AHRQ's toolkit keeps the clinical post-fall assessment and the root-cause review as different activities, and the chart should not carry the incident narrative or speculation about blame. Whether the chart cites the report number is a local risk-management decision. The injury diagnosis and present-on-admission status that coders need come from the provider's note; the fall's time and circumstances come from yours. The incident report page covers the safety document itself. Pitfall: The pre-fall score copied forward after the event, the incident report pasted into the progress note, or a post-fall entry with no revised plan. ### Blank template (copy and adapt) MORSE FALL SCALE DOCUMENTATION BLOCK (risk assessment and linked plan) Date / time: [ ] Setting: [ ] Rater: [ ] Trigger: [admission / transfer / scheduled per policy (name it) / change in condition (state the change) / after a fall] Version: [published scale as built in the EHR / facility variant (name it)] Items that scored (plain words, no weights): [recent or in-facility fall: date and place / more than one active diagnosis: which / walking aid: which / IV line or access: which / observed gait: what you saw / self-appraisal of ability against observed ability] Items that did not score and why: [ ] Total: [ ] of 125 Prior total and date: [ ] Item that changed and why: [ ] Facility risk category: [ ] per [policy name]; categories are policy, not the scale's rule Clinical judgment override: [none / reason] Interventions by scoring item (what is actually in place): prior fall: [rounding interval; call light in reach; teach-back] diagnoses: [medication review requested: by whom, when] walking aid: [aid within reach; assist level for every walk] IV line: [pole placement; tubing managed; disconnection per policy] gait: [gait belt; footwear; bed height; therapy referral] self-appraisal: [supervision plan; toileting schedule; who stays] universal: [bed low and locked; path clear; lighting; belongings in reach] Teaching and response: [patient and family teach-back / preferences / declined: offer, reason, education, alternatives, who was notified] Reassessment: [next per policy; earlier for transfer, new sedating or blood pressure medication, new confusion, mobility change, any fall] After a fall (if applicable): [injury check; vitals; provider notified, time; new score; contributing factors; revised plan; safety report filed separately] Communication: [provider, pharmacy, therapy, charge nurse; handoff; transfer summary] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's items, anchors, or point weights. ## Sample Morse Fall Scale documentation (fictional) Scenario: a US community hospital medical unit, where the admitting nurse documents a 77-year-old woman's Morse Fall Scale on the evening of admission for pneumonia, names the items that scored in plain words, attributes the risk category to hospital policy, ties each scoring item to a specific measure, records a declined bed exit alert with the alternatives, and two days later documents a change-of-condition reassessment in which the total fell while the risk rose. All details are fictional. Patient: L.M., 77 · Setting: Community hospital, medical unit; admission assessment and change-of-condition reassessment · Clinician: R. Delgado, RN · Note date: 09/14/2026 Trigger, version, and rater (09/14/2026, 20:10): Admission fall risk assessment under the hospital's fall prevention policy NSG-207, which requires the Morse Fall Scale on admission, every shift, with any change in condition, and after any fall; the interval is hospital policy, not a regulatory requirement. Published scale as built in the electronic record, rated by R. Delgado, RN, at 20:10 on 09/14/2026 after walking L.M. from the stretcher to the bathroom and back with her cane. Admitted from the emergency department at 19:30 with community-acquired pneumonia; no prior score this admission. Items that scored and total: Five of the six items scored. Prior fall: L.M. fell in her kitchen on 08/22/2026 reaching for a cupboard, no injury, confirmed by her daughter. Active diagnoses beyond the admitting one: heart failure, atrial fibrillation on an anticoagulant, and osteoarthritis of both knees. Walking aid: a single-point cane brought from home. Intravenous access: a peripheral line in the left forearm for antibiotics and a diuretic. Observed gait: short, slow steps, trunk bent forward, eyes on the floor, one hand on the cane and the other reaching for the wall on the return, with two stops to steady herself. The self-appraisal item did not score: she says she cannot walk without the cane, asked for help before standing, and described the near-miss that preceded the August fall accurately. Total 85 of 125. Facility risk category: High fall risk under NSG-207, which classifies totals of 45 and above as high risk, following the example bands in the AHRQ toolkit; the threshold is the hospital's policy choice, not a rule from the scale or from a regulator. No clinical judgment override is needed today, because the category and the observed risk agree. The measures below answer the individual items; the policy's high-risk tier is applied without being restated. Interventions by scoring item: Prior fall: hourly rounding on the unit schedule, call light clipped within reach and demonstrated, teach-back completed (below). Multiple diagnoses and medications: pharmacist medication review requested at 20:40 through the admission order set, with attention to the timing of the diuretic and the anticoagulant's bearing on injury risk; attending informed of the August fall during the admission call at 20:55. Walking aid: cane kept at the bedside on her right, one-person standby assist for every walk entered as a nursing intervention, physical therapy evaluation requested for 09/15/2026. Intravenous access: pole kept on the cane side, tubing routed away from her feet, line capped between infusions per policy. Gait: gait belt at the bedside for assisted walks, nonslip socks applied, bed in low position with wheels locked, bathroom path cleared and the night light on. Universal checks completed at 20:20: belongings, water, and glasses within reach; two upper side rails raised at her request for repositioning, per policy, and not a restraint. Bed exit alert offered and declined (below). Teaching, preferences, and the declined alert: L.M. and her daughter (present; health care proxy) were taught the one rule that matters most: call and wait for help before getting up. L.M. repeated it back correctly and agreed. She declined the bed exit alert, which she recalled from an admission last year as noisy and embarrassing when it sounded; the offer, her reason, the education about how the alert is used, and the alternatives were documented, and she accepted the alternatives: the room across from the nurses' station, hourly rounding with a bathroom offer, and her daughter staying until 22:00 each evening. Charge nurse informed of the declined alert at 21:05 per policy. The daughter's presence is recorded as part of the plan and does not change the score. Reassessment plan: Rescore each shift per NSG-207 and immediately with any new sedating, opioid, or blood pressure medication, new confusion, a change in her gait or aid, removal of the line, or any fall. Handoff at 23:00 carried the five scoring items, the measures in place, and the declined alert. Reassessment entry (09/16/2026, 21:30, change in condition): Change-of-condition reassessment by R. Delgado, RN, after two events on the evening shift: the peripheral line was removed at 15:00 when antibiotics were changed to oral, and since a low-dose opioid was started for pleuritic chest pain on the evening of 09/15/2026 L.M. has twice been found standing at the bedside without calling, at 19:40 and 21:10, saying she did not want to bother anyone and thought she could manage. Rescored after walking her to the bathroom with the cane and standby assist: the prior fall, multiple diagnoses, walking aid, and observed gait items still score, gait unchanged from admission; the intravenous access item no longer scores; the self-appraisal item now scores, because her account of what she can do safely no longer matches what was observed. Total 80 of 125, down from 85. The lower total reflects the removed line and not lower risk; the new scoring item is the more dangerous one, and the category remains high risk under NSG-207. Measures revised item by item: line measures discontinued; for the self-appraisal item, the bed exit alert was re-offered and accepted (on and tested at 21:20), bathroom offers scheduled every two hours while awake and at 02:00, her daughter will stay overnight tonight at her own offer, and one-person assist applies to every transfer; attending and pharmacist notified at 21:35 of the timing of the opioid doses and the change in judgment, medication review requested; the other measures continue, as does the therapy plan from the 09/15/2026 evaluation. L.M. re-taught to call first, repeated it back, and agreed to the alert. Reassess at 06:00 and with any further medication change, transfer, or fall. No fall has occurred, so no safety report is filed; if one occurs, the clinical assessment will be charted here and the event reported through the safety reporting system separately. This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No scale items, anchors, or point weights are reproduced. ↑ Back to the template and downloads ### Why this sample works The trigger names the hospital policy and calls its every-shift interval policy, the version and rater are stated, and the rater watched the patient walk before scoring, so the entry reads as an assessment rather than a form. The five scoring items are named in plain clinical facts with the fall date, the aid, the line, and the observed walk, the item that did not score is explained, and the total sits against its maximum, so a reader can reconstruct the number without seeing a weight. The category is attributed to the policy and the threshold to the policy's source, so nobody reads 45 as the scale's rule, and the measures answer the individual items instead of the tier. The declined alert carries the offer, the reason, the education, the alternatives accepted, and the notification, and the daughter's presence is recorded as part of the plan and not as a change in the score. The reassessment explains a total that fell while the risk rose, revises measures item by item, names the medication change and who was told, and states that a fall would generate a new clinical entry and a separate safety report, which is the sequence surveyors and post-fall reviewers look for. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the Morse Fall Scale, and the rules that do exist judge what the score led to. In long-term care, 42 CFR 483.25(d) (LAW, surveyed as F689) requires the facility to ensure that the resident environment remains as free of accident hazards as is possible and that each resident receives adequate supervision and assistance devices to prevent accidents; the interpretive guidance in Appendix PP of the State Operations Manual sends surveyors to whether foreseeable hazards and each resident's own risks were identified and addressed, expects the causes of a fall to be evaluated and the care plan revised afterward, cautions against relying on position-change alarms as the fall prevention plan, and warns that restraints used to prevent falls can cause serious harm. The MDS 3.0 (LAW as to the assessment) collects fall history at admission (J1700), falls since the prior assessment (J1800), and falls by injury level (J1900), and the RAI Manual version 1.20.1 effective October 1, 2025 clarified the definitions of a fall and a major injury; none of it names a screening tool. In hospitals, the Medicare Conditions of Participation require a nursing assessment for every inpatient (42 CFR 482.23) and complete, dated, timed, and authenticated records (42 CFR 482.24) without prescribing a tool or an interval. Under PAYER POLICY, section 5001(c) of the Deficit Reduction Act of 2005 placed Falls and Trauma (fractures, dislocations, intracranial injuries, crushing injuries, burns, and other injuries) among the hospital-acquired condition categories for which Medicare pays no additional amount when the condition was not present on admission, for discharges on or after October 1, 2008, with the present-on-admission indicator reported since October 1, 2007; the determination rests on the provider's injury diagnosis and the documented time of the fall, never on a risk score, and the provision is distinct from the Hospital-Acquired Condition Reduction Program, which scores a patient safety composite and infection measures. CMS added a Hospital Harm: Falls with Injury electronic clinical quality measure (CMS1017) to the Hospital Inpatient Quality Reporting Program for the 2026 reporting year, which counts inpatient falls with moderate or major injury and, again, names no tool. In home health, OASIS collection is required (LAW and PAYER POLICY), and the current OASIS-E2 item set, effective April 1, 2026, contains no fall risk tool item (the former M1910 is gone) but still collects falls since the start or resumption of care at transfer and discharge (J1800 and J1900) and asks in the intervention synopsis (M2401) whether falls prevention interventions were in the plan of care whenever a standardized, validated multi-factor fall risk assessment showed risk, so the agency chooses the tool under its own policy, often one built for community-dwelling patients rather than the inpatient-derived Morse (the OASIS assessment page covers the items). Accreditation is CONVENTION with teeth. The Joint Commission replaced the hospital and critical access hospital National Patient Safety Goals chapter with National Performance Goals on January 1, 2026; under Goal 11, NPG.11.02.01 requires the hospital to assess and manage the patient's risks for falls and to implement fall risk reduction interventions based on the patient population, the setting, and the individual patient's assessed risks. The fall reduction National Patient Safety Goal, NPSG.09.02.01, continues in 2026 for the nursing care center program and the Joint Commission's assisted living and home care programs, with elements of performance that assess risk, implement interventions based on the assessed risk, educate staff, educate the patient or resident and family on individualized strategies, and evaluate the effectiveness of the program; no Joint Commission requirement names the Morse or an interval. AHRQ's Preventing Falls in Hospitals toolkit carries the scale as Tool 3H beside STRATIFY and directs users to plan care from the patient's own risk factors; the VA's National Center for Patient Safety falls toolkit names admission, transfer, change in condition, and after a fall as the triggers and leaves each unit to set its own score parameters; NDNQI and Magnet track falls with injury as a nurse-sensitive indicator and name no tool. In litigation and survey defense (CONVENTION), the questions a reviewer asks are what was known, what was done about it, and what changed after the event, so a chart of repeated totals with a generic precautions phrase is exposed at each step; no published dataset counts how often that pattern is targeted, this page does not claim one, and negligence standards and incident report privilege are state law questions for counsel. Canada and Australia require a validated process and name no tool, and the evidence explains why the items matter more than the total. In Canada, HSO 5060:2025, Preventing Falls and Reducing Injuries from Falls, is a Required Safety Practice of the Health Standards Organization used by Accreditation Canada's programs (CONVENTION with accreditation force), the successor to the Required Organizational Practice that asked organizations to define a fall, screen on admission and after a change in status or a fall, act on identified risk, and review every fall, and it names no instrument; Ontario's long-term care regulation, O. Reg. 246/22 under the Fixing Long-Term Care Act, 2021 (LAW in Ontario long-term care), requires each home to run a falls prevention and management program and to complete a post-fall assessment when a resident falls; and the Registered Nurses' Association of Ontario best practice guideline on preventing falls and reducing injury from falls (fourth edition, 2017; CONVENTION) recommends multifactorial assessment and intervention rather than reliance on a score. In Australia, the NSQHS Comprehensive Care Standard (second edition) requires systems for preventing falls and harm from falls that are consistent with best-practice guidelines, equipment and devices that promote safe mobility, and information for patients, carers, and families (actions 5.24 to 5.26; CONVENTION with accreditation force); the Commission's 2025 falls guidelines for hospitals, residential aged care, and community care advise considering every older person in hospital to be at high risk and building prevention on the person's individual risk factors and a post-fall review rather than on a calculated score; and a third edition of the standards is in development, so no new requirement should be asserted until it is final. The reference points everyone cites bind no one in the three countries but set the direction: NICE NG249 (April 2025), replacing the 2013 guideline, states in recommendation 1.1.1, "Do not use falls risk prediction tools to predict a person's risk of falling," and the World Falls Guidelines (2022) recommend against scored screening tools in hospitals in favor of multifactorial assessment and tailored intervention. On the evidence: Morse's 1989 study reported sensitivity of 78 percent, specificity of 83 percent, a positive predictive value of 10.3 percent, a negative predictive value of 99.3 percent, and inter-rater reliability of .96; at a cutoff of 45, Chow and colleagues (2007) found sensitivity of only 31 percent in 954 Chinese inpatients, Baek and colleagues (2014) found 51 the best cutoff in a Korean electronic record (sensitivity 0.72, specificity 0.91, area under the curve 0.77), Mao and colleagues (2024) found 40 in obstetric and gynecologic wards, and Schwendimann and colleagues (2006) found 55 in a Swiss hospital; Aranda-Gallardo and colleagues (2013) pooled 14 studies, ranked STRATIFY above the Morse on diagnostic odds ratio (7.64 for STRATIFY), and concluded that the instruments behave very differently across populations and settings; Oppegaard and colleagues (2026) found the mean total above 45 in both fallers and non-fallers among 5,004 VA inpatients, Zhang and colleagues (2026) found no significant difference in totals between matched fallers and non-fallers in 4,887 acute care patients, and Jiang and colleagues (2025) found an area under the curve of 0.825 in 206,846 Chinese inpatients, so discrimination runs from poor to good depending on population, workflow, and threshold; Morris and colleagues (2022) found education the only intervention with a significant reduction in hospital falls (rate ratio 0.70) and found risk assessment tools, alarms, and sensors on their own not associated with fewer falls; and Fall TIPS, which links each scoring item to a matched intervention, was associated with 15 percent fewer falls and 34 percent fewer injurious falls in 37,231 patients (Dykes and colleagues, 2020). On versions: Morse's 2006 commentary, The Modified Morse Fall Scale, addressed the spread of locally altered forms, and the medication items, automatic high-risk rules after an in-facility fall, and 51-point bands found in local variants make those forms different instruments whose validity is not inherited from the published scale, so the note names the version or build; the adult scale is not validated in children, where the Humpty Dumpty Falls Scale and similar pediatric tools apply; and a patient who does not walk can score low on the mobility items while remaining at risk during transfers and toileting, which is why the record carries a clinical judgment override beside the number. The Morse Fall Scale is a copyrighted instrument. Janice M. Morse developed it with Robert M. Morse and Suzanne J. Tylko and published it in the Canadian Journal on Aging in 1989, and her book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008) sets out its use, so the item wording, anchors, and weights sit under the author's and the publishers' rights rather than in the public domain. The author's published position, carried on the Fall TIPS FAQ maintained by the Brigham and Women's Hospital team whose toolkit is built on the scale, is that use of the scale is free of charge but permission from the author is required, with requests directed to Dr. Morse at the University of Utah; Morse's 2006 commentary in the International Journal of Nursing Practice objected to modified versions circulating under the scale's name, so a facility that changes items, weights, or response logic has created a local instrument and should say so. No public statement grants blanket permission for websites, calculators, custom forms, or electronic builds, and the many calculators that reproduce the anchors show market practice, not a license; by contrast, the Johns Hopkins Fall Risk Assessment Tool is licensed for a fee by the Johns Hopkins Health System Corporation and the Hendrich II Fall Risk Model is licensed and patented through AHI of Indiana, so a free-with-permission scale, a licensed one, and a patented one sit side by side in the same market. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Morse Fall Scale documentation errors reviewers flag The numbers behind these errors are specific. Morse's development study classified 80.5 percent of patients correctly, with sensitivity of 78 percent, specificity of 83 percent, and a positive predictive value of only 10.3 percent , so about nine of ten patients the scale calls high risk never fall (Morse, Morse, and Tylko, 1989). Among 5,004 VA inpatients , 3.5 percent fell, the mean total of fallers and non-fallers alike sat above the conventional 45, and the total did not usefully separate the two groups while prior falls, an assistive device, an unsteady gait, and impaired self-appraisal still did (Oppegaard and colleagues, 2026); among 4,887 acute care patients , totals did not differ significantly between matched fallers and non-fallers (Zhang and colleagues, 2026), while a study of 206,846 inpatients reported an area under the curve of 0.825 (Jiang and colleagues, 2025). The best cutoff was 51 in a Korean electronic record (Baek and colleagues, 2014), 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and at 45 the scale identified only 31 percent of fallers in 954 Chinese inpatients (Chow and colleagues, 2007). A 2022 systematic review found risk assessment tools, alarms, and sensors on their own not associated with fewer falls in hospitals and education the only intervention with a significant effect (Morris and colleagues), while the Fall TIPS toolkit, which ties each scoring item to a matched measure, was associated with 15 percent fewer falls and 34 percent fewer injurious falls in 37,231 patients (Dykes and colleagues, 2020). No published audit counts how often a chart carries a Morse total with no linked intervention; the F689 survey pathway, the accreditation requirements, and the hospital-acquired condition rules described under compliance considerations are what turn that omission into a deficiency or a payment and defensibility problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Morse Fall Scale documentation reviews: A total with no items named. "Morse 85" tells the next nurse, the surveyor, or the reviewer nothing about whether the risk is a line, a cane, an unsteady walk, or a patient who overestimates herself, and it cannot be checked against the measures. Name the items that scored in plain clinical facts, say which did not score and why, and write the total against its maximum; the instrument's wording and weights stay on the licensed form. A cutoff written as the scale's rule. "High risk per Morse" or "45 is the official cutoff" states a convention as an instrument rule. The 25 and 45 edges are an example in AHRQ's toolkit and a common VA practice, Morse's guidance is to calibrate the cutoff to the unit, the VA's own policy template leaves the number blank for each unit, and published optima run from 40 to 55. Write the facility category with the policy that defines it, and record any clinical judgment override. A score every shift and no intervention behind it. Twelve totals over six days, each followed by "fall precautions maintained," is the pattern post-fall reviewers and surveyors take apart, because it shows that the assessment never changed care. Every-shift scoring is a policy interval, not a rule, and each entry needs at least one sentence linking a scoring item to a measure actually delivered, or a statement that the items and the measures are unchanged since a named entry. The tier picks the bundle. Alarm, sitter, colored bracelet, and low bed applied because the total crossed a line, with no item behind any of them, or a bed alarm and a chair alarm standing in for a plan. The item chooses the measure: a line calls for pole and tubing management, an unsteady gait for assisted walking and footwear, an overconfident patient for supervision and a toileting schedule. CMS long-term care guidance cautions against alarms as the sole intervention, and no authority reads a high total as a reason for restraint. Scored from the chart, not the patient. A gait item rated from yesterday, a self-appraisal item rated from the admitting diagnosis, or a total auto-populated and signed as a fresh assessment. The gait and self-appraisal items are observations of the patient, which AHRQ's toolkit states, and a copied total hides real change while looking like continuity. Watch the patient move, ask what they think they can do, and rate what you saw; if nothing has changed, say that you re-observed it. A changed total unexplained, or a fall that rescores nothing. A total that dropped because a line came out, charted as improvement while new confusion went unrecorded; a post-fall entry that copies the pre-fall score although the fall now scores; or the incident report narrative pasted into the progress note. Explain every item that moved and why, treat a fall as a new assessment with a revised plan, and keep the safety report and the chart as the two different documents they are. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on nursing fall risk assessments and safety documentation. Give it the facts (the trigger and version, which items scored and what each rests on, the total and any prior total, the facility category and policy, the measure in place for every scoring item, the teaching and any refusal with the alternatives, the notifications, and the next reassessment) and it drafts the documentation block: trigger and version, items in plain words with the total, category attributed to policy, an intervention for each scoring item, teaching and response, and the reassessment and communication lines, ready for your review. Cross-check a finished entry for the gaps reviewers flag: a total with no items named, a cutoff written as the scale's rule, a score with no linked intervention, a bundle with no item behind it, items rated from the chart, a changed total left unexplained, or a post-fall entry with a copied score. Turn the entry into the next document: the fall risk line for the handoff or the transfer of care summary, the medication review request for the pharmacist, the care plan problem and goal, or the clinical post-fall entry that stays separate from the safety report, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What do the six Morse Fall Scale items assess, and how is the total scored? The scale scores six weighted items, some with graded responses, for a total of 0 to 125, and a higher total means higher risk. Named as topics in original words, the items ask whether the patient has fallen recently or during this admission; whether more than one active medical diagnosis is present; what the patient relies on to walk, because dependence on an aid or on whatever is within reach marks unsteadiness; whether an intravenous line or access device is in place, because it tethers and encumbers; how the patient actually walks when you watch; and whether the patient's judgment of what they can safely do matches what you observe, the item that flowsheets usually label mental status. The item wording, the response anchors, and the point weights are the copyrighted text of the instrument, so this page does not reproduce or paraphrase them; score from your facility's licensed form or approved electronic build and write the basis for each scoring item in ordinary clinical words. The total is a screening number: the items that scored, not the sum, tell the next nurse what to do. Which Morse Fall Scale score means high risk: 25, 45, or 51? Whichever your facility's policy says, and the note should cite that policy. The bands most facilities use, with 25 and 45 as the edges, appear as an example in AHRQ's Preventing Falls in Hospitals toolkit and reflect common VA practice, but Morse's own guidance is to calibrate the high-risk cutoff to the unit's patients, the VA's falls toolkit policy template leaves the number blank for each unit to determine, and no regulator, accreditor, or payer in the United States, Canada, or Australia has adopted 45 or any other value. Validation studies explain why: the best cutoff was 51 in a Korean hospital's electronic record (Baek and colleagues, 2014), 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and in a 2026 VA study the average total of the patients who did not fall was already above 45. The 51-point high-risk band traces to local and adapted versions, not to Morse's published scale. Document the total, the facility category, and the policy that defines it, record any clinical judgment override, and let the items that scored, not the tier, choose the measures. How often does the Morse have to be repeated? Does CMS or the Joint Commission require it every shift? No regulator sets a Morse interval, and none requires the Morse at all. CMS's long-term care rule requires supervision and assistance devices to prevent accidents and a care plan revised after a fall; the MDS collects fall history and falls with injury; the Joint Commission's 2026 hospital National Performance Goals require assessing and managing fall risk with interventions based on the population, the setting, and the individual patient's assessed risk, and its nursing care center, assisted living, and home care goal NPSG.09.02.01 requires assessment, interventions based on assessed risk, education, and evaluation; the OASIS-E2 item set asks whether falls prevention interventions followed a standardized multi-factor assessment; none names a tool or an interval. The triggers that the VA toolkit, AHRQ, Canadian accreditation practice, and the Australian guideline agree on are admission, transfer, a change in condition (a new sedating, opioid, or blood pressure medication, new confusion, a new or removed line, a change in mobility), and any fall. Every shift is a workflow many hospitals choose; chart it as your policy, name the trigger for each repeat, and make sure each entry links a scoring item to a measure, because a column of totals with no linked action is the defect reviewers find. Does a saline lock count as intravenous access, and how do I score a patient who cannot walk? Follow the definition in your facility's approved build of the scale and chart what is actually present. The access item exists because a line or device tethers and encumbers a patient who tries to get up, and facility builds and training materials commonly treat a capped or locked peripheral device as access in place; do not invent a local exception because nothing is infusing, and if your build reads it differently, say so in the note. For a patient on bed rest or in a wheelchair, the mobility items rate how the patient actually moves, so a patient who does not walk can score low on them while transfers, toileting attempts, and delirium make the real risk high; the VA's falls toolkit notes that wheelchair users score low on those items even though transfers put them at risk. Score the items as observed, then document a clinical judgment override with its reason and the supervision and transfer measures it triggers, so the record shows the risk that the number cannot. Why did the total go down after a fall, or after a new risk appeared, and what do I write? Because two items can move at once and cancel. A fall during the admission makes the prior-fall item score, so all else equal a post-fall total should rise; if it fell, another item must have resolved at the same time, most often a line removed or an aid no longer needed, and the note must say which item changed and why rather than leave a lower number to imply lower risk. The same logic runs the other way: a line removed on the day a new opioid makes a patient overestimate what she can do produces a lower total and a more dangerous patient, as the sample on this page shows. Write the changed items in clinical terms, keep the category attributed to policy, de-escalate only the measures tied to the item that resolved, add the measures the new item calls for, and record the trigger. After a fall, never copy the pre-fall score: perform a new assessment, in which the fall now scores, and attach the revised plan; an unexplained lower post-fall total should send you back to both entries to reconcile them. What do I chart when a patient refuses fall precautions, or when family stays at the bedside? Chart the offer, the patient's stated reason where appropriate and any capacity concern, the education given about the specific risk, the alternatives offered and which were accepted, who was notified under policy, and the plan as revised so it still mitigates the risk as far as it can; do not write that the patient refused fall precautions and stop there. In long-term care, CMS survey guidance expects the resident's choice to be respected while staff work with the resident to understand the choice and build a plan that honors it and mitigates the risk, and a signed waiver does not end the facility's duty; in hospitals the same documentation protects the record in a post-fall review. Family presence is an intervention, not a scored item: no Morse item represents a sitter or a relative, so a daughter staying overnight changes the supervision plan and never the number, and the note should say who is staying, for how long, and what they have agreed to do (call for staff, not lift). The sample on this page shows a declined bed exit alert documented with its alternatives and then accepted two days later when the risk changed. After a fall, what goes in the chart and what goes in the incident report? Two documents, two purposes. The chart carries the clinical event: the time and how the patient was found, the injury check and vital signs, neurological checks where the head may have struck, the provider notification with time and response, imaging or treatment ordered, a new Morse in which the fall now scores, the contributing factors you can observe, the revised measures, the patient's and family's understanding, and the next reassessment. The safety report carries the organization's analysis: the event reconstruction, system factors, and the post-fall huddle or root-cause review, which AHRQ's toolkit treats as a separate activity from the clinical reassessment. Keep the incident narrative and any speculation about blame out of the progress note, and treat whether the note cites the report number as a local risk-management decision, because incident report privilege and negligence standards are state law. For coding, the injury diagnosis and its present-on-admission status come from the provider's documentation and drive the hospital-acquired condition payment provision for Falls and Trauma; your note supplies the time and circumstances. The incident report page covers the safety document itself. Can we put the Morse Fall Scale in our EHR, on a paper form, or on our website? Ask first. The scale is copyrighted: it was published by Morse, Morse, and Tylko in the Canadian Journal on Aging in 1989 and is set out in Morse's book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008), and the Fall TIPS FAQ, maintained by the Brigham and Women's Hospital team whose toolkit is built on the scale, states that use of the Morse Fall Scale is free of charge but permission from the author is required, with requests directed to Dr. Morse at the University of Utah. No public statement grants blanket permission for websites, calculators, custom forms, or electronic builds, and the many pages that reproduce the anchors show market practice rather than a license. Modifying the scale is a second question: Morse's 2006 commentary addressed altered versions circulating under her scale's name, and a form with a medication item added or a rule that any in-facility fall makes the patient high risk is a local instrument whose validity is not inherited, so name it as a variant. For comparison, the Johns Hopkins Fall Risk Assessment Tool may not be used without a paid license from the Johns Hopkins Health System Corporation, and the Hendrich II Fall Risk Model is licensed and patented through AHI of Indiana. What needs no permission is the concept: fields for which items scored and the total, your own documentation prompts in original words, and the structure on this page, which reproduces none of the instrument. Can BastionGPT draft my Morse Fall Scale documentation? Yes. Give it the facts (the trigger and version, which items scored and what each rests on, the total and any prior total, the facility category and policy, the measure in place for every scoring item, the teaching and any refusal with the alternatives, the notifications, and the next reassessment) and it drafts the full entry: trigger and version, items in plain words with the total, category attributed to policy, an intervention for each scoring item, teaching and response, and the reassessment and communication lines, ready for your review. It can also check a finished entry for a total with no items named, a cutoff written as the scale's rule, a score with no linked intervention, a bundle with no item behind it, items rated from the chart, a changed total left unexplained, or a post-fall entry with a copied score. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Morse JM, Morse RM, Tylko SJ, 1989, Canadian Journal on Aging 8(4):366-377, Development of a scale to identify the fall-prone patient (six items; 80.5 percent correctly classified; sensitivity 78 percent, specificity 83 percent, positive predictive value 10.3 percent, negative predictive value 99.3 percent; inter-rater reliability .96); Morse JM, 1997, Preventing Patient Falls, SAGE, and 2008, Preventing Patient Falls, second edition (Springer Publishing; development of the scale and determining level of risk); Morse JM, 2006, International Journal of Nursing Practice 12(3):174-175, The modified Morse Fall Scale ; Fall TIPS, Brigham and Women's Hospital, FAQs (use of the Morse Fall Scale free of charge, permission from the author required) and resources . Validation and cutoffs: Schwendimann R and colleagues, 2006, Age and Ageing 35(3):311-313, evaluation of the Morse Fall Scale in hospitalised patients ; Chow SK and colleagues, 2007, International Journal of Nursing Studies, applicability in Chinese hospital populations (954 patients; sensitivity 31 percent at 45); Baek S and colleagues, 2014, Journal of Clinical Nursing, validity of the Morse Fall Scale implemented in an electronic medical record system (cutoff 51); Mao B and colleagues, 2024, PLoS One, re-evaluating the Morse Fall Scale in obstetrics and gynecology wards (cutoff 40); Kim YJ and colleagues, 2022, Journal of Clinical Nursing, validity of the Morse Fall Scale and the Johns Hopkins Fall Risk Assessment Tool in an acute care setting . Reviews and recent cohorts: Oliver D and colleagues, 1997, BMJ, development and evaluation of STRATIFY ; Oliver D and colleagues, 2004, Age and Ageing, risk factors and risk assessment tools for falls in hospital inpatients ; Aranda-Gallardo M and colleagues, 2013, BMC Health Services Research, instruments for assessing the risk of falls in acute hospitalized patients (14 studies; STRATIFY diagnostic odds ratio 7.64); Matarese M and colleagues, 2015, Journal of Advanced Nursing, systematic review of fall risk screening tools for older patients in acute hospitals ; Oppegaard KR and colleagues, 2026, Journal of Hospital Medicine, Beyond the score: reevaluating the Morse Fall Scale among hospitalized veterans (5,004 inpatients); Zhang Y and colleagues, 2026, American Journal of Nursing, falls in acute care patients: the predictive value of the Morse Fall Scale (4,887 patients); Jiang H and colleagues, 2025, International Nursing Review, a real-world study on the Morse Fall Scale and clinical judgment (206,846 inpatients). Prevention evidence and guidelines: Dykes PC and colleagues, 2020, JAMA Network Open 3(11):e2025889, evaluation of a patient-centered fall-prevention tool kit (37,231 patients; falls rate ratio 0.85; injurious falls 0.66); Morris ME and colleagues, 2022, Age and Ageing, interventions to reduce falls in hospitals: a systematic review and meta-analysis ; Montero-Odasso M and colleagues, 2022, Age and Ageing, World guidelines for falls prevention and management for older adults ; National Institute for Health and Care Excellence, 2025, NG249, falls: assessment and prevention in older people and in people 50 and over at higher risk (recommendation 1.1.1); Hendrich AL, Bender PS, Nyhuis A, 2003, Applied Nursing Research 16(1):9-21, validation of the Hendrich II Fall Risk Model ; AHI of Indiana, Hendrich II Fall Risk Model (license and patents); Johns Hopkins Medicine, Johns Hopkins Fall Risk Assessment Tool (license required). United States, federal: Legal Information Institute, 42 CFR 483.25 (paragraph (d), accidents; surveyed as F689); CMS, MDS 3.0 RAI User's Manual (version 1.20.1, effective October 1, 2025; Section J revisions); CMS, hospital-acquired conditions (Falls and Trauma; Deficit Reduction Act section 5001(c); discharges on or after October 1, 2008); CMS, Hospital-Acquired Condition Reduction Program ; eCQI Resource Center, Hospital Harm: Falls with Injury (CMS1017), 2026 ; CMS, OASIS data sets and OASIS-E2 All Items instrument, effective April 1, 2026 (J1800, J1900, M2401; no fall risk tool item). Toolkits: AHRQ, Preventing Falls in Hospitals: a toolkit for improving quality of care, tools and resources (Tool 3H Morse Fall Scale; Tool 3G STRATIFY; post-fall tools 3N and 3O); VA National Center for Patient Safety, falls toolkit: fall prevention and management policy (unit-determined score parameters; triggers; calibration per Morse's book). Joint Commission: National Performance Goals and National Performance Goals effective January 2026 for the Hospital Program (NPG.11.02.01, assess and manage the patient's risks for falls); National Patient Safety Goals effective January 2026 for the Nursing Care Center Program (NPSG.09.02.01); National Patient Safety Goals . Canada and Australia: Health Standards Organization, HSO 5060:2025, Preventing Falls and Reducing Injuries from Falls (Required Safety Practice); Government of Ontario, O. Reg. 246/22 under the Fixing Long-Term Care Act, 2021 ; Registered Nurses' Association of Ontario, Preventing Falls and Reducing Injury from Falls, fourth edition, 2017 ; Australian Commission on Safety and Quality in Health Care, NSQHS Comprehensive Care Standard, action 5.24 (with actions 5.25 and 5.26), Falls Guidelines for Hospitals, 2025 , and preventing harm from falls with the updated falls guidelines . ## Related templates Braden Scale Documentation Incident & Adverse Event Report Transfer-of-Care Summary OASIS Assessment Documentation Discharge Summary Medication Management Follow-Up Note Care Coordination Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/bishop-score All note templates # Bishop Score Documentation: Components, Version & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Bishop score is a five-component summary of a digital cervical examination (dilation, effacement, station, consistency, and position) proposed by Edward Bishop in 1964 to judge how ready the cervix is for induction. Obstetricians, midwives, and labor and delivery nurses record it before and during cervical ripening and induction of labor. This page covers how to document a Bishop score with its version, examiner, indication, and plan, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The clinician who performs the digital cervical examination: an obstetrician or maternal-fetal medicine physician, an obstetric resident, a family physician who attends births, a nurse-midwife or midwife, or a labor and delivery nurse within unit scope; no license, training requirement, or publisher qualification level applies, and the examiner is named because the findings are examiner-dependent Audience Labor and delivery nurses and the oncoming team choosing the next step, obstetric and midwifery colleagues, anesthesia and neonatal teams reading the induction plan, maternity-unit quality and audit staff, coders reading the indication, and expert reviewers in induction-related claims Typical length 4 to 10 chart lines (indication and gestational age, examination time and examiner, five findings with units, version and total, membranes and method-specific facts, reading and plan, reassessment) · 1 to 3 minutes to record after the examination Format family Examiner-rated cervical readiness score (five findings from one digital examination converted to points and summed; classic total 0 to 13, with length-based modified versions and a three-component simplified version that have different ranges) When it's used Before cervical ripening or induction of labor at any gestation at which induction is planned, on admission for induction, after each ripening interval or device removal, and whenever a change in cervical findings would change the method; carried into the induction order set, the labor record, and perinatal quality audits Standards context Edward Bishop (1964), a public domain method with modified (Burnett 1966; Calder 1974, length-based) and simplified (Laughon 2011) versions; ACOG Clinical Practice Guideline No. 9 (2025), SOGC 432 (2023), NICE NG207 (2021), and RANZCOG and state guidelines set differing favorable conventions; required by no law; described here for documentation, no scoring table reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Bishop score? The Bishop score is a summary of one digital cervical examination. Edward H. Bishop, an obstetrician at Pennsylvania Hospital in Philadelphia, published Pelvic Scoring for Elective Induction in Obstetrics and Gynecology in August 1964 (volume 24, pages 266 to 268), building on his 1955 paper on elective induction; his population was multiparous women at term being considered for elective induction, and he regarded a total above 8 as marking a patient suitable for it. Five findings from the examination are each converted to points and summed: dilation (how open the cervix is, in centimeters), effacement (how much the cervix has thinned, as a percentage), station (where the presenting part sits relative to the ischial spines, on Bishop's scale of thirds from minus 3 to plus 3), consistency (firm through soft), and position (posterior through anterior). In the classic architecture the first three carry up to 3 points each and the last two up to 2 each, for a total of 0 to 13; the breakpoints belong to the published table and are described here only conceptually. The word modified hides a family rather than one instrument. Burnett published an early alternative preinduction score in 1966; the Calder lineage of the 1970s, embedded in UK practice and NICE guidance, replaces percentage effacement with cervical length in centimeters and changes the maximum; Queensland and Western Australian guidelines use a length-based Modified Bishop Score; a widely circulated variant adds or subtracts points for parity, pre-eclampsia, post-dates pregnancy, and prelabor rupture of membranes, which no ACOG, SOGC, NICE, or RANZCOG instrument does; and the simplified Bishop score of Laughon and colleagues (2011), derived from 5,610 nulliparous term inductions in the Consortium on Safe Labor, keeps only dilation, effacement, and station for a range of 0 to 9. Two unit traps follow. A cervix charted as 2 cm long is speaking the length-based dialect, not the percentage one, and a station read on the centimeter scale of minus 5 to plus 5 that ACOG later adopted is not a Bishop third. The load-bearing fact for documentation is that the total means nothing without its version and its convention. ACOG's Clinical Practice Guideline No. 9, Cervical Ripening in Pregnancy (July 2025), which replaced Practice Bulletin 107 of 2009, states that there is no universally accepted definition of an unfavorable cervix, notes that studies have used Bishop thresholds ranging from 3 to 8 to define who benefits from ripening before oxytocin, and calls the score the most frequently used method of judging cervical readiness; the older Practice Bulletin language (6 or less unfavorable, above 8 comparable to spontaneous labor) is history, not current ACOG wording. SOGC Guideline No. 432b (2023) directs cervical ripening when the modified Bishop score is below 7 and No. 432c moves to amniotomy and oxytocin at 7 or more; NICE NG207 (2021) steers method choice at 6 or less against more than 6 and describes 8 or more as a cervix generally ready to dilate; Australian state guidelines set their own conventions on a length-based modified score; and the StatPearls entry that many clinicians read first says plainly that the precise criteria for a favorable or unfavorable cervix have not been established. So the defensible entry records the five findings in the units actually assessed, the version that converted them to points, the total with its maximum, the examiner and time, the indication, and the plan, and attributes any favorable or unfavorable label to the guideline or unit pathway that sets it. Two limits belong in the same entry. The score does not decide whether to induce: the indication does, and no guideline withholds an indicated induction because the cervix is unfavorable. And it was not built or validated as a cesarean-risk predictor: Bishop designed it to select multiparous women for elective induction, Kolkman and colleagues (2013) pooled 40 studies of 13,757 women and found it a poor predictor of the outcome of induced labor at term, and Laughon and colleagues found that a low score identifies who will not deliver vaginally only about 30 percent of the time. The score also has neighbors it must not absorb: the general cervical examination note, which records the same findings without a total; transvaginal ultrasound cervical length, a separate modality with its own evidence; the induction indication and consent record, which owns the decision to induce; the labor progress record once labor is established; and the Apgar score , its obstetric sibling on this site, which describes the newborn after birth rather than the cervix before induction. ## Who uses Bishop score documentation and when Obstetricians, maternal-fetal medicine physicians, obstetric residents, family physicians who attend births, and nurse-midwives and midwives perform the pre-induction cervical examination and chart the score in the United States, Canada, and Australia; labor and delivery nurses perform and chart serial cervical examinations within unit scope and are often the examiner of record after a ripening interval. Those entries are read by the charge nurse and the oncoming team choosing the next step, by anesthesia and neonatal teams reading the induction plan, by maternity-unit quality staff auditing whether a documented score preceded induction (Queensland proposes exactly that audit measure, and the WA Country Health Service requires the current Bishop score, written consent, the indication, and the intended method on every induction request), by coders who take the indication rather than the score, and by expert reviewers in induction-related claims, who read the cervical findings beside the indication, the consent, and the plan. The Bishop page owns one job: the five findings, the version, the total, the examiner, the indication, and the plan, with the reassessment. Neighbors take over from there. A general cervical examination note carries the same findings when no induction decision is pending; transvaginal cervical length belongs to preterm-birth screening and to research comparisons, not to the Bishop field; the induction consent record owns whether to induce; the labor record and partograph own progress once labor is established; a maternal-fetal medicine opinion on the indication or the method belongs in the consultation report ; a move from a birth center or a low-risk unit to a hospital for ripening travels in the transfer of care summary ; the induction course is summarized in the discharge summary ; the visit at which induction was scheduled closes with an after-visit summary ; and a uterine tachysystole event or a retained device that the organization treats as a safety event goes in the incident report . ## How to document a Bishop score in the induction record No law, payer, or professional body prescribes a Bishop score note format; what exists is Bishop's 1964 method, a family of modified and simplified versions, guidelines whose favorable conventions differ, and unit pathways and state policies that expect a documented score before induction. What survives review is an entry that states the indication and gestational age first, names the examiner and time, records the five findings in the units actually assessed, names the version and computes the total against its maximum, carries the membrane status and the method-specific facts the score cannot encode, attributes any favorable or unfavorable reading to its source and ties the plan to the whole clinical picture, and repeats the examination on the method's schedule with both sets of findings kept. Each element below carries the pitfall that most often undermines it. Indication, gestational age, and the decision to induce. Open with why induction is happening and at what gestation, with the dating method, parity, and the date the decision was made: late-term pregnancy at 41 weeks 3 days by a 9-week ultrasound, gestational hypertension at 37 weeks, elective induction at 39 weeks after shared decision-making, or a medically indicated late-preterm delivery under ACOG Committee Opinion 831. The indication decides whether induction is appropriate; the cervical findings decide only how to begin. Every major guideline (ACOG's 2025 ripening guideline, SOGC 432, RANZCOG C-Obs 22, NICE NG207) uses the score to choose or offer ripening after the indication is established, and none makes an unfavorable cervix a reason to withhold an indicated induction. If the indication and the consent discussion live in a separate induction consent note, reference it by date rather than restating it. Pitfall: "Bishop 3, induction deferred" or "Bishop 9, induction scheduled" with no indication in sight, so the score reads as the reason to induce or to wait. Examination time, examiner, and consent to examine. Record the date and time of the digital examination, the examiner's name and role, and that the patient agreed to the examination. The findings are examiner-dependent: Faltin-Traub and colleagues (2004) found two examiners agreed exactly on the total in 28 percent of 156 women at term and within one point in 66 percent, with weighted kappa from 0.35 to 0.69 across pairs, and Dupuis and colleagues (2005) found station errors in 50 to 88 percent of residents' and 36 to 80 percent of attendings' assessments on a birth simulator. That is why the name belongs beside the number. Serial examinations by different clinicians are separate entries with their own times; none is averaged, corrected, or overwritten, and when two examiners disagree across a threshold that would change management, a repeat or senior examination is documented as a third entry rather than as a resolution of the first two. Pitfall: An unattributed score in a flowsheet, or the attending's later number silently replacing the midwife's admission examination. The five findings, in the units actually assessed. Chart the components before the total, each in the unit it was assessed in: dilation in centimeters; effacement as a percentage or, if your unit assesses cervical length, in centimeters, saying which; station on Bishop's scale of thirds (minus 3 to plus 3) or on the centimeter scale of minus 5 to plus 5, saying which; consistency in plain words; position in plain words. Never convert silently. A cervix 2 cm long is a length-based finding that a length-based modified score can use, not a percentage that the classic score can use; a station of plus 1 on the centimeter scale is not a Bishop third, and Dupuis showed station is the component most often misjudged. Digital cervical length is also not transvaginal ultrasound length, which is a separate measurement with better reproducibility (Garcia-Simon and colleagues, 2015) and its own evidence base, so label the source of any length. Pitfall: "Cervix 2 cm long" scored as if it were a percentage of effacement, or a station read on the centimeter scale entered as a Bishop third, so the total is wrong before it is written. Version named and total computed. Write which version converted the findings to points and the total against its maximum: classic Bishop, five components, 0 to 13; a length-based modified score, named for the guideline or author that defines it (Calder; the Queensland Modified Bishop Score; a unit's own build), with its own maximum; the simplified Bishop score of Laughon and colleagues, three components, 0 to 9; or a variant with parity or complication modifiers, named as such because no national guideline instrument uses them. The version is what makes the number portable: a 5 under one architecture and a 5 under another are different facts, and an EHR field labeled Bishop score that accepts both without a version tag is ambiguous by design. Where the flowsheet stores the total, put the version in the comment or the note, and trend only within one version. Pitfall: "Bishop 5" with no version, or a length-based modified total charted against a classic admission baseline as if the cervix had changed. Membranes, fetal status, and method-specific facts. Record what the score cannot encode and the method depends on: membrane status, presentation and estimated fetal weight, the fetal heart rate tracing category before any agent, prior cesarean or other uterine surgery, placental location, cerclage status and its removal, relevant allergies, and any contraindication to a prostaglandin, a balloon, or amniotomy. ACOG's 2025 guideline marks a uterine scar as a contraindication in its ripening-methods table, RANZCOG treats previous uterine surgery as a separate method-selection factor, and NICE names a previous cesarean birth among the circumstances in which a mechanical method is considered; none of that is visible in the total, which is why two patients with identical scores can have different acceptable options. Australian and UK guidance expects a normal tracing before prostaglandin or oxytocin, so the tracing category belongs in the entry. Pitfall: A prior cesarean recorded on the admission history but absent from the ripening entry, so the choice of a balloon over a prostaglandin cannot be reconstructed from the note. Reading under a named convention, and the plan. Write favorable or unfavorable with the source of the threshold: "unfavorable under the unit's induction pathway (classic total of 6 or less)," "below 7, cervical ripening indicated under SOGC 432b," or "6 or less, pharmacologic or mechanical ripening options offered under NICE NG207." Do not attribute a threshold to ACOG, whose 2025 guideline declines to set one. Then state the plan as a decision about the whole picture: the method chosen (a balloon, dinoprostone, misoprostol, a combination, amniotomy and oxytocin, or expectant management), the reasons beyond the score (parity, urgency of the indication, uterine scar, membrane status, the patient's preferences, the setting), the counseling on methods, benefits, and risks that ACOG requires once ripening is deemed necessary, and the consent. Queensland's pathway, which allows amniotomy in selected patients with a low score when there is a clinical reason to avoid ripening, is the reminder that the number is an input, not the order. Pitfall: "Bishop 4, misoprostol" as if the number chose the drug, or "unfavorable (below 6)" asserted as a universal rule with no guideline or pathway named. Reassessment on the method's schedule, and the record. Repeat the examination when the method's protocol says so (after balloon expulsion or at the device's dwell limit, at the interval set for a dinoprostone insert or a misoprostol dose, and whenever a change in findings would change management), and document it as a new entry with its own time, examiner, five findings, version, and total. Describe the change with both sets of findings visible (1 of 13 at 19:40 by the midwife; 7 of 13 at 04:05 by the attending) so that improvement is not reduced to an unexplained jump, and say what the new findings led to. No universal recheck interval exists: NICE asks for reassessment as induction progresses, Queensland's balloon pathway sets reassessment after device removal, and the WA Country Health Service asks the clinician to notify the unit if a pre-induction score changes enough that ripening is no longer needed. Close with where the entry lives (the induction section of the labor record, with the total in the flowsheet field and the version in its comment) and that the patient was told what the score describes. Pitfall: "Bishop improved from 2 to 8" with the baseline findings gone, or a flowsheet field that mixes a classic admission total with a modified reassessment total. ### Blank template (copy and adapt) BISHOP SCORE DOCUMENTATION BLOCK (findings, version, indication, plan) Date: [ ] Setting: [hospital L&D / birth center / outpatient ripening] Examiner (name, role): [ ] Examination time: [ ] Patient agreed to exam: [yes] Indication for induction: [ ] Gestational age: [ ] weeks [ ] days by [dating] Parity: [ ] Decision to induce and consent: [date; induction consent note referenced] Findings, in the units assessed: Dilation: [ ] cm Effacement: [ ] percent OR cervical length: [ ] cm (state which) Station: [ ] on [thirds, -3 to +3 / centimeters, -5 to +5] (state which) Consistency: [firm / medium / soft] Position: [posterior / mid / anterior] Version: [classic Bishop, 0 to 13 / length-based modified, named / simplified Laughon, 0 to 9 / named local variant] Total: [ ] of [maximum] Membranes: [intact / ruptured, time] Presentation: [ ] Estimated fetal weight: [ ] Fetal heart rate tracing before any agent: [category / normal] Prior cesarean or uterine surgery: [none / details] Placental site: [ ] Cerclage: [none / removed, date] Allergies and method contraindications: [ ] Reading: [favorable / unfavorable] under [guideline or unit pathway that sets the threshold; ACOG 2025 sets none] Plan: [balloon / dinoprostone / misoprostol / combination / amniotomy and oxytocin / expectant] Reasons beyond the score: [parity, urgency, scar, membranes, preference, setting] Counseling and consent for the method: [date, time] Reassessment trigger: [balloon expulsion / dwell limit / dose interval / change in findings] Next examination by: [ ] Prior examination(s): [time, examiner, findings, version, total; both entries kept] Record: [induction section of labor record; flowsheet total with version in comment; patient told what the score describes] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces a scoring table or a guideline table. ## Sample Bishop score documentation (fictional) Scenario: a US community hospital labor and delivery unit, where a nulliparous patient at 41 weeks 3 days is admitted for late-term induction; the nurse-midwife examines and charts the five findings, names the version, reads the total under the unit's convention, records the method-specific facts, and places a transcervical balloon with consent, and the attending re-examines after expulsion and documents the change with both examinations kept. All details are fictional. Patient: T.W., 29, G1P0, 41 weeks 3 days · Setting: Community hospital labor and delivery unit; admission for late-term induction · Clinician: M. Okafor, CNM (examining clinician); R. Castellano, MD (attending) · Note date: 09/16/2026 Indication and decision to induce: Late-term pregnancy at 41 weeks 3 days by a 9-week ultrasound; first pregnancy, no prior uterine surgery. Induction was recommended at the 41-week office visit on 09/14/2026 under the practice's late-term pathway, which follows ACOG's guidance on late-term and postterm pregnancy; T.W. chose induction over continued surveillance, and the discussion of alternatives, benefits, and risks is in the induction consent note dated 09/14/2026. Membrane sweep declined. The indication, not the cervical findings, decided that induction would proceed; the findings below decided only how it would begin. Admitted 09/16/2026 at 19:00 for cervical ripening. Examination: Digital cervical examination at 19:40 on 09/16/2026 by M. Okafor, CNM, with T.W.'s verbal agreement, after a 20-minute admission tracing. Findings in the units assessed: dilation 1 cm; effacement 30 percent (assessed as a percentage; this unit does not chart cervical length); station -3 on Bishop's scale of thirds (the unit scores station in thirds for the Bishop entry and charts the centimeter scale separately once labor is established); consistency firm; position posterior. Cephalic presentation confirmed by Leopold maneuvers and bedside ultrasound; membranes intact; estimated fetal weight about 3,600 g. Version, total, and reading: Version: classic Bishop, five components, range 0 to 13. Total 1 of 13, computed by the examiner and confirmed by the attending. Reading: unfavorable under this unit's induction pathway, which treats a classic total of 6 or less as unfavorable for the purpose of offering cervical ripening; the threshold is the unit's convention, recorded as such because ACOG's 2025 guideline states that no universally accepted definition of an unfavorable cervix exists. The total is not a prediction of delivery mode and did not enter the decision to induce. Membranes, fetal status, and method-specific facts: Membranes intact. Fetal heart rate tracing Category I over 20 minutes before any agent (baseline 135, moderate variability, accelerations present, no decelerations); no contractions. No prior cesarean or uterine surgery; placenta fundal-anterior on the 20-week anatomy ultrasound; no cerclage; no known drug allergies; group B streptococcus culture negative on 09/02/2026; blood pressure 118/74, no hypertensive disorder; BMI 31. No contraindication to a transcervical balloon, dinoprostone, or misoprostol identified. Plan tied to the whole picture, with consent: Options discussed by M. Okafor, CNM, with T.W. at 20:00: transcervical balloon, vaginal dinoprostone insert, oral or vaginal misoprostol, and a balloon with misoprostol together; benefits, risks, and expected timelines of each reviewed, consistent with ACOG's 2025 guideline, which treats pharmacologic, mechanical, and combination methods as reasonable choices. T.W. preferred to start with a mechanical method and to stay mobile. Plan: transcervical balloon placed at 20:15 by M. Okafor, CNM; continuous tracing for 30 minutes after placement, then monitoring per the unit's balloon protocol; oxytocin to follow balloon expulsion or removal at the 12-hour dwell limit; amniotomy deferred until the presenting part is well applied. Reasons recorded beyond the score: unfavorable cervix under the unit convention, nulliparity, intact membranes, no uterine scar, no urgency in the indication, and the patient's stated preference. R. Castellano, MD, reviewed the findings and the plan at 20:05 and agreed. Written consent for the balloon signed at 20:10 and filed with the induction consent note. Reassessment after ripening: Balloon expelled spontaneously at 03:50 on 09/17/2026. Repeat digital examination at 04:05 on 09/17/2026 by R. Castellano, MD, with T.W.'s agreement: dilation 3 cm; effacement 60 percent; station -2 on the scale of thirds; consistency medium; position mid. Version classic Bishop; total 7 of 13. Change from the admission examination: 1 of 13 at 19:40 on 09/16/2026 (Okafor) to 7 of 13 at 04:05 on 09/17/2026 (Castellano); both examinations stand as separate entries, the difference reflects ripening observed by two examiners, and no value was corrected. Favorable under the unit convention (above 6). Tracing Category I; irregular contractions every 6 to 10 minutes. Plan: oxytocin by the unit's low-dose protocol started at 04:30; amniotomy when the head is well applied; continuous electronic fetal monitoring; epidural on request. Record and communication: Both examinations are entered in the induction section of the labor record; the flowsheet Bishop score field carries each total with the version (classic, 0 to 13) in its comment, and the five findings sit in the cervical examination rows with the examiner and time. The indication is documented as late-term pregnancy at 41 weeks 3 days, from which coding assigns the diagnosis (O48.0), never from the score; no Joint Commission PC-01 question arises at 41 weeks, and no payer asked for a Bishop value. T.W. was told at each examination what the findings and the total describe, that the total does not predict how she will deliver, and what the next step depends on; documented in the patient communication note. This sample is fictional and for educational purposes. It does not describe a real patient or record; the findings, scores, times, dates, and details are invented to show documentation structure and are not clinical guidance. No scoring table or guideline table is reproduced. ↑ Back to the template and downloads ### Why this sample works The indication, gestational age, dating, and the date and location of the consent discussion come first, so no reader can mistake the score for the reason induction happened or did not. Each examination carries its time, its examiner, and the five findings in the units assessed, with the station scale and the effacement unit stated, so the total can be recomputed and the two examiners' entries compared without guessing. The version is named with its maximum on both entries, the reading is attributed to the unit's convention with ACOG's no-universal-definition position noted, and the total is expressly kept out of the decision to induce and out of any prediction of delivery mode. Membrane status, the tracing category, the absence of a uterine scar, the placental site, cerclage status, and allergies are recorded before the method is chosen, so the choice of a balloon over a prostaglandin can be reconstructed and defended from the note alone. The plan names the method, the reasons beyond the score, the counseling and written consent, and the reassessment trigger, and the post-ripening entry keeps both sets of findings visible, so the change from 1 to 7 is a documented event rather than an unexplained jump. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the rules around the score are real but indirect, and none of them sets a number. No federal or state LAW names the Bishop score, requires it before induction, or attaches a ripening agent to a threshold; ICD-10-CM has no code for a cervical score, and the diagnoses around an induction (post-term or prolonged pregnancy in category O48, failed induction in category O61) are coded from the provider's documented diagnosis and gestational age, never from the score; no CPT or HCPCS code names the Bishop score, and the examination is part of the evaluation or the global obstetric service rather than a separately billable event. Under PAYER POLICY, the early-delivery rules that do exist key on gestational age and medical indication: New York Medicaid pays 100 percent for a cesarean or induction before 39 weeks that is medically necessary and 25 percent for an elective one, identified by condition codes 81 and 82 on facility claims and modifiers U7 and U8 on practitioner claims, and its policy article never mentions the cervix; no national or major-payer primary source located for this page conditions induction authorization on a Bishop number, so a note that says insurance requires a Bishop score needs a named payer and policy behind it. Hospital scheduling forms that require a Bishop field for elective inductions are local CONVENTION and should be cited as the hospital's policy, not as ACOG's or a payer's. The professional guidance is CONVENTION with real force. ACOG Clinical Practice Guideline No. 9, Cervical Ripening in Pregnancy (Obstetrics and Gynecology, July 2025, volume 146, pages 148 to 160), replaced Practice Bulletin 107 of August 2009; it states that there is no universally accepted definition of an unfavorable cervix, that the Bishop score is the most frequently used method of judging cervical readiness, that study populations have been defined by thresholds from 3 to 8, that parity and gestational duration also bear on the decision to ripen, and that once ripening is deemed necessary the patient is counseled on the methods with their benefits and risks, with pharmacologic, mechanical, and combination methods chosen by clinical circumstances, contraindications (a uterine scar among them), preferences, and setting; the Practice Bulletin 107 language that many pages still quote (6 or less unfavorable; above 8 a vaginal-delivery likelihood similar to spontaneous labor) is superseded and should not be presented as current ACOG wording. ACOG Committee Opinion 831 (2021) governs the timing of medically indicated late-preterm and early-term deliveries by indication, which is where the whether-and-when decision lives. The Joint Commission's PC-01 (elective delivery at 37 to less than 39 weeks) was removed from ORYX reporting effective January 1, 2026 while remaining in use for certification; its exclusions run on diagnosis codes, gestational age, labor status, prior uterine surgery, and stillbirth history, and the words Bishop and cervix appear nowhere in its specification, so a unit that says the Joint Commission requires a Bishop score is quoting its own policy. The ARRIVE trial (Grobman and colleagues, New England Journal of Medicine 2018) randomized low-risk nulliparous women to induction at 39 weeks or expectant management, did not use the Bishop score as an eligibility criterion, reported that 63 percent had an unfavorable modified Bishop score (below 5) at randomization, and found cesarean delivery in 18.6 percent of the induction group against 22.2 percent (relative risk 0.84, 95 percent confidence interval 0.76 to 0.93): evidence that an unfavorable cervix does not convert induction into a higher-cesarean strategy, not a rule about documentation. In litigation (CONVENTION, as evidence), induction and informed-consent disputes are common, and no reported decision located for this page makes a Bishop value the standard of care; what an expert reads is the indication, the examination findings with their examiner, the contraindications, the counseling, the plan, and the reassessment, which is the record this page describes. Canada and Australia set the clearest thresholds and the strongest documentation expectations, all of them CONVENTION, and the evidence explains why the components and the examiner matter more than the total. In Canada, no federal or provincial LAW names the score. SOGC's 2023 series (Journal of Obstetrics and Gynaecology Canada, volume 45, issue 1; Robinson and colleagues; replacing Guideline No. 296 of 2013) splits the topic: No. 432a covers general information, No. 432b directs cervical ripening when the modified Bishop score is below 7, and No. 432c takes a score of 7 or more toward amniotomy and oxytocin; provincial perinatal programs and registries (Perinatal Services BC, Alberta Health Services, BORN Ontario) capture cervical assessment in their own data dictionaries, and some provincial pathways use a length-based modified score, so a Canadian entry names the version as carefully as an American one. In Australia, RANZCOG's statement on prostaglandins for induction of labour (C-Obs 22; current version March 2019 with an interim update of February 2023; listed review date March 2024, now passed, so check for a successor before relying on it) directs that the method be chosen with regard to the favourability of the cervix (a cervical or Bishop score), parity, the indication and the likelihood of fetal compromise, the woman's preferences, and previous uterine surgery, and sets no numerical threshold; Queensland's Clinical Guideline on induction of labour uses a length-based Modified Bishop Score, recommends the score to inform the choice of method, allows amniotomy in selected patients with a low score when there is a clinical reason to avoid ripening, and proposes auditing whether a modified Bishop score was documented in the record before induction; the WA Country Health Service Induction of Labour Policy (version 3.04, published July 31, 2026) requires the requesting clinician to supply the indication, written consent (no consent, no induction), the current Bishop score, and the intended method, states that a score below 8 may likely require cervical ripening, and requires the clinician to notify the maternity unit if the score changes so that ripening is no longer needed; New South Wales issued its own induction policy in 2025; and the national Pregnancy Care Guidelines and the ACSQHC National Core Maternity Indicators track induction without a Bishop threshold. The United Kingdom's NICE NG207 (November 2021) is the reference many Canadian and Australian units read beside their own: 6 or less steers toward pharmacologic ripening with mechanical methods in specified circumstances, more than 6 toward amniotomy and oxytocin, 8 or more generally indicates a cervix ready to dilate, and the score is reassessed as induction progresses. On the evidence: Faltin-Traub and colleagues (2004) found exact agreement between two examiners on the total in 28 percent of 156 women at term and agreement within one point in 66 percent, with weighted kappa of 0.69, 0.54, and 0.35 across three examiner pairs; Dupuis and colleagues (2005) found numerical station errors in 50 to 88 percent of residents' and 36 to 80 percent of attendings' assessments on a birth simulator; Garcia-Simon and colleagues (2015) found ultrasound cervical length more reproducible between a consultant and a first-year resident than the Bishop score in 120 women, with intraclass correlations above 0.9 for the ultrasound measurement; Teixeira and colleagues (2012) pooled 59 studies and found higher scores associated with vaginal delivery and shorter induction-to-delivery intervals, with summary odds ratios from 1.98 to 5.48 depending on the cutoff used; Kolkman and colleagues (2013) pooled 40 studies of 13,757 women and found sensitivity and specificity pairs for cesarean delivery of 47 and 75 percent, 61 and 53 percent, and 78 and 44 percent at scores of 4, 5, and 6; Laughon and colleagues (2011) found positive predictive values for vaginal delivery of 87.7 percent for a simplified score above 5 and 87.0 percent for a classic score above 8, with negative predictive values of only 31.3 and 29.8 percent; the Cochrane review by Ezebialu and colleagues (2015) found two trials of 234 women comparing the Bishop score with transvaginal ultrasound, no superiority of either method on the main outcomes, and more frequent misoprostol use in the ultrasound arm; and a 2026 scoping review of 38 studies in nulliparous inductions (Félix and colleagues) confirmed the score's bedside utility while finding that predictive performance favored combined models and ultrasound parameters over the score alone. Those numbers are the reason to chart components with an examiner, to trend only within one version, and to keep the total out of any prediction about an individual birth. The Bishop score is in the public domain as a method: Edward H. Bishop published it in Obstetrics and Gynecology in 1964, a scoring method is not copyrightable under 17 U.S.C. 102(b), no trademark attaches to the eponym, and any induction record, EHR field, or web tool may implement the five components and a total in its own words and layout, naming the version it implements. The 1964 article and its table as published remain the journal publisher's expression: ACOG's 2025 guideline credits Wolters Kluwer Health for permission to adapt Bishop's table, which is why this page describes the components and reproduces no table. The modified versions belong to their authors' papers (Burnett, 1966; Calder, Embrey, and Hillier, 1974) and the simplified score to Laughon and colleagues (2011); the tables and flowcharts in ACOG Clinical Practice Guideline No. 9, SOGC Guideline No. 432, NICE NG207, RANZCOG C-Obs 22, and the Queensland and Western Australian documents are those bodies' copyright and are linked, never reproduced. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Bishop score documentation errors reviewers flag The numbers behind these errors are specific. Two examiners agreed exactly on the Bishop total in only 28 percent of 156 women at term and within one point in 66 percent, with weighted kappa of 0.35 to 0.69 across examiner pairs (Faltin-Traub and colleagues, 2004); station, one of the five components, was misjudged in 50 to 88 percent of residents' and 36 to 80 percent of attendings' simulator assessments (Dupuis and colleagues, 2005); across 40 studies and 13,757 women, a score of 6 predicted cesarean delivery with sensitivity of 78 percent and specificity of 44 percent (Kolkman and colleagues, 2013); a classic score above 8 carried a positive predictive value for vaginal delivery of 87.0 percent but a negative predictive value of 29.8 percent in 5,610 nulliparous inductions (Laughon and colleagues, 2011); and a meta-analysis of 59 studies found the association with vaginal delivery real but its size dependent on which cutoff a study chose (Teixeira and colleagues, 2012). ACOG's 2025 guideline states that no universally accepted definition of an unfavorable cervix exists. No published audit counts how often an induction record carries a total without its components, version, or examiner; the guideline, quality-audit, and litigation realities described under compliance considerations are what turn those omissions into a problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Bishop score documentation reviews: A total with no components, version, or examiner. "Bishop 5" tells the next examiner nothing about which findings produced it, whether it was a classic, length-based, or simplified total, or who examined and when, and it cannot be recomputed or compared. Chart the five findings in their units, the version with its maximum, the examiner, and the time, every time. Units and scales mixed. A cervix charted as 2 cm long scored as if it were a percentage of effacement, or a station read on the centimeter scale entered as a Bishop third, produces a wrong total before interpretation begins. State the unit and the scale for each finding, apply the version that uses those units, and never convert silently between length and percentage. A threshold asserted as universal, or attributed to the wrong body. "Below 6 requires ripening," "8 predicts vaginal birth," or Practice Bulletin 107 language quoted as current ACOG guidance. ACOG's 2025 guideline sets no universal cutoff, SOGC uses below 7, NICE uses 6 or less, and Australian states set their own on a modified score; write the convention with its source, or the unit's pathway as the unit's. The number charted as the order. "Bishop 4, misoprostol" with no membrane status, tracing, uterine history, preference, or counseling in the entry reads as if the score chose the drug. Every guideline treats the score as one input beside parity, indication, contraindications, and preference; record those and the consent, and state why the method fits the whole picture. Serial examinations overwritten, averaged, or reduced to a delta. A second examiner's total replacing the first, two disagreeing scores averaged into one, or "improved from 2 to 8" with the baseline findings gone. Interobserver agreement is only fair to substantial, so each examination stands as its own entry with time, examiner, findings, version, and total, and change is described with both sets visible. The score standing in for the indication, or a requirement invented for it. "Bishop 3, induction deferred" against a valid indication, "Bishop 9" as the reason to schedule, or a note claiming that the Joint Commission or the payer requires the score. The indication decides whether to induce; PC-01 and the payer policies located for this page never mention the cervix; a local requirement is cited as local policy. Write the indication first and the score after it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on labor and delivery and induction documentation. Give it the facts (the indication and gestational age with the dating method, the examination time and examiner, the five findings in the units assessed, the version, membrane status and the tracing category, prior uterine surgery and any contraindication, the method chosen and the reasons beyond the score, the consent discussion, and the reassessment trigger) and it drafts the documentation block: indication and decision, examination with examiner and time, findings with units, version and total, method-specific facts, the reading attributed to the unit's convention, the plan with its reasons and consent, and the reassessment entry, ready for your review. Cross-check a finished induction record for the gaps reviewers flag: a total with no components, version, or examiner; length and percentage or thirds and centimeters mixed; a threshold asserted with no source; the score charted as the order with no contraindications or preferences recorded; a second examination overwriting the first; or the score standing in for the indication. Turn the entry into the next document: the induction course paragraph of a discharge summary, the handoff line for the oncoming team with both examinations and the next reassessment trigger, or a plain-language explanation for the patient of what the findings describe and what the total does not predict, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What are the five components of the Bishop score, and how is it totaled? Five findings from one digital cervical examination are each converted to points and summed: dilation (how open the cervix is, in centimeters), effacement (how much it has thinned, as a percentage in the classic score), station (where the presenting part sits relative to the ischial spines, on Bishop's scale of thirds from minus 3 to plus 3), consistency (firm through soft), and position (posterior through anterior). In the classic version dilation, effacement, and station carry up to 3 points each and consistency and position up to 2 each, for a total of 0 to 13; Bishop's population was multiparous women at term being considered for elective induction, and he regarded a total above 8 as suitable for it. Modified versions that score cervical length in centimeters instead of percentage effacement have their own maximums, and the simplified score of Laughon and colleagues keeps only dilation, effacement, and station for a range of 0 to 9. This page describes the components conceptually and reproduces no scoring table; score from your unit's pathway or guideline, and chart the five findings in their units before the total. What Bishop score counts as favorable or unfavorable? There is no universal answer, and the note should say whose answer it is using. ACOG's Clinical Practice Guideline No. 9 (July 2025), which replaced Practice Bulletin 107, states that there is no universally accepted definition of an unfavorable cervix and that studies have used thresholds from 3 to 8 to define who benefits from ripening before oxytocin; the older Practice Bulletin language of 6 or less unfavorable and above 8 comparable to spontaneous labor is no longer current ACOG wording. SOGC Guideline No. 432b (2023) directs cervical ripening when the modified Bishop score is below 7, and No. 432c moves to amniotomy and oxytocin at 7 or more. NICE NG207 (2021) steers method choice at 6 or less against more than 6 and describes 8 or more as a cervix generally ready to dilate. Queensland and Western Australian guidelines set their own conventions on a length-based modified score, and the WA Country Health Service writes that a score below 8 may likely require ripening. Write "unfavorable under [guideline or unit pathway]" with the threshold that pathway sets, never a bare "unfavorable," and never attribute a cutoff to ACOG. Original, modified, or simplified: which version do I chart, and what if the cervix is described in centimeters of length? Chart the version your unit's induction pathway actually uses, and name it. In a US unit built on the classic five-component score, write "classic Bishop, 0 to 13" with the five findings; in a Canadian service following SOGC or an Australian service following Queensland, the locally correct instrument is a modified score, and "modified" must be qualified, because the label covers several nonidentical systems: the Calder lineage scores cervical length in centimeters instead of percentage effacement, Queensland's Modified Bishop Score is its own length-based build, and a widely circulated variant adds or subtracts points for parity, pre-eclampsia, post-dates, and prelabor rupture of membranes, which no national guideline instrument does. The simplified score of Laughon and colleagues (2011) is labeled as such and runs 0 to 9. When the examination is recorded as cervical length ("cervix 2 cm long"), that is a length-based finding: record it as length by digital examination, apply a length-based version if your pathway names one, and never convert it to a percentage to feed the classic score. Keep digital length distinct from transvaginal ultrasound length, which is a different measurement, and keep the station scale explicit, because Bishop's thirds and the centimeter scale of minus 5 to plus 5 are not interchangeable. Does a low Bishop score mean I must give a ripening agent, or that induction should wait? Neither. The indication decides whether to induce, and no guideline withholds an indicated induction because the cervix is unfavorable; the score decides only whether ripening is offered and helps choose how. ACOG's 2025 guideline sets no universal cutoff and, once ripening is deemed necessary, directs counseling on pharmacologic, mechanical, and combination methods chosen by clinical circumstances, contraindications, preferences, and setting. SOGC uses below 7 and NICE uses 6 or less to steer toward ripening, which is an operational convention, not a drug order. Queensland's pathway allows amniotomy in selected patients with a low score when there is a clinical reason to avoid ripening, and RANZCOG makes parity, the indication, the woman's preferences, and previous uterine surgery co-equal inputs with cervical favourability. A prior cesarean is the clearest illustration: two patients with the same score have different acceptable methods because the scar is external to the score. Document the indication, the reading under a named convention, the method-specific facts, and the plan with its reasons; do not write the number as the reason. Does the Bishop score predict cesarean delivery? No, and the record should not imply it. Bishop built the score in 1964 to select multiparous women at term for elective induction, not as an individualized cesarean-risk model; cesarean prediction is a later use that the evidence supports poorly. Kolkman and colleagues (2013) pooled 40 studies of 13,757 women and concluded that the score "should not be used to decide whether to induce labor or not," with sensitivity and specificity for cesarean delivery of 47 and 75 percent at a score of 4 and 78 and 44 percent at a score of 6. Laughon and colleagues (2011) found that a classic score above 8 predicted vaginal delivery with a positive predictive value of 87.0 percent, but that a low score identified who would not deliver vaginally with a negative predictive value of only 29.8 percent. Teixeira and colleagues (2012) found across 59 studies that higher scores are associated with vaginal delivery and shorter intervals, which is a population association, not a forecast for one patient. In the ARRIVE trial, 63 percent of low-risk nulliparous women had an unfavorable modified score at randomization, and induction at 39 weeks still lowered the cesarean rate (18.6 against 22.2 percent). A 2026 scoping review of 38 nulliparous studies found the score's stand-alone predictive performance limited compared with combined and ultrasound models. Write what the findings describe and what was planned; leave delivery mode out of it. Two of us got different Bishop scores. Whose goes in the chart, and how do I document the change after ripening? Both, as separate entries, each with its examiner, time, five findings, version, and total. There is no evidence-based rule for averaging two totals, and overwriting one with the other destroys the record of what each examiner found. Disagreement is expected: Faltin-Traub and colleagues (2004) found exact agreement on the total in 28 percent of paired examinations and agreement within one point in 66 percent, with weighted kappa from 0.35 to 0.69, and Dupuis and colleagues (2005) found station misjudged in a third or more of assessments even by attendings. When the two readings straddle a threshold that would change management, a repeat or senior examination is documented as a third entry with its own attribution. After ripening, the reassessment follows the method's protocol (balloon expulsion or dwell limit, the dose interval for a prostaglandin, and any change in findings that would change management; NICE asks for reassessment as induction progresses, and no universal recheck clock exists), and the entry describes the change with both sets of findings visible, so that a move from 1 to 7 reads as ripening observed by two named examiners rather than as an unexplained jump. Is a Bishop score required before every induction, by law, by payers, or by the Joint Commission? Not by law, not by any payer policy located for this page, and not by the Joint Commission; where it is required, the requirement is a guideline body's or an institution's. No US, Canadian, or Australian statute names the score. The Joint Commission's PC-01 elective delivery measure keys on gestational age, diagnosis codes, labor status, prior uterine surgery, and stillbirth history, never on cervical status, and it was removed from ORYX reporting effective January 1, 2026 while remaining in use for certification. Payer early-delivery rules such as New York Medicaid's key on medical necessity and gestational age and do not mention the cervix. What does expect a score is CONVENTION: Queensland proposes auditing whether a modified Bishop score was documented before induction, the WA Country Health Service requires the current Bishop score on every induction request beside the indication, written consent, and the intended method, SOGC and NICE structure their pathways around it, and ACOG calls it the most frequently used method of judging readiness without mandating it. So chart a score when your pathway expects one, cite the pathway or policy that expects it, and never write that a regulator or insurer requires it unless you can name the document. Can we put the Bishop score in our EHR, on an induction form, or in a free web calculator? The method, yes; a published table, no. A scoring method is not copyrightable under US law (17 U.S.C. 102(b) excludes procedures, processes, and systems), and Canadian and Australian copyright law draw the same line between an idea and its expression, so a hospital record, an EHR build, or a free calculator may implement the five components and a total in its own words and layout. What it may not do is copy a particular published expression: Bishop's 1964 article and its table remain the journal publisher's, and ACOG's 2025 guideline credits Wolters Kluwer Health for permission to adapt that table; the tables and flowcharts in the ACOG, SOGC, NICE, RANZCOG, and Queensland documents belong to those bodies. Two design points matter more than the rights question. Store the raw findings with their units, a version tag, and the computed total, not the total alone, because a field labeled Bishop score that accepts classic and length-based totals interchangeably is ambiguous by design. And label the version in the interface (classic 0 to 13, a named modified build, or simplified Laughon 0 to 9) so that clinicians in different units do not trend incompatible numbers. Nothing on this page or in its downloads reproduces a scoring table. Can BastionGPT draft my Bishop score documentation? Yes. Give it the facts (the indication and gestational age with the dating method, the examination time and examiner, the five findings in the units assessed, the version, membrane status and the tracing category, prior uterine surgery and any contraindication, the method chosen and the reasons beyond the score, the consent discussion, and the reassessment trigger) and it drafts the full entry: indication and decision, examination with examiner and time, findings with units, version and total, method-specific facts, the reading attributed to the unit's convention, the plan with its reasons and consent, and the reassessment entry, ready for your review. It can also check a finished induction record for a total with no components, version, or examiner, mixed units or scales, a threshold with no source, the score charted as the order, a second examination overwriting the first, or the score standing in for the indication. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Bishop EH, 1964, Obstetrics and Gynecology 24:266 to 268, Pelvic scoring for elective induction (the original method; the published article and table remain the publisher's expression and are not reproduced); Burnett JE Jr, 1966, Obstetrics and Gynecology 28(4):479 to 483, Preinduction scoring: an objective approach to induction of labor ; Calder AA, Embrey MP, Hillier K, 1974, Journal of Obstetrics and Gynaecology of the British Commonwealth 81:39 to 46, Extra-amniotic prostaglandin E2 for the induction of labour at term (the length-based lineage); Laughon SK, Zhang J, Troendle J, Sun L, Reddy UM, 2011, Obstetrics and Gynecology 117(4):805 to 811, Using a simplified Bishop score to predict vaginal delivery (5,610 nulliparous inductions; range 0 to 9; predictive values); Wormer KC, Bauer A, Williford AE, StatPearls, updated July 2024, Bishop Score (the thirds station scale against ACOG's centimeter scale; criteria not established). American College of Obstetricians and Gynecologists, Committee on Clinical Practice Guidelines-Obstetrics, July 2025, Obstetrics and Gynecology 146(1):148 to 160, Cervical Ripening in Pregnancy: ACOG Clinical Practice Guideline No. 9 (replaces Practice Bulletin No. 107 of August 2009; no universally accepted definition of an unfavorable cervix; thresholds of 3 to 8 in studies; the most frequently used method; counseling on methods; the Bishop table adapted with permission from Wolters Kluwer Health); ACOG Committee on Obstetric Practice and SMFM, 2021, Obstetrics and Gynecology 138(1):e35 to e39, Medically Indicated Late-Preterm and Early-Term Deliveries: ACOG Committee Opinion No. 831 (timing by indication). Society of Obstetricians and Gynaecologists of Canada, 2023, Journal of Obstetrics and Gynaecology Canada 45(1), Robinson D, Campbell K, Hobson SR, MacDonald WK, Sawchuck D, Wagner B: Guideline No. 432a: Cervical Ripening and Induction of Labour, General Information (pages 35 to 44.e1), Guideline No. 432b: Cervical Ripening (pages 56 to 62.e1; modified Bishop score below 7), and Guideline No. 432c: Induction of Labour (pages 70 to 77.e3; 7 or more); National Institute for Health and Care Excellence, November 2021, Inducing labour, NICE guideline NG207 (6 or less against more than 6; 8 or more; reassessment as induction progresses). Australia: RANZCOG, Use of prostaglandins for induction of labour (C-Obs 22) (current version March 2019, interim update February 2023, review due March 2024; method selection factors including favourability of the cervix, parity, indication, preferences, and previous uterine surgery); Queensland Clinical Guidelines, Induction of labour (Modified Bishop Score; pre-induction documentation as a quality measure); WA Country Health Service, July 31, 2026, Induction of Labour Policy, version 3.04 (indication, written consent, current Bishop score, and intended method on every request; a score below 8 may likely require cervical ripening; notify the unit if the score changes). United States measures, coding, and payer policy: The Joint Commission, PC-01 Elective Delivery, measure information form, v2026B (removed from ORYX effective January 1, 2026, still in use in Certification; exclusions by diagnosis code, gestational age, labor status, prior uterine surgery, and stillbirth history; no cervical element); CMS and NCHS, ICD-10-CM Official Guidelines for Coding and Reporting, FY 2027 (effective October 1, 2026; no code for a cervical score; bare code numbers only on this page); New York State Medicaid, Obstetrical deliveries prior to 39 weeks gestation (100 percent for medically necessary, 25 percent for elective; condition codes 81, 82, 83; modifiers U7, U8, U9; no cervical criterion). Reliability: Faltin-Traub EF, Boulvain M, Faltin DL, Extermann P, Irion O, 2004, European Journal of Obstetrics and Gynecology and Reproductive Biology 112:178 to 181, Reliability of the Bishop score before labour induction at term (156 women; exact agreement 28 percent; within one point 66 percent; weighted kappa 0.69, 0.54, 0.35); Dupuis O and colleagues, 2005, American Journal of Obstetrics and Gynecology 192(3):868 to 874, Birth simulator: reliability of transvaginal assessment of fetal head station (32 residents, 25 attendings); Garcia-Simon R and colleagues, 2015, Journal of Obstetrics and Gynaecology Research 41:377 to 382, Cervix assessment for the management of labor induction: reliability of cervical length and Bishop score determined by residents (120 women). Prediction: Kolkman DG, Verhoeven CJ, Brinkhorst SJ, van der Post JA, Pajkrt E, Opmeer BC, Mol BW, 2013, American Journal of Perinatology 30:625 to 630, The Bishop score as a predictor of labor induction success: a systematic review (40 studies, 13,757 women); Teixeira C, Lunet N, Rodrigues T, Barros H, 2012, Archives of Gynecology and Obstetrics 286(3):739 to 753, The Bishop Score as a determinant of labour induction success: a systematic review and meta-analysis (59 studies); Ezebialu IU, Eke AC, Eleje GU, Nwachukwu CE, 2015, Cochrane Database of Systematic Reviews CD010762, Methods for assessing pre-induction cervical ripening (two trials, 234 women); Félix TMJ, Sim-Sim MMF, Ferrão ACC, Zangão MOB, 2026, Healthcare 14:2242, The Bishop Index and birth outcomes in nulliparous women undergoing labor induction: a scoping review (38 studies); Grobman WA and colleagues, 2018, New England Journal of Medicine 379:513 to 523, Labor induction versus expectant management in low-risk nulliparous women (ARRIVE; cesarean 18.6 against 22.2 percent, relative risk 0.84). Rights: US Copyright Office, Circular 33, Works not protected by copyright (ideas, procedures, methods, and systems); Legal Information Institute, 17 U.S.C. 102, subject matter of copyright . ## Related templates Apgar Score Documentation Transfer-of-Care Summary Consultation Response & Report Discharge Summary Patient Instructions & After-Visit Summary Incident & Adverse Event Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/mews-score All note templates # MEWS and NEWS2 Documentation: Score, Trigger, Escalation & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 MEWS (Modified Early Warning Score, Subbe and colleagues, 2001) and NEWS2 (National Early Warning Score 2, Royal College of Physicians, 2017) are aggregate track-and-trigger scores that turn a set of vital signs into a total and an escalation prompt. Ward and step-down nurses, rapid response teams, and hospitalists use them to recognize deterioration and summon help. This page covers how to document the score and the escalation it triggered, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Registered nurses on wards, step-down units, and emergency departments at every observation set, with the nursing assistants and technicians who take vital signs under their oversight, and the rapid response, critical care outreach, and hospitalist clinicians who record the response; no license, training certificate, or publisher qualification level applies Audience Relieving nurses at handoff, rapid response and critical care outreach teams, hospitalists and covering physicians, intensivists deciding on transfer, sepsis coordinators and abstractors, patient safety and quality staff reviewing deterioration events, and later the peer reviewers, surveyors, and attorneys who reconstruct an escalation timeline from the chart Typical length 2 to 8 chart lines per observation set (time, each parameter, oxygen device and flow, score and version, any single-parameter trigger, the escalation with times and response) · an observation set and its scoring take about 3 to 5 minutes, longer when an escalation is documented Format family Aggregate weighted vital-sign score (each parameter earns points by how far it sits from a reference range, the points are summed, and an extreme value in any one parameter is its own trigger; the Subbe MEWS uses five parameters, NEWS2 six plus a weighting for supplemental oxygen) When it's used Every scheduled observation set on acute wards and step-down units, admission and post-operative observations, emergency department triage and boarding, any clinical concern between scheduled sets, sepsis screening prompts, rapid response and outreach reviews, and every handoff and transfer that carries a deterioration status Standards context MEWS: Subbe and colleagues (QJM, 2001), building on Morgan and colleagues (1997), no single rights holder and many local variants; NEWS2: Royal College of Physicians (December 2017), copyright with free reproduction on stated conditions, mandated only in NHS England; no US, Canadian, or Australian law names a score; described here for documentation, no chart reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What are MEWS and NEWS2? Early warning scores answer a problem the resuscitation literature has documented for decades: patients who arrest or need unplanned intensive care on a general ward usually show abnormal vital signs for hours first, and the signs are often charted without anyone acting on them. Morgan, Williams, and Wright described an aggregate early warning score for that purpose in 1997, and Subbe, Kruger, Rutherford, and Gemmel validated a Modified Early Warning Score in QJM in 2001 on 709 medical emergency admissions to a UK district general hospital, scoring five parameters (systolic blood pressure, heart rate, respiratory rate, temperature, and the AVPU level of consciousness) and finding that a total of 5 or more was associated with death (odds ratio 5.4), intensive care admission (10.9), and high-dependency unit admission (3.3). That paper is the reference, and it is all there is: MEWS was never trademarked, licensed, or standardized, hospitals changed the weights, added oxygen saturation, supplemental oxygen, or urine output, and set their own triggers, and the systematic review by Gerry and colleagues (BMJ, 2020) of 95 early warning score studies found the scores developed mostly in the United States (38 percent) and the United Kingdom (29 percent) and every study at high risk of bias. So "MEWS" in a US chart names a family of local instruments, and a 5 on one hospital's chart is not a 5 on another's. The Royal College of Physicians answered the same variation problem for England with the National Early Warning Score in 2012 and its update, NEWS2, in December 2017: six physiological parameters (respiration rate, oxygen saturation, systolic blood pressure, pulse, level of consciousness or new confusion, and temperature) each earn points according to how far they sit from a reference range, any supplemental oxygen adds a fixed weighting, and the points are summed to an aggregate. NEWS2 added a second oxygen saturation scale for patients with confirmed hypercapnic respiratory failure, added new confusion to the consciousness item so that AVPU became ACVPU, reordered the chart to the ABCDE sequence, and tied a score of 5 or more to the question of sepsis. The chart, the scoring table, and the response table are the RCP's and are described here, not reproduced. The load-bearing fact for documentation is that both scores have two triggers and one purpose. The triggers are the aggregate total, read against bands the RCP defines for NEWS2 (an aggregate of 5 or more is the key threshold for an urgent clinical response and 7 or more calls for an emergency response) and each hospital defines for its own MEWS variant, and an extreme value in any single parameter, which NEWS2 treats as a separate prompt for urgent review even when the total is low; the RCP stopped weighting that single red score as the equivalent of an aggregate of 5 in 2017 after an analysis found that escalating on every single-parameter score of 3 would raise workload by about 40 percent while adding about 3 percent to the detection of adverse outcomes. The purpose is a timely response, and that is where charts fail. Trinkle and Flabouris (Resuscitation, 2011) reviewed 575 cardiac arrests, emergency team calls, and unplanned intensive care admissions and found documented calling criteria with no call in the preceding 24 hours in 22.8 percent; Boniatti and colleagues (Critical Care Medicine, 2014) found 21.4 percent of emergency team calls delayed, with 30-day mortality of 61.8 percent against 41.9 percent when the call was timely; Tirkkonen and colleagues (2020) found criteria met without a call in the four hours before 17 percent of 5,568 rapid response calls across Australian and Finnish hospitals. The failure runs the other way too: respiratory rate is the parameter most often left unrecorded (Cretikos and colleagues, MJA, 2008), and in a Korean cohort 45.3 percent of ward patients who arrested still had a low MEWS eight hours beforehand (Kim and colleagues, 2015). A defensible record therefore needs complete observations, a timed response, and the name and version of the score, so a reviewer knows which chart the numbers belong to. The boundaries are firm: the RCP states that NEWS2 should not be used in children under 16 or in pregnancy, so pediatric early warning scores (PEWS) and maternity early warning systems (MEOWS in the UK, the Maternal Early Warning Criteria in the US) are separate instruments; qSOFA and SIRS are sepsis criteria rather than track-and-trigger systems; proprietary indices such as eCART, the Epic Deterioration Index, and the Rothman Index are continuous EHR-based models rather than bedside aggregates; and the consciousness item is AVPU or ACVPU, not the Glasgow Coma Scale , which is reached for when a fuller neurological assessment is needed, while sedation depth belongs to the RASS and delirium screening in the ICU to the CAM-ICU . ## Who uses MEWS and NEWS2 documentation and when Bedside nurses on medical, surgical, and step-down units generate most of the record: the observation set at the interval the chart or order sets, the score the chart or EHR computes, and the escalation when a trigger is met, with nursing assistants and technicians taking many of the vital signs and the registered nurse owning the score and the call. Emergency nurses score at triage and again while patients board; post-anesthesia and admission nurses set the baseline the ward will trend against. Rapid response and critical care outreach teams read the trajectory, document their review with its arrival time, and write the plan and the new observation interval; hospitalists and covering physicians record the response they gave to the call, and intensivists the transfer decision. Sepsis coordinators read the same sets when they abstract a sepsis case, and patient safety and quality staff reconstruct the timeline after an arrest, an unplanned ICU admission, or a fall in the incident report and the peer review that follows. The score travels: a transfer-of-care summary and a discharge summary should carry the last sets with their times and the escalations they prompted, and a responder's assessment often lands in a consultation report . Neighbors win in four places: a pediatric early warning score under 16, a maternity early warning system in pregnancy, a dedicated sepsis screen inside a sepsis pathway, and the GCS when a consciousness change needs more than an ACVPU letter. ## How to document MEWS and NEWS2 in the chart No US, Canadian, or Australian law, payer, or accreditor prescribes an early warning score note format; what exists is a copyrighted NHS instrument with its own recording recommendations, a family of local MEWS variants with hospital policies behind them, accreditation standards that require recognition and response without naming a score, and a literature showing that the chart most often fails at the call, not the number. What survives review is an entry that names the score and version in use, records every parameter with its time, states the total and any single-parameter trigger, reads the trend and the clinical concern, documents the escalation as a timeline, records the response and the reassessment interval, and handles modified criteria, treatment limits, refusals, and late entries honestly. Each element below carries the pitfall that most often undermines it. Which score, which version, which oxygen scale. Name the instrument as your facility has built it: NEWS2 as published, or the facility's MEWS variant with its version date and the policy that defines its triggers, and say whether the total was computed by the EHR or by hand. A MEWS total means nothing without that, because the weights and the trigger differ between hospitals, and the RCP's own FAQ states that any adaptation of the chart means it can no longer be called NEWS. For NEWS2, record which oxygen saturation scale is in use: Scale 2 is only for patients confirmed to have hypercapnic respiratory failure on blood gas analysis on this or a prior admission, the decision belongs to a competent clinical decision maker and must be recorded in the clinical notes (recommendation 26), and the unused scale is crossed out across a paper chart (recommendation 28). The RCP's March 2020 guidance acknowledges that COPD patients without confirmed hypercapnia are often monitored on Scale 2 and that the decision is commonly undocumented, and states that both are out of line with NEWS2; a Scale 2 entry with no recorded decision is exactly that. Pitfall: "MEWS 4" with no version, in a hospital whose chart weights differently from the one the reader trained on, or a Scale 2 patient with no note of who decided and why. The complete observation set, with times. Chart every parameter the score needs with the clock time it was taken: respiratory rate counted for a full minute rather than estimated or copied forward, oxygen saturation with the delivery device and flow rate (the RCP asks for the device and the liters per minute on the chart), systolic blood pressure, heart rate, temperature, and consciousness as AVPU or, for NEWS2, ACVPU, with new confusion assessed by asking staff or family whether the patient is more confused than usual. A total built on an incomplete set is a different number from the one the chart promises: the RCP's 2020 guidance says a score that cannot be completed for lack of equipment is still calculated and documented as incomplete, and a measurement that cannot be obtained despite the equipment should itself trigger an immediate response. Record the pre-admission baseline where it matters (chronic hypoxemia on home oxygen, a resting tachycardia, chronic confusion), because the score is not adjusted for it except through the documented mechanisms in the last element. Pitfall: A respiratory rate of 18 copied across six sets, an SpO2 charted without the oxygen it was measured on, or "alert" ticked without anyone asking about new confusion. The total and any single-parameter trigger. Write the total, the parameters that contributed to it, and, separately, any single parameter at its extreme score, because the two triggers are read differently: an aggregate at or above the threshold your chart sets calls for the response your policy attaches to that band, while an extreme single parameter with a low total is, in NEWS2, its own prompt for urgent review by a clinician competent in acute illness, and in most MEWS policies a call regardless of the sum. New confusion is the trap: under NEWS2 it scores 3 on its own, and if it is unclear whether the confusion is new the RCP says to assume it is until confirmed otherwise (recommendations 30 and 31). The response bands themselves are policy, not physiology: the RCP report says the clinical response should be agreed locally, and no US authority sets any band, so state the trigger as your policy defines it and cite the policy. Pitfall: A total of 3 charted as "no action" while one parameter sat at its maximum, or a newly confused patient scored as alert because he was awake. Trend and clinical concern. Compare each set with the ones before it and say what the trend is doing: a rise from 0 to 4 across three sets on a post-operative day is a finding even when no band has been crossed, and the RCP built the chart to be read as a trend. Record clinical concern as a trigger in its own right, with what prompted it (appearance, pain, the family's report that the patient is not himself), because the RCP states that concern about a patient's condition should always override the score if the attending professional considers escalation necessary (recommendations 4 and 19), and NHS England reports that 82 percent of Martha's Rule calls related to acute deterioration would not have triggered escalation through the early warning score. When nursing judgment increases the observation frequency ahead of the chart, write that it did and why. Pitfall: Three rising sets charted without comment because each was still "low," or a worried nurse's concern left out of the record because the number did not justify a call. The escalation as a timeline. This is the record that protects the patient. For each contact write who was called by name and role, by what route (page, phone, secure message, in person), at what clock time, what was communicated (SBAR or the local structure, including the score, its trend, and the specific request), and when and how the response came. If there was no response, chart the repeat attempt with its time and the move up the chain of command your policy sets: charge nurse, rapid response, the attending, the medical director. The RCP's 2020 guidance describes the structured documentation it expects after a trigger: the time of escalation, the time and grade of the clinical response, and the clinical assessment and plan, including the individualized trigger for further response. A rapid response activation gets the same treatment: time called, time each responder arrived, interventions with times, disposition, and the event record your system uses. Pitfall: "MD aware" or "hospitalist notified" with no time, no name, no content, and no response, which reads to a reviewer as no call at all. Response, orders, and the reassessment interval. Record the responder's assessment and orders with their times, the new observation frequency and who ordered it, and then the observations at that frequency, each with its time, because an order for hourly observations followed by a four-hour gap is among the most common findings in deterioration reviews. State whether the score improved after the intervention and, if it did not, the repeat escalation with its time: the RCP recommends at least hourly monitoring for a medium score or a single red parameter until the patient is reviewed and a plan of care documented, continuous monitoring at 7 or more, and, in its 2020 guidance, assessment within 60 minutes for a new score of 5 or 6, within 30 minutes for 7 or more, and senior review within 60 minutes when a high score does not improve; NSW's Between the Flags system requires a Clinical Review within 30 minutes of a Yellow Zone observation and a Rapid Response if that review cannot be obtained in time. Your policy's numbers may differ; cite them and meet them. Pitfall: "Obs hourly" ordered at 12:16 with the next set at 16:00, or a score that stayed above the trigger after the bolus with nobody called a second time. Modified criteria, treatment limits, refusal, late entries. Four situations change what the score means, and each needs its own sentence. Modified or altered calling criteria (a patient-specific threshold for a chronic derangement) are written with the authorizing clinician, the reason, and a review date; NSW policy requires exactly that, and the RCP's 2020 guidance describes an individualized escalation threshold set on initial and daily assessment by a competent clinical decision maker for patients with chronic respiratory failure. Treatment limits: a do-not-resuscitate order limits resuscitation and nothing else, so scoring and escalation continue unless a senior clinician has documented a different plan after discussion with the patient or family, which the RCP frames as a recorded decision that routine scoring is not appropriate (recommendation 21), and the RCP also asks that reasons not to act on a score be recorded in the notes; write the plan, not the assumption. Refusal: chart what was declined, the patient's capacity to decline, what was offered instead, and who was told. Late entries: label them as late, record the time you are writing and the time of the event, and never backdate; the College of Nurses of Ontario's revised Documentation standard (in force February 1, 2026) asks for the date and time care was provided and the date and time it was recorded, with a late entry identified as such, and regulators everywhere treat a backdated entry as an altered record. Pitfall: A DNR read as "do not call," altered criteria with no author or review date, or a 12:06 page typed at 12:45 and time-stamped 12:06. ### Blank template (copy and adapt) MEWS / NEWS2 OBSERVATION AND ESCALATION BLOCK Date: [ ] Time of set: [ ] Setting: [ward / step-down / ED / other] Nurse: [ ] Score in use: [NEWS2 as published / facility MEWS variant: version date, policy] Computed by: [EHR / hand] NEWS2 oxygen scale: [Scale 1 / Scale 2: decided by whom, date, hypercapnic failure confirmed on blood gas] Parameters (each with time): RR [ ] counted for a full minute SpO2 [ ] on [room air / device, flow] SBP [ ] HR [ ] Temp [ ] Consciousness [A / C new confusion / V / P / U] Baseline that matters: [chronic hypoxemia / resting tachycardia / chronic confusion / none] Total: [ ] Contributing parameters: [ ] Single parameter at maximum: [which / none] Incomplete set: [no / yes: which parameter and why] Trend: [prior sets with times and totals; direction] Concern: [none / what prompted it] Trigger status: [below policy trigger / aggregate trigger met / single-parameter trigger met / concern] Policy cited: [name, version] Frequency raised on nursing judgment: [no / yes: why] Escalation timeline: [time] contacted [name, role] by [page / phone / secure message / in person] content: [SBAR: score, trend, request] response: [time, content] [time] repeat attempt or next level: [name, role, route] response: [time] [time] rapid response called arrivals: [name, time] [name, time] Response: [assessment, orders, interventions with times] Sepsis pathway: [not started / started, time] New observation interval: [ ] ordered by [ ] Disposition: [stays / step-down / ICU, time] Reassessment: [time, total] [time, total] Improved: [yes / no] Repeat escalation: [time, to whom] Modified calling criteria: [none / threshold, authorizing clinician, reason, review date] Treatment limits: [full treatment / plan documented by, date; a DNR is not do-not-escalate] Refusal: [none / what declined, capacity, alternative offered, who informed] Late entry: [none / written at (time), describes (time)] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the NEWS2 chart, scoring table, or response table. ## Sample early warning score documentation (fictional) Scenario: a US community hospital medical-surgical ward, post-operative day 1 after a laparoscopic sigmoid colectomy in a 71-year-old man with COPD on home oxygen; the ward nurse documents the facility's MEWS variant by name and version, an individualized calling criterion written at admission, a rising trend with her own concern, an aggregate and a single-parameter trigger, the rapid response activation as a timeline, the sepsis pathway, the reassessments, and the transfer, all with clock times. All details are fictional. Patient: H.B., 71 · Setting: Medical-surgical ward, community hospital; post-operative day 1 after laparoscopic sigmoid colectomy · Clinician: A. Delgado, RN · Note date: 09/17/2026 Score in use, baseline, and morning set (07:10): Facility MEWS (hospital policy NUR-114, version March 2026): the five Subbe parameters plus oxygen saturation and a weighting for supplemental oxygen, with the consciousness item charted as ACVPU; the total is computed by the EHR from the entered values and checked by hand. Under NUR-114 the triggers are an aggregate of 5 or more, any single parameter at its maximum score, an oxygen requirement above the patient's documented criterion, or nurse concern, each of which requires a call; this is hospital policy, not a national rule. Baseline that matters: COPD on 2 L/min home oxygen with a usual resting SpO2 of 90 to 92 percent. On 09/16/2026 the admitting hospitalist, R. Ostrowski, MD, documented an individualized calling criterion (call for an SpO2 below 88 percent or an oxygen requirement above 3 L/min; other parameters unchanged), reason chronic hypoxemia, review date 09/18/2026. 07:10 set: respiratory rate 16, counted for a full minute; SpO2 92 percent on 2 L/min by nasal cannula; systolic blood pressure 128; heart rate 84; temperature 37.1 C; alert, and his wife, asked whether he is more confused than usual, says no. Total 1, contributed by the supplemental oxygen weighting; no single parameter at maximum; below every policy trigger. Overnight sets at 23:00 and 03:00 were 0 and 1 on the same chart. Rising trend, nurse concern, and first call (11:05 to 11:30): 11:05 set: respiratory rate 22, counted; SpO2 91 percent on 2 L/min; systolic 112; heart rate 104; temperature 38.1 C; alert, no new confusion. Total 4 per the facility chart, contributed by respiratory rate, heart rate, temperature, and oxygen; no single parameter at maximum; below the aggregate trigger of 5. Trend: 0, 1, 1, 4 across the sets since 23:00, rising on three parameters on a post-operative day. Concern: new abdominal pain he rates 6 of 10, distinct from the incision pain, and he looks unwell to me; concern recorded as a trigger in its own right under NUR-114. Observations increased to every 30 minutes on nursing judgment. 11:10 charge nurse P. Ibarra, RN, informed in person. 11:12 hospitalist Dr. Ostrowski paged with SBAR: post-operative day 1, MEWS 4 rising from 1 over four hours, new abdominal pain, temperature 38.1, request bedside review within 30 minutes. 11:20 Dr. Ostrowski returned the page by phone: orders for a complete blood count, lactate, and blood cultures, acetaminophen, and a bedside review promised by 12:00; orders entered 11:24, labs drawn 11:30. Aggregate and single-parameter triggers met, rapid response timeline (12:00 to 12:09): 12:00 set: respiratory rate 26, counted; SpO2 89 percent on 2 L/min, oxygen increased to 4 L/min under the ward oxygen protocol with SpO2 92 percent at 12:05; systolic 96; heart rate 118; temperature 38.4 C; new confusion: his wife reports he has not been making sense since about 11:45 and he could not name the hospital, charted as C on ACVPU. Total 8. Single parameter at maximum: consciousness (new confusion). Triggers met under NUR-114: aggregate 5 or more, single parameter at maximum, and oxygen requirement above the individualized criterion of 3 L/min. 12:02 rapid response activated by phone through the operator (logged by the operator at 12:02). 12:03 charge nurse at the bedside. 12:06 rapid response nurse K. Yusuf, RN, CCRN, arrived; SBAR given: MEWS 8 from 4 at 11:05, new confusion, respiratory rate 26, systolic 96, temperature 38.4, post-operative day 1 colectomy, COPD on home oxygen, hospitalist aware since 11:12 and en route. 12:08 respiratory therapist L. Moreau, RRT, arrived. 12:09 Dr. Ostrowski at the bedside. Response, orders, reassessment, and disposition (12:09 to 13:30): Responder assessment and orders are documented by Dr. Ostrowski in the rapid response event record; this entry carries the times. Sepsis screen positive under the facility protocol (suspected infection with an aggregate trigger and new confusion); sepsis pathway started, with time zero recorded by the team at 12:15. Lactate drawn 12:14 and blood cultures 12:16 as the pathway requires; 30 mL/kg crystalloid bolus started 12:20; first antibiotic dose administered 12:40 and recorded in the medication administration record; oxygen titrated to an SpO2 target of 88 to 92 percent per Dr. Ostrowski's order because of his COPD. Surgeon M. Chen, MD, notified by Dr. Ostrowski at 12:25; CT of the abdomen ordered. New observation interval ordered: every 15 minutes during the response, then every 30 minutes until transfer. 12:30 set: respiratory rate 24; SpO2 92 percent on 4 L/min; systolic 104 after 500 mL; heart rate 110; temperature 38.4 C; still C. Total 7; single-parameter trigger persists. 13:00 set: respiratory rate 22; SpO2 93 percent on 4 L/min; systolic 108; heart rate 102; temperature 38.0 C; still C. Total 6; improving but still above the aggregate trigger, so the team at the bedside was told at 13:02 that the score remained above trigger, which is the repeat escalation NUR-114 requires. Disposition: step-down unit bed accepted by intensivist S. Raman, MD, at 13:10; transferred at 13:30 with the rapid response nurse. Handoff, treatment limits, and record (13:30 to 13:50): Handoff to the step-down nurse in SBAR at 13:30 with the sequence in one line: facility MEWS totals 0 (23:00), 1 (03:00), 1 (07:10), 4 (11:05), 8 (12:00), 7 (12:30), 6 (13:00); new confusion since about 11:45; rapid response 12:02 to 13:30; sepsis pathway time zero 12:15, antibiotic 12:40, bolus complete 13:05; oxygen 4 L/min with an SpO2 target of 88 to 92 percent; the individualized calling criterion of 09/16/2026 carried forward with its 09/18/2026 review date, to be reset by the step-down team. Code status: full code, confirmed with the patient at admission; no treatment limits, so scoring and escalation continue without modification. No refusal of observations at any point. Wife updated at the bedside at 13:40. The rapid response event record was completed by K. Yusuf, RN. This narrative was entered at 13:50 as a late entry describing events from 11:05 to 13:30; the times in it are event times taken from the observation flowsheet, the paging log, and the operator's call log, and the flowsheet values were entered as each set was taken. This sample is fictional and for educational purposes. It does not describe a real patient or record; the values, times, dates, policy names, and details are invented to show documentation structure and are not clinical guidance. No scoring chart, threshold table, or response table is reproduced. ↑ Back to the template and downloads ### Why this sample works The instrument is named with its version and policy, the total is shown with its contributing parameters and any single parameter at maximum, and the triggers are cited to the policy that defines them, so a reader from another hospital knows which chart the numbers belong to. Every set carries a clock time, a respiratory rate that was counted, and an SpO2 with its device and flow, and the pre-admission baseline and the individualized calling criterion are written with their author, reason, and review date. The trend is read across sets and the nurse's concern is recorded as a trigger in its own right before any band was crossed, which is the RCP's rule that concern overrides the score put into practice. The escalation is a timeline: each contact has a name, role, route, clock time, SBAR content, and response, and the rapid response activation has its call time and each arrival time, so no reviewer has to infer whether or when the call was made. The response has orders and interventions with times, the ordered interval is met, the persistent trigger produces a documented repeat escalation, the code status is stated so a limit is not assumed, and the narrative is labeled as a late entry with both times. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names MEWS, NEWS2, or any early warning score, and the rules that bind sit one step from the bedside. Under LAW, the Medicare Conditions of Participation require a nursing assessment and care plan for every inpatient (42 CFR 482.23) and a medical record whose entries are complete, dated, timed, and authenticated and that contains the vital signs and other information needed to monitor the patient's condition (42 CFR 482.24); neither names a score, a chart, a threshold, or an observation frequency. Under ACCREDITATION STANDARD, the Joint Commission requirement that the hospital recognizes and responds to changes in a patient's condition, long numbered PC.02.01.19 and descended from the 2008 National Patient Safety Goal on summoning help, appears in the National Performance Goals effective January 2026 as NPG.01.05.02 with the same wording: its element of performance asks the hospital to develop and implement written criteria describing early warning signs of a change or deterioration in a patient's condition and the appropriate action to take, with documentation required, and its note states that hospitals are not required to create rapid response or medical emergency teams to meet it. So a MEWS chart, a NEWS2 build, a single-parameter calling list, or a proprietary index all satisfy the accreditor if the criteria are written, followed, and documented; none is mandated, and no numeric threshold is. State law adds nothing score-specific: the state provisions that exist concern who may summon help, not which score triggers it. Under PAYER POLICY, CMS's sepsis bundle measure SEP-1, a chart-abstracted Hospital Inpatient Quality Reporting measure that entered the Hospital Value-Based Purchasing program with fiscal year 2026, keys its clock to its own abstraction definition of severe sepsis presentation and names no early warning score, although an aggregate trigger is the usual prompt for the sepsis screen on a ward; the Surviving Sepsis Campaign's 2021 guideline recommends against qSOFA as a single screening tool compared with SIRS, NEWS, or MEWS (a strong recommendation on moderate-quality evidence), which is the closest any guideline used in US hospitals comes to endorsing these scores. Under CONVENTION, the AHRQ Patient Safety Network primer describes rapid response systems by their afferent (calling criteria) and efferent (response) limbs and lists a staff member's significant concern among the standard calling criteria; the FDA has cleared only two general early warning tools as devices, the Rothman Index and eCART, and Edelson and colleagues (JAMA Network Open, 2024) reported that the widely deployed Epic Deterioration Index has not been formally vetted while eCART discriminated deterioration within 24 hours better than NEWS or MEWS across 362,926 encounters at seven hospitals (areas under the curve 0.895, 0.829, and 0.757); Kaiser Permanente's Advance Alert Monitor, a continuous EHR-based model with a nurse-led response, was associated with lower 30-day mortality across 19 hospitals (adjusted relative risk 0.84; Escobar and colleagues, NEJM, 2020), and a NEWS best-practice alert at two Duke hospitals changed neither ICU transfer nor death and was generally ignored by frontline nurses (Bedoya and colleagues, 2019). The practical consequence for the chart is that the US landscape is locally modified MEWS variants, single-parameter calling lists, and proprietary indices side by side, so the entry must say which instrument and which version produced the number and which policy defines its trigger; malpractice reviews of failure to rescue turn on that timeline, and no national claims database publishes an early-warning-score-specific figure. Canada and Australia name no score in law, and the UK-origin rules everyone cites bind no one outside the NHS. In Canada, no federal or provincial LAW names an early warning score; Accreditation Canada's Qmentum program assesses organizations against Health Standards Organization standards, and the instrument is a provincial or institutional choice (CONVENTION): Nova Scotia Health's NEWS2 policy CL-SR-040 (approved February 20, 2025, effective August 25, 2025) applies NEWS2 to admitted adults 16 and over, uses a children's early warning system under 16, excludes pregnancy, makes the patient's documented goals of care and level of intervention direct the escalation, requires an authorized prescriber's order for SpO2 Scale 2, and leaves the escalation pathway to each facility's work instruction; the Hamilton Early Warning Score, a Canadian instrument, discriminated in-hospital mortality among 5,491 rapid response patients at least as well as NEWS2 (areas under the curve 0.76 and 0.72; Fernando and colleagues, Critical Care, 2019); other systems run local MEWS variants, so a Canadian chart names its instrument the same way a US chart does. The College of Nurses of Ontario's revised Documentation practice standard, approved by its Council on September 18, 2025 and in force since February 1, 2026, sets the timing and late-entry expectations described above (REGULATORY STANDARD for Ontario nurses). In Australia, the NSQHS Recognising and Responding to Acute Deterioration Standard (second edition; ACCREDITATION STANDARD binding on accredited services) requires an individualized monitoring plan for each patient, observations documented and tracked graphically over time, and escalation protocols that specify the criteria for escalating care, without naming a score; the states supply the charts. NSW's Between the Flags program, underpinned by policy directive PD2025_014 and embedded in 225 acute public facilities, uses Yellow and Red Zone criteria on standard observation charts, requires a Clinical Review within 30 minutes of a Yellow Zone observation and a Rapid Response when that review cannot be obtained in time or a Red Zone criterion is met, and requires altered calling criteria to be authorized and time-limited; Queensland's Q-ADDS combines an aggregate with single-parameter triggers and, in a comparison on 224,912 US admissions, discriminated deterioration about as well as NEWS (areas under the curve 0.71 and 0.72) and better than Between the Flags (0.64), with eCART highest (0.76) (Campbell and colleagues, Resuscitation, 2020). The UK-origin rules are the most-cited external schedule: NHS England has endorsed NEWS2 as the early warning system for acute hospitals in England, NICE CG50 (2007) asked for track-and-trigger systems, a written monitoring plan, and observations at least every 12 hours, and the RCP's recommendations set 12-hourly monitoring at a score of 0, 4- to 6-hourly at 1 to 4, at least hourly at 5 to 6 or with a single red score until the patient is reviewed and a plan documented, and continuous monitoring at 7 or more, with the clinical response agreed locally. Two boundaries hold everywhere. Version: NEWS2 (December 2017, with a March 2020 implementation guidance note) remains the current RCP instrument as of September 2026, it differs from the 2012 NEWS in its second oxygen scale, its new-confusion item, and the weighting of a single red score, so NEWS and NEWS2 are not interchangeable in a chart, and a MEWS variant is named by the hospital that built it. Population: the RCP states that NEWS2 should not be used in children under 16 or in pregnancy, and its FAQ allows use up to 20 weeks with an obstetric score after that; pediatric wards use a PEWS, NHS England's national PEWS charts (November 2023) come in four age bands, and the largest PEWS trial, EPOCH (21 hospitals in 7 countries), did not reduce mortality (Parshuram and colleagues, JAMA, 2018); maternity units use MEOWS in the UK and the Maternal Early Warning Criteria (Mhyre and colleagues, 2014) in the US. Performance cautions belong in the note's humility, not its format: NEWS discriminated death, ICU admission, and cardiac arrest better than 33 other systems in 35,585 admissions (Smith and colleagues, 2013), yet NEWS2's second oxygen scale did not improve discrimination and worsened it in 48,898 admissions at risk of type 2 respiratory failure (Pimentel and colleagues, 2019), and Gerry and colleagues (2020) rated every early warning score study they reviewed at high risk of bias. Those numbers are the reason the entry records the complete set, the trend, and the response rather than trusting the total. The Modified Early Warning Score was published by Subbe, Kruger, Rutherford, and Gemmel in QJM in 2001, building on the early warning score of Morgan, Williams, and Wright (1997); no single rights holder controls it, no license governs it, and the many local variants belong to the hospitals that built them, which is why a MEWS chart should cite Subbe 2001 and its own policy. NEWS2 is copyright Royal College of Physicians 2017: the RCP states that there is no copyright restriction on NEWS2 provided the RCP is acknowledged with its specified wording, the material is not modified or amended in any way, and the charts are reproduced in colour from the high-quality versions on its site, and its FAQ adds that any adaptation means the chart can no longer be called NEWS. That permission covers reproducing the RCP's own chart on those terms; it is not a license to alter it, and this page describes the chart and reproduces none of it. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common MEWS and NEWS2 documentation errors reviewers flag The numbers behind these errors are specific. Trinkle and Flabouris reviewed 575 cardiac arrests, emergency team calls, and unplanned ICU admissions at one Australian hospital and found documented calling criteria with no call in the preceding 24 hours in 22.8 percent ; patients whose criteria were documented across more than one time period without a call had hospital mortality of 52.5 percent against 31.9 percent for a single period (Resuscitation, 2011). Boniatti and colleagues found 21.4 percent of 1,148 medical emergency team patients had a delayed call, with 30-day mortality of 61.8 percent against 41.9 percent when the call was timely (Critical Care Medicine, 2014), and Tirkkonen and colleagues found criteria met without a call in the four hours before 17 percent of 5,568 rapid response calls across three Australian and Finnish hospitals, with delayed calls for respiratory criteria associated with higher mortality (odds ratio 1.71; Resuscitation, 2020). The other half of the problem is measurement: among 501 ward cardiac arrests in a Korean cohort, 45.3 percent of patients still had a low MEWS eight hours before the arrest and the score rose beforehand in only 46.8 percent (Kim and colleagues, PLoS One, 2015), and Cretikos and colleagues wrote in 2008 that respiratory rate in particular is often not recorded. Implementation does not fix documentation: a NEWS best-practice alert at two Duke hospitals produced no change in ICU transfer or death (adjusted hazard ratios 0.94 and 0.90) and was generally ignored by frontline nursing staff (Bedoya and colleagues, 2019), and Gerry and colleagues rated all 95 early warning score studies they reviewed at high risk of bias (BMJ, 2020). No published audit counts how often a chart carries a total with no instrument or version, a copied respiratory rate, or a contact with no time; the accreditation, coroner, and malpractice reviews described under compliance considerations are what turn those omissions into a finding. The BastionGPT Clinical Advisory Board sees the same errors most often in early warning score documentation reviews: A total with no instrument behind it. "MEWS 4" in a hospital whose chart weights differently from the reader's, NEWS and NEWS2 used as one word, or a Scale 2 patient with no recorded decision. Name the score, the version date, and the policy that defines the trigger, say whether the EHR or a hand computed the total, and for NEWS2 record the oxygen scale and who decided it. An incomplete or copied observation set. A respiratory rate of 18 carried across six sets, an SpO2 charted without the oxygen it was measured on, or a temperature left blank and the total scored as if complete. Count the rate for a full minute, chart the device and flow, mark an incomplete set as incomplete with the reason, and treat a parameter that cannot be obtained despite the equipment as a trigger in itself. The single-parameter trigger and new confusion missed. A total of 3 charted "no action" while one parameter sat at its maximum, or a patient scored as alert because his eyes were open while his wife said he was not making sense. Write any extreme single parameter separately from the total, ask about new confusion at every set, chart C when the answer is yes or unknown, and cite the policy response for each trigger. A contact with no time, name, content, or response. "MD aware" or "hospitalist notified" as the whole escalation. Chart each contact as a timed line with the name and role, the route, the SBAR content including the score and its trend, and the response with its time; when there is no response, chart the repeat attempt and the move up the chain of command, each with its time. The ordered interval not met, or no second call. "Obs hourly" at 12:16 followed by a set at 16:00, or a score that stayed above the trigger after the fluids with nobody called again. Record the ordered frequency and who ordered it, chart every set at that frequency with its time, state whether the score improved, and document the repeat escalation with its time and recipient when it did not. Limits, criteria, refusals, and late entries by assumption. A DNR read as do-not-call, altered calling criteria with no author or review date, a refusal recorded as "declined vitals" with nothing else, or a 12:45 entry time-stamped 12:06. Write the treatment plan with its author and date and keep scoring under it, record who authorized any modified threshold and when it expires, chart the refusal with capacity and the alternative offered, and label a late entry with the time written and the time of the event. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on nursing observation and rapid response documentation. Give it the facts (the score and version in use and the policy that sets its triggers, each parameter with its clock time and the oxygen device and flow, the total and any single parameter at maximum, the prior sets, your concern and what prompted it, each contact with its time, route, content, and response, the responder's orders and the new interval, the reassessments, and any altered criteria, treatment limits, refusal, or late-entry timing) and it drafts the entry: the instrument named, the set with times, the total with its triggers and the policy cited, the trend and concern, the escalation as a timeline, the response and reassessment, and the closing items, ready for your review. Cross-check a finished note or flowsheet narrative for the gaps reviewers flag: a total with no instrument or version, a respiratory rate copied forward, an SpO2 without its oxygen, a single-parameter trigger or new confusion missed under a low total, a contact with no time, name, content, or response, an ordered interval not met, or a treatment limit assumed rather than documented. Turn the episode into the next document: the deterioration lines of a transfer-of-care or discharge summary with the last sets and the escalations they prompted, the timeline paragraph of an incident report after an arrest or unplanned ICU admission, or a plain-language update for a family, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How are MEWS and NEWS2 scored, and how do they differ? Both are aggregate weighted scores. Each routinely measured parameter earns points according to how far it sits from a reference range, the points are summed, and the total is read against bands that decide the response; an extreme value in any single parameter is also a trigger on its own. The Subbe MEWS (2001) scores five parameters: systolic blood pressure, heart rate, respiratory rate, temperature, and the AVPU level of consciousness. Most hospitals that call their chart MEWS have changed the weights or added oxygen saturation, supplemental oxygen, or urine output, so there is no single MEWS and the chart's own policy is the reference. NEWS2 (Royal College of Physicians, December 2017) scores six physiological parameters (respiration rate, oxygen saturation, systolic blood pressure, pulse, consciousness with new confusion, and temperature) and adds a fixed weighting for any supplemental oxygen; it also offers a second oxygen saturation scale for patients with confirmed hypercapnic respiratory failure, and it differs from the 2012 NEWS in that scale, in the new-confusion item, and in no longer weighting a single red score as the equivalent of an aggregate of 5. The RCP's scoring chart and response table are its copyrighted materials and are not reproduced here; score from the published chart or your facility's build. Neither score is for children under 16 or for pregnancy, where a PEWS and a maternity early warning system apply. What score triggers a rapid response or an escalation call? Whatever your policy says, and the policy should be cited in the note, because no US, Canadian, or Australian authority sets a number. For NEWS2 the RCP recommends four trigger levels: a low aggregate of 1 to 4, which a competent registered nurse assesses to decide whether monitoring or care should change; a single red score, an extreme value scoring 3 in any one parameter, which should prompt urgent review by a clinician competent in acute illness; a medium aggregate of 5 or 6, the key threshold for an urgent clinical response and the point at which the RCP says to think of sepsis in a patient with possible infection; and a high aggregate of 7 or more, which calls for emergency assessment by a team with critical care competencies and usually transfer to a higher level of care. The RCP adds that the clinical response should be agreed locally, that its 2020 guidance expects assessment within 60 minutes for a new 5 or 6 and within 30 minutes for 7 or more, and that a single red score no longer carries the weight of an aggregate of 5, because escalating on every one would have raised workload by about 40 percent for about 3 percent more detected adverse outcomes. For MEWS there is no national band: Subbe and colleagues found a total of 5 or more associated with death and intensive care admission in their 2001 validation, hospitals commonly trigger at 4 or 5 or on any single parameter at its maximum, and each hospital's policy decides. In the United States the Joint Commission requires written criteria and an appropriate action, not a particular score. Two things trigger regardless of the number: an extreme single parameter, and clinical concern, which the RCP says should always override the score. What do I document when I escalate, and what if the doctor does not respond? Document the escalation as a timeline, one line per contact: the clock time, who you contacted by name and role, the route (page, phone, secure message, in person), what you communicated (the score, its trend, the parameters driving it, your concern, and the specific request, in SBAR or your local structure), and the time and content of the response. If there is no response within the interval your policy sets, chart the repeat attempt with its time and then the move up the chain of command your policy defines (charge nurse, rapid response team, attending physician, medical director), each with its time; a call to the rapid response team because the primary clinician could not be reached is charted as exactly that. The RCP's 2020 guidance describes the structured record it expects after a trigger: the time of escalation, the time and grade of the clinical response, and the clinical assessment and plan, including the individualized trigger for further response. NSW's Between the Flags system makes the fallback explicit: if a Clinical Review is not obtained within 30 minutes of a Yellow Zone observation, a Rapid Response must be called. Once responders arrive, record each arrival time, the interventions with times, the disposition, and the new observation interval, and complete the rapid response event record your system uses; if the episode ends in harm, the incident report is built from this timeline. The one entry that fails every review is "MD aware" with no time, name, content, or response. How do I document NEWS2 oxygen Scale 2, a chronically abnormal baseline, or altered calling criteria? Each is a documented clinical decision, not a nursing adjustment. Scale 2 on the NEWS2 chart is only for patients confirmed to have hypercapnic respiratory failure on blood gas analysis on this or a prior admission and requiring supplemental oxygen; the RCP recommends a prescribed saturation target of 88 to 92 percent for them, states that the decision to use Scale 2 should be made by a competent clinical decision maker and recorded in the patient's clinical notes, and asks that the unused scale be clearly crossed out across a paper chart. Its 2020 guidance acknowledges that COPD patients without confirmed hypercapnia are often put on Scale 2 and that the decision is commonly undocumented, and calls both departures from NEWS2; Nova Scotia Health's 2025 NEWS2 policy goes further and requires an authorized prescriber's order for Scale 2. For a patient whose baseline is chronically abnormal on either scale, the RCP's answer is an individualized escalation threshold determined on initial and daily assessment by a competent clinical decision maker, and NSW's Between the Flags program calls the same thing altered calling criteria, which must be authorized and carry a review date and time. A US hospital running a MEWS variant has no RCP-sanctioned override, so the same elements come from local policy: chart the clinician who authorized the modified criterion, the reason, the new threshold, and the review date, keep scoring every parameter as measured, and note that the criterion, not the chart, decided whether a call was due. Never lower a score to fit a baseline; change the documented trigger instead. How do I chart new confusion, and is the consciousness item the same as the GCS? NEWS2 assesses consciousness as ACVPU: alert, new confusion, responds to voice, responds to pain, unresponsive. New confusion, which the RCP defines to include disorientation, delirium, or any acute reduction in the Glasgow Coma Scale, scores 3 on its own, a red single-parameter trigger for urgent assessment (recommendations 29 and 30), and if it is unclear whether the confusion is new or the patient's usual state, the RCP says to assume it is new until confirmed otherwise (recommendation 31). Its 2020 guidance suggests asking staff or relatives whether the patient is more confused than usual, and it expects the responding clinician's assessment, including a delirium screen, to decide the ongoing trigger threshold for that patient. Chart the letter, the source of the baseline comparison (a relative, the previous shift, the admission note), and what the patient could not do. A patient with chronic confusion who is at baseline does not score as C; a patient who is awake but newly disoriented does, and that is the case most often missed. The consciousness item is not the Glasgow Coma Scale : the RCP says the two should be used alongside each other, and the GCS is the tool for the fuller neurological assessment a consciousness change calls for; sedation depth belongs to the RASS , and a formal delirium screen in critical care to the CAM-ICU , with delirium itself a provider diagnosis. Does a DNR order mean I stop scoring or stop calling, and what do I chart if the patient refuses observations? No. A do-not-resuscitate order limits resuscitation and nothing else; a patient with a DNR can still have a treatable pneumonia, a bleed, or sepsis, and the score and the call continue unless a senior clinician has documented a different plan. The RCP frames that plan as a recorded decision: when a clinical team decides that routine scoring is not appropriate, for example on an end-of-life care pathway, the decision should be discussed with the patient or family and recorded in the clinical notes (recommendation 21), and the RCP separately asks that reasons not to act on a score be recorded in the notes. Nova Scotia Health's NEWS2 policy makes the patient's documented goals of care and level of intervention direct the escalation, which is the same idea in Canadian form. So the entry names the plan (full treatment, ward-based treatment without transfer, comfort-focused care) with its author and date, and continues to chart observations at the frequency the plan sets; a DNR read as do-not-call is an error reviewers find quickly. For a refusal, document what was declined and when, the patient's capacity to decline and understanding of the risk, what you offered instead (a later attempt, a partial set, a different device), and who you told, then escalate the missing data if it matters clinically; the RCP's 2020 guidance says a set that cannot be completed is still scored and documented as incomplete, and a parameter that cannot be obtained despite the equipment should itself trigger an immediate response. Is NEWS2 required in the United States, Canada, or Australia, and does any law name a score? No law in any of the three names MEWS, NEWS2, or any early warning score, and NEWS2 is mandated only in NHS England, which has endorsed it as the early warning system for its acute hospitals. In the United States the Medicare Conditions of Participation (42 CFR 482.23 and 482.24) require nursing assessment and a complete, dated, timed record (LAW); the Joint Commission requires the hospital to recognize and respond to changes in a patient's condition, with written criteria describing early warning signs and the action to take, and states that rapid response teams are not required to meet it (ACCREDITATION STANDARD, carried into the National Performance Goals effective January 2026 as NPG.01.05.02); CMS's SEP-1 sepsis measure names no score (PAYER POLICY); and which score, which version, and what threshold are hospital policy (CONVENTION). In Canada, no federal or provincial law names a score, and the instrument is a provincial or institutional choice: Nova Scotia Health adopted NEWS2 by policy in 2025, and other systems use the Hamilton Early Warning Score or local MEWS variants. In Australia, the NSQHS Recognising and Responding to Acute Deterioration Standard requires individualized monitoring plans, track-and-trigger observation charts, and escalation protocols that specify criteria, without naming a score (ACCREDITATION STANDARD), and the states supply the charts: NSW's Between the Flags Yellow and Red Zones under policy directive PD2025_014, Queensland's Q-ADDS, and their counterparts elsewhere. The documentation consequence is the same everywhere: name the instrument and version, cite the policy that defines the trigger, and write the escalation as a timed record, because every one of these standards judges the response, not the arithmetic. Can we print the NEWS2 chart, build it into our EHR, or publish a MEWS calculator? NEWS2: yes, on the RCP's terms, which are precise. The RCP states there is no copyright restriction on NEWS2 provided the material is acknowledged with its specified wording (reproduced from the RCP's 2017 report), is not modified or amended in any way, and the charts are reproduced in colour from the high-quality versions on its site; the 2017 report's copyright page says the same, and its FAQ adds that any adaptation means the chart can no longer be called NEWS. So an unaltered NEWS2 chart printed in colour with the acknowledgment, or the unaltered scoring built into an EHR with attribution, sits inside the permission, while a chart with a changed parameter, weight, or colour scheme is a local instrument that must carry its own name. MEWS: there is no license to obtain and none to rely on, because no single rights holder controls it; the Subbe scoring scheme is a published journal method, hospitals have adapted it freely for twenty-five years, and the honest course is to name your variant, cite Subbe and colleagues (2001), and version it. Calculators: computing an aggregate from entered values is arithmetic, and public calculators do it for both scores; reproducing the RCP's chart image or colour table is the restricted act, and copying another hospital's variant table copies that hospital's document. This page describes both instruments and reproduces no chart, scoring table, or response table. Can BastionGPT draft my MEWS and NEWS2 documentation? Yes. Give it the facts (the score and version in use and the policy that sets its triggers, each parameter with its clock time and the oxygen device and flow, the total and any single parameter at maximum, the prior sets, your concern and what prompted it, each contact with its time, route, content, and response, the responder's orders and the new interval, the reassessments, and any altered criteria, treatment limits, refusal, or late-entry timing) and it drafts the full entry: the instrument named, the set with times, the total with its triggers and the policy cited, the trend and concern, the escalation as a timeline, the response and reassessment, and the closing items, ready for your review. It can also check a finished note for a total with no version, a copied respiratory rate, an SpO2 without its oxygen, a missed single-parameter trigger, an "MD aware" with no time or response, an ordered interval not met, or a DNR read as do-not-call. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Morgan RJM, Williams F, Wright MM, 1997, Clinical Intensive Care 8:100, an early warning scoring system for detecting developing critical illness (abstract); Subbe CP, Kruger M, Rutherford P, Gemmel L, 2001, QJM 94(10):521-526, validation of a modified Early Warning Score in medical admissions (709 admissions; odds ratios for death 5.4, ICU admission 10.9, HDU admission 3.3 at a total of 5 or more); Smith GB, Prytherch DR, Meredith P, Schmidt PE, Featherstone PI, 2013, Resuscitation 84(4):465-470, the ability of the National Early Warning Score to discriminate patients at risk (33 other systems; 35,585 admissions); Gerry S and colleagues, 2020, BMJ 369:m1501, early warning scores for detecting deterioration: systematic review and critical appraisal (95 studies; 38 percent developed in the US, 29 percent in the UK; all at high risk of bias); Pimentel MAF and colleagues, 2019, Resuscitation 134:147-156, a comparison of NEWS and NEWS2 (251,266 admissions; 48,898 at risk of type 2 respiratory failure). Royal College of Physicians, National Early Warning Score (NEWS) 2 (no copyright restriction with the specified acknowledgment, no modification, charts in colour; NHS England endorsement; accessed September 2026); RCP, December 2017, NEWS2: standardising the assessment of acute-illness severity in the NHS, updated report of a working party (ISBN 978-1-86016-682-2; copyright page; recommendations 1 to 46, including 2, 4, 9 to 12, 19 to 21, 24 to 33, and 38 to 40; the single-red-score workload analysis; FAQ 7 on pregnancy and 18 on adaptation); RCP, March 2020, NEWS2 additional implementation guidance (individualized thresholds; new confusion question; escalation documentation; Scale 2 misuse; incomplete parameters); NHS England, Martha's Rule (143 phase 1 sites from May 2024; phase 2 from April 2025; 82 percent of deterioration calls would not have triggered the early warning score); NHS England, November 2023, national paediatric early warning system charts (four age bands); NICE, 2007, CG50, acutely ill adults in hospital (track and trigger; monitoring plan; observations at least every 12 hours). Afferent limb failure and measurement: Trinkle RM, Flabouris A, 2011, Resuscitation 82(7):810-814, documenting rapid response system afferent limb failure and associated patient outcomes (575 events; 22.8 percent; mortality 52.5 against 31.9 percent for multi-period against single-period failure); Boniatti MM and colleagues, 2014, Critical Care Medicine 42(1):26-30, delayed medical emergency team calls and associated outcomes (1,148 patients; 21.4 percent delayed; 61.8 against 41.9 percent 30-day mortality); Tirkkonen J and colleagues, 2020, Resuscitation 156:6-14, afferent limb failure revisited (5,568 patients; 17 percent within 4 hours; respiratory criteria odds ratio 1.71); Kim WY and colleagues, 2015, PLoS One 10(6):e0130523, modified early warning score changes prior to cardiac arrest in general wards (501 arrests; 45.3 percent low at 8 hours); Cretikos MA, Bellomo R, Hillman K, Chen J, Finfer S, Flabouris A, 2008, Medical Journal of Australia 188(11):657-659, respiratory rate: the neglected vital sign . Implementation and proprietary indices: Bedoya AD and colleagues, 2019, Critical Care Medicine 47(1):49-55, minimal impact of implemented early warning score and best practice alert (85,322 patients; hazard ratios 0.94 and 0.90); Edelson DP, Churpek MM, Carey KA, and colleagues, 2024, JAMA Network Open 7(10):e2438986, early warning scores with and without artificial intelligence (362,926 encounters; eCART 0.895, NEWS 0.829, MEWS 0.757; Rothman Index and eCART the only FDA-cleared general early warning tools; Epic Deterioration Index not formally vetted); Escobar GJ and colleagues, 2020, New England Journal of Medicine 383(20):1951-1960, automated identification of adults at risk for in-hospital clinical deterioration (19 hospitals; adjusted relative risk 0.84); Parshuram CS and colleagues, 2018, JAMA 319(10):1002-1012, the EPOCH randomized clinical trial (21 hospitals, 7 countries; no mortality reduction); Mhyre JM and colleagues, 2014, Obstetrics and Gynecology 124(4):782-786, the maternal early warning criteria . United States: The Joint Commission, National Performance Goals effective January 2026 for the hospital program (NPG.01.05.02: the hospital recognizes and responds to changes in a patient's condition; written criteria for early warning signs and the action to take; rapid response teams not required) and National Performance Goals overview ; eCFR, 42 CFR 482.23 nursing services and 42 CFR 482.24 medical record services ; CMS, Hospital Value-Based Purchasing Program (SEP-1 in the program from fiscal year 2026 under the FY 2024 IPPS final rule); AHRQ Patient Safety Network, rapid response systems primer (afferent and efferent limbs; calling criteria including staff concern; 2008 National Patient Safety Goal); Evans L and colleagues, 2021, Intensive Care Medicine 47(11):1181-1247, Surviving Sepsis Campaign international guidelines 2021 (against qSOFA as a single screening tool compared with SIRS, NEWS, or MEWS). Australia: Australian Commission on Safety and Quality in Health Care, Recognising and Responding to Acute Deterioration Standard (monitoring plans; graphic tracking of observations; escalation protocols that specify criteria; no score named); NSW Clinical Excellence Commission, Between the Flags overview (policy directive PD2025_014; 225 facilities; Yellow and Red Zones; Clinical Review within 30 minutes; Rapid Response); Campbell V, Conway R, Carey K, and colleagues, 2020, Resuscitation 153:28-34, predicting clinical deterioration with Q-ADDS compared to NEWS, Between the Flags, and eCART (224,912 admissions; areas under the curve 0.71, 0.72, 0.64, 0.76). Canada: Nova Scotia Health, National Early Warning Score 2 (NEWS2) policy CL-SR-040 (approved February 20, 2025; effective August 25, 2025; adults 16 and over; children's early warning system under 16; not in pregnancy; goals of care direct escalation; authorized prescriber order for Scale 2); Fernando SM, Fox-Robichaud AE, Rochwerg B, and colleagues, 2019, Critical Care 23:60, prognostic accuracy of the Hamilton Early Warning Score and NEWS2 among patients assessed by a rapid response team (5,491 patients; 0.76 against 0.72); College of Nurses of Ontario, Council approves revised Documentation practice standard (approved September 18, 2025; effective February 1, 2026; late entries defined) and Documentation practice standard . ## Related templates Glasgow Coma Scale (GCS) Documentation RASS Score Documentation CAM-ICU Documentation Incident & Adverse Event Report Transfer-of-Care Summary Discharge Summary Consultation Response & Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/pressure-injury-staging All note templates # Pressure Injury Staging Documentation: Stage, POA, Healing & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 Pressure injury staging classifies a pressure injury by the deepest tissue exposed or damaged, using the NPIAP system of four numbered stages plus unstageable and deep tissue pressure injury, last revised in 2016. Wound care and bedside nurses stage the wound, the provider documents the diagnosis, and coders, surveyors, and payers read both entries against each other. This page covers how to document the stage and its present-on-admission status, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Registered nurses at the bedside, in long-term care, and in home health, who describe and stage the wound; wound, ostomy, and continence nurses who confirm difficult stages; physicians, advanced practice providers, and podiatrists who document the diagnosis and present-on-admission status; no license or publisher qualification level applies to staging itself Audience Wound care teams, attending physicians and hospitalists, clinical documentation integrity specialists and coders, quality and patient safety staff, MDS and OASIS coordinators, CMS and state surveyors, accreditation reviewers, payers, and expert reviewers in pressure injury claims Typical length 6 to 14 chart lines per wound (location and laterality, stage with the date first identified, present-on-admission status, measurements, wound bed, exudate, edges and periwound, pain, device or mucosal qualifier, photograph, treatment, reassessment) · 5 to 10 minutes by hand after the wound has been examined Format family Observer-rated anatomic classification of an existing wound (Stage 1 to Stage 4 by the deepest tissue exposed, plus unstageable and deep tissue pressure injury as categories rather than intermediate stages; medical device related injuries staged by depth; mucosal membrane injuries not staged) When it's used Admission and transfer skin inspections, first identification of any skin change over a pressure point or under a device, weekly wound reassessment, change in wound status or debridement, MDS and OASIS assessment time points, discharge and transfer summaries, and quality or incident review of a hospital-acquired injury Standards context NPIAP staging system (first issued 1989, revised 2007 and 2016) with the NPIAP, EPUAP, and PPPIA International Guideline; the stage definitions and illustrations are NPIAP copyright used only by permission; CMS coding, HAC, MDS, and OASIS rules govern how the stage is reported; described here for documentation, no definitions reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is pressure injury staging? Pressure injury staging is the classification of a pressure injury by the deepest layer of tissue that has been exposed or damaged. The system in use across the United States, Canada, and Australia is the one maintained by the National Pressure Injury Advisory Panel (NPIAP, known until 2019 as the National Pressure Ulcer Advisory Panel), first issued at a 1989 consensus conference, revised in 2007 when the unstageable and deep tissue categories were added, and revised again at an April 2016 consensus conference (Edsberg and colleagues, Journal of WOCN, 2016), when the term injury replaced ulcer, Arabic numerals replaced Roman numerals, the word suspected was dropped from the deep tissue category, and medical device related and mucosal membrane pressure injuries were defined. The published stage definitions and the illustrations that accompany them are NPIAP intellectual property and are not reproduced here; in original words, the categories run as follows. In a Stage 1 injury the skin surface is unbroken, but an area over a pressure point has changed color and the change does not fade when you press on it and release; in darker skin it may show as a difference in tone, warmth, or firmness rather than redness. In a Stage 2 injury the upper layers of skin are gone and the shallow base is still living skin tissue, pink to red in color, with no fat showing; a blister filled with clear fluid, intact or broken, also belongs here. In a Stage 3 injury the wound passes through the whole thickness of the skin so that the fat beneath it can be seen, sometimes with rolled edges, dead tissue, or pockets extending sideways or downward, but nothing deeper than fat is exposed. In a Stage 4 injury the wound has passed through skin and fat to structures such as muscle, tendon, ligament, cartilage, or bone, which can be seen or felt at the base. An unstageable injury is a full-thickness wound whose base is hidden under dead tissue, so the true depth cannot be assigned until enough of that tissue is removed; it is a depth not yet known, not a stage beyond 4. A deep tissue pressure injury is a deep, bruise-like, purple or maroon area or a blood-filled blister over intact or broken skin, marking damage that began in the deeper tissue before the surface changed; it may resolve or open to reveal its real depth. A medical device related injury is caused by tubing, a cannula, a collar, a cast, a mask, or another device, usually takes the device's shape, and is staged by depth like any other, except that an injury on a mucous membrane (nose, mouth, urethra, or other mucosa) has no skin layers to grade and receives no numeric stage. Three facts carry the documentation. First, staging classifies existing damage; it does not predict risk, so it sits beside, and never replaces, the risk assessment on the Braden Scale page and the skin inspection that finds the wound in the first place. Second, the stage records the deepest damage the wound ever reached and is never reduced as the wound heals: a healing wound fills its defect with granulation and scar tissue rather than regrowing the muscle, fat, or dermis it lost, which is why the NPUAP's 2000 position statement rejected reverse staging and why the MDS and OASIS instructions still forbid it; progress is described by the wound's own characteristics or a validated healing tool, and a healed wound is a healed injury at the stage it reached. Third, the stage and the diagnosis come from different people. The ICD-10-CM Official Guidelines let a coder take the pressure ulcer stage from the documentation of a clinician who is not the patient's provider, and note that a nurse often documents it, while the pressure ulcer diagnosis itself must come from the provider, and conflicting entries have to be resolved by a query; the coding category L89 also still carries the older ulcer terminology, so the 2016 change to injury has not reached the codes or the MDS and OASIS item names. Everything else on this page follows from those three facts, including the present-on-admission determination that decides payment and public reporting and the qualifiers that keep a device or mucosal injury from being misread. The OASIS assessment page covers the home health items that receive the stage; the risk-tool total belongs to the Braden page; and a hospital-acquired injury that a facility treats as an event is reported through the incident report as a separate document from the wound record. ## Who uses pressure injury staging documentation and when Bedside nurses on medical, surgical, orthopedic, and critical care units document the first finding at the admission or transfer skin inspection, at the unit's scheduled reassessment, and whenever a device site or pressure point changes, and the admission entry is the one that later decides whether an injury was present on admission. Wound, ostomy, and continence nurses confirm the stage where the depth is uncertain, measure and photograph on a schedule, and describe healing at the weekly reassessment most facilities adopt. Long-term care nurses stage for MDS 3.0 Section M and the F686 care plan, and home health nurses for the OASIS pressure injury items at start of care, resumption, follow-up, transfer, and discharge. Hospitalists, attending physicians, advanced practice providers, and podiatrists write the diagnosis, the site, the etiology, and the present-on-admission status that a coder needs, often in a consultation report when a wound service is involved. Clinical documentation integrity specialists and coders read the two sets of entries against each other and query when they disagree; quality and patient safety staff, MDS and OASIS coordinators, surveyors, and accreditation reviewers read the same record for hospital-acquired injuries. The staging page owns one job: the stage, the present-on-admission determination, the wound description, the device or mucosal qualifier, the photograph conventions, and healing described without reverse staging. Neighbors take over from there: risk belongs to the Braden Scale , the injury travels in the transfer of care summary and the discharge summary , the interdisciplinary plan sits in the care coordination note , the home health item set is on the OASIS assessment page, and a facility-acquired injury treated as an event goes through the incident report . ## How to document pressure injury staging in the chart No regulator prescribes a staging note format; what exists is a copyrighted classification, coding rules that decide who may document what, payment and reporting rules that turn on the present-on-admission determination, and facility policies that set the reassessment interval and the photography rules. What survives review is an entry that names the site with its side and the etiology, states one stage or category with the date it was first seen and the rater, records the present-on-admission determination on the evidence of the admission inspection, measures and describes the wound by a stated method, carries the device or mucosal qualifier, documents the photograph and its consent, and describes healing by the wound's characteristics while the stage stays where it was. Each element below carries the pitfall that most often undermines it. Location, laterality, and etiology. Name the anatomic site and the side (left ischial tuberosity, right lateral heel, sacrum at midline), place it on the body map, and state why it is a pressure injury: the location over a bony prominence or under a named device, the shape, and the history of pressure or shear. Say what it is not when a look-alike is possible: moisture-associated skin damage and incontinence-associated dermatitis are moisture injuries and are not staged; skin tears are mechanical trauma; arterial, venous, and diabetic foot ulcers have their own classification systems and their own codes. One wound per entry, each with its own stage, because the coder assigns a code for every pressure injury the patient has. Pitfall: "Sacral wound" with no side or map reference, or a moist perineal skin loss from incontinence staged as a Stage 2 pressure injury. Stage or category, date first identified, and rater. Write one stage or category for the wound with the date and time it was first identified and who staged it, and describe the depth in plain words that a reader could check against the wound (skin intact, skin loss with the base still skin, fat visible, structure visible or palpable, base hidden by dead tissue, dark intact discoloration). Unstageable and deep tissue pressure injury are categories, not intermediate steps: an unstageable wound is a full-thickness wound of unknown depth, and a deep tissue pressure injury may resolve or open to reveal its stage. A mucosal membrane injury is documented as a pressure injury and given no numeric stage. Then make sure the provider's note carries the diagnosis for the same wound, the same site, and the etiology, because the coder may take the stage from your entry but must take the diagnosis from the provider. Pitfall: A stage in the nursing flowsheet with no provider diagnosis anywhere in the record, or a provider note reading only decubitus ulcer with no site while the nurse has staged three wounds. Present-on-admission determination and its timing. Record the time of the inpatient admission order and the time of the admission skin inspection, and state whether the wound was present at that inspection, because CMS defines present on admission as present at the time the order for inpatient admission occurs, and conditions that develop in the emergency department, in observation, or in outpatient surgery before that order count as present on admission. There is no 24- or 48-hour window in that definition: a facility's rule that an injury found within a set number of hours is presumed present is a local convention, and the guidelines say there is no required timeframe by which a provider must document a condition as present on admission. When the admission inspection was late or incomplete, say so, document the earliest reliable finding, and leave the determination to the provider, who may record it as undetermined; do not assert present on admission from a later finding. Pitfall: "POA" charted on day four with no admission inspection to support it, or a unit's 24-hour presumption written into the note as a CMS rule. Measurements, wound bed, exudate, edges, and periwound. Measure length head to toe and width perpendicular to it in centimeters, depth at the deepest point with a probe, and any undermining or tunneling by clock-face position and distance, and name the method so the next measurement is comparable. Describe the wound bed by tissue type with percentages that add to 100 (granulation, slough, eschar, and any exposed structure), the exudate by amount, type, and odor, the edges (attached, rolled, undermined), the surrounding skin (intact, red, macerated, indurated, warm), and pain at rest and with care. Depth is the least reliable dimension and photographs distort it, so the probe measurement and the tissue description are what a reviewer trusts; write the numbers, not moderate or large. Pitfall: "Approximately 3 cm, moderate depth," percentages that add to 130, or a wound bed described only as "see photo." Device-related and mucosal qualifiers. When a device caused the injury, name the device (nasal cannula, endotracheal tube holder, cervical collar, cast edge, pulse oximeter probe, non-invasive ventilation mask), state the fit and the time it had been in place, and record what changed (repositioned, padded, resized, removed) alongside the stage by depth. When the injury is on mucous membrane, write pressure injury of the named mucosal site, device related, not stageable, and describe it in words and measurements; MDS and OASIS instructions do not stage mucosal injuries either. The qualifier is what tells a quality reviewer that the prevention bundle for the device, not the mattress and turning schedule, was the gap. Pitfall: "Stage 2 pressure injury, nasal septum" with a numeric stage on mucosa, or a device injury with the device never named. Photograph and consent. No federal rule requires a photograph of every pressure injury; photography follows facility policy and wound society guidance and is a health record entry like any other, so the photograph is protected health information and needs the consent your policy and state law require. Chart that the photograph was taken, the consent obtained (written, verbal, or under a general admission consent, as policy allows), the date and time, that a ruler and patient identifier were in the frame, and where the image is stored, and describe the wound in words in the same entry, because a photograph supports the written description and never replaces it. A dated photograph from the admission inspection is the strongest possible support for a present-on-admission determination. Pitfall: An undated, unlabeled image in a media tab with no consent noted, or a note that reads only see photo where the description should be. Healing, reassessment, and the nurse-provider match. Describe progress by the wound's own characteristics (smaller measurements, less dead tissue, more granulation, less exudate, edges advancing) or by a validated healing tool such as the PUSH tool, and keep the stage where it was: a Stage 4 wound that is filling in is a healing Stage 4, a closed wound is a healed injury at that stage, and a wound that reopens at the same site resumes its prior stage. When an unstageable wound is debrided to a visible base, document the date and the stage revealed, and when a deep tissue pressure injury opens, document the date and the stage revealed; neither is a downgrade or an upgrade, it is new information with a date. Set the reassessment interval your policy requires (weekly is the common convention), and before signing check that the provider's current note still names the same wound, site, and etiology, because the coder will query any conflict. Pitfall: "Stage 4, now Stage 2, improving," or a wound care note that describes a healing injury while the provider note still reads unstageable three weeks after debridement. ### Blank template (copy and adapt) PRESSURE INJURY DOCUMENTATION BLOCK (one wound per block) Date / time: [ ] Setting: [ ] Rater: [ ] Trigger: [admission or transfer inspection / first identification / scheduled reassessment per policy (name it) / change in wound / debridement] Location + laterality: [site, side, body map reference] Etiology: [pressure or shear over bony prominence / under device (name it) / not a pressure injury: MASD, IAD, skin tear, arterial, venous, diabetic] Stage or category: [1 / 2 / 3 / 4 / unstageable (base hidden) / deep tissue pressure injury / mucosal, not stageable] Depth in plain words: [ ] First identified: [date, time, by whom] Prior stage assigned: [same / n/a] Present on admission: [yes / no / undetermined by provider] Admission order: [date, time] Admission skin inspection: [date, time, findings at this site] Evidence: [ED or transfer record, photograph] Measurements (cm, method stated): length [ ] width [ ] depth [ ] undermining / tunneling: [clock position, distance / none] Wound bed: [granulation __% / slough __% / eschar __% / structure visible] Exudate: [amount, type, odor] Edges: [ ] Periwound: [ ] Pain: [ ] Device or mucosal qualifier: [device named, fit, hours in place, action taken / mucosal site, not stageable / none] Photograph: [taken, date, time, ruler and identifier in frame / not taken] Consent: [written / verbal / per policy] Stored: [ ] Healing description (stage unchanged): [improving / stalled / deteriorating; what changed; PUSH or other tool result if used] Provider diagnosis on record: [note date; same wound, site, etiology, POA status / conflict: query sent date] Plan: [dressing and frequency; offloading and surface; nutrition; referral; next reassessment date; handoff, transfer summary, MDS or OASIS] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the NPIAP stage definitions or illustrations. ## Sample pressure injury staging documentation (fictional) Scenario: a US community hospital medical unit, where a wound, ostomy, and continence nurse completes the weekly reassessment of a sacral pressure injury found on the admission skin inspection of an 81-year-old woman admitted through the emergency department a week earlier; the entry states the stage once with the present-on-admission evidence, measures by a stated method, records the photograph and consent, describes healing without changing the stage, and confirms that the provider's diagnosis matches. All details are fictional. Patient: L.B., 81 · Setting: Community hospital, medical unit; weekly wound reassessment · Clinician: P. Adeyemi, RN, CWOCN · Note date: 09/16/2026 Trigger and location: Weekly wound reassessment under the hospital's skin integrity policy (NUR-207), day 7 after admission. Single wound: sacrum, midline, centered 2 cm above the coccyx, body map reference S-1. Etiology pressure and shear: L.B. was found to have spent most of two days in a recliner at home before the emergency department visit, and the wound sits over the sacral prominence with no device involved. Perineal skin is intact with no moisture damage; this is not incontinence-associated dermatitis. No other pressure injury on today's head-to-toe inspection; bilateral heels, ischial tuberosities, trochanters, scapulae, elbows, occiput, and the oxygen tubing contact points behind both ears intact. Stage and present-on-admission determination: Stage 3 pressure injury, first identified 09/09/2026 at 15:10 by the admitting nurse and confirmed by this writer at 09:30 on 09/10/2026: full-thickness skin loss with fat visible across the base, no muscle, tendon, or bone seen or felt with the probe, base fully visible with no covering dead tissue, so the depth could be assigned. Present on admission. The inpatient admission order was written 09/09/2026 at 14:35; the admission skin inspection at 15:10 documented this wound at this site with a photograph, and the emergency department nursing record from 09/09/2026 at 11:50 notes a sacral open area with a dressing applied at home. The stage assigned today is the stage assigned at first identification; nothing has been re-staged. Measurements and wound description: Length 3.4 cm head to toe, width 2.3 cm perpendicular, depth 0.7 cm at the deepest point by cotton-tipped probe (09/09/2026: 3.8 by 2.6 by 0.9 cm). Undermining none today (09/09/2026: 0.6 cm from the 3 to the 5 o'clock position); no tunneling. Wound bed 85 percent red granulation, 15 percent thin yellow slough at the 6 o'clock edge, no eschar, no exposed structure (09/09/2026: 60 percent granulation, 40 percent slough). Exudate small, serous, no odor (09/09/2026: moderate, serous). Edges attached and flat, no rolling. Periwound skin intact, faint pink for about 1 cm that fades on pressure, no maceration, warmth, or induration; light skin tone. Pain 2 of 10 at rest and 4 of 10 during the dressing change by numeric rating, premedicated per order. Photograph and consent: Photograph taken today at 09:40 with the disposable ruler and the patient label in the frame, matching the admission photograph of 09/09/2026 at 15:15 for angle and distance; written consent for wound photography signed by L.B. on 09/09/2026 and scanned to the consents tab; images stored in the wound media folder of the electronic record. The written description above stands on its own. Healing description: Improving: surface area down about 12 percent, depth down 0.2 cm, undermining resolved, granulation up from 60 to 85 percent, exudate reduced, edges attached. PUSH tool 3.0 total 9 today against 12 on 09/09/2026 (surface area 6 to 10 square centimeters range, small exudate, granulation tissue), with attribution to the NPUAP version in the flowsheet. The stage is unchanged and will remain unchanged as the wound closes; when it closes it will be recorded as healed at the stage first assigned, never as a lower stage. Provider documentation, plan, and reassessment: Attending's history and physical of 09/09/2026 and progress note of 09/16/2026 both document a sacral pressure injury at this stage, present on admission, with the same site and etiology; nursing and provider entries agree and no query is needed. Braden total documented separately today (13, with mobility and moisture the low subscales) and the prevention plan continues: pressure redistribution foam mattress, repositioning every two hours agreed with L.B., heels floated, chair time limited to one hour with a cushion, incontinence brief changed on schedule with barrier cream. Dressing: calcium alginate to the wound bed with a silicone foam cover, changed every three days and as needed; dietitian consult of 09/10/2026 continues with a protein supplement twice daily. Next reassessment 09/23/2026 or sooner for increased size, new dead tissue, odor, or fever. Wound description and the present-on-admission status entered in the transfer of care summary for the planned home health referral; skin integrity team informed at the 09/16/2026 meeting. This sample is fictional and for educational purposes. It does not describe a real patient or record; the measurements, dates, and details are invented to show documentation structure and are not clinical guidance. No NPIAP stage definitions or illustrations are reproduced. ↑ Back to the template and downloads ### Why this sample works The site carries its side and a body map reference, the etiology is argued rather than assumed, and the look-alike is ruled out, so a coder, a surveyor, or an expert reviewer can find the wound and accept the diagnosis. The stage is stated once, with the date and time first identified, the rater, and a plain-words depth description that anyone can check against the wound; nothing has been re-staged. The present-on-admission determination rests on the admission order time, the admission inspection time, the emergency department record, and a dated photograph, which is exactly the evidence chain a hospital-acquired condition review asks for. Measurements name their method, the wound bed percentages add to 100, and the baseline values sit beside today's, so healing is shown by the wound's characteristics and a validated tool while the stage stays fixed. The provider's notes are checked against the nursing entry and found to agree, the device and mucosal question is answered, the photograph has consent and storage, and the reassessment date and handoff close the loop. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the classification is a convention, and the rules around it are law and payer policy. Under LAW, the ICD-10-CM Official Guidelines for Coding and Reporting (FY 2026, effective October 1, 2025), whose use is required of HIPAA-covered entities as part of the adopted code set, state in Section I.B.14 that the pressure ulcer stage is one of the few code assignments that may rest on documentation by clinicians other than the patient's provider, that a nurse often documents the pressure ulcer stages, that the associated diagnosis must be documented by the patient's provider, and that conflicting documentation, from the same clinician or different clinicians, calls for a query to the provider. Section I.C.12.a adds the rules coders apply to what you write: codes in category L89 identify site and stage, and as many are assigned as the patient has pressure ulcers; the unstageable code is for a wound whose stage cannot be clinically determined, for example under eschar or after a graft, and is not the code for a missing stage, which is unspecified; if the stage of an unstageable wound is revealed after debridement during the encounter, only the revealed stage is coded; a wound completely healed at admission is not coded; a healing wound is coded to its documented stage; a wound present at admission that progresses to a higher stage takes two codes, the admission stage and the highest stage reached; and pressure-induced deep tissue damage has its own codes. Appendix I defines present on admission as present at the time the order for inpatient admission occurs, counts conditions that develop in the emergency department, observation, or outpatient surgery before that order as present on admission, offers the indicators Y, N, U (documentation insufficient), and W (provider unable to clinically determine), and states that there is no required timeframe by which a provider must identify or document a condition as present on admission; any 24- or 48-hour admission window is therefore a facility convention or a quality-measure specification, not the payment definition. Under PAYER POLICY, section 5001(c) of the Deficit Reduction Act of 2005 placed stage 3 and stage 4 pressure ulcers on the Medicare hospital-acquired condition list, so that for discharges since October 1, 2008 a hospital receives no higher payment for such an ulcer coded as not present on admission, with N and U treated as not present and Y and W as present; the AHRQ Patient Safety Indicator PSI 03 (v2025) counts stage 3, stage 4, or unstageable pressure ulcers coded as secondary diagnoses not present on admission per 1,000 medical and surgical discharges of adults, excludes stays under three days, obstetric cases, severe burns, and exfoliative skin disorders, and excludes events where a deep tissue injury or unstageable ulcer was present on admission at the same site, so a documented admission finding protects the hospital's PSI 90 composite in the Hospital-Acquired Condition Reduction Program. In long-term care, 42 CFR 483.25(b)(1), surveyed as F686 (LAW), requires that a resident who enters without pressure ulcers does not develop them unless clinically unavoidable and that a resident who has them receives the care needed to promote healing and prevent new ones, and MDS 3.0 Section M (items M0210 and M0300 A through G, each with a present-at-entry column) records unhealed injuries by stage and category under RAI Manual instructions that never reverse stage, that report an unstageable wound debrided to a visible base at the stage revealed while keeping its present-at-entry status, and that treat a numerically staged wound that becomes covered by dead tissue during the stay as unstageable and not present at entry. In home health, the OASIS item set (items M1306, M1311, M1322, and M1324) reports unhealed injuries by stage at each time point, does not reverse stage, and, per CMS guidance, reports a closed injury that reopens at its worst prior stage (LAW as to collection, PAYER POLICY as to its use in payment and star ratings). CONVENTION covers the rest: weekly wound reassessment, photography under facility policy and wound society guidance, NDNQI prevalence surveys that separate hospital-acquired from community-acquired injuries, and Joint Commission expectations that risk is assessed and acted on. In litigation the record is read backward from the injury, and the Supreme Court's 2023 decision in Health and Hospital Corporation of Marion County v. Talevski, which held that residents may enforce certain Nursing Home Reform Act rights under 42 U.S.C. 1983, has raised the stakes on long-term care documentation; no published study measures how often the stage, the present-on-admission status, or the provider diagnosis is missing from audited wound records, and this page treats that gap as expert consensus rather than a statistic. Canada and Australia use the same classification with different reporting machinery, and the evidence explains why the plain-words depth description matters everywhere. In Canada, no federal or provincial statute names the NPIAP system; hospital coders work in ICD-10-CA, which keeps ulcer-based terminology in its L89 category and carries a Canada-specific code for suspected deep pressure-induced tissue damage under CIHI coding standards (LAW as to the mandated classification, CONVENTION as to the clinical description); CIHI's hospital harm measure includes a pressure ulcer group and its long-term care indicators track worsened pressure ulcers from interRAI assessments; Accreditation Canada has carried pressure injury prevention as a Required Organizational Practice, and the current wording should be confirmed in the standards set your organization is surveyed against (CONVENTION with accreditation force); the Registered Nurses' Association of Ontario best practice guideline and Wounds Canada's recommendations carry the clinical standard. In Australia, the NSQHS Comprehensive Care Standard (second edition) requires pressure injury prevention and wound management systems consistent with best-practice guidelines, comprehensive skin inspections within best-practice time frames, and patient information and equipment (actions 5.21 to 5.23; CONVENTION with accreditation force); the ACSQHC hospital-acquired complications list includes pressure injury, specified as stage 3, stage 4, unstageable, and suspected deep tissue injury, and feeds activity-based funding adjustments (PAYER POLICY); and coders record whether a condition arose during the episode with a condition onset flag under the Australian Coding Standards rather than a US-style POA indicator (LAW as to the classification). On the evidence: in an Australian tertiary hospital, Fulbrook and Lovegrove (2023) found that of 6,186 pressure injuries validated over four years, the category had been reported correctly in 67.3 percent, with moderate agreement (kappa 0.567); Karadag, Cakar, and Demir (2024) found near-perfect agreement among three nurse academicians staging 694 photographs (kappa 0.842 to 0.937), with the most frequent disagreements between deep tissue pressure injury and Stages 1 and 2; Kottner and colleagues' 2009 systematic review of 24 interrater studies found the literature too heterogeneous to recommend any classification system, and their home care study found 96 percent agreement on classification with most disagreements at the shallowest category; Oozageer Gunowa and colleagues (2018) reviewed 11 studies and concluded that people with darker skin tones are more likely to develop higher-stage injuries, most plausibly because early damage is not recognized. Those numbers are the reason to describe depth and tissue in words the next rater can verify, to assess darker skin by tone, warmth, and firmness rather than redness, and to treat the first stage assigned as a checked finding rather than a label. On versions: the 2016 revision is the current staging system; the International Guideline of NPIAP, EPUAP, and PPPIA reached its fourth edition, released online in chapters from 2025 and presented as the current 2026 guideline, with the 2019 edition remaining the reference for material not yet reissued, so a policy should name the edition it follows. The pressure injury stage definitions and the illustrations that accompany them are the copyrighted intellectual property of the National Pressure Injury Advisory Panel (formerly the National Pressure Ulcer Advisory Panel), revised in 2016; NPIAP's licensing page states that its materials may not be used without permission, that permission is requested through the NPIAP Product Permission Use Request Form and reviewed by the NPIAP Executive Committee, that an approved user signs a usage agreement and pays a licensing fee where applicable, that the materials may be used only after a signed contract is received, and that every use must carry the attribution "Used with permission from the National Pressure Injury Advisory Panel, copyright 20XX" together with a non-endorsement disclaimer in publications and electronic materials. The 2016 article by Edsberg and colleagues in the Journal of Wound, Ostomy and Continence Nursing is the peer-reviewed record of the revision and is available through PubMed Central, while the definitions and images themselves remain NPIAP property; the International Guideline belongs to its partner organizations NPIAP, EPUAP, and PPPIA; and the CMS coding guidelines, MDS, and OASIS instruments are US government works. Describing the depth of a wound in your own words, as this page does, is not a use of NPIAP materials, and facilities that want the definitions or illustrations in a policy, an electronic record, or a public tool should route the request through NPIAP. BastionGPT is not affiliated with, or endorsed by, any of these publishers. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common pressure injury staging documentation errors reviewers flag The numbers behind these errors are specific. In one Australian tertiary hospital, 6,186 validated pressure injuries over four years had been reported at the correct category only 67.3 percent of the time, with moderate agreement between reporting nurses and validators (kappa 0.567; Fulbrook and Lovegrove, 2023). Three nurse academicians staging 694 photographs reached near-perfect agreement (kappa 0.842 to 0.937) and still disagreed most often between deep tissue pressure injury and Stages 1 and 2 (Karadag, Cakar, and Demir, 2024). A systematic review of 24 interrater studies found the evidence too heterogeneous to recommend any classification system (Kottner and colleagues, 2009), and duplicate assessments in 691 home care clients agreed 96 percent of the time with most disagreements at the shallowest category. A review of 11 studies concluded that people with darker skin tones are more likely to develop higher-stage injuries, with early damage going unrecognized as the likely reason (Oozageer Gunowa and colleagues, 2018). No published audit counts how often a chart carries a nursing stage with no provider diagnosis, a present-on-admission claim with no admission inspection, or a reverse-staged wound; the coding, hospital-acquired condition, and survey rules described under compliance considerations are what turn those omissions into a payment or deficiency problem. The BastionGPT Clinical Advisory Board sees the same errors most often in pressure injury staging documentation reviews: Reverse staging. "Stage 4, now Stage 2, improving," or a closed wound recorded as Stage 1. The stage records the deepest damage the wound reached and never moves down; a healing wound fills with granulation and scar tissue, not the muscle, fat, or dermis it lost, which is the physiological basis of the NPUAP's 2000 position and of the MDS and OASIS rules. Describe healing by measurements and tissue, or with a validated healing tool, and write healed at the stage first assigned when it closes. A nursing stage with no provider diagnosis, or conflicting entries with no query. The flowsheet carries a stage, the provider's notes never mention the wound, or the nurse writes Stage 3 while the provider writes Stage 2 and nobody asks. The coder may take the stage from nursing documentation but must take the diagnosis from the provider and must query a conflict; a wound that reaches discharge with no provider diagnosis is uncoded, and a conflict left standing is the first thing a hospital-acquired condition review reads. Check the provider's current note before you sign. Present on admission asserted without evidence, or a local window written as a CMS rule. "POA" entered on the day the wound was first noticed, with an admission skin inspection that was late, incomplete, or silent about the site, or a note that a wound found within 24 hours is present on admission "per CMS." CMS defines present on admission by the time of the inpatient admission order and sets no timeframe; the evidence is the admission inspection, the emergency department record, and a dated photograph. When those are missing, document the earliest reliable finding and let the provider record the status as undetermined. Look-alikes staged as pressure injuries, and a blister misfiled. Incontinence-associated dermatitis charted as a Stage 2 pressure injury, a skin tear staged, a diabetic foot ulcer entered as a heel pressure injury, or a blood-filled blister called Stage 2 when a blood-filled blister signals a deep tissue pressure injury and only a clear fluid blister belongs at Stage 2. Each look-alike has its own etiology, classification, and codes; state the etiology and what you ruled out, and describe the blister's contents. Unstageable or deep tissue injury misread as a severity, or its reveal charted as a downgrade. "Unstageable, worse than Stage 4," or "debrided, downgraded to Stage 3." An unstageable wound is a full-thickness wound of unknown depth and a deep tissue pressure injury is damage that has not yet declared itself; when debridement or evolution reveals the base, document the date and the stage revealed as new information, and know that the coder assigns only the revealed stage for a debrided unstageable wound and both stages for a wound that progressed. Description gaps: unmethodical measurements, unnamed devices, staged mucosa, unconsented photographs. "About 3 cm, moderate depth," wound bed percentages that add to 130, a device injury with the device never named, a numeric stage on the nasal septum or another mucosal site, or an undated image with no consent recorded and "see photo" in place of a description. Measure by a stated method, make the percentages add to 100, name the device and the action taken, write not stageable for mucosa, and give every photograph a date, a ruler, an identifier, a consent line, and a written description beside it. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on wound care and skin integrity documentation. Give it the facts (the site and side, the etiology and any look-alike ruled out, the depth you saw in plain words, the date and time first identified and by whom, the admission order and inspection times with the emergency department or transfer record, the measurements and method, the wound bed percentages, exudate, edges, periwound, and pain, any device or mucosal detail, the photograph and consent, the baseline values, and the provider's current diagnosis) and it drafts the documentation block: location and etiology, one stage or category with its date, the present-on-admission determination with its evidence, the wound description, the qualifiers, the photograph line, the healing description with the stage unchanged, and the provider match and plan, ready for your review. Cross-check a finished entry for the gaps reviewers flag: a stage that moved down, a nursing stage with no provider diagnosis or a conflict with no query, a present-on-admission claim with no admission inspection behind it, a moisture lesion or skin tear staged as pressure, a numeric stage on mucosa, an unnamed device, percentages that do not add to 100, or a photograph with no date, consent, or written description. Turn the entry into the next document: the wound and present-on-admission line for the transfer of care summary or discharge summary, the query language for a provider whose note disagrees, the MDS Section M or OASIS pressure injury item support, or the plain-language explanation for a patient or family, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions What has to be in a pressure injury staging note? Seven things, for each wound separately. The anatomic site with its side and a body map reference, and the etiology, with any look-alike (moisture damage, skin tear, arterial, venous, or diabetic ulcer) ruled out in a phrase. One stage or category, the date and time it was first identified, who staged it, and the depth in plain words a reader could check against the wound. The present-on-admission determination with its evidence: the admission order time, the admission skin inspection time and finding, and the emergency department or transfer record. Measurements by a stated method with undermining and tunneling by clock face, the wound bed by tissue percentages that add to 100, exudate, edges, periwound skin, and pain. The device or mucosal qualifier where one applies. The photograph with its date, consent, and storage, beside a written description. And a healing description that uses the wound's characteristics or a validated tool while the stage stays where it was, followed by a check that the provider's note still names the same wound, site, and etiology. The blank template on this page carries those fields; the sample shows them filled in. Can a nurse assign the stage, or does a physician have to document it? And why does the code still say ulcer? A nurse can stage, and the provider must diagnose. Section I.B.14 of the ICD-10-CM Official Guidelines lists the pressure ulcer stage among the few code assignments that may rest on documentation by clinicians other than the patient's provider, observes that a nurse often documents the pressure ulcer stages, and then requires that the associated diagnosis be documented by the patient's provider; if the documentation conflicts, from the same clinician or different clinicians, the provider is queried. So a complete record needs both entries for the same wound, site, and etiology, and a wound with a nursing stage but no provider diagnosis cannot be coded at all. The word ulcer survives because the coding system did not follow the 2016 NPIAP change: category L89 identifies site and stage with stages 1 to 4, deep tissue pressure injury, unspecified stage, and unstageable, and the MDS and OASIS item names likewise keep ulcer or ulcer/injury. Write pressure injury in your clinical entry if your facility has adopted the NPIAP term, expect the coder to translate it, and never treat a provider's pressure ulcer and your pressure injury as a conflict; the conflict that matters is a different site, a different stage, or a different etiology. What does present on admission mean for a pressure injury, and is there a 24-hour window? Present on admission means present at the time the order for inpatient admission occurs; that is the definition in Appendix I of the ICD-10-CM Official Guidelines, and it counts conditions that develop in the emergency department, in observation, or in outpatient surgery before the order as present on admission. There is no 24- or 48-hour window in the definition, and the guidelines state that there is no required timeframe by which a provider must identify or document a condition as present on admission, because some conditions take days to recognize. A hospital rule that an injury found within a set number of hours is presumed present is a local convention, and quality measures may use their own time-based specifications; neither is the payment definition. The stakes are the Medicare hospital-acquired condition policy, under which stage 3 and stage 4 pressure ulcers coded as not present on admission (indicators N or U) earn no higher payment, and AHRQ's PSI 03, which counts stage 3, 4, or unstageable ulcers not present on admission. The evidence that carries the determination is the admission skin inspection with its time, the emergency department or transfer record, and a dated photograph; when the inspection was late or silent about the site, document the earliest reliable finding and the reason, and leave the provider to record Y, N, U, or W. My patient's Stage 4 injury is granulating and getting smaller. What stage do I chart now? Stage 4, described as healing. The stage records the deepest damage the wound reached, and a healing wound fills its defect with granulation and scar tissue rather than regrowing the muscle, fat, or dermis it lost, so a shallower-looking wound is not a shallower injury. The NPUAP said so in its 2000 position statement (in its words, "once a Stage IV always a Stage IV"), the RAI Manual for MDS 3.0 and the OASIS guidance forbid reverse staging, and the ICD-10-CM guidelines code a healing wound to its documented stage. Chart progress by what changed: length, width, and depth by the same method, undermining and tunneling, tissue percentages, exudate, edges, and pain, or a validated healing tool such as the PUSH tool, whose three parameters (surface area, exudate amount, tissue type) were built to replace reverse staging. When the wound closes, record it as healed at the stage it reached; if it reopens at the same site, it resumes that stage. If your electronic record forces a stage field at every entry, enter the original stage and use the free text for the healing description. How do I document an unstageable injury that is debrided, or a deep tissue pressure injury that opens up? As new information with a date, never as a downgrade or an upgrade. An unstageable wound is a full-thickness wound whose base is hidden, so when debridement exposes the base you document the date, what was revealed, and the stage now assignable, keeping the original finding and its date in the record; the ICD-10-CM guidelines then assign only the code for the stage revealed after debridement, and the RAI Manual keeps the wound's present-at-entry status at the stage revealed if it was present when the resident entered. A deep tissue pressure injury is damage that has not yet declared itself: chart the discoloration or blood-filled blister at first identification, reassess on schedule, and if the skin opens to expose the base, document the date and the stage revealed while the present-on-admission status stays as it was set at first identification. Two coding details to know without doing the coder's job: a deep tissue pressure injury has its own code family, and a wound that was present at admission and progressed to a higher stage during the stay takes two codes, the admission stage and the highest stage reached. Write the history in dated lines so both readings are possible from your entry. Is a blister a Stage 2? And how do I chart moisture damage, skin tears, device injuries, and mucosal injuries? A blister filled with clear fluid over a pressure point, intact or broken, is documented at Stage 2; a blood-filled blister signals a deep tissue pressure injury and is documented as one. Moisture-associated skin damage, including incontinence-associated dermatitis, is a moisture injury, is not a pressure injury, and is never staged; describe it, name the cause, and treat the cause, and when a wound has both moisture and pressure components say which one you are staging and why. A skin tear is mechanical trauma with its own classification, and arterial, venous, and diabetic foot ulcers have their own systems and codes; a heel ulcer in a patient with diabetes is a pressure injury only if the history and location say so. A medical device related pressure injury is staged by depth like any other, with the device named, its fit and hours in place recorded, and the action taken; that qualifier is what tells a reviewer the gap was the device bundle rather than the mattress. A pressure injury on mucous membrane (nasal septum, lip, tongue, urethra, and similar sites, almost always from a device) is documented as a pressure injury, described in words and measurements, and given no numeric stage, because mucosa has none of the skin layers the stages describe; MDS and OASIS do not stage it either. Do I need a photograph of every pressure injury, and what consent applies? No national rule requires one; photography follows your facility's policy and wound society guidance, and where it is done it is a health record entry. A wound photograph is protected health information, so it needs the consent your policy and state law require (written consent is the common convention, some organizations rely on a specific clause in the admission consent, and a patient may decline), and it belongs in the record, not on a personal phone. Chart that the photograph was taken, the date and time, that a ruler and patient identifier were in the frame, the consent obtained, and where the image is stored, and write the description in words in the same entry, because photographs flatten depth and shift color and a reviewer needs the measurements and tissue percentages beside the image. The one photograph that earns its place every time is the admission inspection photograph of any wound present, dated and labeled, because it is the clearest possible evidence for a present-on-admission determination and it anchors every later comparison. Can we put the NPIAP stage definitions or illustrations in our policy, our EHR, or on our website? Only with NPIAP's permission. The stage definitions and illustrations are the intellectual property of the National Pressure Injury Advisory Panel, and its licensing page states that NPIAP materials may not be used without permission: you complete the NPIAP Product Permission Use Request Form, the request is reviewed by the NPIAP Executive Committee, an approved user signs a usage agreement and pays a licensing fee where applicable, and the materials may be used only after a signed contract is received, with the attribution "Used with permission from the National Pressure Injury Advisory Panel, copyright 20XX" and, for publications and electronic materials, NPIAP's non-endorsement disclaimer. That applies to a policy document, an electronic record build, a pocket card, and a public web page alike, and being free, educational, or non-commercial does not create a license. What needs no permission is a description in your own words of what each stage looks like in terms of depth and tissue, a numeric stage field, and the documentation structure on this page; NPIAP's peer-reviewed 2016 article by Edsberg and colleagues can be cited as the source of the system. Wound photographs of your own patients are your records, subject to consent, not NPIAP's. Can BastionGPT draft my pressure injury staging documentation? Yes. Give it the facts (the site and side, the etiology and any look-alike ruled out, the depth you saw in plain words, the date and time first identified and by whom, the admission order and inspection times with the emergency department or transfer record, the measurements and method, the wound bed percentages, exudate, edges, periwound, and pain, any device or mucosal detail, the photograph and consent, the baseline values, and the provider's current diagnosis) and it drafts the full entry: location and etiology, one stage or category with its date, the present-on-admission determination with its evidence, the wound description, the qualifiers, the photograph line, the healing description with the stage unchanged, and the provider match and plan, ready for your review. It can also check a finished entry for a stage that moved down, a nursing stage with no provider diagnosis, a present-on-admission claim with no admission inspection behind it, a moisture lesion staged as pressure, a numeric stage on mucosa, an unnamed device, percentages that do not add to 100, or a photograph with no date, consent, or description. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: National Pressure Injury Advisory Panel, licensing content (materials may not be used without permission; Product Permission Use Request Form; Executive Committee review; usage agreement and fee; required attribution and non-endorsement disclaimer; accessed September 2026); Edsberg LE, Black JM, Goldberg M, McNichol L, Moore L, Sieggreen M, 2016, Journal of Wound, Ostomy and Continence Nursing 43(6):585-597, Revised National Pressure Ulcer Advisory Panel pressure injury staging system (April 2016 consensus conference; injury for ulcer; Arabic numerals; suspected removed from deep tissue pressure injury; definitions not reproduced here); National Pressure Ulcer Advisory Panel, 2000, The Facts about Reverse Staging in 2000 (position statement; PUSH tool); NPIAP, EPUAP, and PPPIA, International Pressure Injury Guideline (fourth edition released online in chapters; presented as the current 2026 guideline; 2019 edition linked for reference). CMS, ICD-10-CM Official Guidelines for Coding and Reporting, FY 2026 (Section I.B.14, documentation by clinicians other than the patient's provider, pressure ulcer stage among the exceptions, nurse often documents the stages, diagnosis by the provider, query on conflict; Section I.C.12.a, pressure ulcer stage codes, unstageable, healed, healing, evolving stage, pressure-induced deep tissue damage; Appendix I, present on admission definition, indicators, and no required timeframe; bare code numbers only); CMS, hospital-acquired conditions (Deficit Reduction Act section 5001(c); stage III and IV pressure ulcers; discharges on or after October 1, 2008); CMS, Hospital-Acquired Condition Reduction Program . Agency for Healthcare Research and Quality, August 2025, Patient Safety Indicator 03, Pressure Ulcer Rate, technical specification v2025 (stage 3 or 4 or unstageable, secondary diagnosis not present on admission, per 1,000 discharges of adults; exclusions for stays under three days, obstetric discharges, severe burns, exfoliative disorders, and deep tissue injury or unstageable ulcer present on admission at the same site). CMS long-term care and home health: eCFR, 42 CFR 483.25 quality of care (paragraph (b)(1), pressure ulcers; F686); CMS, MDS 3.0 Resident Assessment Instrument manual (Section M, items M0210 and M0300 A through G; present at entry; no reverse staging; unstageable wounds debrided to a visible stage); CMS, OASIS data sets (items M1306, M1311, M1322, M1324; current item set and guidance manual); CMS, State Operations Manual Appendix PP guidance for nursing homes (F686 interpretive guidance; avoidable and unavoidable). Reliability and skin tone evidence: Fulbrook P, Lovegrove J, 2023, Journal of Clinical Nursing, reporting accuracy of pressure injury categorisation in an acute tertiary hospital: a four-year analysis (6,186 injuries; 67.3 percent correct; kappa 0.567); Karadag A, Cakar V, Demir AS, 2024, Journal of Tissue Viability 33(4):786-791, an inter-assessor reliability study on the categorization and staging of pressure injuries (694 photographs; kappa 0.842 to 0.937); Kottner J, Raeder K, Halfens R, Dassen T, 2009, Journal of Clinical Nursing 18(3):315-336, a systematic review of interrater reliability of pressure ulcer classification systems (24 studies); Kottner J, Halfens R, Dassen T, 2009, International Journal of Nursing Studies 46(10):1307-1312, interrater reliability of Braden risk and pressure ulcer classification in home care (96 percent agreement; kappa 0.81 and 0.79); Oozageer Gunowa N, Hutchinson M, Brooke J, Jackson D, 2018, Journal of Clinical Nursing 27(17-18):3266-3275, pressure injuries in people with darker skin tones: a literature review (11 studies). Professional guidance: Wound, Ostomy, and Continence Nurses Society, wocn.org (pressure injury and wound photography guidance); AHRQ, Preventing Pressure Ulcers in Hospitals toolkit ; NDNQI pressure injury prevalence methodology as described by Press Ganey (hospital-acquired versus community-acquired classification). Canada: Registered Nurses' Association of Ontario, assessment and management of pressure injuries for the interprofessional team, third edition (2016); Wounds Canada, best practice recommendations ; Canadian Institute for Health Information, hospital harm results and ICD-10-CA and CCI classifications ; Accreditation Canada, Required Organizational Practices (confirm the current pressure injury prevention wording in the applicable standards set). Australia: Australian Commission on Safety and Quality in Health Care, NSQHS Comprehensive Care Standard (actions 5.21 to 5.23, preventing and managing pressure injuries) and hospital-acquired complications list (pressure injury: stage 3, stage 4, unstageable, suspected deep tissue injury, as both research briefs report; page unreachable on September 5, 2026); Independent Health and Aged Care Pricing Authority, ihacpa.gov.au (Australian Coding Standards, condition onset flag; activity-based funding adjustments for hospital-acquired complications). ## Related templates Braden Scale Documentation OASIS Assessment Documentation Incident & Adverse Event Report Transfer-of-Care Summary Discharge Summary Care Coordination Note Consultation Response & Report Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/cam-icu All note templates # CAM-ICU Documentation: RASS Gate, Features, Result & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The CAM-ICU (Confusion Assessment Method for the Intensive Care Unit) is a bedside delirium screen, adapted by Ely and colleagues in 2001 from Inouye's Confusion Assessment Method, that assesses four features once a RASS shows the patient is arousable. ICU and step-down nurses use it every shift to flag delirium for the provider who diagnoses it. This page covers how to document CAM-ICU results, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it Bedside ICU and step-down nurses at each screen, with intensivists, hospitalists, and advanced practice providers documenting the diagnosis when one is made; no license, training certificate, or publisher qualification level applies, and the rights holder permits implementation and clinical use without written permission Audience Relieving nurses at handoff, intensivists and hospitalists deciding whether delirium is present, pharmacists reviewing sedatives and other deliriogenic medications, respiratory therapists timing awakening and breathing trials, coders and documentation integrity staff who take the diagnosis from the provider's note, quality staff counting ICU Liberation assessments, and later reviewers Typical length 2 to 5 chart lines per screen (RASS with time, features assessed, result and subtype or the reason it could not be done, contributors and action) · the screen itself takes about 2 minutes at the bedside Format family Clinician-administered delirium screen with a three-way result (positive, negative, or unable to assess) built on four features and gated by a sedation scale; nonverbal responses allow use in intubated patients When it's used Every shift and with any change in mental status in adult ICU patients, before and after spontaneous awakening trials, in step-down units that keep ventilated or recently extubated patients, in trauma and neurological ICUs with the caveats their protocols state, and, in the pCAM-ICU and psCAM-ICU versions, in children's ICUs Standards context Ely and colleagues (Critical Care Medicine and JAMA, 2001) from Inouye's CAM (1990); copyright 2002 E. Wesley Ely and Vanderbilt University, unrestricted for clinical use with the copyright line; named by the SCCM PADIS guideline and the ICU Liberation bundle, mandated by no law; described here for documentation, no worksheet or stimuli reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the CAM-ICU? The Confusion Assessment Method for the Intensive Care Unit is a bedside delirium screen that E. Wesley Ely and colleagues at Vanderbilt adapted from Sharon Inouye's Confusion Assessment Method (Annals of Internal Medicine, 1990) so that it could be used in patients who cannot speak, including patients on a ventilator. Two validation papers appeared in 2001: in Critical Care Medicine, 38 medical and coronary ICU patients and 293 paired daily evaluations, with two study nurses reaching sensitivities of 95 and 96 percent and specificities of 93 percent against a delirium expert applying DSM-IV criteria; and in JAMA, 111 mechanically ventilated patients and 471 paired evaluations, with sensitivities of 100 and 93 percent, specificities of 98 and 100 percent, a kappa of 0.96 between the two nurses, and a screen that took about two minutes. The screen keeps the CAM's four-feature logic. Feature 1 asks whether the patient's mental status has changed acutely from baseline or fluctuated over the past 24 hours. Feature 2 asks whether the patient can attend, tested with a short auditory task the patient answers by hand squeeze, with a visual recognition task as the backup when the auditory task cannot be used. Feature 3 is altered level of consciousness, read straight from the current RASS: any value other than zero counts as present. Feature 4 is disorganized thinking, tested with a few simple yes-or-no questions and a simple command. The screen is positive when Features 1 and 2 are both present and either Feature 3 or Feature 4 is present; when Feature 1 or Feature 2 is absent the screen cannot be positive and the assessment stops, and when Features 1, 2, and 3 are present Feature 4 does not need to be done. Features 3 and 4 swapped numbers when the materials were revised in 2014 (altered consciousness became Feature 3) with no change in content, so older pocket cards and flowsheets label them the other way round. The worksheet, the flowsheet, the letters, pictures, questions, and commands, and the training manual are Vanderbilt's copyrighted materials and are not reproduced here; the CIBS Center's site carries the revised flowsheet and worksheet (the 2023 files), the training manual (updated August 2016), the FAQ (revised March 2014), and translations in more than 20 languages. The load-bearing fact for documentation is that the CAM-ICU is a screen with three possible results, and only one of them is a delirium finding. Positive means the features of delirium were present at that assessment; negative means they were not; unable to assess means the patient was too deeply sedated or comatose to test content of consciousness, which Vanderbilt's FAQ defines as coma or stupor rather than delirium or normal and restricts to patients at RASS -4 or -5. The RASS therefore comes first at every screen: at the two deepest values the screen is recorded as unable to assess, and most units treat -3 as the deepest value at which it can still be done, some -2. A positive screen is not a diagnosis. Delirium is a medical diagnosis, made and documented by the physician or advanced practice provider from the whole picture (the screen, the sedation and medication history, the examination, and the search for causes), and the ICD-10-CM Official Guidelines assign codes from the provider's documentation, not from a nursing screen. The screen is also weaker at the bedside than in the validation studies: in ten Dutch ICUs, routine nursing CAM-ICUs reached a sensitivity of 47 percent and a specificity of 98 percent against expert assessment (van Eijk and colleagues, 2011), and pooled estimates sit at about 80 to 84 percent sensitivity and 95 to 96 percent specificity, so a negative screen in a patient who looks different to the family or the nurse is a reason to look again, not a reason to stop. Neighbors are charted under their own names: the ICDSC (Bergeron and colleagues, 2001) is an eight-item checklist scored from observations over a period of care rather than at one moment and is the other tool the SCCM guidelines accept; the CAM-ICU-7 (Khan and colleagues, 2017) turns the feature results and the RASS into a 0 to 7 severity score; the pCAM-ICU (Smith and colleagues, 2011) and the psCAM-ICU (Smith and colleagues, 2016) are the pediatric versions; the brief CAM and the Delirium Triage Screen (Han and colleagues, 2013) are the emergency department adaptations; and the Glasgow Coma Scale measures consciousness after brain injury, not delirium. ## Who uses CAM-ICU documentation and when Critical care nurses chart nearly all of it: a screen every shift under the PADIS guideline's once-per-shift convention, another with any change in mental status, and one before and after each spontaneous awakening trial, each with the RASS that gated it. Intensivists, hospitalists, and advanced practice providers read the results on rounds and write the diagnosis when the screen, the history, and the examination support one; pharmacists read a positive screen against the medication record for benzodiazepines, anticholinergics, and opioid dosing; respiratory therapists time the breathing trial to the awakening trial that makes the screen possible; and quality staff count documented assessments for the ICU Liberation bundle. Step-down nurses use it for ventilated or recently extubated patients who have moved out of the ICU, trauma and neurological ICUs use it with the caveat their protocols state for severe brain injury, and children's ICUs use the pCAM-ICU or psCAM-ICU. The result travels: a transfer-of-care summary should carry the last RASS and CAM-ICU with their times and the next screen due, an incident report after a self-extubation or a fall should record the screen results around the event, a capacity evaluation in the ICU starts with whether delirium is present, and a discharge summary should name an ICU delirium episode so that cognition is followed after discharge. Neighbors win in four places: the ICDSC when the unit has standardized on it, the brief CAM or the 4AT in the emergency department and on general wards where the CAM-ICU was not the validated choice, a cognitive screen such as the Mini-Cog or MoCA once delirium has cleared and the question becomes baseline cognition, and the CIWA-Ar when the question is alcohol withdrawal severity rather than whether delirium is present. ## How to document CAM-ICU results in the chart No law, payer, or guideline prescribes a CAM-ICU note format; what exists is a copyrighted screen whose authors recommend documenting it in the hourly nursing flowsheet and, where there is room, at the feature level, a guideline convention of once per shift and as needed, bundle metrics that count documented assessments, and unit protocols that fix the terms and the interval. What survives review is an entry that records the RASS with its time before the screen, names the baseline and its source, records each feature assessed and where the assessment stopped, states the result in the unit's fixed terms with the time and a reason whenever it could not be done, adds the subtype and the contributors reviewed when the screen is positive, records the action and the notification, and leaves the diagnosis to the provider's note while the screen trends toward resolution. Each element below carries the pitfall that most often undermines it. RASS first, with its time. Open every entry with the RASS value and clock time that gated the screen, and with the stimulus if the value is ambiguous (the RASS page covers that assessment). At -4 or -5 the patient is comatose or too deeply sedated to test content of consciousness: record the CAM-ICU as unable to assess with that reason and set the next attempt, usually after the awakening trial or when the RASS lightens. At -3 and lighter the screen proceeds; Vanderbilt's FAQ notes that most sites use -3 as the lower border and some -2, so name the unit's border in the protocol rather than in each entry. Chart the two values as a pair every time, because Feature 3 is read from that RASS and a reviewer cannot judge a result without it. Pitfall: A CAM-ICU negative at 06:00 beside a RASS of -5 at 06:00, or "UTA" on an awake patient who was simply hard to test. Baseline mental status and Feature 1. Feature 1 needs a baseline to compare against: write where it came from (family or caregiver, the referring unit, a prior admission, the primary care record) and what it was, including known dementia, hearing or vision loss, language, and a stroke that has left a new permanent baseline. Then state whether there has been an acute change from that baseline or fluctuation over the past 24 hours, drawing on the night entries, handoff, and the family's account; a family report of confusion the nurse did not see counts and is charted as collateral. Feature 1 evaluates the day and Feature 3 the moment: a patient who fluctuated overnight and is now alert at RASS 0 is Feature 1 present and Feature 3 absent, and the entry should read that way. Pitfall: No baseline anywhere, so a patient with dementia is charted as acutely changed, or a night of fluctuation is erased by a calm 08:00 assessment. The features assessed, and where the assessment stopped. Record each feature assessed as present or absent, and say which were not assessed because the rule made them unnecessary: if Feature 1 or Feature 2 is absent the screen stops and is negative; if Features 1, 2, and 3 are present the screen is positive without Feature 4; Feature 4 is needed only when Features 1 and 2 are present and the patient is at RASS 0. For Feature 2 note which attention task was used and, if the visual backup was needed, why (hearing loss, language); for Feature 4 note when the questions alone were scored because the patient could not move an arm. Most flowsheets store only the overall result; Vanderbilt's FAQ says feature-level documentation helps compliance and accuracy and gives chart review something to audit, and the note of which features were tested is what makes a positive reproducible. Pitfall: A bare "CAM+" with no feature results, or all four features charted as tested on a patient whose screen should have stopped at Feature 2 absent. Result, time, subtype, and the reason when it could not be done. Write the result in the unit's fixed vocabulary (positive or negative, yes or no, present or absent, with unable to assess as the third value) and the clock time, and do not mix vocabularies across shifts. Unable to assess always carries its reason: RASS -4 or -5 with the sedation or the coma named, pharmacologic paralysis, or another stated barrier; it is never a synonym for negative and never a way to record a difficult patient. When positive, add the motoric subtype by the RASS the way the unit protocol defines it, usually hypoactive at a negative or zero RASS and hyperactive or mixed at a positive one, because hypoactive delirium is the common form and the one most often missed: in 614 medical ICU patients, purely hyperactive delirium was 1.6 percent of delirium, hypoactive 43.5 percent, and mixed 54.9 percent (Peterson and colleagues, 2006). Chart the prior result and its time beside the new one so the fluctuation is visible. Pitfall: UTA with no reason, or a positive with no subtype, so three quiet days of hypoactive delirium are never named. Contributors reviewed. A positive screen is a prompt to look for causes, and the entry should show that the look happened: sedatives and benzodiazepines on board or given overnight, anticholinergics and other deliriogenic drugs, opioid dosing against the pain score and its tool, sleep and the night's interruptions, oxygenation, infection markers and the team's workup, electrolytes and glucose, restraints, catheters and lines, and whether the hearing aids and glasses are in. Name the pharmacist's review when one was made, and the alcohol or sedative withdrawal question when the history raises it (the CIWA-Ar owns that assessment). The list is the nursing half of a differential the provider completes; it is not a diagnosis. Pitfall: A positive screen followed by "delirium precautions" with nothing reviewed, or a contributor list that repeats unchanged for four shifts. Action, notification, and the family. Record what was started or changed: reorientation, sleep protection, mobility with the therapist, devices removed, family presence, the medication changes made under orders, and the awakening and breathing trials that day, which are the bundle elements a positive screen is meant to trigger. Record who was notified, at what time, and what they decided, including a decision to make no drug change. No guideline requires a medication in response to a positive screen: the SCCM PADIS guideline (2018) suggests against routine antipsychotics for treatment, and the 2025 focused update was unable to recommend for or against them, so any antipsychotic given is charted against a specific indication such as distressing or dangerous agitation, with the order and the response, rather than against the screen result. Note the family conversation and the information given. Pitfall: A positive screen with no action and no notification time, or an antipsychotic charted as the response to a positive CAM-ICU with no indication stated. Diagnosis, trend, and handoff. Keep the two documents in their lanes. The nursing entry reports the screen; the provider's progress note states the diagnosis (delirium, its likely contributors, and the plan) and is the note a coder can use, because the ICD-10-CM Official Guidelines base code assignment on the provider's documentation and list no delirium screen among the exceptions that allow another clinician's note. Chart the results serially with times so the course is visible, repeat the screen when a UTA patient lightens, and, if the unit follows Vanderbilt's convention that a patient is free of delirium after 24 hours of negative screens, say so with the times. At handoff and transfer carry the last RASS and CAM-ICU with their times, the subtype, the contributors still open, the medication changes, and the next screen due; the transfer-of-care summary is where that line lives. Pitfall: A nursing note that reads "patient is delirious" as a diagnosis, or a handoff of "CAM negative" with no time, no RASS, and no next screen due. ### Blank template (copy and adapt) CAM-ICU DOCUMENTATION BLOCK (screen entry with gate and context) Date: [ ] Time: [ ] Setting: [ICU / step-down / trauma or neuro ICU / PICU] Nurse: [ ] Version: [CAM-ICU / pCAM-ICU / psCAM-ICU] Unit terms: [positive, negative, UTA / yes, no, UTA] RASS at the screen: [ ] Time: [ ] (if -4 or -5: CAM-ICU unable to assess, see below) Baseline mental status: [ ] Source: [family / referring unit / prior record / primary care] Known dementia, hearing or vision loss, language, prior stroke: [ ] Feature 1, acute change or fluctuation over 24 hours: [present / absent] Evidence: [night entries, handoff, family account] Feature 2, inattention: [present / absent] Task used: [auditory / visual backup, reason] Feature 3, altered level of consciousness (current RASS other than 0): [present / absent] Feature 4, disorganized thinking: [present / absent / not needed] [questions alone: reason] Result: [positive / negative / unable to assess] Time: [ ] Unable to assess, reason: [RASS -4 or -5 with sedation or coma named / paralysis / other] Subtype if positive (by RASS, per unit protocol): [hypoactive / hyperactive / mixed] Prior result and time: [ ] Pattern over 24 hours: [ ] Contributors reviewed: [sedatives, benzodiazepines / anticholinergics / opioids and pain score with tool / sleep / oxygenation / infection / electrolytes, glucose / restraints, catheters, lines / hearing aids, glasses] Pharmacist review: [who, time, findings] Action: [reorientation / sleep protection / mobility / devices removed / family presence / medication changes under orders / awakening and breathing trial today] Notification: [provider, time, decision] Family: [present, informed, baseline confirmed] Provider diagnosis: [documented in the provider's note / not established] (the screen is not a diagnosis) Next screen due: [ ] Handoff line: [last RASS and CAM-ICU with times, subtype, open contributors, medication changes, next due] Nurse signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the worksheet, flowsheet, stimuli, or training manual. ## Sample CAM-ICU documentation (fictional) Scenario: a US community teaching hospital's surgical ICU, day two after an emergency laparotomy in a 72-year-old man who is still ventilated; the day nurse charts an unable-to-assess entry at deep sedation, a positive screen after the awakening trial with the features assessed in sequence, the contributors reviewed with the pharmacist, the notification, the intensivist's separate diagnosis, an evening reassessment, and the run of negative screens that follows. All details are fictional. Patient: H.B., 72 · Setting: Surgical ICU, community teaching hospital; day 2 after emergency laparotomy · Clinician: M. Okafor, RN, CCRN · Note date: 09/16/2026 Baseline, gate, and an unable-to-assess entry (06:10): CAM-ICU as built in the unit flowsheet (results recorded as positive, negative, or UTA), assessed by M. Okafor, RN, on 09/16/2026. Baseline from his daughter by phone at 06:30 and from the admission history: independent at home, managing his own medications and finances, no diagnosis of dementia, wears hearing aids in both ears (at home overnight, brought in this morning), reading glasses. Night entries: CAM-ICU negative at RASS 0 at 20:15 on 09/15/2026; RASS -1 to -2 after midnight; RASS +2 at 02:10 while pulling at his gastric tube, then RASS -3 at 02:40 after a propofol bolus, with no screen attempted between; propofol increased at 03:30 for ventilator dyssynchrony. 06:10: RASS -4, no eye opening to his name, movement to physical stimulation only. CAM-ICU: unable to assess, reason deep sedation on propofol 40 mcg/kg/min; not recorded as negative. CPOT 0. Next attempt set for after the spontaneous awakening trial planned with the respiratory therapist (K. Brandt, RRT) at 08:30, under the unit protocol (SICU-DEL-02), which sets a screen every shift, with any change in mental status, and after each awakening trial. Awakening trial, the gate passed, and the screen (09:40): Propofol stopped at 08:30 for the awakening trial; fentanyl 50 mcg/h continued. 09:35: RASS -1, opens his eyes to his name and holds contact briefly, then drifts; no physical stimulation needed, so the screen proceeds. Hearing aids in place from 09:00. Feature 1 present: acute change from an independent, oriented baseline, and fluctuation over the past 24 hours (the 02:10 agitation, the sedated interval, the current drowsiness). Feature 2 present: inattention on the auditory attention task by hand squeeze; the visual backup was not needed with the hearing aids in. Feature 3 present: current RASS -1. Feature 4 not assessed: not needed once Features 1, 2, and 3 were present. Result at 09:40: CAM-ICU positive. Subtype recorded as hypoactive under the unit protocol (positive screen at a negative RASS, quiet, little spontaneous movement). Prior result: UTA at 06:10; last completed screen before that: negative at RASS 0 at 20:15 on 09/15/2026. Contributors reviewed and actions (09:45 to 11:30): Pain: CPOT 3 with grimacing on turning at 09:45, fentanyl 25 mcg bolus under the pain order, CPOT 1 at 10:15. Sedation: propofol off since 08:30 and not restarted (RASS -1 to 0 through the morning against the ordered target of -1 to 0 written 09/15/2026); midazolam 2 mg intravenous given three times overnight for agitation per the medication administration record; diphenhydramine 25 mg on the bedtime order given at 22:30. Pharmacist review (R. Delgado, PharmD) at 10:20: midazolam and diphenhydramine flagged as deliriogenic, with a recommendation to the intensivist to discontinue both and to treat agitation with pain reassessment first. Sleep: interrupted every hour overnight per the flowsheet. Oxygenation: SpO2 94 to 97 percent on FiO2 0.40. Infection: afebrile, white cell count and cultures reviewed by the team on rounds; sodium 133 and glucose 162 noted for the team. Devices: urinary catheter day 2, reviewed for removal; two peripheral lines. Actions: hearing aids and glasses in; whiteboard updated with the date, the place, and the daughter's visit time; blinds opened; reorientation at each contact; physical therapy sat him at the edge of the bed at 11:10; daughter at the bedside from 11:00 and told what a positive screen means and does not mean. Spontaneous breathing trial started 09:50 by K. Brandt, RRT, and passed at 10:20; extubation decision deferred to rounds. Notification and the provider's diagnosis: Intensivist (Dr. S. Nakamura) notified at 09:50 of the positive screen, the RASS, the features, and the pharmacist's findings; at 10:40 she discontinued midazolam and diphenhydramine, ordered melatonin at bedtime, kept the sedation target at -1 to 0 with no continuous sedative, and asked for the screen to be repeated on the evening shift. Her progress note for 09/16/2026 documents the diagnosis, hypoactive delirium, postoperative and multifactorial, with sedative exposure, sleep disruption, and pain named as contributors, and the plan above. This nursing entry reports the screen and the actions; the diagnosis, and the code that follows from it (F05), come from her note and not from the flowsheet. No antipsychotic was ordered or requested; the intensivist recorded that the indication for one was absent. Evening reassessment (20:30): Extubated at 15:40 after rounds, on nasal cannula. RASS 0 at 20:30, awake and calm, no sedative running. Feature 1 present: fluctuation over the past 24 hours. Feature 2 present: inattention on the auditory task, hearing aids in. Feature 3 absent: RASS 0. Feature 4 assessed because Feature 3 was absent: present, with disorganized answers to the questions and the command not followed. Result at 20:30: CAM-ICU positive, hypoactive by the protocol's RASS rule (RASS 0, quiet). CPOT 1; melatonin given at 21:00; lights dimmed and care clustered for the night; daughter reoriented him before leaving at 21:30. Night nurse to screen again with any change in behavior and at the start of the next shift. Resolution and handoff (09/17/2026 to 09/18/2026): 09/17/2026 08:15: RASS 0; Feature 1 present (fluctuation within the past 24 hours); Feature 2 absent, so the screen stopped: CAM-ICU negative. 20:20: RASS 0; Feature 1 present; Feature 2 absent: negative. 09/18/2026 08:05: RASS 0; Feature 1 absent (no change or fluctuation in the past 24 hours); screen stopped: negative, and 24 hours of negative screens recorded, the unit's convention for delirium resolved, with the intensivist documenting resolution in the 09/18/2026 progress note. Screening continues every shift while he remains in the unit. Handoff line at 18:45 on 09/18/2026: RASS 0 at 18:30; CAM-ICU negative at 08:05, next due 20:00; hypoactive delirium 09/16/2026 to 09/17/2026, resolved per the provider's note; midazolam and diphenhydramine discontinued, melatonin nightly; hearing aids and glasses at the bedside; daughter visits at 11:00 and 19:00. Every result in this record sits beside the RASS and the time that gated it, and no result was entered as negative when the screen could not be done. This sample is fictional and for educational purposes. It does not describe a real patient or record; the values, times, doses, dates, and details are invented to show documentation structure and are not clinical guidance. No worksheet, stimuli, or training manual text is reproduced. ↑ Back to the template and downloads ### Why this sample works Every result sits beside the RASS and time that gated it, and the 06:10 entry is unable to assess with its reason rather than negative, so the day can be reconstructed and the ICU Liberation count is honest. The baseline and its source are named before Feature 1 is called present, and the night's fluctuation is carried into the morning screen instead of being lost at a calm assessment. Each feature assessed is recorded with where the assessment stopped (Feature 4 not needed in the morning, needed in the evening at RASS 0), which is what makes both positives reproducible and auditable. The contributors are specific (named sedatives, a pain score with its tool, sleep, devices, sensory aids), the pharmacist's review is timed, and the actions are the bundle elements a positive screen exists to trigger, with the notification and the provider's decisions timed. The nursing entry reports the screen while the provider's note carries the diagnosis and the code, no antipsychotic is charted against a screen result, and the handoff carries the last RASS and result with times, the next screen due, and the medication changes. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the CAM-ICU or sets a delirium screening interval, and the rules that bind sit one step from the screen. Under LAW, the Medicare Conditions of Participation require a medical record with nursing notes, medication records, and the information needed to monitor the patient's condition, with entries complete, dated, timed, and authenticated (42 CFR 482.24), and a nursing assessment and care plan for every inpatient (42 CFR 482.23); a CAM-ICU entry meets them when a reader can see what was assessed, at what RASS, at what time, with what result and response. The FY 2026 ICD-10-CM Official Guidelines (Section I.B.14) base code assignment on the documentation of the patient's provider and list the exceptions where another clinician's note may be used (body mass index, pressure ulcer stage, coma scale, the NIH stroke scale, and a few others); a delirium screen is not among them, so the diagnosis behind F05 has to be in the physician's or advanced practice provider's note, and a nursing flowsheet reading positive does not create it. Under PAYER POLICY, CMS's Age Friendly Hospital Measure, adopted in the FY 2025 inpatient prospective payment rule for the Hospital IQR Program beginning with the calendar year 2025 reporting period and the FY 2027 payment determination, is a structural attestation measure with five domains; its Frailty Screening and Intervention domain asks hospitals to attest that they screen patients 65 and older for mentation (delirium and cognitive impairment), mobility, and malnutrition with validated instruments and act on positive screens, and the first attestations were due through the Hospital Quality Reporting system by May 15, 2026 (CMS specifications and attestation guide, July 2025). It names no instrument and sets no interval; failure to report carries the IQR consequence and the attestation is the deliverable, so a unit's CAM-ICU process is evidence for the attestation rather than a measure in its own right, and no CMS national coverage rule, NCCI edit, or Hospital IQR outcome measure keys payment to a CAM-ICU result. The American College of Surgeons Geriatric Surgery Verification program carries delirium screening for older surgical patients as a program standard (CONVENTION with verification force); a June 2026 study in the Journal of the American College of Surgeons found 94.3 percent of older surgical patients screened at verified hospitals against 52.5 percent elsewhere, with similar positive rates, which is the measurable cost of not screening. Under CONVENTION, the SCCM PADIS guideline (Devlin and colleagues, 2018) states as a good practice statement that critically ill adults should be regularly assessed for delirium using a valid tool, names no frequency, and suggests against routine antipsychotics for prevention and for treatment (conditional, low quality evidence); Vanderbilt's FAQ carries the guideline convention as every shift (every 8 to 12 hours) and as needed, with a screen before and after each awakening trial because Patel and colleagues (2014) found that delirium which resolved within two hours of stopping sedatives, 12 percent of 102 patients, carried a prognosis like no delirium while persistent delirium did not. The 2025 focused update (Lewis and colleagues) was unable to recommend for or against antipsychotics for treatment, and the MIND-USA trial behind that shift randomized 566 delirious patients across 16 centers, 89 percent of them hypoactive, and found median days alive without delirium or coma of 7.9 with haloperidol, 8.7 with ziprasidone, and 8.5 with placebo (P 0.26) (Girard and colleagues, 2018). The SCCM ICU Liberation data collection manual (2025) turns the convention into a metric: D element compliance is at least two documented delirium assessments per 24 hours with the CAM-ICU or ICDSC, a finding of unable to assess is considered aligned with the goal, and patients in coma or at RASS -4 or -5 are excluded, which is the definition a quality team should cite before deciding whether a UTA counts as screened, because other programs define it differently. Vanderbilt's own trauma delirium guideline (revised June 2026) states that delirium is a clinical diagnosis made by the team from the injuries, GCS, RASS, and CAM-ICU together, that the CAM-ICU cannot be assessed at RASS -4 or -5 and should only be reported as unable to assess, and that in severe traumatic brain injury the screen is not solely diagnostic. The Joint Commission has no CAM-ICU-specific standard. Three habits deserve the same label: a once-daily screen as the standard, a sedated patient charted as negative, and an antipsychotic as the automatic response to a positive screen are unit habits, and no authority above states any of them. Canada and Australia name a tool only by example and set no interval, and the version and accuracy boundaries hold everywhere. In Canada, no federal or provincial LAW names the CAM-ICU; the Canadian Coalition for Seniors' Mental Health national guidelines on the assessment and treatment of delirium (2006, with a 2014 update, and a 2010 guideline for delirium at the end of life) are CONVENTION for older adults, and no delirium-specific Accreditation Canada Required Organizational Practice was located for this page. The clearest critical care guidance is provincial: Critical Care BC's ICU Liberation bundle (Provincial Health Services Authority, 2025; CONVENTION) states that delirium should be routinely monitored in all ICU patients with a validated tool, the CAM-ICU or the ICDSC, at least once a shift and more often with any change in mental status, that delirium is reported on daily interdisciplinary rounds with the medications and environmental factors that may be contributing, that the first step in a positive patient is to identify the cause before any drug, and that no data support routine antipsychotics for prevention. Health Standards Organization's HSO 11001:2026 Critical Care Services was in public review as a draft dated March 2026 as of September 2026 and asks for validated assessment tools with reassessment intervals defined by the organization, naming no delirium tool, so it should be cited as a draft. In Australia the frame is firmer. The ACSQHC Delirium Clinical Care Standard (2016, revised September 2021, with assessments against the revised standard from 1 September 2022; CONVENTION with accreditation force) carries eight quality statements, asks for a validated tool rather than naming one, and states that routine antipsychotic use is not recommended; action 5.29 of the NSQHS Comprehensive Care Standard (an accreditation requirement for Australian hospitals) requires routine screening for cognitive impairment with a validated tool for patients 65 and over, for patients of any age at risk of delirium, and when family or carers raise concerns, with results documented and communicated, and states that a positive score on a screening tool is a prompt for further assessment, not a diagnosis. Delirium is hospital-acquired complication 11 on the national list, with a 2019 to 2020 rate of 35.7 per 10,000 admissions quoted by the Commission; IHACPA has applied a funding reduction to episodes with a hospital-acquired complication since July 2018 (PAYER POLICY), and its 2026 to 2027 pricing documents state that under clause A90 of the Addendum to the National Health Reform Agreement 2026 to 2031 the complication penalties are shadow priced rather than applied while the Commission reviews the safety and quality measures, so an Australian ICU should describe the current year as shadow priced rather than penalized. Three boundaries apply in every country. Version: the pCAM-ICU (Smith and colleagues, 2011) was validated in 68 children aged at least 5 years, chronologically and developmentally, across 146 paired assessments (sensitivity 83 percent, specificity 99 percent, kappa 0.96), and the psCAM-ICU (Smith and colleagues, 2016) in 300 children aged 6 months to 5 years across 530 paired assessments (sensitivity 75 percent, specificity 91 percent, kappa 0.79), so a children's ICU names the version and does not import adult conventions; the CAM-ICU-7 is a severity score, not the screen; and the brief CAM and Delirium Triage Screen belong to the emergency department (Han and colleagues, 2013: 406 patients, the triage screen 98 percent sensitive, the brief CAM 84 percent sensitive and 96 percent specific in physicians' hands). Accuracy: research nurses reached sensitivities of 93 to 100 percent in the 2001 studies, pooled estimates are 80.0 percent sensitivity and 95.9 percent specificity across nine studies and 969 patients (Gusmao-Flores and colleagues, 2012) and 0.84 and 0.95 across 29 studies (Chen and colleagues, 2021, who found accuracy shaped by the share of hypoactive delirium, ICU type, and mechanical ventilation), and routine nursing screens in ten Dutch ICUs reached 47 percent sensitivity with 98 percent specificity (van Eijk and colleagues, 2011), so a negative screen rules out less than the validation papers suggest, and the ICDSC (Bergeron and colleagues, 2001; eight items, predicted sensitivity 99 percent and specificity 64 percent in the original study) trades in the opposite direction. Population: the FAQ states the screen is valid with and without dementia but harder, and asks the family whether the patient could do the tasks at baseline; it was validated in 129 post-stroke patients (sensitivity 76 percent, specificity 98 percent) and the authors ask neurological ICUs to read a positive as symptoms of delirium rather than definitely delirium; a blind patient or one who cannot move an arm is scored on the questions alone for Feature 4; and a language barrier is met with one of the more than 20 translations or charted as the reason the screen could not be completed, never as negative. The CAM-ICU and its educational materials are copyrighted: the copyright line is Copyright © 2002, E. Wesley Ely, MD, MPH and Vanderbilt University, all rights reserved, and it appears on the revised flowsheet and worksheet as well as the 2016 training manual and the 2014 FAQ. Vanderbilt's Critical Illness, Brain Dysfunction, and Survivorship Center states that it has deliberately made the materials unrestricted in terms of use, asks that the copyright line appear at the bottom of pocket cards and other educational materials, does not require a written letter of permission for implementation and clinical use, and asks anyone who wants to use the materials for other uses to contact [email protected] for permission; as of September 2026 its public pages publish no separate terms for EHR builds, publications, or commercial products, so those are the other uses that go to that address rather than rights this page can describe. The RASS printed on the CAM-ICU pocket card is Virginia Commonwealth University's copyrighted scale under its own terms, and the original CAM belongs to Dr. Inouye and the Hospital Elder Life Program. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common CAM-ICU documentation errors reviewers flag The numbers behind these errors are specific. In the validation studies, two research nurses reached sensitivities of 100 and 93 percent with specificities of 98 and 100 percent across 471 paired evaluations in 111 ventilated patients (Ely and colleagues, JAMA 2001); in routine practice across ten Dutch ICUs, nursing screens reached a sensitivity of 47 percent with a specificity of 98 percent after 101 of 282 patients were excluded as comatose (van Eijk and colleagues, 2011), and pooled estimates sit at 80.0 percent sensitivity and 95.9 percent specificity (Gusmao-Flores and colleagues, 2012) and 0.84 and 0.95 (Chen and colleagues, 2021). Implementation studies show that completion and accuracy are different problems: at two medical centers, compliance with the CAM-ICU was 90 and 84 percent across 711 patients and 4,163 patient-days, with kappas of 0.92 and 0.75 against reference raters (Pun and colleagues, 2005); in a trauma ICU, nurses completed it in 84 percent of evaluations with a kappa of 0.77 overall and 0.62 in ventilated patients against 1,011 expert assessments (Soja and colleagues, 2008); and in a surgical ICU, a one-week nursing education campaign cut inappropriate unable-to-assess ratings from 32 to 19 percent , with ventilated patients far more likely to receive one (Swan, 2014). In 614 medical ICU patients, purely hyperactive delirium made up 1.6 percent of delirium, hypoactive 43.5 percent, and mixed 54.9 percent (Peterson and colleagues, 2006), which is why an unnamed subtype is usually a missed hypoactive case. No published audit counts how often a CAM-ICU entry lacks its RASS, its time, or its feature results; the bundle metrics and coding rules described under compliance considerations are what turn those omissions into a quality or coding finding. The BastionGPT Clinical Advisory Board sees the same errors most often in CAM-ICU documentation reviews: A negative charted at a RASS that made the screen impossible. A CAM-ICU negative beside a RASS of -4 or -5, or a screen charted at a time with no RASS at all. At the two deepest values the result is unable to assess with the reason, never negative; chart the RASS and time first, and when the patient lightens after the awakening trial, screen and chart again. Unable to assess as a convenience. UTA on an awake but slow, deaf, intubated, or non-English-speaking patient, with no reason and no retry. Vanderbilt restricts UTA to coma and stupor and warns that staff overuse it; intubation is not a reason (the screen was built for ventilated patients), hearing loss calls for the hearing aids or the visual task, a language barrier for a translation or an interpreter, and every UTA carries its reason and a next attempt. A result with no feature trail. A bare "CAM+" or "CAM-" that cannot be reproduced: no features listed, no note of where the assessment stopped, no baseline source for Feature 1. Record each feature assessed as present or absent, which were not needed, the attention task used, and the baseline and its source, so a reviewer can see why the result is what it is. Fluctuation lost, or baseline confused with change. A night of agitation and a sedated interval followed by a calm 08:00 screen charted with Feature 1 absent, or a patient with known dementia charted as acutely changed because no one asked the family. Feature 1 evaluates the past 24 hours against a stated baseline and Feature 3 evaluates the moment; write both and name the baseline source. Positive screen, nothing follows. A positive with no subtype, no contributors reviewed, no action, and no notification time, or an antipsychotic given as the reflex response. Name the subtype by the RASS, list what was checked, record the bundle measures started, the medication changes made under orders, and who was told when; no guideline requires a drug for a positive screen, and PADIS suggests against routine antipsychotics. The screen charted as the diagnosis, or the diagnosis nowhere. A nursing note that reads "patient is delirious" as a diagnosis, or a positive screen for three days with no provider documentation of delirium, so the record carries a nursing finding and no medical diagnosis for the coder or the discharge summary. The nurse charts the screen; the provider documents the diagnosis and its contributors; the two should agree, and the handoff should carry both. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on critical care nursing and delirium screening documentation. Give it the facts (the RASS and time, the baseline and its source, each feature assessed and where the assessment stopped, the result and subtype or the reason it could not be done, the contributors reviewed, the actions taken, who was notified and what they decided, and the prior result) and it drafts the entry: gate, features, result, contributors, action, notification, and the handoff line, ready for your review. Cross-check a finished flowsheet narrative or note for the gaps reviewers flag: a negative at an unassessable RASS, a UTA with no reason, a result with no feature trail, fluctuation lost at a calm morning, a positive with nothing after it, or a nursing note that states the diagnosis. Turn the shift into the next document: the delirium lines of a transfer-of-care summary, the timeline of an incident report after a self-extubation, or a plain-language explanation for a family of what a positive screen means and does not mean, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the CAM-ICU scored, and what makes a screen positive? The CAM-ICU is not added up; it is a decision built from four features once the patient is arousable. Feature 1 asks whether mental status has changed acutely from baseline or fluctuated over the past 24 hours. Feature 2 asks whether the patient can pay attention, tested with a short auditory task the patient answers by squeezing the assessor's hand, with a visual recognition task as the backup for patients who cannot hear the task. Feature 3 is altered level of consciousness and is read straight from the current RASS: any value other than zero counts as present. Feature 4 is disorganized thinking, tested with a few simple yes-or-no questions and a simple command. The screen is positive when Features 1 and 2 are both present and either Feature 3 or Feature 4 is present. Only the features needed to reach the answer have to be done: if Feature 1 or Feature 2 is absent the screen is negative and stops, and if Features 1, 2, and 3 are present the screen is positive without Feature 4, which is needed only when the patient is at RASS 0. The letters, pictures, questions, and commands, and the error thresholds for Features 2 and 4, are on Vanderbilt's copyrighted worksheet and flowsheet and are not reproduced here; train from the current materials on icudelirium.org, because Features 3 and 4 swapped numbers in the 2014 revision and older cards label them the other way round. Which RASS values allow a CAM-ICU, and how do I chart unable to assess? Chart the RASS with its time first, every time, and then the screen. At RASS -4 or -5 the patient is comatose or too deeply sedated to test content of consciousness, and Vanderbilt's FAQ is explicit that the result is unable to assess, which means coma or stupor rather than delirium or normal, and that UTA should be used only for comatose patients; the SCCM ICU Liberation manual and Vanderbilt's own trauma guideline treat -4 and -5 the same way. At -3 the FAQ says most patients can still be rated and most sites use -3 as the lower border, with some choosing -2, so the border is a unit protocol decision that belongs in the protocol, not in each entry. Any lighter value clears the gate, including agitated values. Write UTA with its reason (RASS -4 on propofol, coma, pharmacologic paralysis) and the next attempt, usually after the awakening trial; never as a synonym for negative, and never for an awake patient who is slow, deaf, intubated, or speaks another language, because those are testing problems with solutions, not coma. A one-week education campaign in one surgical ICU cut inappropriate UTA ratings from 32 to 19 percent, and ventilated patients were the ones most often mislabeled (Swan, 2014). Whether a UTA counts as a completed screen for your quality report depends on the program: the ICU Liberation manual considers a UTA finding aligned with its goal and excludes patients in coma or at RASS -4 or -5, while other programs define it differently, so cite the definition you use. The RASS page covers the assessment that produces the gating value. How often should the CAM-ICU be documented? Is there a regulatory interval? No law or accreditor in the United States, Canada, or Australia sets a CAM-ICU interval, and neither does PADIS: the 2018 guideline states as good practice that critically ill adults should be regularly assessed for delirium using a valid tool and leaves the frequency to the unit. The working convention, carried in Vanderbilt's FAQ from the 2013 PAD guideline, is every shift (every 8 to 12 hours) and as needed, with more frequent screens when the patient's condition changes; Vanderbilt also recommends screening before and after each spontaneous awakening trial, because delirium that clears within two hours of stopping sedatives carries a different prognosis from delirium that persists (Patel and colleagues, 2014). Critical Care BC says at least once a shift and more often with any change in mental status; Vanderbilt's trauma guideline says each shift with the results reported on rounds; the SCCM ICU Liberation data manual counts at least two documented assessments per 24 hours as compliant. Once-daily screening is a habit, not a standard, and it misses fluctuation by design. Whatever interval your protocol sets, cite the protocol in the record as policy, chart the RASS with each screen, and add an unscheduled screen whenever behavior changes, because the value of the tool is serial. Is a positive CAM-ICU a diagnosis of delirium, and who documents the diagnosis? No. A positive screen means the features of delirium were present at that assessment; delirium itself is a medical diagnosis, made by the physician or advanced practice provider from the screen, the history, the examination, the medication record, and the search for causes, and documented in the provider's note. Vanderbilt's trauma guideline puts it plainly: delirium is a clinical diagnosis made by the team from the injuries, GCS, RASS, and CAM-ICU together, and in severe brain injury the screen is not solely diagnostic; the FAQ asks neurological ICUs to read a positive as symptoms of delirium rather than definitely delirium. The coding rules follow the same line: the ICD-10-CM Official Guidelines base code assignment on the provider's documentation and list the exceptions where another clinician's note may be used (body mass index, pressure ulcer stage, coma scale, NIH stroke scale, and a few more), and a delirium screen is not one of them, so F05 comes from the provider's note and never from the flowsheet. The nursing entry therefore reports the screen result, the features, the contributors reviewed, the actions, and the notification, and says that the provider was informed; it does not read "patient is delirious." If you believe the patient is delirious and the provider disagrees, chart your objective findings and the conversation, repeat the screen, and ask for the reassessment; a screen result recorded as observed is not a claim about the diagnosis. Where the diagnosis is made, the discharge summary should carry it so that cognition is followed after the ICU. How do I chart the CAM-ICU for a patient who is intubated, paralyzed, blind, unable to move an arm, or does not speak my language? Intubation is never a reason to skip it: the screen was built for ventilated patients, the attention task is answered by hand squeeze, the questions by nodding or squeezing, and the command is nonverbal, so chart the screen as usual and note intubated for context. Under neuromuscular blockade the patient cannot respond at all, so chart unable to assess with the reason (pharmacologic paralysis, the agent, and its start time), carry the last result before the blocker with its time, and screen again when the block has worn off and the RASS is -3 or lighter; the same logic applies to the RASS under paralysis, which the RASS page covers. A patient who is blind or cannot move an arm is scored on the Feature 4 questions alone, per Vanderbilt's FAQ, and a patient who cannot hear the auditory task uses the visual backup; write which task was used and why. For a language barrier, use one of the more than 20 translations of the worksheet if the unit has it, or an interpreter for the questions, and if neither is possible chart that the screen could not be completed and the reason, never negative. In every one of these cases the entry names the barrier, what was done about it, and when the next attempt is due. What do I document when the screen is positive: subtype, contributors, actions, and medication? Four things, in the order the bundle expects. Subtype: name it by the RASS the way your protocol defines it, usually hypoactive at a negative or zero RASS and hyperactive or mixed at a positive one; hypoactive delirium is the common form and the one most often missed, and in 614 medical ICU patients purely hyperactive delirium was 1.6 percent of delirium (Peterson and colleagues, 2006). Contributors: list what you reviewed, named sedatives and benzodiazepines, anticholinergics, opioid dosing against a pain score with its tool, sleep and its interruptions, oxygenation, infection markers, electrolytes and glucose, restraints, catheters and lines, and whether hearing aids and glasses are in, and note the pharmacist's review when made; withdrawal is assessed under its own tool (the CIWA-Ar for alcohol). Action: reorientation, sleep protection, mobility, devices removed, family presence, the medication changes made under orders, and the awakening and breathing trials that day, with the provider notified at a stated time and their decision recorded. Medication: no guideline requires a drug for a positive screen; PADIS 2018 suggests against routine antipsychotics for prevention and for treatment, the 2025 focused update was unable to recommend for or against them for treatment, and MIND-USA found no difference in days alive without delirium or coma among haloperidol, ziprasidone, and placebo in 566 patients (Girard and colleagues, 2018), so an antipsychotic, when given, is charted against a specific indication such as distressing or dangerous agitation with the order and the response, never against the screen result. Then screen again at the interval the protocol sets and whenever behavior changes. Can I use the CAM-ICU on the step-down unit, the ward, in the emergency department, or in children? Use the version validated for the setting and name it. The adult CAM-ICU was validated in ICU patients, including ventilated ones, and has emergency department validation (Han and colleagues, Academic Emergency Medicine, 2014); a step-down unit that keeps ventilated or recently extubated patients can use it with the same gate and terms, and Vanderbilt supports delirium assessment throughout the hospital while recommending setting-appropriate tools. In the emergency department the two-step Delirium Triage Screen and brief CAM (Han and colleagues, 2013: 406 older patients, the triage screen 98 percent sensitive as a rule-out, the brief CAM 84 percent sensitive and 96 percent specific in physicians' hands) were built from the CAM-ICU for that setting; on general wards the CAM, the 3D-CAM, the 4AT, and the Nursing Delirium Screening Scale are the usual choices, and Australian hospitals meeting action 5.29 use whichever validated tool the service has adopted. For children, the pCAM-ICU covers children at least 5 years old both chronologically and developmentally (Smith and colleagues, 2011) and the psCAM-ICU covers infants and children from 6 months to 5 years (Smith and colleagues, 2016), so the choice between them follows developmental as well as chronological age; the SCCM ICU Liberation manual also accepts the Cornell Assessment of Pediatric Delirium. Whatever you use, the same documentation rules hold: the arousal check first, the features assessed, the result with its time and reason, and the diagnosis left to the provider. Can we build the CAM-ICU into our EHR, print it on our flowsheet, or put it on our website? For your own clinical use, yes, on Vanderbilt's stated terms. The CAM-ICU and its educational materials are copyrighted (Copyright 2002, E. Wesley Ely, MD, MPH and Vanderbilt University, all rights reserved), and the Critical Illness, Brain Dysfunction, and Survivorship Center states that it has deliberately made them unrestricted in terms of use, does not require a written letter of permission for implementation and clinical use, and asks that the copyright line appear at the bottom of pocket cards and other educational materials. A hospital's own flowsheet, pocket card, or training deck sits inside that permission with the copyright line on it. For anything else the site says one thing: for other uses, contact [email protected] for permission. As of September 2026 Vanderbilt publishes no separate terms for EHR vendors, publications, commercial products, or public web tools, and it has not said in terms whether a health system's own EHR build counts as implementation, so a cautious team confirms with the center and keeps the copyright line in the build, while a vendor product, a paid app, a public calculator, or a published reproduction of the worksheet is an other use to clear with the center first. Free, attributed, or non-commercial does not by itself create a license, and the CAM-ICU is not public domain. Two neighbors have their own owners: the RASS on the pocket card is Virginia Commonwealth University's copyrighted scale with its own free-use terms, and the original CAM belongs to Dr. Inouye's Hospital Elder Life Program. This page and its templates stay on the right side of that line: the feature logic in original words, and no worksheet, stimuli, thresholds, or manual text. Can BastionGPT draft my CAM-ICU documentation? Yes. Give it the facts (the RASS and time, the baseline and its source, each feature assessed and where the assessment stopped, the result and subtype or the reason it could not be done, the contributors reviewed, the actions taken, who was notified and what they decided, and the prior result) and it drafts the full entry: gate, features, result, contributors, action, notification, and the handoff line, ready for your review. It can also check a finished note for a negative at an unassessable RASS, a UTA with no reason, a result with no feature trail, fluctuation lost at a calm morning, a positive with nothing after it, or a nursing note that states the diagnosis. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Vanderbilt Critical Illness, Brain Dysfunction, and Survivorship (CIBS) Center, monitoring delirium in the ICU (copyright 2002 E. Wesley Ely and Vanderbilt University; unrestricted implementation and clinical use with the copyright line; other uses to [email protected] ; once per shift per PADIS 2018; revised flowsheet and worksheet; training manual updated August 2016; accessed September 2026); CIBS Center, CAM-ICU frequently asked questions (revised March 2014; RASS -4 and -5 not tested and most sites use -3 as the border; UTA means coma or stupor and is overused; features in any order and only as needed; every shift and as needed; screen before and after awakening trials; documentation in the hourly flowsheet with feature-level detail where possible; over 20 languages; dementia, stroke, blind and paralyzed patients); CIBS Center, pediatric care (pCAM-ICU at least 5 years, psCAM-ICU under 5 years). Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI, 1990, Annals of Internal Medicine 113(12):941-948, clarifying confusion: the confusion assessment method (the four-feature algorithm); Ely EW, Margolin R, Francis J, and colleagues, 2001, Critical Care Medicine 29(7):1370-1379, evaluation of delirium in critically ill patients: validation of the CAM-ICU (38 patients, 293 paired evaluations; sensitivities 95 and 96 percent, specificities 93 percent; kappa 0.84 and 0.79); Ely EW, Inouye SK, Bernard GR, and colleagues, 2001, JAMA 286(21):2703-2710, delirium in mechanically ventilated patients: validity and reliability of the CAM-ICU (111 patients, 471 paired evaluations; sensitivities 100 and 93 percent, specificities 98 and 100 percent; kappa 0.96; about two minutes). Bedside accuracy, implementation, and unable-to-assess ratings: van Eijk MM, van den Boogaard M, van Marum RJ, and colleagues, 2011, American Journal of Respiratory and Critical Care Medicine 184(3):340-344, routine use of the CAM-ICU: a multicenter study (10 ICUs, 282 patients, 101 comatose excluded; sensitivity 47 percent, specificity 98 percent); Gusmao-Flores D, Salluh JI, Chalhub RA, Quarantini LC, 2012, Critical Care 16(4):R115, CAM-ICU and ICDSC for the diagnosis of delirium: systematic review and meta-analysis (9 studies, 969 patients; 80.0 and 95.9 percent); Chen TJ and colleagues, 2021, International Journal of Nursing Studies 113:103782, diagnostic accuracy of the CAM-ICU and ICDSC: a bivariate meta-analysis (29 and 12 studies; 0.84 and 0.95 against 0.83 and 0.87; moderators); Pun BT, Gordon SM, Peterson JF, and colleagues, 2005, Critical Care Medicine 33(6):1199-1205, large-scale implementation of sedation and delirium monitoring in the ICU (711 patients, 4,163 patient-days; compliance 90 and 84 percent; kappa 0.92 and 0.75); Soja SL, Pandharipande PP, Fleming SB, and colleagues, 2008, Intensive Care Medicine 34(7):1263-1268, implementation, reliability testing, and compliance monitoring of the CAM-ICU in trauma patients (84 percent completion; 1,011 expert assessments; kappa 0.77, 0.75 in brain injury, 0.62 ventilated); Swan JT, 2014, American Journal of Critical Care 23(1):60-69, decreasing inappropriate unable-to-assess ratings for the CAM-ICU (surgical ICU; 32 to 19 percent); Patel SB, Poston JT, Pohlman A, Hall JB, Kress JP, 2014, American Journal of Respiratory and Critical Care Medicine 189(6):658-665, rapidly reversible, sedation-related delirium versus persistent delirium in the ICU (102 patients). Variants, subtypes, and comparators: Smith HA, Boyd J, Fuchs DC, and colleagues, 2011, Critical Care Medicine 39(1):150-157, the pediatric CAM-ICU (68 children at least 5 years, 146 paired assessments; 83 and 99 percent; kappa 0.96); Smith HA, Gangopadhyay M, Goben CM, and colleagues, 2016, Critical Care Medicine 44(3):592-600, the preschool CAM-ICU (300 children 6 months to 5 years, 530 paired assessments; 75 and 91 percent; kappa 0.79); Khan BA, Perkins AJ, Gao S, and colleagues, 2017, Critical Care Medicine 45(5):851-857, the CAM-ICU-7 delirium severity scale (0 to 7; 518 patients; alpha 0.85; mortality odds ratio 1.47); Peterson JF, Pun BT, Dittus RS, and colleagues, 2006, Journal of the American Geriatrics Society 54(3):479-484, delirium and its motoric subtypes: a study of 614 critically ill patients (mixed 54.9, hypoactive 43.5, hyperactive 1.6 percent); Han JH, Wilson A, Vasilevskis EE, and colleagues, 2013, Annals of Emergency Medicine 62(5):457-465, the delirium triage screen and the brief CAM (406 patients); Bergeron N, Dubois MJ, Dumont M, Dial S, Skrobik Y, 2001, Intensive Care Medicine 27(5):859-864, the Intensive Care Delirium Screening Checklist (eight items; predicted sensitivity 99 percent, specificity 64 percent). Guidelines and the bundle: Devlin JW, Skrobik Y, Gelinas C, and colleagues, 2018, Critical Care Medicine 46(9):e825-e873, clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption (PADIS) , with the SCCM guideline page (good practice statement on regular assessment with a valid tool; conditional suggestions against routine antipsychotics); Lewis K, Balas MC, Stollings JL, and colleagues, 2025, Critical Care Medicine 53(3):e711-e727, focused update to the PADIS guideline (unable to recommend for or against antipsychotics for treatment); Girard TD, Exline MC, Carson SS, and colleagues, 2018, New England Journal of Medicine 379(26):2506-2516, haloperidol and ziprasidone for treatment of delirium in critical illness (MIND-USA) (16 centers; 566 patients; 7.9, 8.7, and 8.5 days; P 0.26); Society of Critical Care Medicine, ICU Liberation Bundle data collection manual (D element: at least two delirium assessments per 24 hours with the CAM-ICU or ICDSC; unable to assess aligned with the goal; coma and RASS -4 or -5 excluded); Pun BT, Balas MC, Barnes-Daly MA, and colleagues, 2019, Critical Care Medicine 47(1):3-14, caring for critically ill patients with the ABCDEF bundle: results of the ICU Liberation Collaborative in over 15,000 adults . United States: CMS, July 2025, Age Friendly Hospital Measure specifications (Hospital IQR Program; calendar year 2025 reporting period; five domains including Frailty Screening and Intervention), with the American College of Surgeons, CMS Age Friendly Hospital Measure (screening for cognitive impairment including delirium, mobility, and malnutrition) and June 10, 2026 release, GSV hospitals screen nearly all older surgical patients for delirium (94.3 against 52.5 percent); eCFR, 42 CFR 482.24 medical record services and 42 CFR 482.23 nursing services ; CMS, ICD-10-CM Official Guidelines for Coding and Reporting, FY 2026 (Section I.B.14, documentation by clinicians other than the patient's provider; bare code numbers only); Vanderbilt University Medical Center Trauma and Surgical Critical Care, June 2026, trauma delirium management guideline (clinical diagnosis by the team; GCS, RASS, and CAM-ICU each shift; unable to assess at RASS -4 or -5; not solely diagnostic in severe brain injury). Canada: Canadian Coalition for Seniors' Mental Health, delirium clinical guidelines and 2014, guideline update on the assessment and treatment of delirium (2006 national guideline, 2010 end-of-life guideline); Critical Care BC, Provincial Health Services Authority, 2025, ICU Liberation bundle elements A to F (element D: CAM-ICU or ICDSC at least once a shift and with any change in mental status; report on daily rounds; identify the cause first; no data for routine antipsychotic prevention); Health Standards Organization, March 2026, HSO 11001:2026 Critical Care Services, public review draft (validated assessment tools; intervals defined by the organization). Australia: Australian Commission on Safety and Quality in Health Care, Delirium Clinical Care Standard (2016, revised September 2021; eight quality statements; assessments from 1 September 2022; routine antipsychotics not recommended; delirium as a hospital-acquired complication at 35.7 per 10,000 admissions in 2019 to 2020); ACSQHC, NSQHS Comprehensive Care Standard, action 5.29 (screening for patients 65 and over and any patient at risk of delirium with a validated tool; a positive score is a prompt for further assessment, not a diagnosis); ACSQHC, hospital-acquired complications (delirium, HAC 11); Independent Health and Aged Care Pricing Authority, March 2026, National Pricing Model 2026 to 27: hospital acquired complications technical specifications (funding adjustment since July 2018; shadow priced for 2026 to 27 under clause A90 of the Addendum to the NHRA 2026 to 31). ## Related templates RASS Score Documentation Glasgow Coma Scale (GCS) Documentation CIWA-Ar Protocol Documentation Transfer-of-Care Summary Mini-Cog Score Documentation MEWS & NEWS2 Early Warning Score Documentation Discharge Summary Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/wong-baker-faces All note templates # Wong-Baker FACES Documentation: Self-Report, Reassessment & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The Wong-Baker FACES Pain Rating Scale is a self-report pain intensity tool on which a patient aged three or older chooses the face that matches their pain, recorded on a 0 to 10 metric. Pediatric, emergency, perioperative, and medical-surgical nurses use it when a patient cannot use a numeric scale; it is never rated by an observer. This page covers how to document a FACES rating, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The patient rates; the nurse, physician, dentist, or other clinician who administered the scale records it. Registered nurses at the bedside, in emergency departments, procedure areas, and clinics do most of the charting; no publisher qualification level or training requirement applies, and organizations obtain their own permission or license from the Foundation for the artwork they display Audience Oncoming nurses and the treating physician or advanced practice provider, pharmacists reviewing analgesic response, anesthesia and procedural sedation teams, discharge and after-visit summary writers, quality staff auditing pain reassessment, accreditation surveyors, and expert reviewers in pain management claims Typical length 3 to 8 chart lines per assessment (scale and reason chosen, self-selected rating with time, location, onset, and character, intervention, timed reassessment on the same scale, function against the goal) · about 1 to 2 minutes to administer and record Format family Patient self-report pain intensity scale (six faces the patient chooses from, recorded on a 0 to 10 metric in steps of two; validated for ages three and older; not an observational or clinician-rated tool) When it's used Emergency department and admission pain screening, pre-procedure and post-procedure pain assessment, post-operative and medical-surgical reassessment after analgesia, pediatric and adult primary and urgent care visits, dental and outpatient procedures, and any patient who cannot use a numeric scale but can understand the task and choose Standards context Wong and Baker (Pediatric Nursing, 1988; created 1983); artwork copyrighted and the name a registered trademark of the Wong-Baker FACES Foundation, licensed by use category; required by no law, accreditor, or payer in the United States, Canada, or Australia; described here for documentation, no artwork or instructions reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the Wong-Baker FACES Pain Rating Scale? The Wong-Baker FACES Pain Rating Scale is a self-report tool for pain intensity: the patient looks at six faces and chooses the one that matches how much pain they have, and the clinician records the number that belongs to the chosen face on a 0 to 10 metric in steps of two. Donna Wong, a nursing author, and Connie Baker, a child life specialist, created it in 1983 at Hillcrest Medical Center in Tulsa to help hospitalized children communicate pain, collected the validation data at Hillcrest, St. Francis Hospital, and the University of California Davis Medical Center, and published the comparison study in Pediatric Nursing in 1988 (volume 14, pages 9 to 17); the scale first appeared in the textbook Whaley and Wong's Nursing Care of Infants and Children. The original version was scored 0 to 5, and the Foundation now uses the 0 to 10 metric so that the values sit on the same range as the numeric rating scale, although, as the sections below explain, they are not the same values. The Foundation states that the scale is used around the world with people ages three and older, not only children, that it is a self-assessment tool for physical pain only, that it may not be modified or colored, and that it is not to be used by a third person, whether a parent, a caregiver, or a health professional, to assess the patient's pain. The artwork and the scripted instructions are the Foundation's copyrighted material and are not reproduced or paraphrased here. The load-bearing fact for documentation is that the scale is self-report or it is nothing. A chart entry that reads as if the nurse read the patient's face against the pictures records something the Foundation says the tool must never be used for, and it records a different construct from the one the validation studies measured. Three consequences follow. First, the patient must be able to understand the task and choose, which is why the Foundation excludes unresponsive patients and why the note has to say that self-report was obtained. Second, the rating belongs to the patient even when the nurse disagrees with it: the behavioral observation goes in the record beside the rating, never in place of it. Third, a FACES value and a numeric rating scale value are not interchangeable even though both run 0 to 10, so the scale name travels with every number and the two are never plotted as one trend. The Faces Pain Scale-Revised, owned by the International Association for the Study of Pain, is a different faces scale with different anchors and its own rights terms, and the observational tools used when a patient cannot self-report (the FLACC for young children, the CPOT and the PAINAD for adults) rate behavior rather than report; the sedation and arousal assessment that decides whether an intensive care patient can self-report at all belongs to the RASS page, the delirium screen that follows it to the CAM-ICU page, and the responsiveness assessment in trauma and neurologic care to the Glasgow Coma Scale page. ## Who uses Wong-Baker FACES documentation and when Pediatric and emergency nurses chart it at triage and after every intervention for children who can choose a face but cannot yet use numbers reliably, and for older children who prefer it. Post-anesthesia, day surgery, and medical-surgical nurses use it after procedures and operations for adults who cannot use a numeric scale because of language, literacy, hearing, or early cognitive change, and dental and outpatient procedure staff use it before and after local anesthesia and minor procedures. Primary and urgent care clinicians record it at acute visits for injuries, otitis, sore throat, and abdominal pain. Anesthesia and procedural sedation teams read the entries to judge analgesic response; pharmacists read them when reviewing opioid orders; discharge writers carry the last rating and the scale name into the after-visit summary and the discharge summary ; quality staff audit whether every intervention was followed by a timed reassessment; and surveyors read the same entries against the organization's own pain assessment criteria. The FACES page owns one job: how to record a patient-selected rating so that the scale, the self-report, the context, the response, and the reason for the tool choice are all visible. Neighbors take over at the edges: when a patient cannot self-report, the nurse switches to an observational tool and documents why, the intensive care gate for that decision is the RASS , and the nursing admission bundle that often sits on the same flowsheet includes the Braden Scale and the Morse Fall Scale ; the pain regimen and the scale in use travel in the transfer of care summary at handoff. ## How to document a Wong-Baker FACES assessment in the chart No regulator, accreditor, or payer prescribes a FACES note format or requires the scale by name; what exists is the Foundation's use conditions (self-report, physical pain, ages three and older, no observer rating), the Joint Commission's requirement that an organization define its own criteria to screen, assess, and reassess pain consistent with the patient's age, condition, and ability to understand, and unit policies that set reassessment timing. What survives review is an entry that names the scale and the reason it was chosen, states that the patient selected the rating, describes the pain in the patient's words, sets a function goal, records the intervention with its time, dose, and route, reassesses on the same scale at a time that fits the route, keeps observations and proxy impressions separate from the rating, and never mixes FACES values with numeric-scale values in one trend. Each element below carries the pitfall that most often undermines it. Scale selection and the reason for it. Open with the tool and why: "Wong-Baker FACES used; patient is six, understood the task, and chose readily; could not explain the numeric scale," or, for an adult, "FACES used because the patient reads little English and declined the number scale; interpreter present." The Foundation validates the scale for ages three and older, but the evidence for self-report in three- and four-year-olds is weak, the verbal numeric scale is supported from about age six, and cognitively impaired older adults often manage a faces scale better than numbers, so the reason is a clinical judgment worth one clause. If the patient could not understand or respond, record that self-report was not obtainable and name the observational tool used instead. Pitfall: A rating with no scale named, or a scale chosen by habit (FACES for every child, numbers for every adult) with nothing in the note to show the patient could actually use it. Self-report, stated as such. Write the rating as the patient's choice with the scale beside it: "Wong-Baker FACES 8, patient selected at 16:25." The word selected, or pointed to, or chose, is what tells the next reader that the tool was used as designed. Never rate the patient yourself by comparing their face to the pictures; the Foundation states the scale is not to be used by a third person, parents or clinicians included. If a parent or caregiver offers an impression, record it as a parent impression in its own clause, never as the patient's rating, because parent and child ratings diverge and a proxy number masquerading as self-report cannot be trusted later. Pitfall: "Appears to be a 6 on FACES," or a parent's guess entered in the pain score field as if the child had chosen it. Location, onset, character, and what changes it. A rating is one dimension of pain; the assessment is the rest. Record where it hurts (with laterality), when it started and what started it, the patient's own descriptive words, what makes it worse and better, and the pattern since the last assessment. For a young child, take the words the child uses and add what the parent reports about onset and behavior at home, labeled as parent history. The Joint Commission's hospital standard asks organizations to define criteria for assessment consistent with age, condition, and ability to understand, and every published criteria set includes location and character; a bare number satisfies none of them. Pitfall: "Pain 8/10 FACES" with no location, onset, or character, so the entry cannot be compared with the next one or defended as an assessment. Comfort and function goal. Set a goal in terms of what the patient needs to be able to do and write it down: hold the arm still for splinting and imaging, take a deep breath and cough after surgery, sleep through the night, walk to the bathroom, tolerate the dental impression. The Joint Commission's reassessment element asks for progress toward pain management goals including functional ability and gives deep breathing, turning in bed, and walking as its examples; the AHRQ patient safety chapter calls this the comfort-function goal. Agree the goal with the patient or family where possible and state who agreed it. Pitfall: "Goal: pain less than 4" as the only target, with nothing about what the patient should be able to do and no record that anyone agreed it. Intervention with time, dose, and route. Record what was done and when: the analgesic with dose, route, and time; the nonpharmacologic measures (immobilization, ice, elevation, positioning, distraction, a parent at the bedside); and, when nothing was given, the reason (patient declined, rating below the agreed threshold, awaiting the provider). The intervention is what the reassessment will be judged against, so the times must be real and the dose must match the medication administration record. Pitfall: "Medicated per order" with no drug, dose, route, or time, or an intervention charted at a time that conflicts with the administration record. Reassessment, timed to the route, on the same scale. Reassess at a time that fits the drug's onset and your unit policy, and say so: the AHRQ patient safety chapter gives 15 to 30 minutes after an intravenous opioid and 45 to 60 minutes after an oral opioid or nonopioid as the working intervals, and no regulator sets one. Use the same scale, have the patient choose again, and record the new value with the time, the function result against the goal, side effects, and, when an opioid was given, a sedation score such as the POSS, because sedation precedes respiratory depression. A sleeping patient is not reassessed with the scale: describe the sleep and the breathing, do not enter a score, and reassess when the patient wakes. Pitfall: "Reassessed, pain improved" with no time, no scale, and no value, or a sleeping child entered as 0. Trend integrity, tool changes, and handoff. Keep FACES values in a FACES trend. A 6 chosen on the faces and a 6 spoken on the numeric scale are different measurements, and averaging or plotting them together produces a line that means nothing; when the tool changes because the patient's ability changed, start a new labeled trend and write the reason. When self-report is no longer obtainable, switch to an observational tool, name it, and record why, following the hierarchy in the ASPMN position statement (attempt self-report, consider causes of pain, observe behavior, obtain proxy report, consider an analgesic trial). Carry the scale name into the handoff, the transfer summary, and the discharge plan so that the next clinician keeps the same tool, and tell informatics if the electronic record maps both scales into one pain score field. Pitfall: A pain graph that mixes FACES and numeric values, or a handoff that says pain 4 with no scale, so the next shift switches tools and the trend breaks. ### Blank template (copy and adapt) WONG-BAKER FACES PAIN DOCUMENTATION BLOCK (patient self-report only) Date / time: [ ] Setting: [ ] Clinician: [ ] Encounter: [ED / admission / procedure / post-operative / medical-surgical / primary or urgent care / dental] Ability to self-report: [confirmed: understood the task and chose a face / not obtainable: observational tool used instead (name it) and the reason] Scale chosen and why: [Wong-Baker FACES because ... / numeric scale not usable because ... / patient preference] Language / interpreter: [ ] Prior tool in this record: [same scale continued / tool changed, new trend, reason] Rating: Wong-Baker FACES [ ] (0 to 10 metric), patient selected at [time] Parent or caregiver impression (if recorded, labeled as such): [ ] Location and laterality: [ ] Onset and cause: [ ] Duration / pattern: [ ] Character (patient's words): [ ] Worse with: [ ] Better with: [ ] Comfort and function goal: [what the patient needs to be able to do; agreed with] Intervention: [drug, dose, route, time / nonpharmacologic measures / none, reason] Reassessment at [time, chosen to the route per unit policy]: Wong-Baker FACES [ ], patient selected Goal: [met / not met] Side effects: [ ] Sedation score if opioid given: [ ] Observed behavior (recorded beside the rating, never in place of it): [ ] Sleeping at reassessment: [describe sleep and breathing; no score; reassess when awake] Communication: [provider notified, time / handoff / discharge plan names the scale] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the faces artwork or the Foundation's instructions. ## Sample Wong-Baker FACES documentation (fictional) Scenario: a community hospital pediatric emergency department, where a six-year-old with a forearm injury cannot use the numeric scale but chooses a face readily; the nurse records why the Wong-Baker FACES scale was chosen, the patient-selected rating with the time, the pain description in the child's words, a function goal, an oral analgesic with its dose and time, a reassessment timed to the route on the same scale, the observed behavior and the parent's impression kept separate from the rating, and a discharge plan that names the scale. All details are fictional. Patient: T.M., 6 · Setting: Pediatric emergency department, community hospital · Clinician: K. Alvarez, RN · Note date: 09/12/2026 Scale selection and self-report: T.M., age 6, brought by his mother after a fall from a scooter at about 15:40 on 09/12/2026, arrived 16:10, triaged 16:20. Asked to rate his pain on the department's 0 to 10 numeric scale, he could not say what a number would mean. Offered the Wong-Baker FACES scale, he understood the task on the first explanation and pointed to a face without prompting, so FACES was chosen over the numeric scale; an observational tool (FLACC) was not used because self-report was obtainable. English speaking, no interpreter needed. Rating: Wong-Baker FACES 8, patient selected at 16:25. His mother's comment that the pain is worse than he is showing is recorded here as a parent impression, not as the rating. Pain assessment: Location: right forearm, dorsal aspect, about two finger widths above the wrist, which he points to with his left hand; no other painful area reported and none found on examination. Onset: at the fall, about 45 minutes before triage, landing on the outstretched right hand. Character in his words: it stings and it hurts when I wiggle it. Worse with movement and with touch over the area; better when he holds the arm against his chest. No prior injury to the arm, no analgesic given at home today, no known allergies, weight 21.0 kg on the department scale. Mother's history, labeled as such: he was upset at the scene, has not used the hand since, and ate a snack in the car. Goal and intervention: Comfort and function goal agreed with T.M. and his mother at 16:30: able to hold the arm still for imaging and splinting and to tell us when it hurts. Ibuprofen 210 mg (10 mg per kg) oral suspension given at 16:35 under the department's pediatric analgesia protocol order, documented on the medication administration record at the same time. Arm placed in a padded temporary splint and elevated on a pillow at 16:38; wrapped ice pack applied over the splint; mother at the bedside with a tablet for distraction. Emergency physician notified of the 16:25 rating and the injury at 16:28. Radiographs of the right forearm and wrist obtained at 16:50; he held still for imaging. Reassessment: At 17:25, 50 minutes after the oral dose (department policy sets 45 to 60 minutes for oral analgesics; the interval is policy, not a regulation), asked to choose again on the same scale: Wong-Baker FACES 4, patient selected at 17:25. Observed behavior, recorded beside the rating and not in place of it: watching a video, arm resting quietly in the splint, no guarding when the ice pack was repositioned, answering questions readily. Goal met so far: held still for imaging without distress. No nausea, no vomiting, no rash. No opioid was given, so no sedation score is required. Mother's observation that he seems much better is recorded as a parent observation. Result, plan, and communication: Radiographs read by the emergency physician at 17:35: buckle fracture of the distal right radius, no displacement, no growth plate involvement. Short arm splint applied by the orthopedic technician at 17:40; T.M. held still and reported no increase in pain during application. Pre-discharge rating on the same scale: Wong-Baker FACES 2, patient selected at 18:05. Discharged with his mother at 18:10. Plan in the after-visit summary: ibuprofen 210 mg by mouth every 6 to 8 hours with food as needed for up to 3 days, splint care and elevation, no sports or scooter until cleared, return precautions for numbness, swelling, color change, or pain that wakes him or is not relieved by the dose; orthopedic clinic appointment 09/17/2026. The summary names the Wong-Baker FACES scale as the tool used today so that the clinic continues it. All three ratings are charted in the FACES trend only; no numeric-scale value was entered. Findings and the two reassessments reviewed with the emergency physician at 17:38 and reported at the 19:00 handoff. This sample is fictional and for educational purposes. It does not describe a real patient or record; the ratings, dates, and details are invented to show documentation structure and are not clinical guidance. No scale artwork or instructions are reproduced. ↑ Back to the template and downloads ### Why this sample works The scale is named beside every rating and each rating is written as the patient's choice with a time, so no reader can mistake it for an observer rating or for a numeric-scale value. The reason for choosing FACES over the numeric scale and over an observational tool is one clause built from what the child actually did, which is the age-and-ability judgment the Joint Commission criteria and the Foundation's use conditions both call for. Location, onset, character, and aggravating and relieving factors are in the child's words with the parent's history labeled as such, so the entry is an assessment rather than a number. The goal is functional and agreed, the intervention carries dose, route, time, and a matching administration record, and the reassessment is timed to the route on the same scale with the observed behavior recorded beside the rating, never in place of it. All three values stay in one FACES trend, no numeric value is entered, the sedation-score question is answered explicitly, and the discharge plan carries the scale name to the next clinician. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: no LAW names the Wong-Baker FACES scale, and the rules that do exist judge whether an assessment led to a response. Under LAW, the Medicare Conditions of Participation for hospitals require the nursing staff to develop and keep current a nursing care plan for each patient that reflects the patient's goals and needs (42 CFR 482.23(b)(4)) and require every medical record entry to be legible, complete, dated, timed, and authenticated, with the record describing the patient's progress and response to medications and services (42 CFR 482.24(c)); neither names a pain tool. California's Health and Safety Code section 1254.7 (LAW, for California-licensed facilities) makes pain an item that must be assessed as a condition of licensure, performed in a consistent manner that is appropriate to the patient and noted in the patient's chart; AB 1048, effective January 1, 2018, removed the earlier requirement that pain be assessed at the same time as vital signs. The Joint Commission's hospital standard PC.01.02.07 (CONVENTION with accreditation force, effective January 1, 2018, and for ambulatory care, critical access hospitals, and office-based surgery from January 1, 2019) requires the organization to define criteria to screen, assess, and reassess pain that are consistent with the patient's age, condition, and ability to understand (EP 1), to screen for pain at emergency department visits and at admission (EP 2), and to reassess and respond to pain through evaluation and documentation of the response to interventions, progress toward pain management goals including functional ability, side effects, and risk factors for adverse events (EP 7); its rationale states that using numerical pain scales alone to monitor pain is inadequate and that pediatric patients and intensive care patients who cannot self-report require alternative tools. The Joint Commission names no instrument and sets no reassessment interval, and its own clarifications matter for the habits this page corrects: in April 2016 it stated that it does not endorse pain as a vital sign, that the standards never required treating pain to a number or to zero, and that the requirement to assess pain in all patients was removed from hospital accreditation in 2009; a 2017 JAMA account by the Joint Commission's David Baker records that the 2001 implementation example calling pain a fifth vital sign had been deleted from the manual by 2004. Under PAYER POLICY, CMS removed the pain management questions from the HCAHPS survey for discharges from October 1, 2019, so no patient-experience score now turns on pain ratings, and no Medicare payment rule names a pain scale. The rest is CONVENTION: the 15 to 30 minute reassessment after an intravenous opioid and 45 to 60 minutes after an oral analgesic come from drug onset and the AHRQ patient safety chapter, not from any regulation; "medicate above 4" appears in no standard; and Mularski and colleagues found across 600 visits at a Veterans Affairs medical center that measuring pain as the fifth vital sign did not by itself improve the quality of pain management, which is the evidence that a documented rating matters only when it is followed by a documented response. In survey and claims review (CONVENTION), the retrospective question is whether a rating led to an intervention and a timed reassessment on the same scale; no published dataset counts how often pain reassessment gaps appear in nursing negligence claims, and this page does not claim one. Canada and Australia require assessment and name no tool, and the age and comparability evidence explains why the reason for the tool choice belongs in the note. In Canada, no federal or provincial LAW names the scale. CAN/HSO 13200:2023, Pediatric Pain Management, the national standard co-developed by the Health Standards Organization and Solutions for Kids in Pain (CONVENTION with standards force), applies to hospital inpatient, procedural, and outpatient care from birth to 19 years, sets out 34 criteria, calls for consistent documentation of pain assessment results and the care plan in the child's health record and for support that maximizes the child's ability to self-report, and states that it does not prescribe a particular approach; Accreditation Canada's Qmentum standards address pain assessment, and the wording your organization is surveyed against should be confirmed in its own standards set; the Registered Nurses' Association of Ontario's best practice guideline Pain: Prevention, Assessment, and Management reached its fourth edition in February 2025. In Australia, no LAW names the scale; the Australian Commission on Safety and Quality in Health Care's Opioid Analgesic Stewardship in Acute Pain Clinical Care Standard (2022) carries a quality statement on acute pain assessment among its nine statements, the NSQHS Comprehensive Care Standard requires screening and comprehensive assessment processes without naming a pain tool (CONVENTION with accreditation force), and the ANZCA and Faculty of Pain Medicine text Acute Pain Management: Scientific Evidence, fifth edition (2020), is the evidence reference. On the evidence: the Foundation states the scale is for ages three and older, and a 2017 systematic review (von Baeyer and colleagues) found no evidence that three-year-olds and only weak evidence that four-year-olds can use published self-report tools validly, with IASP commentary placing strong evidence at five and older; Tsze and colleagues (2018) found the verbal numeric scale strong for ages 6 to 17 and not strong at 4 and 5, and von Baeyer and colleagues (2009) tentatively supported the numeric scale from age 8; Birnie and colleagues' 2019 systematic review strongly recommended the NRS-11, the Faces Pain Scale-Revised, and the Coloured Analogue Scale for acute pain, rated the Wong-Baker scale well established but gave it only a weak recommendation, and recommended no measure for children under six; Tomlinson and colleagues (2010) reviewed 14 faces scales, found four adequately supported, found that children preferred the Wong-Baker scale when given a choice, and named the confounding of pain intensity with affect as its disadvantage, the effect Chambers and colleagues demonstrated in 1998, 1999, and 2005, where scales with emotion-laden end anchors produced higher ratings than scales with neutral anchors; Garra and colleagues (2010) found agreement with a visual analogue scale of rho 0.90 in 120 emergency patients with a median age of 13, which supports validity at the group level and says nothing about whether an individual's FACES 6 equals a numeric 6; and Ware and colleagues (2006) found that among 68 older minority adults the cognitively intact preferred the numeric scale while cognitively impaired, African American, and Hispanic participants preferred the Faces Pain Scale-Revised. Two version facts belong in the record: the original scale was scored 0 to 5 and current use is on the 0 to 10 metric, so an old form and a new one do not read the same; and the Foundation permits no modification and no added color, so a locally altered version is not the validated scale and should not be charted under its name. Wong-Baker FACES is a registered trademark of the Wong-Baker FACES Foundation (US Reg. No. 4221357, registered October 9, 2012 in International Class 010 for pain rating scales and charts for diagnostic use in medical examinations, renewed March 3, 2022, with the word FACES disclaimed apart from the mark), and the faces artwork is copyrighted: the Foundation's required caption reads "Copyright 1983, Wong-Baker FACES Foundation, www.WongBakerFACES.org. Used with permission." with an Elsevier credit added in publications because the scale was originally published in Whaley and Wong's Nursing Care of Infants and Children. The Foundation, established in 2009 to protect the integrity of the scale, licenses it by use category (personal, healthcare organization, healthcare provider, healthcare educator, healthcare student, school nurse, commercial, publishing, research, and paramedic); its pages state that licensing is not required for every situation, that personal permission is granted through a web form, that use in facilities, commercial products, or publications will likely need a licensing agreement and fee, with organizations asked to request a quote, that no modifications may be made and no colors added, and that the scale is a self-assessment tool for physical pain only, never for unresponsive patients and never for a third party to rate someone else's pain. No blanket free clinical-use or electronic-record grant is published on those pages as of September 2026, so a facility form, an EHR build, an app, or a public web page that displays the faces should rest on the Foundation's written permission or license; this page displays none of the artwork and quotes none of the instructions. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common Wong-Baker FACES documentation errors reviewers flag The numbers behind these errors are specific. In the systematic review of 14 faces scales, children preferred the Wong-Baker scale when given a choice, and the same review named the confounding of pain intensity with affect as its disadvantage ( Tomlinson and colleagues, 2010 ); in Chambers and Craig's 1998 experiment, children using a scale with an emotion-laden no-pain anchor gave significantly higher scores to no-pain vignettes than children using a neutral-anchored scale ( Pain, 1998 ), and the same effect held for children's and parents' ratings after minor surgery ( Chambers and colleagues, 2005 ). The 2019 quality assessment of pediatric self-report measures strongly recommended the NRS-11, the Faces Pain Scale-Revised, and the Coloured Analogue Scale for acute pain, rated the Wong-Baker scale well established but only weakly recommended, and recommended no measure for children under six ( Birnie and colleagues ); a 2017 systematic review found no evidence that three-year-olds and weak evidence that four-year-olds can use any published self-report scale validly ( von Baeyer and colleagues ); and the verbal numeric scale was strong for ages 6 to 17 but not for 4 and 5 in emergency department data ( Tsze and colleagues, 2018 ). Agreement between the Wong-Baker scale and a visual analogue scale was rho 0.90 in 120 emergency patients ( Garra and colleagues, 2010 ), a group-level result that does not make an individual FACES value equal to a numeric one. No published audit counts how often a chart entry omits the scale name or records a FACES rating as an observer's judgment; the Foundation's use conditions and the Joint Commission criteria described under compliance considerations are what turn those omissions into a defensibility problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Wong-Baker FACES documentation reviews: The clinician chooses the face. "Appears to be a 6 on FACES," a face picked by looking at the patient, or a rating entered for a sedated or unresponsive patient. The Foundation states that the scale is not to be used by a third person and never with unresponsive patients, and observer use measures something other than the validated self-report. Write the rating as the patient's choice with the time; when self-report is not obtainable, switch to an observational tool, name it, and record why. A number with no scale, or two scales in one trend. "Pain 6" with no tool named, a FACES 6 plotted beside a numeric 6, or a graph that averages them. The values share a range and nothing else; group-level agreement between the scales does not make an individual value interchangeable. Name the scale beside every number, keep FACES values in their own trend, and start a new labeled trend with the reason when the tool changes. Used below the validated age, or a proxy rating charted as self-report. FACES attempted in a two-year-old, a rating recorded for a child who never showed they understood the task, or a parent's impression entered in the pain score field. The Foundation validates the scale for ages three and older, the evidence for self-report is weak at three and four, and parent and child ratings diverge. Record that the patient understood and chose, label any parent or caregiver impression as such, and use an observational tool when the child cannot self-report. A rating with no assessment behind it. "FACES 8" with no location, onset, character, or aggravating factor, and no goal. The Joint Commission asks organizations to define assessment criteria consistent with age, condition, and ability to understand, and a bare number meets none of them. Record where, when, what it is like in the patient's words, what changes it, and what the patient needs to be able to do. Intervention without a timed reassessment on the same scale. "Medicated per order, will reassess," a 30-minute rule applied to every route, or a sleeping patient entered as 0. Reassessment timing is convention drawn from drug onset (about 15 to 30 minutes after an intravenous opioid, 45 to 60 after an oral analgesic) and from unit policy, not regulation. Record the time, the same scale, the patient-selected value, the function result, side effects, and a sedation score when an opioid was given; describe sleep rather than scoring it. The self-report overwritten by the observer's impression. A child chooses the top of the scale while playing and the nurse charts a 2 because he looked comfortable, or a note reads pain 8 but appears comfortable and the field says 4. The rating is the patient's report at that moment and stays as given; the behavioral observation is recorded beside it, the discrepancy is noted, and the clinical response is decided on both. Emotion-laden anchors and a child's fear can push a choice upward, which is a reason to document both, not to substitute your judgment for the report. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on pain assessment and nursing documentation. Give it the facts (the scale used and why, that the patient chose the face and when, the pain location, onset, character, and what changes it, the function goal, the intervention with dose, route, and time, the reassessment value and time on the same scale, the observed behavior and any parent impression, and the plan) and it drafts the documentation block: scale and reason, patient-selected rating, assessment in the patient's words, goal, intervention, timed reassessment, observations kept separate, and the trend and communication lines, ready for your review. Cross-check a finished entry for the gaps reviewers flag: a rating with no scale named, an observer-rated or proxy value written as self-report, a number with no location or onset, an intervention with no timed reassessment, a sleeping patient scored, or FACES and numeric values mixed in one trend. Turn the entry into the next document: the pain line of the handoff or transfer summary with the scale named, the home pain plan for an after-visit summary, or the reassessment audit note for a quality review, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions How is the Wong-Baker FACES scale scored, and is it a 0 to 5 or a 0 to 10 scale? The patient chooses one of six faces and the clinician records the number that belongs to it on a 0 to 10 metric in steps of two, so a valid entry is 0, 2, 4, 6, 8, or 10. The original 1980s version was scored 0 to 5; the Foundation now uses the 0 to 10 metric so that the values sit on the same range as the numeric rating scale, which is why older forms and some study reports still show 0 to 5. The Foundation states that the scale is for people ages three and older, that it measures physical pain only, and that it may not be modified or colored. The artwork and the scripted instructions are the Foundation's copyrighted material and are not reproduced here; administer the scale from the Foundation's licensed material, and in the chart record the scale name, the value, that the patient selected it, and the time. Who chooses the face, the patient or the nurse? Can I use FACES for a nonverbal or sedated adult? The patient, always. The Foundation states that the scale is a self-assessment tool, that it is not to be used by a third person, whether a parent, a caregiver, or a health professional, to assess the patient's pain, and that it should never be used with unresponsive patients or to compare the patient's face to the scale. A nonverbal adult who understands the task and can point can use it; an adult who is sedated, delirious, or otherwise unable to understand cannot, and the ASPMN position statement's hierarchy applies: attempt self-report, consider conditions likely to cause pain, observe behavior with a validated observational tool (the CPOT in intensive care, the PAINAD in advanced dementia, the FLACC in young children), obtain a proxy report labeled as such, and consider an analgesic trial. Document the switch and the reason in the note. In critical care the arousal assessment that decides whether self-report is possible at all is the RASS , and a positive CAM-ICU is a reason to doubt that a chosen face means what it would in an alert patient. What ages is the FACES scale validated for, and when should I use the numeric scale instead? The Foundation states ages three and older, adults included. The evidence is more graded than that: a 2017 systematic review found no evidence that three-year-olds and only weak evidence that four-year-olds can use any published self-report scale validly, IASP commentary places strong evidence at five and older, and the 2019 quality assessment of pediatric measures recommended no measure for children under six. For numbers, Tsze and colleagues found the verbal 0 to 10 scale strong for ages 6 to 17 and not strong at 4 and 5, and von Baeyer and colleagues tentatively supported it from age 8. In practice: offer the numeric scale to a school-age child who can explain what a number would mean and use FACES when they cannot; use FACES or the Faces Pain Scale-Revised for younger children who understand the task; use an observational tool below that; and for adults offer numbers first and move to a faces scale when language, literacy, hearing, or cognitive change gets in the way, remembering that the cognitively impaired older adults in Ware and colleagues' study preferred a faces scale. Whatever you choose, write the reason, and keep the same tool for the same patient unless their ability changes. How do I document a FACES rating so that it stands up to review? Seven elements, one or two chart lines each: the scale named and the reason it was chosen over a numeric or observational tool; the rating written as the patient's choice with the time (Wong-Baker FACES 8, patient selected at 16:25); location, onset, character in the patient's words, and what makes it worse or better; a comfort and function goal agreed with the patient or family; the intervention with drug, dose, route, and time, or the nonpharmacologic measures used; a reassessment timed to the route on the same scale, again patient selected, with the function result and side effects; and a trend that contains only FACES values, with any parent impression or behavioral observation recorded beside the rating rather than in place of it. The sample on this page shows all seven in an emergency department visit, and the blank template carries them as fields. Is a FACES 6 the same as a 6 on the numeric rating scale? Can I trend them together? No, and no. Both scales run 0 to 10, and at the group level they agree well (Garra and colleagues found rho 0.90 against a visual analogue scale in 120 emergency patients), but a FACES value is a choice among six pictures with emotional cues at the anchors and a numeric value is a count the patient generates, and no authority treats them as interchangeable for an individual. Chambers and colleagues showed that faces scales with emotion-laden anchors produce higher ratings than neutral scales, which is the kind of systematic difference that makes a mixed trend misleading. Keep FACES values in their own trend, name the scale beside every number, and when the tool changes because the patient's ability changed, start a new labeled trend and write why. If your electronic record maps both scales into one pain score field, raise it with informatics, because that field produces exactly the mixed trend this page warns against. Does any regulator require the FACES scale, a 30-minute reassessment, or treating a score above 4? None of the three. In the United States, the Medicare Conditions of Participation require a current nursing care plan and complete, dated, timed, authenticated records and name no pain tool (LAW); California requires licensed facilities to assess pain in a consistent manner appropriate to the patient and note it in the chart, and since 2018 no longer ties that assessment to vital signs (LAW, California only). The Joint Commission's PC.01.02.07 requires an organization to define its own criteria to screen, assess, and reassess pain consistent with the patient's age, condition, and ability to understand, and to document the response to interventions and progress toward functional goals; it names no tool and no interval, it stated in 2016 that it does not endorse pain as a vital sign and never required treating pain to a number, and it removed the requirement to assess pain in all patients from hospital accreditation in 2009 (CONVENTION with accreditation force). CMS removed the pain questions from the HCAHPS survey for discharges from October 1, 2019 (PAYER POLICY). In Canada, CAN/HSO 13200:2023 asks for consistent documentation of pediatric pain assessment and does not prescribe an approach; in Australia, the 2022 opioid stewardship clinical care standard carries a quality statement on acute pain assessment and the NSQHS standards name no tool. The 15 to 30 minute and 45 to 60 minute reassessment windows are conventions from drug onset in the AHRQ patient safety chapter, and the idea that a rating above 4 must be medicated appears in no standard. Follow your policy, cite it in the note, and name the trigger for each reassessment. The child chose the top of the scale but is playing, the parent gives a different number, or the child is asleep at reassessment. What do I chart? Chart each as what it is. The child's choice is the self-report and stays in the record as chosen: "Wong-Baker FACES 10, patient selected at 14:05," followed by the observation in its own clause: "playing with blocks, moving freely, no guarding." Do not replace the rating with your reading of the behavior; children choose the top of the scale for fear as well as pain, and the emotion-laden anchors push some choices upward, which is a reason to record both and decide the response clinically. A parent's number is a proxy impression: write it as one ("mother estimates his pain as moderate") and never in the patient's pain score field. A sleeping child cannot choose a face, so no value is entered; describe the sleep and the breathing, note the time, and reassess with the same scale when the child wakes. When an opioid was given, sleep is also the moment to record a sedation score such as the POSS, because sedation precedes respiratory depression and a quiet child is not proof of comfort. Can we print the faces on our forms, put them in our EHR, or show them on our website? Only with the Foundation's permission or license. The artwork is copyrighted, Wong-Baker FACES is a registered trademark of the Wong-Baker FACES Foundation, and the Foundation's pages state that use in facilities, commercial products, or publications will likely need a licensing agreement and fee, that healthcare organizations request a quote through its licensing form, that personal permission is granted through a separate form, that no modifications may be made and no colors added, and that reproductions carry its copyright caption. Its pages also say that licensing is not required for every situation, but they publish no blanket free grant for facility forms, electronic record builds, apps, or websites, and the fact that colorized copies circulate online is not permission. What needs no license is the concept: a field for the value, your own documentation prompts in original words, and the structure on this page. If you need a faces scale you can reproduce without a fee, the Faces Pain Scale-Revised is owned by the International Association for the Study of Pain, which states that no permission is required for clinical, educational, or research use provided it is not modified, with commercial use licensed separately; it is a different scale with different anchors, so name it and do not trend it against Wong-Baker values. Can BastionGPT draft my Wong-Baker FACES documentation? Yes. Give it the facts (the scale used and why, that the patient chose the face and when, the pain location, onset, character, and what changes it, the function goal, the intervention with dose, route, and time, the reassessment value and time on the same scale, the observed behavior and any parent impression, and the plan) and it drafts the full entry: scale and reason, patient-selected rating, assessment in the patient's words, goal, intervention, timed reassessment, observations kept separate from the rating, and the trend and communication lines, ready for your review. It can also check a finished entry for a rating with no scale named, an observer-rated or proxy value written as self-report, a number with no location or onset, an intervention with no timed reassessment, a sleeping patient scored, or FACES and numeric values mixed in one trend. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Wong-Baker FACES Foundation, home page (self-assessment tool; ages three and older; not for use by a third person), FAQ (never with unresponsive patients or for third-party assessment; no modification or added color; organizations may need a license), terms of use (facilities, commercial products, and publications will likely need a licensing agreement and fee; required copyright caption), licensing dashboard (ten use categories; licensing not required for all situations), publishing use (bibliography citation; no modifications without permission), and history (created 1983 at Hillcrest Medical Center, Tulsa; original 0 to 5 scoring; Foundation established 2009); all accessed September 2026. United States Patent and Trademark Office, TSDR record for WONG-BAKER FACES, Reg. No. 4221357 (serial 85543668; filed February 15, 2012; registered October 9, 2012; International Class 010; FACES disclaimed; renewed March 3, 2022); Wong DL, Baker CM, 1988, Pediatric Nursing 14(1):9-17, pain in children: comparison of assessment scales . The Joint Commission, R3 Report Issue 11, August 29, 2017, pain assessment and management standards for hospitals (PC.01.02.07 EP 1, EP 2, and EP 7; effective January 1, 2018), and R3 Report Issue 16, June 25, 2018, pain assessment and management standards for office-based surgeries (ambulatory care, critical access hospital, and office-based surgery standards effective January 1, 2019); American Hospital Association News, April 27, 2016, Joint Commission seeks to clarify misperceptions of its pain standards ; Baker DW, 2017, JAMA 317(11):1117-1118, history of The Joint Commission's pain standards . Cornell Legal Information Institute, 42 CFR 482.23 nursing services and 42 CFR 482.24 medical record services ; CMS, August 2, 2019, FY 2020 IPPS final rule fact sheet (pain management questions removed from the HCAHPS survey beginning with October 1, 2019 discharges); California Legislative Information, Health and Safety Code section 1254.7 (as amended by AB 1048, effective January 1, 2018). Age and validity evidence: Tomlinson D, von Baeyer CL, Stinson JN, Sung L, 2010, Pediatrics 126(5):e1168-e1198, a systematic review of faces scales for the self-report of pain intensity in children ; Garra G and colleagues, 2010, Academic Emergency Medicine 17(1):50-54, validation of the Wong-Baker FACES Pain Rating Scale in pediatric emergency department patients ; Birnie KA and colleagues, 2019, Pain 160(1):5-18, recommendations for selection of self-report pain intensity measures in children and adolescents ; von Baeyer CL and colleagues, 2017, Journal of Pain 18(9):1017-1026, systematic review of self-report measures of pain intensity in 3- and 4-year-old children ; Tsze DS, von Baeyer CL, Pahalyants V, Dayan PS, 2018, Annals of Emergency Medicine 71(6):691-702, validity and reliability of the Verbal Numerical Rating Scale for children aged 4 to 17 years ; von Baeyer CL and colleagues, 2009, Pain 143(3):223-227, three new datasets supporting use of the NRS-11 for children ; Ware LJ, Epps CD, Herr K, Packard A, 2006, Pain Management Nursing 7(3):117-125, evaluation of four pain scales in older minority adults . Affect critique and the Faces Pain Scale-Revised: Chambers CT, Craig KD, 1998, Pain 78(1):27-37, an intrusive impact of anchors in children's faces pain scales ; Chambers CT and colleagues, 1999, Pain 83(1):25-35, a comparison of faces scales: children's and parents' ratings , and 2005, Clinical Journal of Pain 21(3):277-285, faces scales for postoperative pain intensity after minor surgery ; Hicks CL and colleagues, 2001, Pain 93(2):173-183, the Faces Pain Scale-Revised ; International Association for the Study of Pain, Faces Pain Scale-Revised (no permission required for unmodified clinical, educational, or research use; commercial use licensed) and children reporting pain (von Baeyer and Jaaniste; evidence by age). Assessment when self-report is not obtainable and reassessment conventions: Herr K and colleagues, 2024, Pain Management Nursing 25(6):569-570, ASPMN 2024 position statement: pain assessment in the patient unable to self-report , and 25(6):551-568, clinical practice recommendations ; Merkel SI and colleagues, 1997, Pediatric Nursing 23(3):293-297, the FLACC behavioral scale ; Wells N, Pasero C, McCaffery M, 2008, in Patient Safety and Quality: An Evidence-Based Handbook for Nurses (AHRQ), improving the quality of care through pain assessment and management (reassessment intervals by route; comfort-function goal; self-report over behavior); Mularski RA and colleagues, 2006, Journal of General Internal Medicine 21(6):607-612, measuring pain as the 5th vital sign does not improve quality of pain management ; American Academy of Pediatrics, 2001, Pediatrics 108(3):793-797, the assessment and management of acute pain in infants, children, and adolescents , and Fein JA and colleagues, 2012, Pediatrics 130(5):e1391-e1405, relief of pain and anxiety in pediatric patients in emergency medical systems . Canada and Australia: Health Standards Organization, CAN/HSO 13200:2023 Pediatric Pain Management (hospital settings; birth to 19 years; does not prescribe a particular approach) and Solutions for Kids in Pain, executive summary (34 criteria; documentation of pain assessment results in the child's health record; support for self-report); Registered Nurses' Association of Ontario, Pain: Prevention, Assessment, and Management, fourth edition (February 2025); Australian Commission on Safety and Quality in Health Care, Opioid Analgesic Stewardship in Acute Pain Clinical Care Standard (2022; quality statement on acute pain assessment); ANZCA and Faculty of Pain Medicine, Acute Pain Management: Scientific Evidence, fifth edition (2020). ## Related templates RASS Score Documentation CAM-ICU Documentation Glasgow Coma Scale (GCS) Documentation Braden Scale Documentation Morse Fall Scale Documentation Patient Instructions & After-Visit Summary Transfer-of-Care Summary Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/template/oasis-assessment All note templates # OASIS Documentation: Time Points, Item Support & Sample Note Reviewed by the BastionGPT Clinical Advisory Board · Updated September 2026 The OASIS (Outcome and Assessment Information Set) is the assessment item set CMS requires of Medicare home health agencies, in force as OASIS-E2 since April 1, 2026 and collected at start of care, resumption, recertification, transfer, and discharge. Home health nurses and therapists complete it, and its responses drive PDGM payment, star ratings, HHVBP, and survey findings. This page covers how to document a defensible OASIS, with a fictional sample. Copy blank template Download free PDF PDF · 2 pages Editable DOCX Free to use and share. No signup required. Already have session bullets or a transcript? Generate a structured draft with BastionGPT — you review and sign it. Who writes it The assessing clinician: a registered nurse when skilled nursing is ordered, a physical therapist or speech-language pathologist in a therapy-only case, and since January 1, 2022 an occupational therapist when OT is ordered with PT or SLP and no nursing; licensed practical and vocational nurses, therapy assistants, social workers, and aides may contribute through collaboration but cannot own the assessment Audience Agency clinical managers and OASIS review or quality assurance staff, coders and billers who derive the HIPPS code, the physician or allowed practitioner who signs the plan of care, CMS and state surveyors, Medicare Administrative Contractors and other medical reviewers, iQIES and the quality measures built from it, and families reading Care Compare Typical length 400 to 900 words of supporting visit narrative beside the item set at start of care, resumption, and recertification, a few lines at transfer and death at home · 60 to 120 minutes in the home at start of care (clinical team estimate) Format family Federally required structured assessment item set (CMS data specifications, encoded and transmitted through iQIES) collected at fixed time points and read together with the clinician's narrative and the plan of care; a data set with coding conventions, not a scored scale When it's used Start of care within 5 calendar days of the SOC date, resumption of care within 2 calendar days of return home, recertification on days 56 to 60 of each 60-day period, other follow-up after a significant change in condition, and transfer, death at home, and discharge within 2 calendar days; all payers since July 1, 2025, except patients under 18, maternity-only cases, and personal care or chore-only cases Standards context Required by the Medicare home health Conditions of Participation (42 CFR 484.55 and 484.45) and interpreted through the CMS OASIS-E2 Guidance Manual and quarterly Q&As, effective April 1, 2026; a CMS instrument in the public domain, United States only; described here for documentation, item text and response options not reproduced On this page: Template & downloads What it is How to write it Example Compliance Audit errors US, Canada & Australia FAQs ## What is the OASIS assessment? The Outcome and Assessment Information Set is the standardized set of patient assessment items that every Medicare home health agency must collect, encode, and transmit to CMS as part of the comprehensive assessment required by the Conditions of Participation. It was developed in the 1990s by the Center for Health Services and Policy Research at the University of Colorado for the agency now called CMS, became a condition of participation in 1999, and has been revised in numbered versions since: OASIS-C (2010), C1, C2, D (2019), D1 (2020), E (January 1, 2023), E1 (January 1, 2025), and OASIS-E2, which took effect on April 1, 2026 for every assessment with a start of care, resumption of care, recertification, transfer, or discharge date on or after that day and is the version in force as of September 2026 (OMB control number 0938-1279, expiration December 31, 2028); no later version has been announced. Named here by item number and topic only, the E2 item set opens with administrative and time point items (M0080 assessing discipline, M0090 date the assessment was completed, M0100 reason for assessment, M0102 and M0104 ordered and referral dates) and then runs through sensory status (B0200 hearing, B1000 vision, A1110 language), cognition and mood (the C0200 to C0500 mental status interview, C1310 signs of delirium, the D0150 mood interview), functional abilities and goals (GG0130 self-care and GG0170 mobility), health conditions (M1400 shortness of breath, M1033 risk of hospitalization), skin (M1306 and M1311 pressure ulcer and injury items), medications (M2001 drug regimen review, M2003 follow-up, M2005 at transfer or discharge, M2020 oral medication management, N0415 high-risk drug classes), the legacy activity items M1800 to M1870, and the transfer and discharge items J1800 and J1900 (falls) and M2401 (intervention synopsis). The E2 revision replaced the gender item with A0810, replaced the A1250 transportation items with A1255, removed the O0350 vaccination item, added the hearing, vision, and language items at resumption of care, and dropped four social-need items finalized a year earlier before they were ever collected. The item text, response options, and skip logic are CMS's, are read from the instrument and the Guidance Manual, and are not reproduced on this page. The load-bearing fact for documentation is that OASIS is a data set with coding conventions, not a scale, and every response has to be traceable to something the assessing clinician observed or gathered within the assessment window. Two conventions do most of the work and are routinely confused. The legacy items M1800 to M1860 and M1033 are rated on the patient's ability to perform the task safely on the day of assessment, which the Guidance Manual defines as the 24 hours before the visit plus the time in the home, with the usual status rule (what is true more than half of that time) when ability varies; these eight items set the PDGM functional impairment level for payment. The GG0130 and GG0170 items code usual performance, based on a functional assessment at or soon after the start or resumption of care and reflecting the patient's baseline before agency services could improve it; these items feed the Discharge Function Score used in the quality reporting program and the expanded HHVBP model, and they do not drive PDGM payment. Layered on both is the one clinician convention: one qualified clinician is responsible for the comprehensive assessment and signs it, but since January 1, 2018 that clinician may gather information from the patient, caregivers, the physician, the pharmacist, and other agency staff who had direct contact with the patient, provided agency policy allows it, the collaboration happens inside the assessment window, and M0090 records the last date information was gathered. Everything else on this page follows from those rules: the windows that bound each time point, the narrative that must support each response, the correction procedures, and the payment, star rating, value-based purchasing, and survey consequences of getting a response wrong. Neighbors own the clinical content the items summarize. The pressure injury staging page owns how a wound is staged and described, while this page owns how M1306 and M1311 must agree with that wound note; the Braden Scale and Morse Fall Scale pages own the risk tools that the current item set no longer names (the M1300 and M1910 risk-tool items were retired in OASIS-D), and the medication management note owns the reconciliation record behind M2001 to M2005 and N0415. ## Who uses OASIS documentation and when Registered nurses complete the start of care and resumption of care assessments whenever skilled nursing is ordered, physical therapists and speech-language pathologists complete them in therapy-only cases, occupational therapists complete them for Medicare patients when OT is ordered together with PT or SLP and nursing is not, and any of those four disciplines may complete the recertification, other follow-up, transfer, death at home, and discharge assessments once the case is open; licensed practical and vocational nurses, physical therapist assistants, occupational therapy assistants, social workers, and aides contribute observations through collaboration and cannot be the assessing clinician. Agency clinical managers and OASIS review staff read the item set against the visit narrative and the plan of care before the record is locked and transmitted, coders and billers take the eight functional items and the diagnoses into the HIPPS code with occurrence code 50 carrying the M0090 date on the claim, and the physician or allowed practitioner signs a plan of care that 42 CFR 484.60 requires to answer the needs the assessment found. Downstream, CMS builds the PDGM functional level, the Quality of Patient Care star rating, the HH QRP measures, and the expanded HHVBP scores from the transmitted responses; state surveyors reconcile M0030, M0090, M0102, M0104, the visit dates, and the transmission date against the Conditions of Participation; Medicare Administrative Contractors and the Review Choice Demonstration contractors compare functional responses with the notes in medical review; and the Office of Inspector General compares OASIS with claims. The OASIS page owns one job: the time point and its window, who assessed and who collaborated, the responses and the narrative that supports each, the two functional conventions, the medication items, and the correction and transmission trail. Neighbors take over from there: the care coordination note carries the interdisciplinary plan the assessment feeds, the transfer of care summary and the discharge summary carry the patient out of the episode alongside the transfer and discharge OASIS, a fall during the episode is reported through the incident report while J1800 and J1900 record it at transfer or discharge, and the medication management note holds the reconciled list the drug regimen review items summarize. ## How to document an OASIS assessment in the chart CMS prescribes the item set, the data specifications, the time points, and the coding conventions; it does not prescribe the narrative that sits beside them, and that narrative is what a surveyor, a medical reviewer, or a quality auditor reads when a response is questioned. What survives review is a record in which the time point and its window are provable from dates, the assessing clinician and any collaborators are named and the M0090 date is set correctly, every response that drives payment or a measure traces to a dated observation, the day-of-assessment and usual-performance conventions are kept apart, the drug regimen review timing is documented, and the plan of care, the QA review, any correction, and the transmission all reconcile. Each element below carries the pitfall that most often undermines it. Time point, window, and who assessed. Open with the time point (M0100) and the dates that prove its window: the referral date (M0104), any physician-ordered start date (M0102), the start of care date (M0030), the visit dates, and M0090. Under 42 CFR 484.55 the initial assessment visit falls within 48 hours of the referral, within 48 hours of the patient's return home, or on the ordered date, and the comprehensive assessment is finished no later than 5 calendar days after the start of care; the Guidance Manual sets resumption of care within 2 calendar days of the facility discharge or of learning the patient is home, recertification on days 56 to 60 of the 60-day period, other follow-up within 2 calendar days of a significant change, and transfer, death at home, and discharge within 2 calendar days of the event or of learning about it, with no visit required for transfer or death at home. Name the assessing discipline (M0080) and, at start of care, the reason that discipline was eligible: a registered nurse when nursing is ordered, PT or SLP in a therapy-only case, OT only when PT or SLP is also ordered and nursing is not. Pitfall: A start of care with no referral date, so the 48-hour visit cannot be shown, or "OASIS must be done on the first visit" written as a rule when the CoP allows 5 calendar days. One clinician, collaboration, and M0090. State who owns the assessment and who contributed. One qualified clinician is responsible for the comprehensive assessment and signs it, and the signature attests that the responses reflect the patient's status as assessed and supported in the record. Since January 1, 2018 that clinician may gather information from the patient, caregivers, the physician, the pharmacist, and other agency staff who had direct contact with the patient, when agency policy allows it and the input arrives inside the assessment window; agencies may also keep the older practice of limiting the OASIS to the assessing clinician's own findings. Record the contributors and their dates the way your policy requires, and set M0090 to the last date information used to complete the assessment was gathered and documented, which may be a later visit than the first; an assessment may span more than one visit inside the window. A correction made later under the agency's correction policy does not by itself move M0090. Pitfall: A therapist's evaluation pasted in as the functional items with no assessing clinician named, or M0090 set to the first visit while the mobility observations came two days later. Item response and the narrative that supports it. For every response that a measure, the payment grouping, or a survey tag can turn on, the visit narrative must carry the observation behind it: what the patient did, with how much help, using what device, and when. The manual's convention is that current-status items rest on observation and report at the time of the assessment, not on documentation from the hospital or the last certification period. Make the item families agree with the clinical notes that describe the same thing: the M1306 and M1311 skin items with the wound note that stages and measures the injury under the staging conventions , the falls items J1800 and J1900 at transfer and discharge with the fall documentation and the safety report , the M1400 shortness of breath item with the observed activity that produced it, and the interview items (the C0200 to C0500 mental status interview, the D0150 mood interview) with a note that the patient, not a family member, answered. Use a dash only when no information is available, which CMS expects to be rare; where an item carries its own code for a refused, not attempted, or not applicable activity, use that code and write why. Pitfall: A response coded from the hospital discharge summary, a wound item that disagrees with the wound note, or a dash where the item offered a refused or not-attempted code. Functional items M1800 to M1860 and M1033: ability on the day of assessment. These items rate what the patient is able to do safely, not what the patient chose to do or did with a caregiver's habitual help, over the day of assessment: the 24 hours before the visit plus the time in the home. When ability varied across that day, the general convention applies and the response reflects what was true more than half of the time, so write both observations and say which one the response reflects. Because these eight items set the PDGM functional impairment level for both 30-day periods of the certification, the narrative should show the task attempted, the assistance actually required for safety, the device, and any symptom that limited it, in ordinary clinical words rather than in the language of the response options; where the patient refused to attempt a task, record the refusal and the basis for the ability judgment. Pitfall: A patient scored as needing assistance because a daughter always helps, or "independent" for a transfer the clinician never watched. Section GG items: usual performance at baseline. GG0130 and GG0170 code the patient's usual performance across the assessment period, based on a functional assessment at or soon after the start or resumption of care and reflecting the baseline before agency services could improve it; a single best or worst attempt is not the code. Have the patient attempt the activity before instruction that would make the performance more independent, record each attempt with the type and amount of help, and let therapy observations feed the code through collaboration when policy allows. When an activity was not attempted, use the item's own reason code and write the reason (safety, refusal, the environment) rather than a dash, because a dashed item can remove the patient from the Discharge Function Score, the measure that these items feed in the HH QRP and the expanded HHVBP model. Keep the GG narrative separate from the M1800-series narrative: the two conventions can legitimately produce different pictures of the same patient on the same day. Pitfall: GG coded from one assisted walk down the hall, or the M1800-series ability judgment copied into GG as if the conventions were the same. Drug regimen review and the medication items. Document the M2001 drug regimen review at start and resumption of care against the reconciled list in the medication management note : the potential or actual clinically significant issues found, the date and time the physician or allowed practitioner was contacted, and when orders or recommendations were received. The Guidance Manual's standard for M2003 is that the contact and the prescribed or recommended actions were completed by midnight of the next calendar day at the latest, so the timestamps are the evidence, and M2005 carries the same question across the episode at transfer and discharge. N0415 records the high-risk drug classes in the patient's current reconciled regimen even if a dose was held or missed on the day of assessment, which is an exception to the day-of-assessment convention; M2020 rates the patient's ability to take oral medications reliably and safely, and the teaching that follows belongs in the narrative and the plan of care. Pitfall: "Med rec done" with no issue named and no contact time, or an anticoagulant left off N0415 because the patient skipped it yesterday. Plan of care, QA review, corrections, and transmission. Close the loop. The plan of care under 42 CFR 484.60 must specify the care and services that meet the needs identified in the comprehensive assessment, with measurable outcomes and goals, so the problems the responses describe should each have an intervention and a goal in the plan; agency QA review before the record is locked checks item against narrative against plan and returns queries to the assessing clinician. The agency must encode and transmit the completed assessment to CMS within 30 days of completing it, counted from M0090, and the record can be corrected afterward: Chapter 2 of the Guidance Manual provides a modification record for clinical errors and errors in non-key fields and an inactivation record when the event did not occur or a key field is wrong, with 24 months from the assessment target date to submit, modify, or inactivate. Corrections follow the agency's written correction policy, normally by the assessing clinician; a true change in the patient's status after the window closed is new information for the next time point, not a correction. For a late start of care, record the actual dates, the reason the window was missed, and any agency attestation your policy requires; CMS publishes no late-attestation form. Pitfall: A plan of care with a wound order and no wound response, or "the OASIS was transmitted so it cannot be fixed" written into a QA note when the 24-month correction window is open. ### Blank template (copy and adapt) OASIS DOCUMENTATION BLOCK (supporting record for one time point) Date / time: [ ] Agency and setting: [ ] Assessing clinician + discipline (M0080): [ ] Item set version: [OASIS-E2, effective April 1, 2026] Payer: [Medicare / Medicaid / other payer; all-payer collection since July 1, 2025 unless exempt] Time point (M0100): [SOC / ROC / recertification / other follow-up / transfer, not discharged / transfer, discharged / death at home / discharge] Window evidence: referral date (M0104) [ ] ordered start date (M0102) [ ] SOC date (M0030) [ ] visit dates [ ] M0090 date completed [ ] window met: [yes / late: reason and agency attestation per policy] Eligibility of the assessing discipline at SOC: [RN, nursing ordered / PT or SLP, therapy only / OT with PT or SLP ordered and no nursing] Collaboration (agency policy allows: yes / no): [contributor, discipline, date, what was contributed; input received inside the window; M0090 reflects last date] Item responses with supporting narrative (plain clinical words, dated observation): sensory, cognition, mood interviews: [patient answered / refused: item code used] M1400 shortness of breath: [activity observed and response] M1800 to M1860 and M1033 (ability, day of assessment = 24 hours before the visit plus the visit; usual status if it varied): [task, assist, device, symptom] GG0130 self-care and GG0170 mobility (usual performance at baseline, before agency services; each attempt recorded; not-attempted reason code, no dash): [ ] skin M1306 / M1311 agree with the wound note dated: [ ] falls J1800 / J1900 (transfer and discharge) agree with the fall record: [ ] Drug regimen review: M2001 issues found [ ] contact date/time [ ] orders received [ ] (by midnight of the next calendar day) N0415 classes in the reconciled regimen [ ] M2020 oral medication ability [ ] M2005 at transfer or discharge [ ] Dashes used: [none / item and reason; CMS expects dashes to be rare] Plan of care linkage (42 CFR 484.60): [each identified need has an intervention and a measurable goal; sent for signature on ...] QA review: [date; queries and how resolved; corrected before lock under the agency correction policy; M0090 unchanged unless new information was gathered] Transmission: [iQIES date, within 30 days of M0090] Correction after acceptance: [modification / inactivation record, date, reason; 24-month window] Next time point due: [recertification days 56 to 60 / other] Clinician signature / credentials: Date: Download the template (PDF) PDF · 2 pages Editable DOCX Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces OASIS item text or response options. ## Sample OASIS documentation (fictional) Scenario: a US home health agency's start of care for a 79-year-old woman discharged from a hospital heart failure admission, with skilled nursing and physical therapy ordered; the registered nurse documents the window with dates, completes the assessment across two visits with the physical therapist's collaborative input, sets M0090 to the last date information was gathered, supports every functional response with an observation under the correct convention, times the drug regimen review, and records the QA review, the plan of care linkage, and the transmission. All details are fictional. Patient: H.T., 79 · Setting: Home health agency, start of care after hospital discharge (heart failure) · Clinician: A. Brennan, RN · Note date: 09/17/2026 Time point, window, and assessing clinician: Start of care time point, OASIS-E2 item set, traditional Medicare. Referral received from the hospital case manager on 09/14/2026 at 13:20 (M0104), the day of discharge; no physician-ordered start date (M0102 not applicable). Initial assessment visit 09/15/2026 at 10:30 by A. Brennan, RN, within 48 hours of the 09/14/2026 return home; start of care date (M0030) 09/15/2026. Skilled nursing and physical therapy are ordered, so a registered nurse is the assessing clinician (M0080). The comprehensive assessment was completed across the 09/15/2026 and 09/17/2026 visits, inside the 5-calendar-day window that closes 09/20/2026. Face-to-face encounter documented by the hospitalist on 09/13/2026, within 90 days before the start of care. Collaboration and M0090: Collaboration is permitted under agency policy HH-OASIS-04. J. Okoro, PT, evaluated H.T. on 09/16/2026 and reported gait distance, stair attempt, and transfer observations to the assessing RN by secure message the same day; the daughter (primary caregiver) described the home routine on 09/15/2026; the reconciled medication list was confirmed with the pharmacy by telephone on 09/15/2026. The RN considered that input, re-observed transfers and a walk on 09/17/2026, and selected the responses. Contributors are logged in the assessment record per policy. M0090 (date assessment completed) is 09/17/2026, the last date information used to complete the assessment was gathered and documented; the RN's signature attests that the responses reflect H.T.'s status as assessed and supported in this record. Functional items M1800 to M1860 and M1033: ability on the day of assessment: Observed on 09/17/2026 between 09:40 and 10:50, with the preceding 24 hours by report. Ambulation (M1860): walked 25 feet from the recliner to the bathroom with a rolling walker and contact-guard assist, two standing rest breaks, short of breath while talking by the end; she reports the same distance and breaks on 09/16/2026. Bathing (M1830): needs a person present to get in and out of the tub shower with a shower chair and to wash below the knees; washes upper body herself once seated. Lower body dressing (M1820): needs help with compression stockings, socks, and shoes; upper body dressing (M1810): independent, slow. Toilet transferring (M1840): uses the grab bar and the walker, no hands-on help, observed once. Bed to chair transfer (M1850): stands with one-person standby assist and a verbal cue to lock the walker before turning. Grooming (M1800): independent once supplies are within reach. Shortness of breath (M1400): occurs with walking about 20 feet and with lower body dressing, not at rest, not while seated talking. Risk of hospitalization (M1033): supported in the history by two hospital admissions since March 2026, nine scheduled medications, and the exhaustion she describes since discharge. Each response entered on the item set matches the observation written here, and no response was taken from the hospital discharge summary. Section GG: usual performance at baseline: GG0130 self-care and GG0170 mobility were coded on usual performance across the two visits and the PT's 09/16/2026 observations, before the benefit of agency services: no teaching or new technique was introduced before the attempts that were coded. Where attempts differed, usual performance was used and each attempt is listed: on 09/16/2026 she walked 40 feet once with the PT before needing to sit, while the other three observed walks across the two visits stopped at 20 to 25 feet with contact guard, so the walking distance items reflect the shorter distance with contact-guard assistance. Sit to stand and chair to bed transfers needed the same standby assistance and verbal cue on every attempt. Stairs were not attempted for safety on either visit (four exterior steps, no rail; dyspnea at 20 feet), and the stair items carry the item's own not-attempted reason code rather than a dash, with the reason written here. Tub transfer and shower use were observed with the daughter present at the 09/15/2026 visit. These baseline codes are what the Discharge Function Score will be measured against at discharge, so nothing was rounded toward independence. Drug regimen review and medication items: Drug regimen review (M2001) completed 09/15/2026 against the reconciled list of nine scheduled medications (medication management note dated 09/15/2026). Potential clinically significant issue: two loop diuretics, the furosemide started in the hospital and the torsemide from before admission, both present in the home and both taken on 09/14/2026 evening. Cardiology office contacted 09/15/2026 at 11:50; order received 16:40 the same day to stop torsemide and continue furosemide, and the daughter removed the torsemide from the organizer at 17:10 with the RN on the phone. M2003 is therefore documented as contact and prescribed actions completed well before midnight of the next calendar day. N0415 records the anticoagulant (apixaban) in her reconciled regimen although the 09/16/2026 evening dose was missed; the class is coded from the current regimen, not from the day's intake. Oral medication management (M2020): she needs the daughter to fill the weekly organizer and a reminder for evening doses; she takes morning doses correctly from the organizer. Teaching on the single diuretic, daily weights, and the evening-dose alarm is in the plan of care. Other items, plan of care, QA review, and transmission: Skin: head-to-toe inspection on 09/15/2026 and 09/17/2026 found intact skin at every pressure point and device site, so the M1306 pressure injury item and the integumentary narrative agree; a Braden assessment was completed under agency policy since the item set names no risk tool. Falls: one fall in July 2026 without injury, from the history and the daughter's account, recorded in the hospitalization risk factors and in the agency's multifactor fall risk assessment; J1800 and J1900 will be completed at transfer or discharge. Interview items: H.T. answered the mental status and mood interviews herself on 09/15/2026 with the daughter out of the room. Plan of care drafted 09/17/2026 from these findings under 42 CFR 484.60, with goals for walking distance, independent lower body dressing, daily weight logging, and diuretic adherence, each tied to a nursing or therapy intervention, and sent to the cardiologist for signature on 09/18/2026. Agency OASIS review on 09/18/2026 (M. Salcedo, RN, QA) compared items, narrative, and plan and returned three queries (the walking distance code, the stair reason code, and the missing pharmacy contact time); the assessing RN corrected the two documentation gaps and confirmed the code before the record was locked, under agency correction policy, and M0090 stayed 09/17/2026 because no new patient information was gathered. Transmitted to iQIES on 09/21/2026, four days after M0090 and inside the 30-day limit; the transmission confirmation is filed with the assessment. Recertification window falls on days 56 to 60 of the period and is flagged on the agency calendar. This sample is fictional and for educational purposes. It does not describe a real patient or record; the dates, observations, and details are invented to show documentation structure and are not clinical guidance. No OASIS item text or response options are reproduced. ↑ Back to the template and downloads ### Why this sample works The time point and every window are provable from dates that reconcile: referral, return home, initial visit, start of care date, the two assessment visits, M0090, the face-to-face encounter, and the transmission, which is exactly the trail a surveyor reconstructs. The assessing clinician, the eligibility of her discipline, the agency policy that permits collaboration, each contributor with a date, and the M0090 rule are stated, so the two-visit assessment reads as one clinician's responsibility rather than a merged record. Each M1800-series and M1033 response has a dated observation in plain clinical words, with the caregiver's habitual help and the patient's own choices separated from her ability, and the eight payment items rest on what the nurse watched. Section GG is coded on usual performance at baseline with every attempt listed, the not-attempted stairs carry a reason code instead of a dash, and the GG picture is kept apart from the M1800-series picture of the same walk. The drug regimen review shows the issue, the contact time, and the order time against the midnight-next-day standard, N0415 was not filtered by a missed dose, and the QA review, plan of care linkage, correction before lock, and iQIES transmission close the loop without moving M0090. Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript. Generate a note from bullets ## Documentation and compliance considerations United States: the item set is LAW, and most of what happens to a response afterward is PAYER POLICY. Under LAW, 42 CFR 484.55 requires the initial assessment visit within 48 hours of the referral or of the patient's return home, or on the physician-ordered start date; a comprehensive assessment completed no later than 5 calendar days after the start of care; a review of all medications; and updates at least in the last 5 days of every 60 days, within 48 hours of the patient's return from a hospital admission of 24 hours or more, and at discharge. It names the registered nurse as the initial assessor when nursing is ordered and lets the appropriate rehabilitation professional conduct the visit and complete the assessment in a therapy-only case, with the occupational therapy allowance that took effect January 1, 2022. 42 CFR 484.45 requires the agency to encode and electronically transmit each completed assessment to the CMS system within 30 days of completing it, and 42 CFR 484.60 requires an individualized plan of care that specifies the care and services needed to meet the needs identified in the comprehensive assessment, with measurable outcomes and goals. The OASIS-E2 Guidance Manual and CMS's quarterly Q&As carry the sub-regulatory interpretation (the time point table with its 2-calendar-day windows for resumption, transfer, death at home, and discharge and its recertification days 56 to 60, the day-of-assessment and usual-performance conventions, the one clinician convention and its 2018 collaboration expansion, and the Chapter 2 correction policy), and surveyors cite comprehensive-assessment deficiencies under the 484.55 tags. All-payer collection is LAW by rule: since July 1, 2025, OASIS is collected and submitted for every patient beginning skilled home health services regardless of payer, except patients under 18, maternity-only cases, and personal care or chore-only cases, and non-Medicare data enter the annual payment update calculation for patients with a start of care M0090 date on or after January 1, 2027. Under PAYER POLICY, the HH QRP requires a 90 percent submission compliance rate or the agency loses 2 percentage points from its market basket update; PDGM builds the functional impairment level for each 30-day period from M1800, M1810, M1820, M1830, M1840, M1850, M1860, and M1033 alone, with the claim's occurrence code 50 carrying the M0090 date of the assessment that supplied them, and the CY 2027 proposed rule (fact sheet released July 1, 2026) updates the functional point tables, proposes no new OASIS items, and proposes revised HH QRP submission deadlines and calendar-year reporting beginning with the CY 2027 program year; the Quality of Patient Care star rating uses seven measures (timely initiation of care, improvement in ambulation, bed transferring, bathing, shortness of breath, and management of oral medications, and the claims-based within-stay potentially preventable hospitalization measure that replaced the acute care hospitalization measure), needs 20 complete quality episodes per measure and 5 of the 7 measures to report, and refreshes quarterly; the expanded HHVBP model adjusts payment by up to 5 percent in either direction, paid its first adjustment in 2025 on 2023 performance, and for CY 2026 uses six OASIS-based measures (discharge function score, improvement in dyspnea, improvement in management of oral medications, and the newly added improvement in bathing, upper body dressing, and lower body dressing) beside claims-based potentially preventable hospitalization, discharge to community, and Medicare spending per beneficiary measures and the HHCAHPS survey, the total normalized composite measures having ended after 2024. Medical review is PAYER POLICY with teeth: the Review Choice Demonstration runs in Illinois, Ohio, Texas, North Carolina, Florida, and Oklahoma (added December 1, 2023), was extended five years from June 1, 2024 with the minimal-review choice removed, and its reviewers, like the Targeted Probe and Educate and Supplemental Medical Review contractors, read the functional responses against the visit notes; the face-to-face encounter under 42 CFR 424.22 (LAW) must occur within 90 days before or 30 days after the start of care, and the comprehensive assessment is where homebound status and skilled need are usually evidenced. CONVENTION covers the rest: the visit narrative that supports each response, the QA review before the record is locked, an agency correction policy that names who corrects when the assessing clinician is unavailable, and the late-start attestation some agencies require; none of these is a CMS form, and CMS publishes none. The version line and the borders are part of the record, and the evidence explains why the narrative matters. OASIS-E2 is the item set and Guidance Manual in force since April 1, 2026, an off-cycle change from OASIS-E1 (January 1, 2025) and OASIS-E (January 1, 2023); the E2 instrument carries OMB control number 0938-1279 with an expiration date of December 31, 2028, the CMS OASIS Data Sets and User Manuals pages list no later version, and the CY 2027 proposed rule proposes no new items, so a note written in late 2026 names OASIS-E2 and watches the CMS pages for a draft successor rather than assuming one. The instrument is United States only: Canada's home care programs assess with interRAI Home Care and Australia's aged care system uses its own national assessment tools, and OASIS has no standing in either, so this page and its templates apply to US home health agencies alone. On the evidence: in the original reliability study, two clinicians independently assessed 66 home health patients and inter-rater reliability was excellent (kappa above 0.80) for many items and substantial (kappa above 0.60) for most (Hittle and colleagues, Home Health Care Services Quarterly, 2003); Madigan and Fortinsky (The Gerontologist, 2004) examined inter-rater reliability of the item set, Tullai-McGuinness, Madigan, and Fortinsky (Home Health Care Services Quarterly, 2009) tested the validity of OASIS items against established instruments, and O'Connor and Davitt (Home Health Care Services Quarterly, 2012) reviewed the validity and reliability literature; neither research pass for this page located peer-reviewed inter-rater reliability data for the Section GG items in home health, a gap to remember when GG codes are audited. The Office of Inspector General found that 55 percent of the falls with major injury and hospitalization it identified in Medicare claims were not reported on the associated OASIS assessments, with worse reporting for Black, Hispanic, and Asian patients and by for-profit agencies (OEI-05-22-00290, September 5, 2023), and CMS concurred with all four recommendations; MedPAC's March 2026 report to Congress repeats the Commission's concern about the validity of the provider-reported patient assessment information used for case mix, reports a 21.2 percent fee-for-service Medicare margin for freestanding agencies in 2024, and recommends a 7 percent reduction to the 2026 base payment rate for 2027. Those findings are why a response with no dated observation behind it is the defect every reviewer looks for first. OASIS is a Centers for Medicare & Medicaid Services instrument. The item set, the data specifications, and the Guidance Manual are published by CMS as works of the United States government, which federal copyright law places in the public domain, and the instrument's original developers at the Center for Health Services and Policy Research in Denver stated, in a permission notice that CMS now archives on its OASIS Data Sets page, that OASIS is their intellectual property, may not be copyrighted by any other party, and may be freely copied and incorporated into assessment forms and software by home care providers, organizations, and individuals with no royalty or use fee, with an acknowledgement of authorship expected. Reproducing the items in an agency's own forms or electronic record is therefore permitted, while submission to CMS must follow the CMS data specifications through iQIES, and the screening tools embedded in the item set carry their own status: the PHQ-2 to 9 mood interview is free to use, the mental status interview is a government-developed instrument, and the delirium item derives from the Confusion Assessment Method, whose rights belong to its author. BastionGPT is not affiliated with, or endorsed by, the Centers for Medicare & Medicaid Services. This page reproduces no test items, stimuli, norms, or scoring materials. ↑ Back to the template and downloads ## Common OASIS documentation errors reviewers flag The numbers behind these errors are specific. In the reliability study run by the instrument's developers, two clinicians independently assessed 66 home health patients and inter-rater reliability was excellent (kappa above 0.80) for many items and substantial (above 0.60) for most, which means a minority of items disagree between trained raters before any workflow pressure is applied (Hittle and colleagues, 2003). The Office of Inspector General found that 55 percent of the falls with major injury and hospitalization it identified in Medicare claims were missing from the associated OASIS assessments, that for-profit agencies and the care of Black, Hispanic, and Asian patients showed worse reporting, and that the agencies with the lowest published fall rates reported falls least often (OEI-05-22-00290, September 5, 2023). MedPAC's March 2026 report to Congress repeats its concern about the validity of provider-reported patient assessment information used for case mix, beside a 21.2 percent fee-for-service Medicare margin for freestanding agencies in 2024 and a recommended 7 percent rate reduction for 2027. The HH QRP threshold is 90 percent submission compliance with a 2 percentage point penalty, the Review Choice Demonstration in six states reads functional responses against visit notes claim by claim, and every accepted record can be modified or inactivated for 24 months. No published audit counts how often a chart carries an OASIS response with no supporting observation; the survey, payment, and medical review rules described under compliance considerations are what turn that omission into a deficiency or a denial. The BastionGPT Clinical Advisory Board sees the same errors most often in OASIS documentation reviews: A response with nothing behind it. An ambulation response that says the patient needs help while the visit note says she walked to the door to let the nurse in, a skin item that disagrees with the wound note, a fall in the history that never reaches J1800 at discharge, or a mood interview answered by the daughter. Every response a measure or the payment grouping can turn on needs a dated observation in ordinary clinical words in the same record, and the item families must agree with the clinical notes that describe the same thing. Ability, performance, and usual performance confused. The M1800-series items rated on what the patient does with a caregiver's habitual help rather than what she is able to do safely on the day of assessment; GG coded from a single best or worst attempt instead of usual performance at baseline; or the M-item judgment copied into GG as if the conventions were the same. Write the attempts, name the convention each response follows, and let the two pictures differ when the rules make them differ. Windows that cannot be proven, and M0090 in the wrong place. No referral date in the record, so the 48-hour initial visit is unprovable; "OASIS on the first visit" written as a rule when the CoP allows 5 calendar days; M0090 set to the first visit while the mobility observations came two days later; the 30-day transmission clock run from the visit date instead of M0090; or a recertification outside days 56 to 60 with no explanation. Record M0104, M0102, M0030, every visit date, M0090, and the transmission date so a surveyor can reconcile them. Collaboration misread in both directions. A therapist's evaluation pasted in as the functional items with no assessing clinician named, a licensed practical nurse or therapy assistant signing as the assessor, or the opposite error: GG items dashed because the nurse believed nobody else's observations could be used. Since January 1, 2018 the assessing clinician may use input from the patient, caregivers, the physician, the pharmacist, and agency staff who had direct contact with the patient when policy allows and the input falls inside the window; name the contributors and keep one signature. Dashes and copied hospital data. A dash where the item offered a refused, not attempted, or not applicable code, which can drop the patient from the measure the item feeds; a response taken from the hospital discharge summary instead of the home visit; or a not-attempted activity with no reason written. The manual expects dashes to be rare and current-status items to rest on observation at the time of the assessment; use the item's own code and say why. Corrections and the plan of care mishandled. "The OASIS was transmitted so it cannot be fixed" in a QA note; a key field edited instead of the record being inactivated and resubmitted; a change in the patient's status after the window closed retro-coded into the accepted record; a wound order in the plan of care with no skin response, or a mobility problem in the items with no goal in the plan; or a late start of care with no reason recorded. Chapter 2 allows modification and inactivation for 24 months under the agency's correction policy, 42 CFR 484.60 ties the plan to the assessment, and a late assessment needs the real dates and the reason, not a backdated M0090. How BastionGPT helps BastionGPT is specifically trained, tuned, and clinically tested on home health assessments and plan of care documentation. Give it the facts (the time point and the dates that prove its window, the assessing discipline and why it was eligible, each collaborator with a date and what they contributed, the observations behind each functional response under the convention that applies, the drug regimen review issue with its contact and order times, the skin and fall findings, the plan of care goals, the QA queries, and the transmission date) and it drafts the supporting record: window statement, collaboration and M0090 line, item-by-item narrative in plain clinical words, Section GG baseline with attempts listed, medication items with timestamps, and the plan of care, QA, and transmission lines, ready for your review. Cross-check a finished record for the gaps reviewers flag: a response with no observation behind it, ability and usual performance confused, a window that cannot be proven from the dates, M0090 on the wrong visit, a collaborator unnamed, a dash where a code exists, a skin or fall item that disagrees with the clinical note, or a plan of care that does not answer the assessment. Turn the record into the next document: the QA query to the assessing clinician, the plan of care problem and goal list for the physician's signature, the functional summary for the physical therapist or the transfer of care summary, or the modification record justification for the agency correction log, ready to confirm against the record. See how clinicians use it day to day on the AI medical notes page. Many BastionGPT users report saving more than 90 minutes per day on documentation. HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign. Start your free trial How we clinically test ## Frequently asked questions Which OASIS version is in force, and what changed in OASIS-E2? OASIS-E2, effective April 1, 2026 for every assessment with a start of care, resumption of care, recertification, transfer, or discharge date on or after that day, an off-cycle change from OASIS-E1 (January 1, 2025) and OASIS-E (January 1, 2023). The CMS OASIS Data Sets and User Manuals pages carry the E2 instruments, the final E2 Guidance Manual, and the E2 Q&As, and the instrument carries OMB control number 0938-1279 with an expiration date of December 31, 2028; no later version has been announced, and the CY 2027 proposed rule proposes no new items. Described at the topic level, E2 replaced the gender item with A0810, replaced the A1250 transportation items with A1255, removed the O0350 vaccination item when its measure was retired, added the B0200 hearing, B1000 vision, and A1110 language items at resumption of care, revised the J1900 fall-injury descriptions, and dropped four social-need items that had been finalized a year earlier before they were ever collected. Many training materials and vendor pages still call E1 current; name E2 in your policy and your record, and read the E2 manual, not a cached E1 copy, for item guidance. What are the OASIS time points and completion windows, and does every one need a visit? Eight reasons for assessment sit under M0100 in six time points: start of care, resumption of care, follow-up (recertification, and other follow-up after a significant change), transfer to an inpatient facility (with or without discharge from the agency), death at home, and discharge. The Conditions of Participation set the outer limits: the initial assessment visit within 48 hours of the referral or of the patient's return home, or on the physician-ordered date; the comprehensive assessment completed no later than 5 calendar days after the start of care; updates in the last 5 days of every 60 days, within 48 hours of return from a hospital admission of 24 hours or more, and at discharge. The Guidance Manual's time point table fills in the rest: resumption of care within 2 calendar days of the facility discharge date or of learning the patient is home, recertification on days 56 to 60 of the current 60-day period, other follow-up within 2 calendar days of a significant change in condition, and transfer, death at home, and discharge within 2 calendar days of the event or of learning about it. Transfer and death at home are completed from the record and do not require an in-person visit; every other time point does. The 30-day transmission clock under 42 CFR 484.45 runs from M0090, the completion date, not from the visit. Write the dates that prove each window into the record, because a surveyor reconciles M0104, M0102, M0030, the visit dates, M0090, and the iQIES date. Who can complete the OASIS? Can a therapist do the start of care, and can an LPN or a PTA help? The assessing clinician must be a registered nurse, physical therapist, speech-language pathologist, or occupational therapist, and which one depends on the orders. When skilled nursing is ordered, the registered nurse conducts the initial assessment visit and completes the start of care comprehensive assessment. In a Medicare therapy-only case the physical therapist or speech-language pathologist may do both, and since January 1, 2022 an occupational therapist may do both when the referral includes PT or SLP along with OT and no nursing; OT alone does not establish initial Medicare eligibility, so an OT-only referral cannot open a Medicare case, although a patient who qualified through nursing, PT, or SLP keeps eligibility if OT becomes the only remaining discipline. Once the case is open, any of the four qualified disciplines may complete the recertification, other follow-up, resumption, transfer, death at home, and discharge assessments. Licensed practical and vocational nurses, physical therapist assistants, occupational therapy assistants, social workers, and aides may not be responsible for the comprehensive assessment or sign it, but under the collaboration convention their direct observations may inform the assessing clinician's responses when agency policy allows and the input arrives inside the window. Write the assessing discipline in M0080 and, at start of care, the reason it was eligible. What does the one clinician rule mean now, and how do I document collaboration and M0090? One qualified clinician is responsible for the comprehensive assessment and signs it, and the signature attests that the responses reflect the patient's status as assessed and supported in the record. What changed on January 1, 2018 is that the assessing clinician may collect data for any or all OASIS items with input from the patient, caregivers, and other health care personnel, including the physician, the pharmacist, and other agency staff who had direct in-person contact with the patient or another means of gathering information (a monitoring device, a photograph, a phone call), provided agency policy allows it and the collaboration happens inside the assessment timeframe and consistent with item guidance; the clinician weighs the input and selects the response. Agencies may instead keep limiting the OASIS to the assessing clinician's own findings, and either way the agency is responsible for tracking who contributed. M0090 is the last date information used to complete the assessment was gathered by the assessing clinician and documented, so a two-visit start of care carries the second visit's date, and a later correction under the agency's policy does not by itself change it. Document the policy that permits collaboration, each contributor with the date and what they contributed, the assessing clinician's own re-observation where one occurred, and the M0090 rule applied, as the sample on this page does. My patient did a task once with help and once alone. How do I score the M1800-series items and the GG items? Under two different rules, and the record should show both. The legacy items M1800 to M1860 rate the patient's ability to perform the task safely on the day of assessment, defined as the 24 hours before the visit plus the time in the home; the general convention for status that varies is to report what is true more than half of that period, and ability is not willingness or the help a caregiver habitually gives. The GG0130 and GG0170 items code usual performance across the assessment period, from a functional assessment at or soon after the start or resumption of care that reflects the baseline before agency services could improve it, so a single best attempt and a single worst attempt are both wrong codes; if the only help needed was retrieving a device, the manual treats that as setup assistance. Write each attempt with the help required, name the convention behind each response, and expect the two sections to describe the same walk differently when the rules make them differ. When an activity was not attempted, use the item's own reason code and write the reason rather than a dash, and record a refusal as a refusal. Collaboration lets the therapist's observations inform both sections when policy allows. Which OASIS items drive PDGM payment, the star rating, and HHVBP? Different items for different programs, which is why a myth persists that GG drives payment. PDGM's functional impairment level for each 30-day period comes from eight items, M1800, M1810, M1820, M1830, M1840, M1850, M1860, and M1033, taken from the most recent start of care, resumption, recertification, or other follow-up assessment before the claim period, with occurrence code 50 on the claim carrying that assessment's M0090 date; the GG items play no part in the payment grouping, and the CY 2027 proposed rule keeps that design while updating the point tables. The Quality of Patient Care star rating uses seven measures: timely initiation of care, improvement in ambulation, bed transferring, bathing, shortness of breath, and management of oral medications, and the claims-based within-stay potentially preventable hospitalization measure that replaced acute care hospitalization; an agency needs 20 complete quality episodes per measure and 5 of the 7 measures, refreshed quarterly. The expanded HHVBP model, which adjusts payment by up to 5 percent in either direction with the first adjustment paid in 2025 on 2023 performance, uses six OASIS-based measures for CY 2026: the discharge function score built from the GG items, improvement in dyspnea, improvement in management of oral medications, and the newly added improvement in bathing, upper body dressing, and lower body dressing, beside the claims-based and HHCAHPS measures. The practical point is that the same functional responses reach three programs by different routes, and a response that cannot be supported is exposed in all of them. Can an OASIS be corrected after it is transmitted, and what do I do about a late start of care? Yes, for 24 months, and the myth that a submitted OASIS is frozen costs agencies real corrections. Chapter 2 of the Guidance Manual describes the automated correction policy in iQIES: a modification record corrects clinical errors and errors in non-key fields in a record that has been accepted, and an inactivation record is used when the event did not occur or any key field is wrong, because key fields cannot be edited and the record must be inactivated and resubmitted. Since January 1, 2020 agencies have had 24 months from the assessment target date to submit, modify, or inactivate. Corrections follow the agency's written correction policy and are normally made by the assessing clinician, with a supervisor or QA staff correcting under the policy when that clinician is unavailable and the reason documented; a genuine coding error is corrected whenever it is found, while a true change in the patient's status after the window closed is information for the next time point, not a correction of the last one, and M0090 does not necessarily change when an error is fixed. Before public reporting, agencies also get a preview period to review and correct. For a late start of care, record the actual visit and completion dates, set M0090 to the real completion date, document why the 5-day window was missed, and add whatever attestation your agency policy requires; CMS publishes no late-attestation form, and a backdated M0090 is a falsified record rather than a late one. Does OASIS apply to non-Medicare patients, and can we put the items in our own forms and EHR? Yes to both. Since July 1, 2025, after a penalty-free phase-in from January 1 to June 30, 2025, OASIS is collected and submitted for every patient who begins skilled home health services regardless of payer, with three exclusions: patients under 18, patients receiving maternity services only, and patients receiving only personal care, housekeeping, or chore services; non-Medicare data begin to count toward the annual payment update for patients whose start of care M0090 date is January 1, 2027 or later. On rights, OASIS is a CMS instrument published as a work of the United States government, and its original developers at the Center for Health Services and Policy Research in Denver granted everyone a free, nonexclusive right to copy the items into printed forms, software, and other products with no royalty, expecting an acknowledgement of authorship; CMS archives that permission notice on its OASIS Data Sets page. So an agency may build the item set into its own assessment form and electronic record, provided the transmitted data follow the CMS data specifications through iQIES, and a vendor's build should carry the current E2 specifications. Two cautions: the embedded screening tools (the PHQ-2 to 9 mood interview, the mental status interview, the delirium item derived from the Confusion Assessment Method) have their own rights positions, and this page and its templates reproduce none of the item text because the documentation guidance, not the instrument, is the point. Can BastionGPT draft my OASIS documentation? Yes. Give it the facts (the time point and the dates that prove its window, the assessing discipline and why it was eligible, each collaborator with a date and what they contributed, the observations behind each functional response under the convention that applies, the drug regimen review issue with its contact and order times, the skin and fall findings, the plan of care goals, the QA queries, and the transmission date) and it drafts the supporting record: the window statement, the collaboration and M0090 line, the item-by-item narrative in plain clinical words, the Section GG baseline with attempts listed, the medication items with timestamps, and the plan of care, QA, and transmission lines, ready for your review. It can also check a finished record for a response with no observation behind it, ability and usual performance confused, a window that cannot be proven, M0090 on the wrong visit, a collaborator unnamed, a dash where a code exists, or a plan of care that does not answer the assessment. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models. ## Primary sources The instrument facts and compliance claims on this page trace to these sources, last verified September 2026: Centers for Medicare & Medicaid Services, OASIS data sets (OASIS-E2 instruments effective April 1, 2026; OMB control number 0938-1279, expiration December 31, 2028; archived right-to-copy notice), OASIS-E2 All Items instrument (item numbers and time points; item text not reproduced), OASIS user manuals , and final OASIS-E2 Guidance Manual (Chapter 1 time point table and general conventions, including the day of assessment definition, the usual status rule, the dash convention, and the one clinician convention with collaboration; Chapter 2 data accuracy and correction, modification and inactivation records, 24 months; Section GG usual performance and baseline instructions; drug regimen review timing); Center for Health Services and Policy Research, Right to copy, reprint, and use OASIS (archived by CMS; free nonexclusive use, no royalty, acknowledgement of authorship expected). CMS OASIS Q&A, August 2017, Expansion of the one clinician convention (effective January 1, 2018; collaborators; M0090 as the last date input was gathered or received; agencies may still limit the OASIS to the assessing clinician's own data); CMS QIES Technical Support Office, OASIS Q&As and iQIES resources (quarterly and category Q&As; key fields and correction details). Legal Information Institute, 42 CFR 484.55, comprehensive assessment of patients (initial assessment visit within 48 hours; completion within 5 calendar days; who may assess; drug regimen review; update in the last 5 days of every 60 days, within 48 hours of return from a 24-hour inpatient stay, and at discharge), 42 CFR 484.45, reporting OASIS information (encode and transmit within 30 days of completing the assessment), 42 CFR 484.60, care planning (plan of care based on the comprehensive assessment; measurable outcomes and goals), and 42 CFR 424.22 (face-to-face encounter within 90 days before or 30 days after the start of care). CMS, Home Health Quality Reporting Program requirements (time points; all-payer phase-in January 1 to June 30, 2025 and mandatory collection from July 1, 2025; exclusions; 90 percent compliance threshold; 2 percentage point reduction; non-Medicare data in the annual payment update from start of care M0090 dates on or after January 1, 2027); CMS, home health quality measures ; CMS, home health star ratings (seven Quality of Patient Care measures; 20 complete quality episodes; 5 of 7 measures; quarterly refresh). CMS, Home Health Patient-Driven Groupings Model (admission source and timing, clinical grouping, functional impairment level, comorbidity adjustment; 432 groups); CGS Administrators, PDGM: functional impairment level and occurrence code 50 (M1800, M1810, M1820, M1830, M1840, M1850, M1860, M1033; occurrence code 50 carries the M0090 date); CMS, November 28, 2025, CY 2026 Home Health Prospective Payment System final rule fact sheet (CMS-1828-F; four social-need items and the vaccination measure removed; HHVBP measures added and removed); CMS, July 1, 2026, CY 2027 Home Health Prospective Payment System proposed rule fact sheet (CMS-1844-P; no new OASIS items; functional impairment level tables updated; HH QRP submission deadlines and calendar-year reporting proposed from the CY 2027 program year). CMS Innovation Center, expanded Home Health Value-Based Purchasing Model (applicable measure set; six OASIS-based measures for CY 2026 including improvement in bathing, upper body dressing, and lower body dressing; claims-based and HHCAHPS measures; total normalized composite measures through CY 2024; payment adjustment of up to 5 percent in either direction; first payment year 2025 on 2023 performance); CMS, Review Choice Demonstration for Home Health Services (Illinois, Ohio, Texas, North Carolina, Florida, Oklahoma from December 1, 2023; five-year extension effective June 1, 2024; minimal-review choice removed). US Department of Health and Human Services, Office of Inspector General, September 5, 2023, Home health agencies failed to report over half of falls with major injury and hospitalization among their Medicare patients (OEI-05-22-00290; 55 percent; disparities by race and ownership; four recommendations, CMS concurred); Medicare Payment Advisory Commission, March 2026, Report to the Congress, chapter 8, home health care services (validity of provider-reported assessment information; 21.2 percent fee-for-service Medicare margin in 2024; recommendation to reduce the 2026 base payment rate by 7 percent for 2027). Reliability and validity: Hittle DF, Shaughnessy PW, Crisler KS, and colleagues, 2003, Home Health Care Services Quarterly 22(4):43-63, a study of reliability and burden of home health assessment using OASIS (66 patients; kappa above 0.80 for many items and above 0.60 for most); Madigan EA, Fortinsky RH, 2004, The Gerontologist, interrater reliability of the Outcomes and Assessment Information Set ; Tullai-McGuinness S, Madigan EA, Fortinsky RH, 2009, Home Health Care Services Quarterly, validity testing the Outcomes and Assessment Information Set ; O'Connor M, Davitt JK, 2012, Home Health Care Services Quarterly, the Outcome and Assessment Information Set: a review of validity and reliability . ## Related templates Pressure Injury Staging Documentation Braden Scale Documentation Morse Fall Scale Documentation Discharge Summary Transfer-of-Care Summary Medication Management Follow-Up Note Care Coordination Note Browse all note templates → Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements. --- > Source: https://bastiongpt.com/company/about About · BastionGPT # The HIPAA-compliant AI built for healthcare. For three years, BastionGPT has helped clinicians, practices, and health systems use generative AI without putting patient data at risk. Built by healthcare technology veterans, trusted by 10,000+ healthcare organizations. Meet the team See our advisors How we build BastionGPT ## Four principles we won't trade off. These aren't posters in a break room. They're how we decide what ships and what doesn't. ### Healthcare or nothing BastionGPT is built exclusively for healthcare. Every model, prompt template, and workflow is tuned for clinical, administrative, and compliance use cases, not retrofitted from a general-purpose chatbot. ### HIPAA-compliant by default A signed BAA is included with every plan. Your data is encrypted, isolated, and never used to train models. Privacy and security are part of the product, not an enterprise upsell. ### Built with clinicians Practicing physicians, psychologists, and nurses on our advisory board review the templates, evaluations, and safety guardrails. If a workflow wouldn't survive contact with a real clinic, it doesn't ship. ### Multi-model, single interface BastionGPT integrates the leading frontier models, including OpenAI, Anthropic, and Google, behind one HIPAA-compliant interface. You get the best of each, without the procurement overhead. Our history ## Five years building healthcare's most trusted AI. Five years of focused engineering. Today, BastionGPT processes 1.3 million clinical documents each month and saves clinicians 15,000 hours every day. 2011 ### Founders lead AI innovation in major health systems Our executives lead the evaluation and implementation of AI technologies for healthcare providers with an annual spend totaling $4.9 billion. 2021 ### Bastion is founded Founded with a bet that generative AI in healthcare would need dedicated, healthcare-only expertise, not a side practice of a generalist software company. 2023 ### BastionGPT launches We ship the first version of BastionGPT, the HIPAA-compliant AI platform with a signed BAA on every plan, built in response to repeated requests for a compliant alternative to ChatGPT. 2024 ### Microsoft Partner status; multi-model integration Admitted to the Microsoft Partner Network for Azure AI deployments. BastionGPT expands to integrate OpenAI, Anthropic, and Google models behind one compliant interface. 2025 ### AI Medical Scribe launches BastionGPT's AI Medical Scribe arrives with up to 4-hour recordings to meet the demands of psychology and long-form clinical sessions, plus best-in-class speaker recognition for more accurate, multi-speaker notes. Same HIPAA-compliant foundation, purpose-built for the way clinicians actually document. 2026 ### 10,000 organizations, one focus Over 10,000 healthcare organizations now use BastionGPT, from solo psychiatrists to integrated delivery networks, all held to the same compliance posture we started with. Leadership ## The team behind BastionGPT. Decades of combined leadership across healthcare systems, enterprise security, and AI, guiding BastionGPT through scale, compliance, and clinical trust. JS ### Josh Spencer CEO Founded Bastion Intelligence and sets overall strategy, architecting BastionGPT's healthcare-exclusive mandate across product, security, and research. Previously spent 13 years at UT Southwestern Medical Center, a $4.9B academic medical center, as CISO and later CTO, leading 120+ technology professionals and earning commendation from HHS. MBA (cum laude), UTSA; AI/ML certificates from Stanford and Microsoft. Adjunct faculty, UT Alliance for Physician Leadership. RH ### Robert Henehan COO Runs administrative operations and enterprise customer relationships at Bastion Intelligence. Previously scaled analytics, EHR, and business intelligence programs across leading health systems, most recently as Director of Analytics and Business Applications at El Camino Health in Silicon Valley. Former Senior Consultant at Nordic Global and Huron; former Project Manager at Epic. BS, Yale; Executive MHSA, University of Michigan. AT ### Amber Turrill CMO Leads brand, demand generation, and communications for Bastion Intelligence and BastionGPT. Brings more than a decade of experience in SEO, paid media, demand generation, and analytics, including Director of Digital Marketing and Demand Gen at SpiderOak, a cybersecurity software company. MA, University of Maryland. NC ### Nicolas Cava CTO Leads engineering, infrastructure, and AI model integration for BastionGPT. Brings 15+ years building and managing product engineering teams at PetalMD (healthtech), Swile (fintech, $70M Series C), and CrakRevenue (adtech/big data), with deep expertise in the organizational and architectural patterns that let companies ship production-quality software at velocity. CS ### Catherine Spencer, RN Head of Customer Experience Leads customer onboarding, billing, and the sales and support experience at Bastion Intelligence, serving the thousands of clinicians and healthcare organizations that rely on BastionGPT. Has built the customer-facing systems and standards supporting a healthcare-only customer base with strict privacy and compliance expectations since 2023. JC ### Joshua Cruz Customer Success Manager Leads product demonstrations, evaluation workflows, and conference engagement for BastionGPT customers. Partners with clinicians and practice leaders to translate platform capabilities into measurable workflow impact, with a background in healthcare AI go-to-market. MZ ### Marie Zhang Head of User Experience Owns the interface design and user experience, site architecture, and marketing web properties for Bastion Intelligence and BastionGPT. Background as a UX and Webflow designer focused on high-converting marketing experiences for SaaS and technology companies. ER ### Evan Rallis Head of Product Advises product strategy, roadmap, and discovery for BastionGPT. Four-time founder and product leader, most recently Head of Product at Grantable (AI-assisted grant writing) and VP of Product at Grata (B2B data platform). Reforge alum with deep experience in product-led growth across B2B and B2C SaaS. NV ### Nirav V. Lead Developer Serves as the primary technical lead for BastionGPT development. Brings a decade of full-stack experience across enterprise financial and healthcare platforms, with published technical writing on modern JavaScript architecture. BE, Computer Engineering. AR ### Amin Raiyani Lead Front End Designer Translates BastionGPT's product into clean, accessible interfaces for healthcare users. Founder and CEO of Amilek Consultancy Solution, with 10+ years designing and shipping web apps, CRMs, and SaaS products. Previously Senior Front End Developer at smartSense Consulting Solutions (2015 to 2023). Deep expertise in Angular, React, and modern design systems. TP ### Ty Pierre Senior Prompt Engineer Optimizes the prompting layer across BastionGPT's integrated OpenAI, Anthropic, and Google models for safety, accuracy, and clinical documentation standards. Works closely with customers to develop both foundational and advanced prompting skills. Focused on the healthcare-specific tuning that differentiates BastionGPT from general-purpose AI, including evidence-based guardrails, clinical templates, and workflow-specific response design. MM ### Michael Maxted, MD Chief Clinical Advisor Advises Bastion Intelligence on clinical workflow safety, physician usability, and real-world deployment of generative AI in surgical and outpatient settings. Board-certified orthopaedic surgeon practicing in Oklahoma City with subspecialty training in sports medicine and arthroscopy. MD, UT Southwestern Medical School; peer-reviewed publications in Arthroscopy Techniques and the Journal of Orthopaedic Case Reports. Clinical Advisory Board ## Practicing clinicians, at the center of every decision. Practicing clinicians across psychology, primary care, pediatrics, hospital medicine, and air medical transport. They review BastionGPT's evaluation rubrics, stress-test our safety work, and tell us when a feature wouldn't survive contact with a real clinic. AD ### Dr. Anna DiNoto, PsyD, LMHC, CFMHE Forensic Psychology · Neurodiversity Clinical psychologist and founder of Sandbox Therapy Group in Monroe, Washington. Expert in autism assessment and neurodivergent populations across the lifespan, with a special focus on neurodivergent females. Advises on documentation standards for forensic evaluations and neurodiversity-informed clinical workflows. Founder, Sandbox Therapy Group; Faculty, The New International University KK ### Kolby Kolbet, RN, MSN, CFRN, FACHE, CMTE Air Medical · Clinical Innovation Chief Innovation Officer at Life Link III, a nonprofit air medical transport consortium serving the upper Midwest. Fifteen years in air medical transport, including a decade as VP and Chief Clinical Officer. Leads integration of ambient AI for clinical documentation in flight medicine and advises on real-time care documentation workflows. Chief Innovation Officer, Life Link III EL ### Emma Levy Educational Psychology · Learning Assessment Clinical psychologist and co-founder of How I Learn, a Wellington-based educational assessment service for children, adolescents, and adults. Specializes in evaluations for learning disorders, giftedness, and intellectual disability. Advises on pediatric and educational documentation accuracy and family-centered communication. Clinical Psychologist, How I Learn, New Zealand AM ### Dr. Adil Manzoor, DO, MBA, MS Primary Care · Pediatrics Internal medicine and pediatrics physician; Medical Director of HouseCall Doctors in New Jersey. Brings clinical experience spanning pediatric and adult ERs, ICU settings, and rehabilitation facilities. Advises on primary care documentation workflows and preventive medicine use cases. Aspen Health Innovators Fellow; Cornell MBA MS ### Dr. Marta M. Shinn, PhD Child Psychology · Assessment Licensed clinical child psychologist and founder of Variations Psychology in Orange County, California. Specializes in independent diagnostic testing, special education planning (IEPs, 504s), and mental health interventions for underserved child populations. Serves as a Psychology Subject Matter Expert for the California Board of Psychology and national licensing bodies. Founder, Variations Psychology; PCIT Trainer, UC Davis KS ### Dr. Kendall Smith, MD Hospital Medicine · Revenue Cycle Board-certified hospitalist and Chief Medical Officer of PayerWatch, with 25 years of experience in EMR technology, medical coding, and denial/appeals management. Designed clinical documentation improvement programs at major health systems. Advises on clinical documentation quality, coding accuracy, and payer compliance use cases. CMO & Chief Physician Advisor, PayerWatch; SFHM RS ### Dr. Rachael St.Claire, PsyD Psychology · Chronic Illness Licensed clinical psychologist in Colorado with over 30 years of experience, including 28 years at Kaiser Permanente Behavioral Health. Specializes in evidence-based approaches (ACT, CBT, exposure therapy) for patients facing serious medical illness, organ transplantation, and autoimmune disorders. Advises on behavioral health documentation and integration with chronic care workflows. Private Practice, Littleton, Colorado 3 yrs BastionGPT in market 10,000+ Healthcare organizations 1.3M Clinical documents produced every month 15,000 Clinician hours saved every day ## Ready to try BastionGPT? Join 10,000+ healthcare organizations using BastionGPT for HIPAA-compliant AI. Free to start, BAA included on every plan. Try BastionGPT free → --- > Source: https://bastiongpt.com/company/ai-principles Trust at our core # Our principles for generative AI in healthcare As pioneers in healthcare technology, we recognize the crucial role we play in incorporating artificial intelligence into healthcare workflows. We firmly support and align to leading AI ethical and safety principles. G20 AI Principles Microsoft Responsible AI Principles NIST AI Risk Management Framework ## BastionGPT's Healthcare Generative AI Principles We have established a complimentary, and more focused set of principles, specific to healthcare generative AI, that act as our north star to guide our innovative journey to elevate patient care and reduce healthcare workforce burnout, focusing unwaveringly on the safety, privacy, and well-being of patients. These principles aim to foster trust and transparency with our diverse community and reflect our steadfast commitment to maintaining the highest standards of integrity, accuracy, and reliability in our services. We encourage other organizations to embrace these guidelines, helping to realize a safe and transformative integration of technology and healthcare, with medical professionals at the forefront, ensuring a human-centric and ethical approach to patient care. ##### Principle 5: Guard Patient Interaction “Generative AI must not directly provide medical advice to patients unless monitored and strictly validated.” Generative AI is advancing rapidly, creating boundless possibilities. However, when it comes to providing medical advice, these solutions should not directly communicate with patients unless the outputs are strictly controlled and thoroughly vetted by both industry experts and the relevant regulatory bodies. The distinct risk of AI hallucinations and biases in healthcare topics necessitates medical professionals, who are trained and experienced, to supervise interactions. Medical professionals are essential in ensuring the accuracy and reliability of the advice given, recognizing the limitations of AI, and preventing potential harm to patients. ##### Principle 4: Uphold Privacy and Security “All personally identifiable information must be kept private and secure.” The influx of insecure AI services, often cheaper and below standard, poses significant risks to the security and privacy of sensitive personal information. The integrity and confidentiality of patient information must never be compromised. All AI healthcare services must align with healthcare standards and comply with regulatory requirements to avoid breaches that could jeopardize patient wellbeing. Establishing robust privacy controls and security measures are critical in maintaining trust and safeguarding patient interests. ##### Principle 3: Prioritize Evidence-Based Medicine “Source data for generative AI must prioritize evidence-based medicine and reputable sources.” Misinformation and biases are prevalent, even in reputable sources across the internet. Generative AI in healthcare must be anchored in evidence-based medicine, relying on trusted, scientific sources to inform its knowledge base. A stringent, proactive approach is essential to mitigate the risk of circulating harmful and inaccurate data points, improving the accuracy, reliability, and safety of the AI-generated outputs. ##### Principle 2: Promote Transparency and Caution “Generative AI must transparently communicate its propensity for errors even when used by medical professionals.” The incredible capabilities of generative AI can lead to a false sense of security and overshadow its inherent limitations and potential for mistakes. It is crucial to maintain transparency about AI’s limitations and encourage both users to approach the AI outputs with caution and critical evaluation. The emphasis should be on ensuring the responsible use of AI, with medical professionals maintaining vigilance and scrutiny to detect any inaccuracies, thereby preventing potential harm to patients. ##### Principle 1: Maintain Human Oversight “A medical professional must mediate the provision of medical advice or information from AI to patients.” The interface between AI and patients, especially in delivering medical advice or information, must be sufficiently supervised by medical professionals unless appropriate safeguards are robustly implemented. These safeguards can include strict governance around AI outputs, ongoing monitoring by healthcare professionals, and rigorous validation of medical accuracy. We must keep healthcare human-centric, with clinicians retaining control to ensure the safety, accuracy, and appropriateness of the advice provided. “Every decision we make echoes in the well-being of a patient. Our commitment is not just to innovation, but to safe and ethically-driven innovation that always places human health and trust at its core.” Josh Spencer BastionGPT Founder --- > Source: https://bastiongpt.com/news/innovation # Healthcare Innovation News and guidance on innovation, new technologies and new approaches with healthcare AI. Healthcare Innovation ### HIPAA-Compliant AI in Power Automate for Healthcare Workflow Automation September 23, 2026 Healthcare Innovation ### Is Otter.ai HIPAA Compliant? September 19, 2026 Healthcare Innovation ### Before You Put PHI Into Claude: What Anthropic's BAA Actually Covers August 19, 2026 Healthcare Innovation ### AI Scribe and Healthcare AI Consent Forms: Free Templates August 16, 2026 Healthcare Innovation ### AI Deskilling in Clinical Training: A Psychologist's Framework for AI Adoption July 6, 2026 Healthcare Innovation ### What Makes an AI HIPAA Compliant? July 3, 2026 Healthcare Innovation ### AI for Urgent Care and Walk-In Clinics: Documentation and Workflow Efficiency June 12, 2026 Healthcare Innovation ### Is Claude HIPAA Compliant? (2026 Guide) April 22, 2026 Healthcare Innovation ### State Laws on Healthcare AI April 15, 2026 Healthcare Innovation ### We Tested 5 AI Platforms for Writing Complex Reports: Why We Switched to BastionGPT January 22, 2026 Healthcare Innovation ### Why Clinical AI Scribes Need True Speaker Recognition December 6, 2025 Healthcare Innovation ### AI For Chiropractors: How Chiropractors Can Use AI November 13, 2025 Healthcare Innovation ### A Psychologist's Guide to Using AI for Clinical Documentation October 27, 2025 Healthcare Innovation ### Is Google Gemini HIPAA Compliant? August 28, 2025 Healthcare Innovation ### AI For Dental Practices: 29 Prompts Dentists Can Use August 4, 2025 Healthcare Innovation ### How Healthcare Administrators Can Use AI June 25, 2025 Healthcare Innovation ### How To Use AI for PT, OT, or Sports Medicine Documentation June 6, 2025 Healthcare Innovation ### Transforming Mental Health Practice: How AI Gave Me Back Time to Bake Bread and Better Serve My Clients June 3, 2025 Healthcare Innovation ### How to Use AI as a School Psychologist May 15, 2025 Healthcare Innovation ### 40+ Ways Psychiatrists & Therapists Use AI - BastionGPT May 9, 2025 Healthcare Innovation ### How I've Leveraged BastionGPT to Transform My Clinical and Business Practice May 6, 2025 Healthcare Innovation ### How Healthcare Legal Professionals & Lawyers Are Using AI April 22, 2025 Healthcare Innovation ### Healthcare AI Prompts: Building Your Clinical Library April 16, 2025 Healthcare Innovation ### How DME Companies Are Using AI for Operational Efficiency April 13, 2025 Healthcare Innovation ### Is ChatGPT HIPAA Compliant? March 26, 2025 Healthcare Innovation ### Is Microsoft Copilot HIPAA Compliant? AI Healthcare Compliance Guide March 24, 2025 Healthcare Innovation ### BastionGPT: The Safe, Compliant, and Credible AI Tool for Psychiatrists March 17, 2025 Healthcare Innovation ### DAP Notes: What They Are and How to Write Them With AI March 4, 2025 Healthcare Innovation ### What Healthcare Data Anonymization is and How to De-identify PHI February 3, 2025 Healthcare Innovation ### HIPAA Compliance Risks with Tokenization and Anonymization in Generative AI December 30, 2024 Healthcare Innovation ### A Guide to AI Scribes in Healthcare December 12, 2024 Healthcare Innovation ### Compliance for Australian Users: BastionGPT and the Australian Privacy Principles (APPs) December 6, 2024 Healthcare Innovation ### Best Practices for Healthcare AI Prompt Engineering November 12, 2024 Healthcare Innovation ### Affiliate Referral Program November 8, 2024 Healthcare Innovation ### How to Stay Compliant Using AI Within Global Regulatory Frameworks October 17, 2024 Healthcare Innovation ### 30 Ways To Use Generative AI in Healthcare August 15, 2024 Healthcare Innovation ### Top Use Cases for the Generative AI in Therapy and Psychiatry June 10, 2024 Healthcare Innovation ### Recognizing and Mitigating Bias in Healthcare AI: A Guide for End Users May 26, 2024 Healthcare Innovation ### Top Use Cases for Generative AI in Healthcare April 24, 2024 Healthcare Innovation ### Healthcare AI Comparison January 23, 2024 Healthcare Innovation ### Comparison of Word Limits in Generative AI October 30, 2023 Healthcare Innovation ### Founder Podcast: Exploring the Digital Battlefield of Healthcare Cybercrime September 2, 2023 Healthcare Innovation ### AI Transcription Services for Medical Professionals August 8, 2023 Healthcare Innovation ### Healthcare: Balancing Risks and Rewards with ChatGPT July 11, 2023 --- > Source: https://bastiongpt.com/post/25-ways-to-use-generative-ai-in-healthcare Healthcare Innovation # 30 Ways To Use Generative AI in Healthcare Written By: Josh Spencer • August 15, 2024 In the ever-evolving landscape of healthcare, medical professionals are continually seeking innovative tools to enhance patient care, streamline administrative processes, and conduct in-depth research. BastionGPT, a generative AI assistant and AI scribe tailored specifically for healthcare professionals, offers a multitude of applications that can transform everyday tasks and improve overall efficiency. We speak with customers every day to understand and optimize their needs. Below are 30 use cases where our customers have told us they are using our service to help them get more done in the day. ## 1. Medical Documentation Assistance ##### Example: A physician needs to document a detailed patient visit. ##### Capabilities: • Draft comprehensive patient visit notes. • Generate summaries of patient interactions and follow-up plans. • Create templates for different types of medical documentation. ##### Detailed Example: A general practitioner uses BastionGPT's AI scribe to draft a detailed note for a patient visit, summarizing the patient’s symptoms, the physical exam findings, diagnostic tests ordered, and follow-up instructions. This draft can be reviewed and edited before being finalized in the electronic health record (EHR). ## 2. Writing Patient Education Materials ##### Example: A nurse educator needs to create a pamphlet on managing hypertension. ##### Capabilities: • Generate clear and concise educational content tailored to patient needs. • Summarize complex medical information in layman's terms. • Develop engaging and informative content for brochures, handouts, and digital content. ##### Detailed Example: A nurse educator uses BastionGPT to draft a pamphlet on hypertension management, including lifestyle changes, medication adherence, and monitoring techniques. The draft is then reviewed for accuracy and clarity before distribution to patients. ## 3. Research Support and Literature Reviews ##### Example: A healthcare researcher needs to compile a literature review on the latest treatments for diabetes. ##### Capabilities: • Summarize key findings from multiple academic articles. • Create cohesive literature reviews highlighting major trends and significant data points. • Assist in identifying gaps in current research. ##### Detailed Example: A researcher uses BastionGPT to summarize findings from recent studies on diabetes treatments. The AI generates a comprehensive review, which includes summaries of clinical trials, new medications, and emerging therapies, providing a solid foundation for further research. ## 4. Patient Communication and Follow-Up ##### Example: A clinic manager needs to send follow-up emails to patients. ##### Capabilities: • Draft personalized follow-up emails based on specific patient interactions and treatment plans. • Generate reminders for appointments and medication refills. • Create clear and empathetic communication tailored to patient needs. ##### Detailed Example: A clinic manager uses BastionGPT to draft follow-up emails for patients who recently had appointments, reminding them of their next steps in their treatment plan. These emails are then reviewed and personalized before being sent out. ## 5. Creating Clinical Guidelines and Protocols ##### Example: A hospital administrator needs to update the clinical guidelines for infection control. ##### Capabilities: • Generate drafts of clinical guidelines based on the latest evidence and best practices. • Summarize complex protocols in an easily understandable format. • Ensure that all relevant regulatory and safety standards are included. ##### Detailed Example: A hospital administrator uses BastionGPT to draft updated infection control protocols, including new sanitation procedures, PPE usage, and patient isolation guidelines. The draft is reviewed by the infection control team before implementation. ## 6. Assisting with Grant Writing and Funding Applications ##### Example: A healthcare researcher needs to write a grant proposal. ##### Capabilities: • Generate drafts of grant proposals and funding applications. • Summarize research objectives, methodologies, and expected outcomes. • Ensure alignment with funding agency requirements. ##### Detailed Example: A researcher uses BastionGPT to draft a grant proposal for a study on the impact of telehealth on patient outcomes. The AI generates sections on the study’s significance, methodology, and potential impact, which are then refined and reviewed before submission. ## 7. Developing Training and Educational Programs ##### Example: A medical educator needs to create a training module on new diagnostic techniques. ##### Capabilities: • Draft comprehensive training materials, including presentations and handouts. • Summarize key learning objectives and instructional content. • Create engaging and interactive educational content. ##### Detailed Example: A medical educator uses BastionGPT to draft a training module on the latest diagnostic techniques in cardiology. The AI generates detailed content, including case studies and practical exercises, which are then reviewed and incorporated into the training program. ## 8. Generating Comprehensive Patient Histories ##### Example: A physician needs to compile a detailed patient history for a complex case. ##### Capabilities: • Generate summaries of patient histories based on available medical records and patient interviews. • Ensure all relevant medical, social, and family history details are included. • Create a cohesive narrative that supports clinical decision-making. ##### Detailed Example: A physician uses BastionGPT to draft a comprehensive patient history for a new patient with multiple chronic conditions. The AI compiles information from past medical records and recent consultations, providing a thorough overview that aids in diagnosis and treatment planning. ## 9. Creating Reports and Presentations for Stakeholders ##### Example: A healthcare executive needs to present quarterly results to the board. ##### Capabilities: • Generate drafts of professional reports and presentations. • Summarize key data points and trends in an accessible format. • Ensure that reports are clear, concise, and visually engaging. ##### Detailed Example: A healthcare executive uses BastionGPT to draft a quarterly report summarizing the hospital’s performance, including patient outcomes, financial metrics, and strategic initiatives. The draft is reviewed and polished before being presented to the board. ## 10. Supporting Telemedicine Consultations Example: A telehealth provider needs to document virtual consultations. Capabilities: • Draft detailed summaries of telehealth consultations, including patient concerns, diagnoses, and treatment plans. • Generate follow-up recommendations and patient communication. • Ensure accurate and comprehensive telehealth records. Detailed Example: A telehealth provider uses BastionGPT to draft summaries of virtual consultations, documenting patient symptoms, the clinician’s observations, and recommended treatments. The drafts are reviewed for accuracy and completeness before being added to the patient’s medical record. ## 11. Developing Personalized Care Plans ##### Example: A nurse practitioner needs to create a personalized care plan for a patient with multiple chronic conditions. ##### Capabilities: • Draft comprehensive care plans tailored to individual patient needs. • Integrate information from various sources, including medical records, patient interviews, and diagnostic tests. • Ensure that care plans are holistic and address all aspects of the patient’s health. ##### Detailed Example: A nurse practitioner uses BastionGPT to draft a personalized care plan for a patient with diabetes, hypertension, and chronic kidney disease. The AI generates a detailed plan that includes medication management, lifestyle modifications, and follow-up appointments. The draft is reviewed and customized before being shared with the patient. ## 12. Streamlining Insurance Documentation ##### Example: A healthcare administrator needs to submit documentation for insurance claims. ##### Capabilities: • Generate drafts of insurance claim forms and documentation. • Ensure that all necessary information is included and formatted correctly. • Reduce the administrative burden on healthcare providers. ##### Detailed Example: A healthcare administrator uses BastionGPT to draft insurance claim forms for a series of patient visits. The payer policies are included in the prompt, and all relevant details, including diagnosis codes and treatment descriptions, are included per the required specifications. The drafts are reviewed and finalized before submission. ## 13. Creating Patient Discharge Summaries ##### Example: A hospitalist needs to prepare discharge summaries for multiple patients. ##### Capabilities: • Draft detailed discharge summaries, including patient instructions and follow-up care. • Summarize key aspects of the hospital stay, treatments received, and outcomes. • Ensure that discharge summaries are clear and comprehensive. ##### Detailed Example: A hospitalist uses BastionGPT to draft discharge summaries for patients being released from the hospital. The AI includes information on medications, follow-up appointments, and self-care instructions. The drafts are reviewed and personalized before being provided to the patients. ## 14. Drafting Informed Consent Forms ##### Example: A surgeon needs to prepare informed consent forms for a new procedure. ##### Capabilities: • Generate drafts of informed consent forms, ensuring all necessary information is included. • Clearly explain the risks, benefits, and alternatives of the procedure. • Ensure that forms are written in patient-friendly language. ##### Detailed Example: A surgeon uses BastionGPT to draft informed consent forms for a new minimally invasive surgery. The AI includes detailed explanations of the procedure, potential risks, and recovery expectations. The drafts are reviewed for accuracy and clarity before being used with patients. ## 15. Creating Marketing Materials for Healthcare Services ##### Example: A healthcare marketing team needs to develop promotional materials for a new health program. ##### Capabilities: • Generate creative and engaging marketing content, including brochures, social media posts, and website copy. • Ensure that content aligns with the desired tone and messaging. • Highlight the benefits and unique features of the healthcare services. ##### Detailed Example: A healthcare marketing team uses BastionGPT to draft a series of social media posts promoting a new wellness program. The AI generates engaging content that highlights the program’s benefits, including improved fitness and mental health. The drafts are reviewed and polished before being published. ## 16. Preparing Case Reports for Medical Conferences ##### Example: A physician needs to prepare a case report for presentation at a medical conference. ##### Capabilities: • Draft detailed case reports, including patient history, diagnosis, treatment, and outcomes. • Summarize key findings and clinical insights. • Ensure that reports are professional and suitable for presentation. ##### Detailed Example: A physician uses BastionGPT to draft a case report on a rare neurological disorder for an upcoming conference. The AI includes detailed patient information, diagnostic tests, and treatment outcomes. The draft is reviewed and refined before being presented. ## 17. Generating Referral Letters to Specialists ##### Example: A primary care physician needs to refer a patient to a specialist. ##### Capabilities: • Draft referral letters that include detailed patient history and current treatment plans. • Ensure that all relevant information is communicated effectively. • Facilitate better-coordinated care between healthcare providers. ##### Detailed Example: A primary care physician uses BastionGPT to draft a referral letter to a cardiologist for a patient with suspected heart disease. The AI includes the patient’s medical history, current symptoms, and previous test results. The draft is reviewed and sent to the specialist. ## 18. Creating Health Promotion Campaigns ##### Example: A public health official needs to design a campaign to promote vaccination. ##### Capabilities: • Generate drafts of campaign text content, including posters, flyers, and social media content. • Ensure that messages are clear, persuasive, and evidence-based. • Tailor content to different audiences and demographics. ##### Detailed Example: A public health official uses BastionGPT to draft materials for a vaccination campaign targeting young adults. The AI generates engaging content that emphasizes the importance and benefits of vaccination. The drafts are reviewed and customized before being distributed. ## 19. Developing Clinical Research Protocols ##### Example: A clinical researcher needs to create a protocol for a new study. ##### Capabilities: • Draft detailed research protocols, including study design, methodology, and ethical considerations. • Summarize the objectives, hypotheses, and expected outcomes of the study. • Ensure that protocols meet regulatory and ethical standards. ##### Detailed Example: A clinical researcher uses BastionGPT to draft a protocol for a study on the effects of a new medication on chronic pain. The AI includes detailed information on the study design, participant recruitment, and patient data collection methods. The draft is reviewed and finalized before submission to the ethics committee. ## 20. Supporting Quality Improvement Initiatives ##### Example: A healthcare quality manager needs to document a quality improvement project. ##### Capabilities: • Generate drafts of quality improvement reports, including project objectives, methodologies, and outcomes. • Summarize key findings and recommendations for future practice. • Ensure that reports are clear, concise, and actionable. ##### Detailed Example: A healthcare quality manager uses BastionGPT to draft a report on a project aimed at reducing hospital readmission rates. The AI includes detailed information on the interventions implemented, data collected, and results achieved. The draft is reviewed and refined before being shared with the hospital’s leadership team. ## 21. Creating Patient Progress Reports ##### Example: A physical therapist needs to document a patient’s progress over the course of treatment. ##### Capabilities: • Generate detailed progress reports summarizing patient improvements and setbacks. • Include objective measurements and subjective patient feedback. • Ensure reports are clear and suitable for sharing with other healthcare providers. ##### Detailed Example: A physical therapist uses BastionGPT to draft a progress report for a patient recovering from a knee injury. The AI includes information on range of motion, strength improvements, and patient-reported pain levels. The draft is reviewed and finalized before being shared with the patient's physician. ## 22. Drafting Medical Research Articles ##### Example: A medical researcher needs to write an article for a peer-reviewed journal. ##### Capabilities: • Generate drafts of research articles, including introduction, methods, results, and discussion sections. • Summarize key findings and their implications for the field. • Ensure that articles meet the standards of academic writing. ##### Detailed Example: A medical researcher uses BastionGPT to draft an article on the outcomes of a clinical trial for a new cancer treatment, utilizing the researchers notes and study documentation. The AI generates detailed sections on the study design, statistical analysis, and clinical implications. The draft is thoroughly reviewed and refined before submission to a journal. ## 23. Creating Patient Intake Forms ##### Example: A clinic manager needs to develop comprehensive patient intake forms. ##### Capabilities: • Draft customized patient intake forms that gather all necessary information. • Include sections for medical history, current medications, and insurance information. • Ensure forms are user-friendly and easy to complete. ##### Detailed Example: A clinic manager uses BastionGPT to create patient intake forms for a new clinic. The AI generates detailed sections for patient demographics, medical history, and consent forms. The drafts are reviewed and finalized before being used in the clinic. ## 24. Developing Health and Wellness Newsletters ##### Example: A wellness coordinator needs to create a monthly newsletter for patients. ##### Capabilities: • Generate engaging and informative newsletter content. • Summarize recent health news, tips, and resources. • Ensure content is relevant and beneficial to the patient population. ##### Detailed Example: A wellness coordinator uses BastionGPT to draft a monthly newsletter that includes tips for managing stress, healthy recipes, and updates on upcoming wellness events. The draft is reviewed and edited before being distributed to patients. ## 25. Preparing Legal Documents for Healthcare Compliance ##### Example: A healthcare compliance officer needs to prepare legal documents for regulatory compliance. ##### Capabilities: • Draft legal documents, including policies, procedures, and compliance reports. • Summarize regulatory requirements and ensure documents meet legal standards. • Ensure clarity and precision in legal language. ##### Detailed Example: A healthcare compliance officer uses BastionGPT to draft a compliance report detailing adherence to new healthcare regulations. The AI includes sections on policy updates, staff training, and audit results. The draft is reviewed by the legal team before submission to regulatory bodies. ## 26. Generating Patient Reminder Letters ##### Example: A dental office manager needs to send reminder letters for upcoming appointments. ##### Capabilities: • Draft personalized reminder letters for patient appointments. • Include details of the appointment, preparation instructions, and contact information. • Ensure that letters are clear and professional. ##### Detailed Example: A dental office manager uses BastionGPT to draft reminder letters for patients with upcoming appointments. The AI includes information on the date and time of the appointment, any necessary preparations, and contact details for rescheduling. The drafts are reviewed and sent to patients. ## 27. Developing Community Health Outreach Materials ##### Example: A public health coordinator needs to create materials for a community health outreach program. ##### Capabilities: • Generate engaging and informative outreach materials, including flyers, posters, and brochures. • Highlight key health messages and resources available to the community. • Ensure materials are culturally sensitive and accessible. ##### Detailed Example: A public health coordinator uses BastionGPT to draft materials for a community health fair. The AI generates content on topics such as vaccination, nutrition, and mental health resources. The drafts are reviewed and customized before being distributed in the community. ## 28. Creating Educational Webinars and Online Courses ##### Example: A medical educator needs to develop content for an online course on chronic disease management. ##### Capabilities: • Draft comprehensive course materials, including lecture notes, slides, and quizzes. • Summarize key concepts and learning objectives. • Ensure content is engaging and suitable for online delivery. ##### Detailed Example: A medical educator uses BastionGPT to draft content for an online course on managing chronic diseases. The AI generates detailed lecture notes, interactive quizzes, and case studies. The drafts are reviewed and finalized before being uploaded to the online learning platform. ## 29. Generating Medication Management Plans ##### Example: A pharmacist needs to create a medication management plan for a patient with multiple prescriptions. ##### Capabilities: • Draft detailed medication management plans, including dosage schedules and potential interactions. • Summarize important information on each medication. • Ensure plans are clear and easy for patients to follow. ##### Detailed Example: A pharmacist uses BastionGPT to draft a medication management plan for a patient taking multiple medications for chronic conditions. The AI includes detailed information on dosages, timing, and potential side effects. The draft is reviewed and customized before being provided to the patient. ## 30. Assisting in the Development of Telemedicine Protocols ##### Example: A telehealth director needs to create protocols for virtual consultations. ##### Capabilities: • Generate detailed telemedicine protocols, including guidelines for patient interactions and documentation. • Summarize best practices for virtual care. • Ensure protocols meet regulatory and safety standards. ##### Detailed Example: A telehealth director uses BastionGPT to draft protocols for virtual consultations, including guidelines for patient identification, privacy measures, and documentation requirements. The drafts are reviewed and finalized before being implemented in the telehealth program. ## Conclusion BastionGPT is a versatile and powerful tool that can significantly enhance the efficiency and effectiveness of healthcare professionals. By automating and streamlining various tasks, BastionGPT allows healthcare providers to focus more on patient care and less on administrative burdens. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/affiliate-referral-program Healthcare Innovation # Affiliate Referral Program • November 8, 2024 ## Transform Healthcare with AI: Join BastionGPT's Referral Program Are you connected to healthcare professionals who could benefit from cutting-edge AI solutions? Turn your network into opportunity by partnering with BastionGPT, the leading AI platform revolutionizing healthcare delivery. ### Quick and Simple: Informal Referral Program Want a hassle-free way to earn rewards? Our informal referral program is perfect for you. Simply share BastionGPT with friends and colleagues, and earn a subscription credit for every new customer you bring on board. Your credit will match the first month's subscription value of the person you refer (for example, $20 for a Professional signup or $45 for a Professional Plus signup). Just send us their email addresses via email or chat, and we'll handle the rest. In addition, we will extend the trial of the person you referred from 7 to 30 days! You can find more information about BastionGPT to share with your colleagues and followers in our Media Kit . ### Maximize Your Earnings: Formal Affiliate Program For those looking to turn their recommendations into a revenue stream, our formal affiliate program offers lifetime commissions and exclusive perks: 8% Lifetime Commission : Earn ongoing commissions on all sales generated through your unique affiliate link, with no earnings cap Extended Trial Period : Automatically give your referrals an exclusive 30-day free trial (instead of the standard 7 days) Custom Branded Links : Personalize your affiliate link with your brand name (e.g., hxxps://bastiongpt.com/companyname) Flexible Payments : Choose from Zelle, Venmo, or ACH transfers ### Your Success is Our Success Our rapidly expanding community of healthcare providers is proof that BastionGPT is meeting a crucial need in the industry. As an affiliate partner, you'll help practitioners: Streamline their workflows Enhance patient care Stay ahead of healthcare innovation Increase practice efficiency ### Getting Started is Easy Complete the 5-minute registration form at https://bastion-intelligence.getrewardful.com Share the link with your potential customers via social media or direct email. Receive payouts once per month once earnings top $20 per month. Tip: Your affiliate link will be https://bastiongpt.com?via=xxxxxx The xxxxxx will be replaced by whatever name you register as a "First Name", which most affiliates will choose their business name. If you would like to change or have additional links, feel free to contact us! ### Ready to Start Earning? Join our referral community today and transform your BastionGPT enthusiasm into rewards. Our support team is here to help you succeed: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ### Did You Already Refer Someone? No worries if you've previously referred customers without an affiliate link! We're happy to set up retroactive referrals – just let us know their email addresses. ### Payment Details Monthly payouts on the first week of each month Minimum payout: $20 (Zelle/Venmo) or $100 (ACH) Tax documentation is required for earnings over $600, or based on other tax requirements in your jurisdiction. You can review the complete referral program terms and payouts in our support center. Together, let's shape the future of healthcare. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/ai-for-pt-ot-sports-medicine Healthcare Innovation # How To Use AI for PT, OT, or Sports Medicine Documentation Written By: Robert Henehan • June 6, 2025 If you work in physical therapy, occupational therapy, or sports medicine, documentation likely consumes more of your day than you'd like. Evaluations, SOAP notes, progress reports, home exercise programs, insurance letters are relentless, and it pulls you away from the patients you're there to help. A 2022 American Physical Therapy Association report found that over 86% of physical therapists agree that excessive documentation contributes to burnout. That number isn't surprising to anyone working in the field. This article walks through how occupational and physical therapists are using AI tools like BastionGPT to draft high-quality clinical documents faster, reduce administrative burden, and reclaim time for patient care without sacrificing accuracy, compliance, or professional judgment. You'll see concrete use cases, real example prompts, and a clear picture of where human oversight fits into the process. ## Benefits of Using AI for Physical Therapy Documentation ### Drafting SOAP Notes, Progress Notes, and SMART Goals Accurate daily notes and periodic progress reports are critical in therapy practice, but writing them consumes a large portion of clinicians’ time. BastionGPT’s medical scribe capabilities can draft structured clinical notes based on prompts or transcribed dictations, helping therapists keep documentation thorough and timely. For example: #### Session-to-SOAP Note Transcription : BastionGPT can convert a therapist’s spoken or rough written notes into a well-formatted SOAP note . For instance, a physical therapist can dictate key observations after a treatment session into BastionGPT (e.g. “Patient reports decreased pain today; performed gait training, balance exercises; noted improved right knee flexion; plan to advance difficulty next visit”). The AI will generate a coherent SOAP note with labeled Subjective, Objective, Assessment, and Plan sections. The clinician then quickly reviews this draft to ensure all details (like any safety concerns or specific metrics) are correct. By offloading the initial writing, therapists save time and reduce after-hours charting, while still maintaining full oversight of the content. Prompt: Turn the following session notes into a well-formatted SOAP note with labeled Subjective, Objective, Assessment, and Plan sections. Use professional, clinical language appropriate for documentation. #### Progress Notes with Objective Outcomes : Generating comprehensive progress reports is faster with AI assistance. For example, a physical therapist needs to document a four-week progress note for a patient rehabbing a knee injury. By inputting the patient’s initial evaluation data and recent session summaries, BastionGPT can draft a progress report that highlights objective improvements (e.g. increased range of motion from 100° to 130°, quadriceps strength from 3/5 to 4/5) and subjective feedback (patient reports decreased pain during stairs). The draft will clearly outline which SMART goals have been met or updated and note any suggested modifications to the treatment plan. The therapist verifies these details and makes any edits before adding the note to the chart. BastionGPT helps you to capture key progress metrics and that the report is formatted cleanly for easy sharing with physicians or insurance as needed. You are a physical therapist. Write a structured progress note based on the following information. Include objective outcomes, subjective feedback, SMART goal updates, and any recommended treatment plan modifications. Format it for professional clinical or insurance documentation #### Treatment Plans and SMART Goal Suggestions : BastionGPT can even help clinicians develop initial plans of care and goals tailored to a patient’s diagnosis. For instance, an occupational therapist evaluating a patient with a stroke can prompt BastionGPT with the patient’s deficits (e.g. left-side weakness, difficulty with dressing) and ask for draft goals and interventions. The healthcare AI might propose SMART goals like “In 6 weeks, patient will independently dress upper body using adaptive techniques, requiring no more than setup assistance” along with suggested interventions (e.g. donning shirt with hemiplegic dressing strategies, using adaptive equipment). Similarly, a sports medicine rehab specialist could request a therapy plan by diagnosis, say, an ACL reconstruction protocol, and BastionGPT would outline a phased rehabilitation program with appropriate milestones (range of motion targets, strengthening progressions, return-to-jog timeline, etc.). The clinician can then adjust these goals and plans to ensure they are realistic and individualized, but the AI-generated draft provides a strong starting point. This speeds up the planning process and helps ensure the goals are specific, measurable, and aligned with best practices. You are an occupational therapist. Create an initial treatment plan with SMART goals and suggested interventions based on the details below. Tailor it to the patient’s diagnosis or condition. By using BastionGPT to draft daily notes, progress summaries, and goal plans, therapy professionals can significantly reduce the time spent starting with a blank page. The AI serves as a scribe producing organized, legible documentation while the clinician remains the final editor, confirming everything is accurate and clinically sound. ## Using AI Transcription and Scribe Tools During Sessions Many clinicians in physical therapy, occupational therapy, and sports medicine face the challenge of documenting care accurately without interrupting patient engagement. BastionGPT’s integrated AI transcription and scribe capabilities offer a powerful solution: clinicians can record visits in real time, upload voice notes after a session or even dictate observations between exercises, then generate structured drafts for review. This makes it easier to stay present during visits while still capturing everything needed for documentation and care planning. #### Post-Visit Voice Notes Transcribed to SOAP : After a therapy session, clinicians often need to summarize what happened: which exercises were done, how the patient responded, what adjustments were made. Rather than typing everything manually, the clinician can dictate a quick voice memo (e.g. “Patient tolerated all three sets of squats with minimal knee discomfort. Advanced to resistance band level 2. Working on proprioception drills; balance improved slightly today. Plan to reassess single-leg stability next week.”). BastionGPT’s transcription feature converts this into accurate text, and then formats it as a SOAP note draft with labeled sections: Subjective, Objective, Assessment, and Plan. The clinician reviews the note, makes any adjustments or additions, and then saves the final version to the EHR. This process reduces after-hours documentation and allows clinicians to focus on patient interaction instead of scrambling to write everything down between appointments. Generate a clinical SOAP note with clearly labeled Subjective, Objective, Assessment, and Plan sections. Use professional, concise language appropriate for inclusion in a therapy documentation system. Ensure the note captures what exercises or interventions were performed, the patient’s response or tolerance, any changes made during the session, and the plan for next steps. #### In-Clinic Dictation for Complex Exercises or Protocols: Some exercises or treatment sequences are too detailed to write up on the fly. With BastionGPT, clinicians can describe what was done aloud (during or after the session) and get a clear, structured write-up of the interventions. For example, an occupational therapist working with a TBI patient might say: “We practiced sequencing a morning routine using task cards. Patient required 3 cues initially, down to 1 by the end. Used a visual schedule and checklist. No fatigue noted; will add toileting sequence next session.” The AI will transcribe the voice note, then draft a clinical summary of the session that the OT can edit and sign off on. This helps ensure key clinical insights aren’t lost, especially when the session involves complex or multi-step interventions that would otherwise take time to write up. You are an OT. Convert the following session dictation into a structured clinical summary. Highlight key interventions, patient response, cues required, and any next steps. #### Exercise Narration to Draft HEP Instructions : BastionGPT can also be used to capture clinicians’ spoken descriptions of exercises as they demonstrate them. For instance, during a sports medicine session, a clinician might say: “Next, we’re doing single-leg hops. Start on your surgical leg, land softly, control the landing. Repeat 10 reps, 2 sets.” BastionGPT transcribes this, then formats it as part of a patient-facing home exercise program (HEP) handout. Clinicians can also batch-record instructions for multiple exercises in one go and have BastionGPT split and label them individually. This ensures HEP instructions are accurate, specific, and consistent with what the patient was actually shown, without requiring extra time typing them after the session. You are a physical therapist. Transcribe and format the following narrated exercises into patient-facing HEP instructions. Write each exercise in step-by-step format, include reps/sets, safety notes, label each clearly, and use language for a 7th-grade reading level. #### Custom Prompts from Audio Recordings : Clinicians can also record a visit or upload a transcript or raw audio recording from a session and pair it with a custom prompt to generate specific clinical content. For example, after recording an exercise-based session, a physical therapist might ask BastionGPT, “Create a progression plan for the next 4 weeks based on the patient’s performance in this session,” or “Draft a patient-facing summary of today’s treatment including safety tips and activity modifications.” Similarly, an occupational therapist could upload an audio evaluation and prompt, “List potential ADL goals and required assistive devices based on observed challenges,” or “Summarize key observations to share with the caregiver.” This flexible, prompt-based workflow allows providers to extract exactly the insights or documentation components they need, from care planning and communication to patient education and interdisciplinary coordination, all while staying in control of the final output. Based on the transcript or audio notes below, generate a [type of document or output] that includes [key elements or goals]. Write in a format suitable for [intended audience: e.g., patient, caregiver, interdisciplinary team, chart]. By integrating transcription into the documentation workflow, BastionGPT allows clinicians to capture detailed information in the moment while preserving their ability to engage directly with patients. It’s especially helpful for those juggling high caseloads or working in fast-paced environments like sports sidelines, inpatient rehab, or busy outpatient clinics. ## Patient Education Handouts, HEP Instructions, and ADL Guidance Educating patients is a big part of PT, OT, and sports medicine, whether it’s providing a home exercise program or tips for daily living. Crafting clear, reader-friendly educational materials can be time-consuming, but BastionGPT helps generate these documents quickly. Clinicians can use the AI to draft patient handouts that explain exercises, precautions, or wellness strategies in plain language, then customize the details. For example: #### Home Exercise Program (HEP) Instructions BastionGPT can produce step-by-step exercise descriptions and safety notes for patients to follow at home. Suppose a physical therapist wants to create a HEP handout after an initial visit for low back pain. The PT can list the exercises (e.g. pelvic tilts, cat-camel stretch, bird-dog) and ask BastionGPT to generate clear instructions for each. The AI might write: “Pelvic Tilt: Lie on your back with knees bent. Gently tilt your pelvis to press your low back into the floor, then release. Reps: 10 slow repetitions, 2 times a day. Tip: Breathe out as you tilt; avoid holding your breath.” It could also include general reminders like warming up and noting any pain. The therapist can review and target the language to their comprehension level adding details related to reading level or targeting to a non-native English speaker before giving the handout to the patient. By automating the first draft, BastionGPT allows clinicians to quickly generate personalized exercise guides that patients can confidently follow at home. You are a physical therapist. Create a patient-friendly Home Exercise Program (HEP) handout for someone with low back pain. Include step-by-step instructions, reps/sets, safety tips, and common modifications for these exercises: pelvic tilts, cat-camel stretch, bird-dog. Target the language for a 6th grade reading level. Add a warm-up reminder and general notes about stopping if pain increases. #### ADL and Activity Modification Guides Occupational therapists can utilize BastionGPT to draft Activities of Daily Living (ADL) guidance for patients and caregivers. For instance, after working with an elderly patient who struggles with bathing and dressing safely, an OT can ask BastionGPT to outline a patient education handout on “Shower Safety and One-Handed Dressing Techniques.” The AI might generate tips such as installing grab bars, using a shower chair, laying out clothing in advance, and step-by-step instructions for putting on a shirt with one arm. It can also emphasize safety precautions (like water temperature checks to prevent burns, or sitting down while dressing to avoid falls). The end result is a clear, helpful guide that the patient (and their caregiver) can reference – created with a fraction of the time it would normally take to write from scratch. You are an occupational therapist. Write a caregiver-friendly guide titled ‘Shower Safety and One-Handed Dressing Techniques’ for an elderly patient recovering from a stroke. Include step-by-step instructions for safe bathing and dressing with one hand, and tips like using a shower chair, laying out clothing in advance, and sitting down while dressing. Emphasize fall prevention and temperature safety. Keep the tone reassuring and easy to follow. #### Condition Explanations and Sensory Strategy Sheets: BastionGPT’s strength in summarization means it can help translate complex medical or therapeutic concepts into layperson terms. Therapists might use it to draft educational content explaining a patient’s condition or coping strategies. For example, a parent of a child with sensory processing challenges might ask their pediatric OT for ideas to help with transitions or meltdowns. The OT can prompt BastionGPT to generate a sensory strategy guide describing techniques like a “sensory corner” with soft items, use of noise-cancelling headphones, or preparatory calming exercises before transitions. The AI-written draft might include why each strategy helps (“deep pressure input can be calming to the nervous system,” etc.) in accessible language. The therapist would review this draft, ensuring the advice aligns with the child’s specific sensory profile and family routines, then provide it as a written resource to the parents. Similarly, a sports medicine clinician could create a one-page info sheet about a common injury (like “Understanding Concussions and Recovery”) for athletes, using BastionGPT to get the medical facts and precautions in plain English. In all cases, the AI aids in producing patient-friendly educational materials that improve understanding and adherence, while the clinician verifies every point for accuracy and relevance. You are an occupational therapist. Draft a sensory strategy handout for the following patient with sensory processing challenges. Include ideas like setting up a calming corner, using noise-canceling headphones, and doing deep pressure input activities before transitions. Explain why each strategy helps in simple language, and suggest 3 calming routines families can try at home. By speeding up the creation of handouts, exercise instructions, and educational sheets, BastionGPT helps clinicians consistently provide high-quality patient education without reinventing the wheel each time. The materials are always reviewed and tailored by the professional, but the heavy lifting of writing paragraphs and formatting instructions is handled by AI, freeing up clinicians to focus on teaching and reinforcing the content during sessions. ## Referral Notes, Discharge Plans, and Team Updates Communicating effectively with other healthcare providers and team members is essential in rehabilitation and sports medicine. Whether you’re sending a referral to a specialist, writing a discharge summary for a patient’s physician, or updating the care team on a client’s status, BastionGPT can assist by generating clear, well-structured drafts of these communications. This helps make sure all the important details are included and formatted professionally, while saving clinicians time. For example: #### Referral Letters to Other Providers When referring a patient to another specialist or requesting additional services, clinicians need to summarize the case succinctly. BastionGPT can help draft referral notes that quickly get to the point. For instance, an occupational therapist might be referring a patient to a neurologist for an evaluation. The OT can input the key background (diagnosis, therapy progress, and specific concerns) and ask BastionGPT to formulate a referral letter. The AI could produce a letter that opens with the patient’s identifying info and diagnosis, then states: “I am referring Mr. Smith for neurological assessment due to observed fine motor coordination decline and episodic hand tremors during OT sessions.” It would list relevant findings (e.g. “Patient has completed 8 weeks of OT for stroke rehab; upper extremity strength and basic ADL function have improved, but new tremors began 2 weeks ago, impacting his progress with writing tasks.”) and clearly request the neurologist’s input. Using AI in this way helps ensure no crucial information is omitted and that the referral is structured in a professional, easy-to-read format for the receiving provider. You are a sports medicine clinician. Draft a professional referral letter to another provider based on the information below. Include patient diagnosis, relevant clinical history, therapy progress, observed concerns or symptoms, and a clear reason for referral. Use a formal tone appropriate for communication between healthcare providers. Include these details for my practice [insert your provider details, address, and dates of service]. #### Discharge Summaries and Plans At discharge or transfer of care, therapists prepare documentation summarizing what was accomplished and what next steps are recommended. BastionGPT can compile a thorough discharge summary by pulling together information from the patient’s notes and outcomes. For example, when a physical therapy patient is ready for discharge, the PT could prompt the healthcare AI with the patient’s final status (e.g. functional scores, remaining limitations) and have it draft the discharge report. The AI-generated summary might include: “Patient has attended 12 PT sessions post-ACL surgery. Strength in the surgical leg improved from 3/5 to 5/5; gait normalized without assistive devices; functional hop test 90% of unaffected side. Patient met all short-term goals and is independent in a gym strengthening program.” It would then list discharge recommendations (like continuing a home exercise routine, follow-up with orthopedic surgeon on a certain date, precautions such as avoiding pivoting sports for 3 more months, etc.). The therapist reviews this carefully against their own notes and makes corrections or additions (for example, ensuring that any special instructions to the patient are clearly stated). The final discharge plan can then be shared with the patient and their physician or attached to the medical record. By automating the collation of progress data and recommendations, BastionGPT supports consistency in discharge documentation and saves the clinician from having to manually summarize weeks or months of therapy. You are a sports medicine clinician. Generate a discharge summary and recommendations based on the information below. Include: Total sessions and duration of care, functional progress and outcome metrics, goals achieved or unmet, discharge recommendations (e.g., home program, follow-up care, precautions), write in a format appropriate for both patient communication and inclusion in the medical record. #### Interdisciplinary Team Reports and Updates In settings where multiple professionals collaborate (rehab hospitals, school-based therapy, sports medicine teams), providing regular updates is important. BastionGPT can assist in drafting team conference notes or progress updates that synthesize information from various entries. Imagine a rehab team meeting where the PT, OT, speech therapist, and physician will review a patient’s progress. A PT could use BastionGPT beforehand to generate a brief report encapsulating the patient’s status: combining the latest PT notes (mobility improvements), OT notes (ADL independence level), and any relevant nursing notes (medical issues). The AI might produce a cohesive update like: “Since last meeting, patient Smith has improved transfer ability (now minimal assist) and can self-feed with adaptive utensils. OT reports progress in dressing (needs help only with buttons), PT notes increased walking distance from 50 to 200 feet with a walker, and speech therapy notes improved word-finding in cognitive-linguistic tasks.” The PT would fact-check each point, adjust any phrasing, and then present this summary at the meeting or send it out as an email update to the team. In a sports medicine context, similarly, an athletic trainer could summarize an injured athlete’s rehab progress for the coaching staff and team doctor using an AI-drafted update. By doing the first pass of consolidating information, BastionGPT helps teams stay informed with accurate, up-to-date reports, while the clinician ensures the final version is correct and comprehensible to the audience. Create a concise interdisciplinary team update using the following session notes from multiple providers. Summarize key progress areas (mobility, ADLs, speech/cognition, medical updates, etc.), and present them in a format suitable for case conferences or email updates to colleagues. Notes from PT, OT, SLP, MD, etc.: [Insert here] Across referrals, discharges, and team communications, BastionGPT acts like a diligent documentation assistant. It gathers the right details and formats them into professional correspondence or summaries, so clinicians spend less time wordsmithing and more time on patient care. Crucially, the therapist or physician always reviews these drafts – the AI’s role is to assist, not to replace the clinician’s judgment in communicating patient information. ## Appeals Letters, Clinical Justifications, and Intake Scripts Administrative documents such as insurance appeals, medical necessity letters, and intake forms are another area where BastionGPT can provide valuable help. Writing a compelling appeal or a thorough justification letter can be tedious and require a formal tone with specific terminology. Likewise, creating standardized intake questionnaires or scripts involves careful thought to cover all relevant information. BastionGPT can draft these materials in a polished format, which the clinician or admin staff can then refine. For example: #### Insurance Appeal and Authorization Letters Therapists unfortunately often need to argue for coverage. For example, when an insurer denies additional therapy visits or a particular treatment. BastionGPT can help craft a professional appeals letter that weaves in the pertinent clinical details and adheres to insurance jargon. Suppose a physical therapist must appeal a denial for more PT sessions for a patient who isn’t yet back to functional independence. The PT can provide BastionGPT with the context (diagnosis, progress made, patient’s remaining deficits, risk of stopping therapy early) and request a draft appeal. The AI might generate a structured letter that opens with patient identifiers and claim info, then states: “I am writing to appeal the decision to deny further physical therapy for [Patient Name].” It would go on to cite medical necessity: “[Patient] has made significant gains (e.g. can now walk 150 ft with a cane, up from 50 ft at start of care) but still cannot safely perform essential tasks like showering or grocery shopping. Clinical evidence and standard guidelines (cited below) indicate additional therapy can reasonably be expected to yield further functional improvement, preventing falls and hospitalizations.” The letter would conclude with a respectful request to approve X more visits, referencing attached physical therapy documentation such as progress reports. By structuring the argument and pulling in the clinical facts, BastionGPT enables clinicians to produce clear, persuasive appeals letters with less manual writing, while still meeting payer requirements and maintaining a professional tone. You are a physical therapist. Write a formal appeal letter to an insurance company requesting continued therapy services based on the following information. The letter should begin by stating the patient’s identifiers and diagnosis, then explain the clinical progress the patient has made, their current functional limitations, and why continued treatment is medically necessary. Include a clear explanation of the risks associated with discontinuing care at this point, such as potential regressions or safety concerns. The tone should remain professional and respectful, referencing attached documentation such as progress notes or outcome measures. #### Clinical Justification Letters for Equipment or Special Services Similar to appeals, therapists often have to write letters justifying a request – for example, a custom wheelchair, a specialized splint, or an atypical therapy approach – to payers or other authorities. BastionGPT can outline these justifications by highlighting the patient’s needs and the benefits of the requested item. For instance, an occupational therapist may need to draft a letter of medical necessity for an augmentative communication device for a child with cerebral palsy. By feeding the AI the child’s profile (diagnosis, communication limitations, trials of device in therapy sessions, results), the OT can prompt BastionGPT to create the first draft. The AI might produce: “[Patient Name] is a 7-year-old with spastic cerebral palsy and significantly impaired oral communication. Despite cognitive understanding, he cannot meet his communication needs through speech. During therapy, he demonstrated the ability to use a speech-generating device to make choices and express basic needs when provided with an iPad communication app on trial.” The draft would explain how the requested device (e.g. a specialty eye-gaze communication system) will improve the patient’s daily functioning (“This device will enable him to participate in school and social interactions by selecting words/sentences with minimal physical effort”) and note the consequences of not having it. It would reference any supporting evaluations or therapist recommendations. The OT would then adjust wording, add any required citations (like a quote from a speech therapist’s report or a specific product model), and ensure the letter matches any formatting guidelines. Using AI in this way ensures the final justification letter is comprehensive and well-organized, increasing the chances of approval. You are an occupational therapist. Please generate a clinical justification letter for a requested item or service based on the following information. The letter should clearly describe the patient’s condition and how their daily functioning is affected. Then, explain how the requested item or service—such as a wheelchair, communication device, or orthotic—addresses those limitations and improves their ability to participate in daily life or therapy. If applicable, include a summary of results from any clinical trials or sessions where the item was used. The letter should also note the likely consequences of not providing the requested support and mention any supporting evaluations or therapist recommendations. The format should follow professional standards for medical necessity documentation. #### Intake Forms and Evaluation Scripts On the front end of care, BastionGPT can assist in developing intake documents or scripts that ensure no important question is missed. Whether you’re setting up a new clinic’s patient intake form or creating a structured interview guide for initial evaluations, the AI can generate a template covering all bases. For example, a sports medicine clinic manager could ask BastionGPT to draft a new patient intake form. The AI would include sections for patient demographics, medical history (prior injuries, surgeries, medications), current complaint description, pain scales, activity level, and insurance information. It might also suggest consent language for treatment and HIPAA acknowledgment. The manager then reviews the form, customizing it to the clinic’s workflow and ensuring clarity (e.g. maybe adding specific questions about athletic participation or removing sections that don’t apply). In another scenario, a physical therapist might use BastionGPT to outline an initial evaluation script or checklist, a series of questions and exam steps to follow for a particular condition, like “low back pain intake.” The AI could list questions about symptom onset, aggravating/alleviating factors, red flag screening (numbness, bowel/bladder changes), and functional impact, followed by exam items to check (posture, range of motion, neurological tests). The end result is a thorough template that helps clinicians gather all necessary information in a consistent way. By letting BastionGPT handle the first draft, clinics can establish comprehensive forms and scripts more quickly, ensuring that important data isn’t overlooked during patient intake or evaluation. Draft a comprehensive new patient intake form template suitable for a [insert clinic type or specialty]. The form should include sections for the patient’s demographic information, contact details, medical history including prior treatments and current medications, and a description of the presenting complaint. It should also gather details about pain levels, functional limitations, and any relevant lifestyle or activity context. Include space for insurance and billing information, and if applicable, add a section for HIPAA consent and acknowledgment of clinic policies. The structure should be clear and easy to follow, allowing staff or patients to complete it efficiently. Please create a structured script or checklist for use during an initial evaluation of a patient with [insert diagnosis or condition]. Begin with appropriate history-taking questions, such as when symptoms began, how they present, and what makes them better or worse. Include prompts to assess the impact of symptoms on the patient’s daily life or functional abilities. Add screening questions to identify any red flags that might require referral or caution, and then describe physical exam components that should be included in the assessment. The format should help ensure consistency across evaluations while allowing clinicians to adapt as needed. In all these administrative writing tasks, BastionGPT serves as a knowledgeable aide. It knows the formal language and structure these documents typically require, allowing clinicians to produce polished drafts without starting from zero. The therapist or clinic staff always finalizes the content, adding any case-specific nuances and confirming accuracy, but the time and mental effort saved in drafting can be significant. With BastionGPT handling the repetitive formality, clinicians can focus on the content that truly needs their expert touch. ## Exercise Descriptions, Protocol Templates, and Motivation Scripts Designing therapy interventions and keeping patients motivated are core aspects of rehabilitation. BastionGPT can help clinicians create the written content around these activities, from detailed exercise descriptions and rehab protocols to motivational scripts that encourage patient adherence. By generating clear instructions and structured plans, the AI frees up clinicians to spend more time actually delivering therapy. Here are a few ways this can be applied: #### Detailed Exercise Descriptions for Programs Therapists often give patients lists of exercises to do (in clinic or as homework) and need to include precise instructions for each. BastionGPT can quickly draft exercise descriptions that explain how to perform each movement correctly. For example, a strength and conditioning coach or PT could use the AI to write out the steps for a set of therapeutic exercises for an athlete recovering from shoulder surgery. For an exercise like a resistance band external rotation, BastionGPT might output: “Start by tucking your elbow at your side, bent 90°. Hold the band with tension across your body. Slowly rotate your forearm outward, keeping your elbow glued to your side, then return to the start.” It would add details on sets/reps and form tips (“Keep your shoulder blade gently squeezed back; stop if you feel sharp pain”). The clinician double-checks these instructions for accuracy and may add any patient-specific adjustments (e.g. a smaller range of motion if needed). The benefit is that nothing is left ambiguous; patients get step-by-step guidance in writing, which can improve exercise technique and safety. BastionGPT can generate such text faster than writing it manually, and it can be reused or adapted for future patients with similar programs. Generate step-by-step written instructions for each exercise listed below, suitable for inclusion in a home exercise program or patient-facing handout. Each description should clearly outline the starting position, movement pattern, key form cues, safety guidance, and suggested sets and repetitions. Use accessible language appropriate for patients with no clinical background, and highlight any adjustments or cautions based on the patient’s presentation. Base the content on the exercises and therapist notes provided. #### Rehabilitation Protocol Templates Creating a standardized protocol for a common injury or surgery can ensure consistency in care, but writing one from scratch is labor-intensive. BastionGPT can serve as a starting point by producing a structured template that clinicians then customize. Consider a sports medicine team developing a functional return-to-play protocol for ACL injuries. By prompting BastionGPT with the general outline (phases of rehab, criteria for progression), the AI can draft a multi-phase protocol document. It might break it down into Phase I (acute post-op, focus on pain/swelling control and range of motion), Phase II (strengthening and balance, criteria to progress like 0°-120° knee motion and no swelling), Phase III (running and agility drills, criteria like 80% leg strength), and so on, up to full return-to-sport criteria. For each phase, it can list example exercises or goals (“single-leg hop distance >85% of uninvolved side” as a discharge criterion, for instance). Similarly, an occupational therapist could generate a protocol template for stroke rehabilitation outlining stages from early mobility and basic self-care training to community reintegration tasks. The clinician team would then review these AI-generated protocols, ensuring they align with the latest clinical guidelines and tailoring them to the patient population or facility resources. With BastionGPT’s draft as a base, the process of developing comprehensive protocols is much faster, and the final document helps to ensure no key component of rehab is forgotten (since the AI can incorporate a wide range of suggested activities and milestones). You are a certified athletic trainer. Draft a structured rehabilitation protocol template for the condition described below, broken into progressive phases. Each phase should include therapy goals, example interventions, and specific progression criteria (such as strength, range of motion, or balance milestones). Organize the protocol to reflect a full recovery pathway from early rehab through return-to-function or return-to-play. This draft will be used as a customizable starting point for clinical teams and should align with standard practice patterns or current rehab guidelines based on the case information provided. #### Patient Motivation and Adherence Scripts Keeping patients engaged and adherent to therapy (especially home programs) sometimes calls for creative communication. Clinicians might use BastionGPT to help draft encouraging messages, scripts, or even role-play scenarios that boost patient motivation. For example, a physical therapist could ask the AI to generate a friendly motivational letter or talking script for a patient who is feeling discouraged about slow progress. The AI might produce a compassionate note along the lines of: “I know rehab can be tough and progress sometimes feels slow, but every small step is getting you closer to your goal of running again. Remember when you started you could barely bend your knee – now you’re riding a stationary bike! Keep up the great work with your exercises; consistency will pay off. We’re in this together, and I’m confident you’ll get back on the field in time.” The therapist would personalize this message, adding specific examples of the patient’s progress and a tone that fits their relationship. Or consider an athletic trainer creating a brief scripted pep talk for a team recovering from injuries – BastionGPT could draft an outline hitting positive affirmations and goal reminders. By using AI to get past “writer’s block,” clinicians can come up with uplifting, tailored communications that help maintain a patient’s drive to stick with therapy. It’s crucial that the clinician deliver these messages (BastionGPT is not interacting with patients directly), but the content generation becomes easier with AI’s help. Draft a structured rehabilitation protocol template for the condition described below, broken into progressive phases. Each phase should include therapy goals, example interventions, and specific progression criteria (such as strength, range of motion, or balance milestones). Organize the protocol to reflect a full recovery pathway from early rehab through return-to-function or return-to-play. This draft will be used as a customizable starting point for clinical teams and should align with standard practice patterns or current rehab guidelines based on the case information provided. From exercise instructions to full-blown rehab protocols and encouraging notes, BastionGPT assists in generating the textual materials that surround therapeutic exercises and patient motivation. The AI’s ability to recall countless exercise techniques and coaching phrases means clinicians can draw on a rich knowledge base when creating these resources. ## Educational and Outreach Materials Clinicians in PT, OT, and sports medicine don’t just communicate with current patients, they also play a role in educating the broader community and engaging in professional outreach. BastionGPT can help draft a variety of high-quality educational and marketing materials, from blog posts and workshop outlines to newsletters and social media content. By offloading some of the writing work to AI, therapy professionals and clinic managers can produce valuable content that highlights their expertise and serves their audience, all while saving time. For example: #### Patient Education Articles and Blog Posts Many therapy clinics maintain blogs or article libraries on their websites covering health tips, injury prevention, and success stories. BastionGPT is well-suited to generate draft articles on such topics. A sports physiotherapist could use it to write a blog post on “Top 5 Exercises to Prevent Knee Injuries for Runners.” The AI would produce a structured piece discussing exercises like single-leg squats, hamstring stretches, and core strengthening, explaining how each helps stabilize the knee. It would likely add general advice about proper warm-ups and gradually increasing mileage. The clinician would review this draft, fact-checking any claims (ensuring consistency with current research or accepted best practices), and could include details to match their clinic’s voice. Or perhaps an occupational therapist wants an article about “Strategies for Ergonomic Work-from-Home Setups” to share with patients. BastionGPT can outline recommendations for chair height, screen position, break schedules, etc. You are a sports medicine clinician. Write a patient-friendly blog post or educational article on the topic described below. Structure the article with a clear introduction, practical sub-sections or tips, and a brief conclusion or call to action. Content should be accurate, accessible to the general public, and aligned with common clinical guidelines. Include basic explanations of any recommended techniques or health concepts, and highlight how readers can apply the information in daily life or therapy. Use the following topic or draft outline to guide the post and ensure relevance to the clinic’s patient population. #### Workshops, Presentations, and Community Outreach Material Therapists frequently engage in community education, such as hosting workshops (e.g. a fall-prevention class at a senior center or a “stretching 101” seminar at a gym) or giving talks at local events. BastionGPT can assist in planning these presentations and creating supporting material. The content could start with patient cases which BastionGPT can help de-identify for the presentation. For example, an occupational therapist preparing a workshop on joint protection techniques for people with arthritis could ask the AI to outline the key points and create a handout. BastionGPT might generate an outline covering topics like “Understanding Arthritis and Joint Stress,” “Principles of Joint Protection” (with sub-points like use of larger joints to carry items, taking frequent rest breaks, using assistive devices), and “Examples of Joint-Friendly Techniques in Daily Tasks.” For each section it could suggest a few bullet points or demonstration ideas. It could also draft a one-page summary handout for participants with the main tips. The therapist would then flesh out the outline with personal anecdotes or specific product demonstrations, and help determine if all suggestions are safe and evidence-based. Similarly, a sports medicine physician asked to give a talk on concussion awareness at a school could have BastionGPT help draft the presentation notes or slides text. By providing a solid outline and content to work from, the AI lets clinicians focus on delivering the material engagingly, rather than on writing every word from the ground up. You are a PT. Develop an outline and supporting content for a presentation or workshop based on the topic provided below and remove any patient identifiers for the included. The output should include a structured overview of key talking points, recommended subtopics or activities, and optional visuals or demonstration ideas. Also generate a one-page participant handout summarizing the main takeaways in clear, non-technical language. The material should be engaging, evidence-based, and suitable for a general audience such as community members, patients, caregivers, or students. Base your content on the themes, setting, or event context described below. #### Newsletters and Informational Emails Clinics often send out newsletters or email updates containing rehab tips, clinic news, or seasonal health advice (for instance, “staying hydrated during summer workouts”). BastionGPT can generate these newsletter sections quickly. A physical therapy clinic’s monthly newsletter, for example, might include a “Exercise of the Month” feature, a short patient success story, and a reminder about flu season precautions. The clinic’s marketing coordinator can use AI to draft each segment: the exercise spotlight might be a paragraph on the benefits of plank exercises for core stability, the success story could be a generalized narrative of a patient (with identifying details removed) regaining function after injury, and the health tip might cover something like proper warm-up routines in cold weather. The AI’s drafts give a nice framework and content which the staff then tweak, perhaps by inserting an actual patient quote (with permission) into the success story, or adjusting the exercise description to be more aligned with the clinic’s philosophy. By semi-automating the writing, the clinic ensures they can consistently put out engaging and educational newsletters without overburdening any team member with writing duties. BastionGPT can even help keep the tone positive and clear, which is great for outreach materials meant for a general audience. You are a sports medicine clinician. Create a newsletter section or informational email based on the theme or clinic update described below. The content should be concise, approachable, and suitable for a general audience. Possible elements may include a seasonal health tip, exercise spotlight, success story (anonymized), or clinic news. Keep the tone friendly and professional, and ensure the message is consistent with the clinic’s values or branding. Use the information provided below to shape the draft and support educational outreach or patient engagement goals. Whether it’s for online education or community outreach, BastionGPT allows therapy professionals to share their knowledge widely with less hassle. The AI can adapt to different formats, be it a detailed article, a bullet-point flyer, or a conversational email – providing a high-quality draft that just needs the clinician’s polishing and personal touch. This means more frequent and effective communication with patients, referral sources, and the public, helping to establish the clinician as a trusted resource while ultimately educating people on therapy and wellness topics. ## Summarizing Recovery Narratives and Emergency Reports Modern rehabilitation and sports medicine often involve dealing with data and complex case histories. Patients might come in with wearable device logs (steps, heart rate, activity levels), or clinicians might need to summarize a patient’s entire recovery journey for reporting purposes. Emergencies or incidents during sessions also require prompt documentation. BastionGPT’s ability to analyze and summarize large volumes of information can be benificial in these scenarios. It can comb through raw data or lengthy notes to produce concise summaries and narratives that the clinician can then verify and use. For example: #### Comprehensive Recovery Narratives Sometimes clinicians need to create a narrative summary of a patient’s rehabilitation journey. For example, for a case conference, a research case study, or an insurance report for extended coverage. BastionGPT can pull together information from multiple documents (evaluation, monthly progress notes, re-assessments, etc.) and draft a cohesive story of the patient’s recovery. Imagine a patient who has been in therapy for a year after a traumatic brain injury, now preparing for discharge. The rehab team wants a document summarizing the care provided and progress made. They can feed BastionGPT key parts of the record: initial functional status, major interventions and their dates, and outcome measures over time. The AI might generate a narrative like: “Over 12 months of rehabilitation, patient J.D. progressed from total dependence in self-care to needing only minimal assistance with dressing and independent feeding. In the first quarter, therapy focused on basic mobility (achieved ability to transfer from bed to chair with moderate assist by month 3). The second quarter saw introduction of cognitive tasks and fine motor training; by month 6 the patient could write short lists and walk 150 ft with a walker. In the later stages of therapy, J.D. worked on higher-level balance (progressing from parallel bars to cane by month 9) and began cooking simple meals with setup help. At discharge, the patient can walk household distances with a cane, perform all ADLs with compensatory strategies, and is set to continue outpatient OT for vocational skills training.” The clinicians involved would check this narrative, ensuring each milestone is correct and adding any qualitative insights (perhaps mentioning specific challenges overcome or key motivational factors). Having this kind of summarized timeline is incredibly useful for communicating the value of therapy to external stakeholders. BastionGPT basically acts as a synthesis tool, weaving many pieces of physical therapy documentation into one readable story that demonstrates patient progress and outcomes. You are an OT. Create a comprehensive recovery summary based on the patient’s full rehabilitation course. The narrative should include the patient’s initial presentation, major milestones achieved, interventions used, and outcome measures at key stages of care. Emphasize functional changes, progress toward goals, and the timeline of improvements across therapy domains. Structure the summary so that it can be shared with external reviewers (such as payers, case managers, or research teams) or used for discharge planning or case study documentation. Use the details provided to guide the draft. #### Incident and Emergency Reports If an unexpected event occurs – such as a patient fall, an injury during a session, or any emergency requiring documentation – BastionGPT can assist in drafting the incident report swiftly and systematically. Writing an incident report often requires a clear recounting of events, involved parties, and actions taken, which can be stressful to compose after a tense situation. For example, say an athletic trainer must document an incident where a high school athlete collapsed on the field from heat exhaustion. The trainer can outline the facts to BastionGPT (time of incident, athlete’s condition, immediate actions like cooling and hydration, EMS called, etc.) and ask for a formal report. The AI might produce: “On July 10, 2025, during soccer practice at approximately 4:30 PM, player A.B. collapsed on the field exhibiting signs of heat exhaustion. The ambient temperature was ~90°F. Athletic Trainer [Name] responded immediately, noting the athlete was conscious but disoriented, with an elevated heart rate. The athlete was moved to shade, given cool fluids, and ice packs were applied to neck and armpits. EMS was activated at 4:40 PM. Vital signs were monitored (BP 100/60, HR 130, RR 22) until paramedics arrived at 4:55 PM. A.B. was transported to XYZ Hospital for further evaluation. Parent was notified at 4:50 PM.” This draft covers the who/what/when and actions in an objective manner. The trainer would review it, ensure all details are accurate and nothing is missing (e.g. maybe add that the coach was also present or that the athlete had no prior history of heat illness), then finalize the report. Similarly, a PT in a clinic documenting a patient’s slip in the bathroom would use AI to get a structured incident report written up quickly. The advantage is that BastionGPT can produce a well-organized, professional report under pressure, which the clinician can then calmly review rather than having to compose from scratch in the aftermath of an emergency. This ensures timely and thorough documentation of critical events, which is important for patient safety follow-up and legal compliance. You are a sports medicine clinician. Draft a formal incident report based on the situation described below. Include the date and time of the event, the individuals involved, the exact location, a brief description of what occurred, the patient’s condition, and all immediate actions taken by clinical staff. Document any vital signs, treatments initiated, external parties notified (e.g., EMS, parents, administrators), and the outcome or follow-up plan. Maintain an objective, factual tone appropriate for internal documentation or risk reporting. Use the information provided to complete the report accurately. By harnessing BastionGPT for data summarization, narrative compilation, and incident reporting, clinicians can more efficiently make sense of complex information and ensure accurate records of unusual events. The AI’s strength lies in handling volume and detail: it can crunch through pages of data or notes to help find the signal in the noise. Human experts must validate and refine the AI’s output, but even as a starting point, these summaries and reports can save significant time. Ultimately, this means professionals spend much less effort on laborious record aggregation and more on analyzing results and improving patient care. ## The Future of AI Documentation in PT, OT, and Sports Medicine From the therapy clinic to the training field, BastionGPT transforms how physical therapists, occupational therapists, and sports medicine professionals handle their documentation and writing tasks. The examples above illustrate that generative AI (when used in a secure, HIPAA-compliant manner) can dramatically boost productivity in drafting clinical notes, patient instructions, letters, and reports. Routine documentation can be completed in minutes with AI assistance, allowing providers to focus more on hands-on patient care and clinical decision-making. BastionGPT also helps ensure consistency and thoroughness, reducing the chance of missing a detail or having subpar documentation, since it can include relevant information and follow standardized formats as prompted. In summary, BastionGPT enables therapy and sports medicine organizations to scale their documentation efforts safely and efficiently, keeping privacy and regulatory requirements front and center. By handling the first draft of everything from SOAP notes to educational brochures, BastionGPT saves valuable time and reduces burnout, all while clinicians keep full control over the content. The result is a win-win: better productivity and documentation quality for the practice, and more attention available for what matters most – delivering excellent patient care. Most users are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/ai-for-urgent-care-walk-in-clinics Healthcare Innovation # AI for Urgent Care and Walk-In Clinics: Documentation and Workflow Efficiency Written By: Josh Spencer • June 12, 2026 Every minute an urgent care provider spends typing is a minute a patient spends waiting. Generative AI can now hand a meaningful share of those minutes back: drafting a SOAP note from the encounter itself, producing discharge instructions in seconds, and turning a 30-second dictation into a complete procedure note. The catch is that most AI tools were never built to handle protected health information. This guide walks through seven ways urgent care and walk-in clinic teams use HIPAA-compliant AI to reduce documentation time without compromising care quality or compliance. Each use case includes a sample prompt you can adapt to your own workflow. ## Why urgent care documentation is a solvable problem The math of clinical documentation is unforgiving. Ambulatory physicians spend an average of 5.8 hours in the EHR for every eight hours of scheduled patient time, according to a study of more than 200,000 physicians published in the Journal of General Internal Medicine and highlighted by the American Medical Association . In a walk-in setting, that burden is compressed: 30 to 50 patients per shift, presenting complaints that swing from a pediatric ear infection to chest pain within the same hour, and rarely a dedicated scribe in the room. What changed is not the workload. It is the tooling. Generative AI now handles the structured, repetitive writing that surrounds clinical judgment, so providers spend their time reviewing and finalizing notes instead of composing every one from a blank screen. BastionGPT is built for exactly this environment: a HIPAA-compliant AI assistant and unlimited AI Scribe designed for healthcare workflows, with a Business Associate Agreement (BAA) included on every plan and zero data selling to AI providers. ## Seven ways urgent care teams use AI ### 1. Drafting SOAP notes at the point of care The SOAP note is the backbone of urgent care documentation, and writing a thorough one for every patient under time pressure is where providers feel the squeeze most. BastionGPT's AI Scribe captures audio from the patient encounter and generates a structured draft SOAP note automatically. You review and edit the draft instead of composing from scratch, so the final note reflects the actual encounter rather than a reconstruction from memory at the end of a shift. Multi-speaker recognition (up to ten speakers) distinguishes provider, patient, and family member voices, which matters in urgent care rooms where a parent or caregiver is often part of the conversation. Sample prompt (note drafting from bullet points): Draft a SOAP note for an urgent care visit. The patient is a 34-year-old male presenting with right ankle pain after a fall playing basketball. No fracture on X-ray. Assessment: Grade 2 lateral ankle sprain. Plan: RICE, NSAIDs, ace wrap, follow up with PCP in 7-10 days, return precautions given. ### 2. Generating discharge instructions in seconds Discharge instructions are among the most repetitive writing tasks in urgent care. They need to be clear enough to follow at home, clinically accurate, and specific to the visit. Describe the condition, the treatment given, and the follow-up plan, and BastionGPT drafts patient-facing language you can review, print, or send. Sample prompt: Write discharge instructions for a patient diagnosed with a urinary tract infection. They were prescribed a 5-day course of nitrofurantoin. Include what to expect, when to take the medication, warning signs to watch for, and when to return to urgent care or go to the ER. The same approach works for wound care, splint care, return precautions after head injury, and post-procedural guidance after laceration repair. ### 3. Documenting procedures by dictation Procedure notes support billing, continuity of care, and occasionally legal or insurance review. Dictating during or immediately after the procedure and letting AI structure the note is faster and more accurate than transcribing from memory hours later. Sample prompt: Write a procedure note for a laceration repair. Location: 3 cm linear laceration to the left forearm, no involvement of tendons or neurovascular structures. Irrigation performed, wound closed with 4-0 nylon simple interrupted sutures x6. Patient tolerated procedure well. Instructions for suture removal in 7-10 days. ### 4. Drafting prior authorization requests and appeals Prior authorizations pull providers and coordinators away from the floor, particularly for imaging, specialty referrals, and higher-cost medications. BastionGPT drafts prior authorization narratives and appeal letters from the clinical picture you describe. That support matters most on nights and weekends, when the full administrative team is off but clinical decisions still need documentation. Sample prompt: Draft a prior authorization request for an MRI of the right knee for a 52-year-old patient with acute knee pain, hemarthrosis on exam, and a mechanism consistent with ligamentous injury. X-ray negative for fracture. PCP referral sent. Insurance is [carrier name]. ### 5. Writing referral letters and consult notes A clear referral letter helps the receiving specialist understand the clinical picture without starting over, and it cuts back-and-forth between offices. BastionGPT produces a professional referral letter in a consistent format from the visit details you provide, in far less time than composing one manually. Sample prompt: Write a referral letter to orthopedics for a patient with a distal radius fracture confirmed on X-ray. The fracture appears minimally displaced. Patient is a 67-year-old female, right hand dominant, mechanism was a fall on outstretched hand. Splinted in the ED, follow-up recommended within 3-5 days. ### 6. Supporting front desk and operations teams Documentation burden in urgent care extends past the exam room. Front desk staff and clinic coordinators use BastionGPT to draft patient notification emails, write and update clinic policies and SOPs, create training materials for new hires, and summarize documents for operational review. Sample prompt: Write a patient communication email letting patients know our urgent care clinic will have extended hours on July 4th weekend. Include our address, hours, walk-in policy, and a note that we can handle non-emergency needs to help reduce ER congestion. ### 7. Creating patient education materials on demand Generic printed handouts rarely reflect what happened in the visit. BastionGPT generates condition-specific, plain-language education materials tailored to the encounter, at the reading level you specify. It also drafts materials in multiple languages on request, which helps in pediatric visits where parents need clear guidance and with patients whose first language is not English. Sample prompt: Write a one-page patient education sheet on managing a second-degree burn at home. Include wound care steps, signs of infection, when to return to urgent care, and over-the-counter pain management options. Write at a 6th-grade reading level. ## Compliance in urgent care: what to look for in an AI tool Urgent care sits at a demanding compliance intersection. Walk-in volume means more patients and more PHI in a compressed timeframe. Staff turnover runs higher than in many traditional practice settings, and multi-location chains add data governance complexity. General-purpose AI tools were not designed for this. Standard ChatGPT does not include a Business Associate Agreement and is not intended for use with protected health information. BastionGPT is HIPAA compliant by design: BAA included on every plan, including individual plans AES-256 encryption at rest and HTTPS/TLS encryption in transit HITRUST and ISO 27001-certified data centers in the USA, Canada, and Australia Chats and transcripts wiped after 30 days by default; you can retain them longer or delete them sooner Prompts, documents, and audio are never shared with OpenAI or sold to other AI providers, and never used for model training For urgent care operators managing multiple locations, the Ultra plan adds custom security configurations, tailored legal terms, and a dedicated support representative: enterprise controls that help standardize AI use across a larger team without creating a compliance patchwork. ## Getting started in an urgent care workflow There is no EMR integration project to schedule. BastionGPT works alongside Epic, Cerner, Meditech, and other major EMR systems through a simple copy-paste or document-upload workflow. Providers run the AI Scribe during encounters or draft notes between patients in the chat interface. Either way, the goal is the same: less time reconstructing documentation after the fact, more time in the room with the next patient. "BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations." Dr. Adil Manzoor, DO, MBA/MS, Internal Medicine and Pediatrics ## Frequently asked questions ### Is BastionGPT HIPAA compliant? Yes. BastionGPT is HIPAA compliant by design, and a Business Associate Agreement (BAA) is included with every plan. It also supports PIPEDA, PIPA, and PHIPA requirements in Canada and APP requirements in Australia. ### Will AI replace clinical judgment in urgent care? No. BastionGPT drafts documentation; providers review, edit, and approve every note before it enters the record. The AI takes on the writing workload that surrounds clinical judgment so providers can spend more time on patient care. ### Can AI scribes be used in urgent care settings? Yes. AI scribes are well-suited to urgent care because of the high visit volume and documentation pressure providers face. BastionGPT's AI Scribe captures audio from patient encounters and generates structured output including transcription, SOAP notes, DAP notes, and custom note formats. Multi-speaker recognition supports rooms where a patient, family member, and provider are all present. ### What kinds of documents can AI help urgent care clinics generate? Urgent care teams use AI to draft SOAP notes, procedure notes, discharge instructions, patient education materials, referral and consult letters, prior authorization requests, and internal operational templates like staff communications and clinic SOPs. Both clinical and administrative staff benefit. ### Does BastionGPT integrate with our EMR? BastionGPT works alongside Epic, Cerner, Meditech, and other major EMR systems using copy-paste or document uploads. No IT setup or integration build is required, so teams can start the same day. ### Is patient data used to train AI models? No. Prompts, uploads, and audio are never shared with OpenAI or sold to other third-party AI providers and are never used for AI training. Chats and transcripts are wiped after 30 days by default, and users keep full ownership of their data. ### How much does BastionGPT cost for urgent care teams? Plans start at $20 per user per month, with a BAA included at every tier and unlimited AI Scribe on all plans. ## Try BastionGPT free for 7 days BastionGPT is available for individual providers and urgent care teams of any size. Plans start at $20/user/month, a BAA is included with every plan, and there are no hidden fees. Start your free 7-day trial . BastionGPT assists healthcare professionals with documentation and administrative tasks. It does not replace professional judgment. Clinicians review all AI-generated output before it becomes part of the medical record. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/ai-healthcare-data-anonymization Healthcare Innovation # What Healthcare Data Anonymization is and How to De-identify PHI Written By: Josh Spencer • February 3, 2025 Secure data sharing is essential for advancing medical research, improving quality of care, and supporting public health initiatives. Healthcare providers need HIPAA-compliant AI solutions that can help them share valuable medical data while ensuring patient confidentiality and regulatory compliance with privacy laws like HIPAA, PIPEDA, and APP. ## What Is Healthcare Data Anonymization? Healthcare data anonymization is the process of removing or obscuring personally identifiable information (PII) and protected health information (PHI) from medical records so individuals cannot be re-identified. While often used interchangeably, anonymization, de-identification, and pseudonymization have distinct meanings: Anonymization: Irreversibly removes all identifying information, making re-identification extremely difficult even with additional data. De-identification: Removes direct identifiers (names, SSNs) per regulatory standards like HIPAA's Safe Harbor method, but may retain some risk of re-identification through linkage attacks. Pseudonymization: Replaces identifiers with artificial substitutes (like codes) that can be reversed with a key, maintaining data utility for internal analysis. In healthcare contexts, PHI (Protected Health Information) under HIPAA includes 18 specific identifiers like names, dates, addresses, medical record numbers, and even IP addresses. PII is broader, covering any data that could identify individuals (email, phone numbers, device IDs). Key regulations include HIPAA (US), APP (Australia), and PIPEDA (Canada), which all require organizations to protect personal data from misuse and unauthorized access. HIPAA offers two de-identification methods: Safe Harbor (removing 18 PHI elements) and Expert Determination (statistical validation of re-identification risk). ## Why Data Anonymization Is Important for Healthcare Organizations? Data anonymization bridges the gap between healthcare data sharing and privacy protection. By leveraging AI-powered de-identification tools, healthcare providers can efficiently remove personally identifiable information (PII) and protected health information (PHI) while maintaining the clinical value of their medical records. ### 1. Advancing Medical Research Through Secure Data Sharing Medical researchers rely heavily on de-identified patient data to study diseases, evaluate treatments, collaborate with colleagues, and develop new therapies. Through AI-driven anonymization, hospitals and clinics can securely share patient records with research institutions without compromising HIPAA compliance. This capability is particularly valuable for rare disease research, where access to diverse patient data drives medical breakthroughs. ### 2. Supporting Mental Health Research and Treatment Mental health professionals face unique challenges when sharing valuable clinical insights while protecting sensitive patient information. Using HIPAA-compliant AI tools, therapists and psychiatrists can efficiently contribute anonymized case studies, treatment outcomes, and therapeutic approaches to advance mental health research and improve patient care. This secure data sharing enables professionals to collaborate on complex cases, develop more effective treatment strategies, and contribute to the growing body of mental health research while maintaining strict patient confidentiality. ### 3. Enhancing Quality Improvement Through Case Analysis Healthcare organizations frequently need to share specific patient cases for peer review and quality improvement initiatives. For example, a hospital's quality improvement team might need to analyze a complex surgical case that had unexpected complications. Using AI-powered anonymization, providers can share detailed case information - including pre-operative assessments, surgical notes, and post-operative care details - while protecting patient privacy. This enables thorough analysis of clinical decision-making and outcomes, facilitating meaningful discussions about potential improvements in care protocols without compromising patient confidentiality. ### 4. Supporting Medical Education and Training Teaching hospitals and medical schools can leverage AI-powered anonymization to create valuable educational resources from real patient cases. By carefully de-identifying patient records, healthcare institutions can build comprehensive case libraries for training medical students, residents, and continuing education programs. This approach maintains the clinical value of real-world examples while ensuring patient confidentiality in educational settings. ### 5. Facilitating Legal Reviews and Fraud Detection Healthcare organizations often need to share patient records for legal reviews, audits, or fraud investigations. AI-driven anonymization helps providers securely share relevant medical documentation with legal teams, auditors, and investigation units while maintaining HIPAA compliance. This enables thorough review processes without compromising patient privacy or violating regulatory requirements. ## The Critical Role of Provider Review in AI Anonymization While AI significantly streamlines the anonymization process, healthcare provider oversight remains essential. Medical professionals must carefully review AI-generated outputs to ensure: Complete removal of all direct and indirect patient identifiers Preservation of clinically relevant information Appropriate handling of unique or complex cases Consistency in de-identification across related documents Verification that anonymized data maintains its utility for intended purposes This human oversight, combined with AI efficiency, creates a robust system for secure data sharing while maintaining professional standards and regulatory compliance. ## BastionGPT: HIPAA-Compliant AI for Healthcare Data Anonymization and De-Identification BastionGPT provides healthcare providers with an advanced AI solution for efficient, accurate data de-identification. Our platform combines generative AI technology with healthcare-specific security features to ensure both data utility and HIPAA compliance. ## How Our AI Anonymization Works AI-Powered PHI Detection: Our system uses large language models to identify both direct and indirect identifiers in electronic medical records (EMR). Contextual Analysis: Healthcare-specific AI algorithms recognize PHI within complex clinical narratives where traditional de-identification tools might fail. Intelligent Data Processing: Rather than simple redaction or masking, our AI implements -sophisticated anonymization techniques that preserve clinical relevance. ## Implementing AI for Medical Records Anonymization Healthcare providers can utilize BastionGPT's HIPAA-compliant AI capabilities using this specialized prompt: You are a medical records specialist, help me anonymize the following patient chart while preserving the medical relevance. Replace all Protected Health Information (PHI) with bracketed placeholders: PHI to remove and replace: Names (patient, family, providers) → [name] Dates → [date] Ages → [age] Sex → [sex] Geographic data smaller than state → [location] Contact info (phone, email, addresses) → [contact] Account/medical record numbers → [ID] Vehicle/device identifiers → [device_ID] URLs/IP addresses → [URL] Biometric identifiers → [biometric_ID] Photos/comparable images → [image] Any other unique identifying characteristics → [identifier] After anonymizing, verify no PHI remains exposed. ### Example of AI-Driven De-identification of Healthcare Data Original Medical Record: "John Doe, a 42-year-old male from Los Angeles, was admitted to Cityview Hospital on March 10, 2023, for a cardiac evaluation. He is an executive at a biotech firm in Silicon Valley. His provider, Dr. Smith, prescribed a stress test." AI-Anonymized Version: "[name], a [age]-year-old [sex] from [location], was admitted to [location] Hospital on [date] for a cardiac evaluation. He works in the biotech industry. His provider, [name], prescribed a stress test." ## The Future of AI in Healthcare Data Security The advancement of medical science depends on our ability to share and analyze healthcare data securely. BastionGPT's HIPAA-compliant AI technology helps healthcare providers contribute to this future while maintaining the highest standards of patient privacy and regulatory compliance. By combining AI-powered medical transcription anonymization with healthcare professional oversight, organizations can confidently share valuable clinical data for research, quality improvement, and public health initiatives. This comprehensive approach ensures that patient privacy remains protected while enabling the collaborative efforts that drive medical progress. Ready to explore how HIPAA-compliant AI can enhance your organization's data security? Start a free 7-day trial of BastionGPT If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/ai-transcription-services-for-medical-professionals Healthcare Innovation # AI Transcription Services for Medical Professionals Written By: Josh Spencer • August 8, 2023 In today's healthcare environment, clinical documentation is of the utmost importance. Medical professionals are burdened with hours of record-keeping and administrative work that can often distract from their core role: patient care. One avenue to explore for minimizing this burden is AI generated transcription services , especially for voice-recorded medical notes, interviews, and consultations. For a broader walkthrough, see our healthcare AI scribe documentation guide . BastionGPT is used by hundreds of medical professionals daily to process transcriptions and will have its built-in audio transcription feature in future versions. In the meantime, the market offers a wide range of transcription options, some specifically tailored for medical professionals. But how do you choose one that's right for you? This article provides a rundown of some of the most popular medical transcription services that we have hands-on experience with, focusing on cost, quality, and HIPAA compliance. If you have others that we should try, or have any feedback on those that we have reviewed, please reach out to us. The author is not sponsored by, nor has received compensation from any of these companies for writing our opinions. We are a partner with Microsoft but have not been compensated for writing our review of any of their services. ## Considerations for Choosing a Service 1. HIPAA Compliance: Ensure the service you choose complies with Health Insurance Portability and Accountability Act (HIPAA) standards to protect patient information. 2. Quality: Higher quality services might reduce the need for editing, thereby saving time in the long run. 3. Cost: Weigh the features offered against the monthly or one-time costs to decide what's best for your organization's budget. 4. Integration: How well does the service integrate with your existing workflow and software systems? 5. Scalability: Can the service handle increasing demands as your practice grows? ## Major Players Microsoft Word Transcription : This is free with an existing license of Microsoft Word. However, it performed the poorest of the the solutions we tested. However, it does now support the ability to label different voices automatically, which you can then assign names to. Microsoft Teams Premium : This requires purchase of a Microsoft subscription that includes Microsoft Teams. If you would like to take advantage of AI summarization, you will need to purchase the Microsoft Teams Premium option, which currently retails for $7 per user per month. Nuance Dragon Professional : This is a long established AI transcription service that provides great transcription, but it can often miss certain medical terminology. However, for a one time cost of $699, we don’t expect the performance of their higher prices options. A free demo is available. For a voice-assistant alternative in the same dictation category, see how BastionGPT compares with Suki . Whisper : This is a publicly available AI transcription model. There are a number of different Apps that provide access to Whisper. The versions that run its AI entirely on your computer and not in the cloud can be HIPAA compliant. The versions that run in the cloud is often not HIPAA compliant. At this time, we do not see an app that we trust enough to recommend. However, if you are technically proficient, or know someone who is, this can be installed on your computer. There are a number of HIPAA compliant cloud versions announced, but as of this writing, we are not aware of any that are on the market. Nabla Copilot : This is a new player to the healthcare AI transcription game. Their services start at $2,300/month. This solution has some promise, and they do claim HIPAA compliance. However, I strongly encourage a thorough review in this case for HIPAA compliance, as some compliance items might not pass muster for specific healthcare organizations. They have a free service tier, but that doesn’t include HIPAA compliance. Deepscribe : This is also a relatively small and young service, but also has some great potential. We will be watching this service closely to see how it performs against some of the giants such as Microsoft Nuance in this space. We do not have official numbers on pricing, but have seen reports of approximate costs coming in at $800/month/user. Nuance DAX Express : This is the best of the best. Microsoft has made a huge investment in this technology, and it shows. However, this also fetches the highest price tag, with pricing reported at approximately $1,800/month/physician. *Updated 3/3/2024* BastionGPT : Due to high user requests, BastionGPT has incorporated transcription at no additional cost. This uses some of the same technology as the Microsoft Nuance DAX Express for healthcare-grade transcription and compliance, but at a fraction of the cost. To test this out today, register for a free 7-day trial of BastionGPT , and click the "Transcription" option in the upper right corner. ## Conclusion Transcription services can significantly reduce the administrative burden on medical professionals. When choosing a service, it's crucial to consider factors like HIPAA compliance, quality, cost, and how well the service fits into your existing workflow, especially for clinical documentation. Choose wisely, and you'll find that the right transcription service can become an invaluable part of your healthcare practice. For a focused comparison, see which is the best AI scribe . ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/apa-guidance-generative-ai Healthcare Innovation # BastionGPT: The Safe, Compliant, and Credible AI Tool for Psychiatrists Written By: Josh Spencer • March 17, 2025 Psychiatrists are understandably cautious about adopting new AI tools, especially when sensitive patient information and clinical decision-making are involved. The American Psychiatric Association (APA) has emphasized “ important guardrails for psychiatrists to consider ” with AI, such as ensuring privacy compliance and evidentiary support. In fact, APA guidance strongly warns clinicians to avoid entering patient data into generic AI apps like ChatGPT unless HIPAA requirements are met. BastionGPT was designed specifically to address these concerns – it aligns with the APA’s AI tool guidelines on safety and ethics, and it passed review from the APA’s evaluation team, earning it a featured spot at the 2025 APA Conference. From HIPAA-compliant data practices to a focus on clinically vetted knowledge, BastionGPT offers psychiatrists a powerful yet user-friendly AI assistant they can trust in practice. ## Aligning with APA Guidelines and Professional Standards APA’s own checklists for evaluating clinical AI ask critical questions: Is the tool safe and effective? Is it privacy-compliant? Does it augment (not replace) clinician judgment? BastionGPT checks all these boxes. It was built in line with APA-recommended principles for augmented intelligence – to “augment human decision-making rather than replace it” while keeping the physician in control. For example, APA policy advises that any AI used with patient data must have proper HIPAA safeguards (including a HIPAA business associate agreement where applicable). BastionGPT is exclusively for healthcare and mental health professionals, a HIPAA Business Associate Agreement is automatically incorporated into every plan, ensuring you maintain compliance. The platform also abides by APA’s ethical stance that AI is “a tool, not a therapy,” meaning the psychiatrist remains the ultimate decision-maker. BastionGPT’s credibility has been affirmed through numerous professional reviews. It underwent scrutiny by the American Psychological Association's review team to be approved for presentation at the 2025 Conference , highlighting that leaders in the field recognize its adherence to safety and quality standards. This kind of vetting – alongside alignment with the APA’s own AI Tool Guide – should reassure psychiatrists and psychologists that BastionGPT has made the proper investments to be a powerful and safe clinical tool. ## HIPAA-Compliant and Secure by Design When it comes to client confidentiality, BastionGPT was built from the ground up with data security in mind. All data exchanged with BastionGPT services is protected under HIPAA compliance , so you can confidently input PHI (Protected Health Information) without violating privacy laws. In practice, this means robust safeguards: encryption, secure servers, and strict data handling policies. BastionGPT uses private instances of leading AI models the data entered into our service is never provided to OpenAI or sold to 3rd parties, ensuring your conversations stay protected. BastionGPT acts as a secure, closed “bastion” for your data – giving you the convenience of AI assistance without compromising on privacy or ethics. ## Built with Clinical Expertise for Accuracy and Trust Another key concern for psychiatrists is whether an AI’s answers can be trusted. General-purpose chatbots may more heavily improvise or provide irrelevant information, which is problematic in a patient care context. BastionGPT tackles this by testing and tuning its AI models to optimize clinical expertise and evidence-based knowledge at its core. The platform is engineered to prioritize medical principles and scientific accuracy from reputable healthcare resources, reducing the risk of the pseudoscience or factual errors often seen in generic AI outputs. In practice, this means BastionGPT is less likely to hallucinate false medical information and more likely to stick to established guidelines and research when assisting you. Crucially, BastionGPT was developed by healthcare experts for healthcare professionals . The leadership team and advisory board includes psychiatrists, AI experts and security specialists, led by a founder with over a decade of experience in healthcare technology and compliance. This real-world clinical insight shaped BastionGPT to fit seamlessly into psychiatric workflows – whether drafting a progress note, transforming a collection of loose session notes into a well-structured summary, or formulating client education materials. Unlike one-size-fits-all AI, BastionGPT won’t balk at medical terminology or mental health topics. It was purpose-built for the language of healthcare and psychiatry, so it understands context and nuance that matter in clinical discussions. This includes the ability to work with violent or sexual content that is forbidden by other AI services. The result is an AI partner that helps elevate the quality of your documentation while giving you more time in the day with your clients and families. Psychiatrists using BastionGPT have the added confidence that thousands of other health professionals and organizations are already trusting it in their daily work. This broad adoption, combined with its APA vendor approval, underscores that BastionGPT has earned trust in the medical community through proven performance and compliance. ## Experience BastionGPT Yourself At the end of the day, the best way to evaluate any tool is to try it in your own practice. BastionGPT offers a risk-free 7-day free trial so you can see firsthand how it handles your clinical documentation, education, or administrative tasks. Take this opportunity to put BastionGPT’s compliance and clinical savvy to the test with your toughest use-cases. You’ll find that having a HIPAA-secure, APA-aligned AI assistant at your side can streamline your workload without ever undermining your professional standards or patient trust. BastionGPT stands out as a rare breed of AI – one that is both powerful and psychiatrists’ practice-ready. It has the endorsement of guidelines and peers, the technical safeguards required in healthcare, and the knowledge base to actually be useful in a mental health setting. If you’ve been looking for a way to work smarter, not harder, this is your chance. Join over 4,000 psychiatrists already benefiting from BastionGPT by starting your free 7-day trial today . ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/app-compliance-australia Healthcare Innovation # Compliance for Australian Users: BastionGPT and the Australian Privacy Principles (APPs) Written By: Josh Spencer • December 6, 2024 BastionGPT (“we, “our”, or “us”) is a healthcare-specific AI service offered by FortaTech Security and designed with Australian privacy, security, and compliance at its core. We protect your privacy in accordance with the Privacy Act 1988 (Cth) and the Australian Privacy Principles (APPs). This is an addendum to our BastionGPT Privacy Policy and Terms of Use . ## Compliance with the 13 Australian Privacy Principles ### 1. Open and Transparent Management of Personal Information We are transparent and open about how we manage your personal information. Our privacy policy outlines how we collect, manage, store, and disclose your data. ### 2. Anonymity and Pseudonymity We allow users to interact anonymously or use pseudonyms wherever possible. We only collect the minimal information necessary to validate your identity to provide our services. You can use whatever information you deem necessary and in alignment with your policies within the tools. ### 3. Collection of Solicited Personal Information We gather only personal information that is necessary for us to provide services: · Contact details (name, e-mail address, phone number) · Personal information (company, title) · Payment information · Data from the use of our services · Any other information you provide directly to us ### 4. Dealing with Unsolicited Personal Information If personal information is collected that we did not solicit in accordance with APP-3, we will destroy or de-identify it if it is lawful and reasonable to do so. ### 5. Notification of the Collection of Personal Information When we collect personal information, we will take reasonable steps to notify you about: · Our identity and contact details. · The entities or types of entities with which we share information. · Information about how you can access and correct personal information. ### 6. Use or Disclosure of Personal Information BastionGPT only uses or discloses customer data based on customer instructions or lawful requests. The use of personal information is strictly limited to providing the requested services, such as generating documentation or analysis. Your data is not used to improve AI models like ChatGPT and Claude. We will only use or disclose personal information for the purpose it was collected or: · If you have consented to share the data. · You would reasonably expect us to disclose the information to provide our services. · As otherwise required by law. ### 7. Direct Marketing BastionGPT only uses your information for direct marketing purposes if we collect the information from you and in relation to our services. You may opt out of receiving direct marketing communications by contacting us directly or using the mechanisms in our communications. ### 8. Cross-border Disclosure of Personal Information We adhere to requirements for data residency, security, transmission, and encryption. User data is stored in Australia for accounts registered in Australia. Regardless of data location, we adhere to stringent data protection standards to comply with cross-border privacy requirements. ### 9. Adoption, Use, or Disclosure of Government Identifiers We do not identify you with government identifiers (such as Tax File Numbers) unless required or authorized by law. ### 10. Quality of Personal Information We aim to maintain the quality of personal information through regular checks to ensure it is accurate, up-to-date, and complete. If your information changes, you can update it directly or contact us for support to ensure accuracy. ### 11. Security of Personal Information We use advanced encryption to protect personal information and restrict access based on the principle of least privilege. Our platform undergoes regular risk assessments, including third-party security reviews and penetration testing, to maintain high-security standards. When your personal information is no longer required, we take reasonable steps to destroy or de-identify it unless legally required to retain it. ### 12. Access to Personal Information You have the right to request your personal information at any time. Please contact our support team, and we will respond to your request within a reasonable timeframe. We will share the information per your request if it is reasonable and practicable to do so in accordance with the APPs. ### 13. Correction of Personal Information If you believe any information is incorrect, you can modify your data to ensure it remains accurate and current. Our customer support team is available to assist you, further supporting the accuracy of the data. E-mail: [email protected] Phone: +1 (214) 619-8696 Mailing Address: 11816 Inwood Rd #3181, Dallas, TX 75244 If you have any questions about adherence or specific legal requirements, you can reach out to our legal team at [email protected] We are here to support your AI journey, and you can schedule time with us to discuss any questions you have around compliance with the APPs. We are committed to resolving all complaints promptly. If you feel our response is inadequate, you can reach out to the Office of the Australian Information Commissioner (OAIC): Website: www.oaic.gov.au Phone: 1300 363 992 Email: [email protected] ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/best-practices-for-healthcare-ai-prompts Healthcare Innovation # Best Practices for Healthcare AI Prompt Engineering Written By: Robert Henehan • November 12, 2024 AI is rapidly transforming how healthcare professionals document, communicate, and make decisions, but the quality of those outcomes depends heavily on how the technology is used. Prompt engineering plays a critical role in guiding AI-generated content to be accurate, structured, and clinically useful. BastionGPT is designed to support high-impact workflows like generating clinical documentation, transcribing visits , care planning, writing patient education, and supporting billing compliance. This guide breaks down best practices for crafting effective healthcare prompts that help clinicians get more reliable, precise, and actionable output. ## Tips for Healthcare AI Prompt Engineering ### Establishing Professional Context The foundation of any medical prompt begins with establishing the appropriate clinical context and expertise level. Define the healthcare professional role explicitly Example: "You are a pediatric neurologist evaluating a complex case of childhood epilepsy" Example: "Act as a healthcare navigator, helping patients understand their insurance options" Specify the level of expertise and target audience Clinical specialists Allied health professionals Patients and caregivers Administrative staff ### Structuring Your Prompt Clear structure ensures consistent and reliable outputs while maintaining healthcare standards, a principle rooted in prompt engineering. Begin with a clear objective statement Example: "Create a patient education handout explaining the basics of type 2 diabetes management" Example: "Analyze these lab results and provide clinical interpretation with recommended follow-up" Use standardized placeholders for variable information Patient demographics: , Clinical data: , Missing information: , Specify desired format and components SOAP note structure Bullet-point summary Decision tree format Table format for medication lists ### Setting Parameters and Requirements Specify language and comprehension level "Write at an 8th-grade reading level for patient instructions" "Use technical medical terminology appropriate for specialist consultation" "Include both generic and brand names for medications" Define content boundaries "Focus only on FDA-approved treatments" "Include contraindications and warning signs" "Exclude experimental procedures" ### Clinical Context and Medical Accuracy Provide relevant clinical information Patient history elements Current symptoms and findings Relevant test results Treatment history Emphasize evidence-based practice "Include only current clinical guidelines" "Reference recent meta-analyses when available" "Highlight strength of evidence for recommendations" ### Quality Assurance Guidelines Request fact-checking parameters "Verify all medication dosages against current guidelines" "Cross-reference recommendations with standard of care" "Flag any areas requiring physician verification" Include safety measures "Add standard medical disclaimers" "Include emergency warning signs" "Note when immediate medical attention is required" ### Final Tips for Success Test and iterate prompts to refine outputs Use BastionGPT to save successful prompts for common tasks Check out the BastionGPT library of prompts to get you started Document any limitations or areas requiring human verification ###### Remember: AI models are tools to augment, not replace, clinical judgment. Always verify generated content against current medical standards and institutional policies. ## Comparing Prompt Examples in Practice ### Poor Prompt Example: "Write about diabetes treatment" ### Better Prompt Example: "You are an endocrinologist creating a comprehensive treatment plan for a newly diagnosed Type 2 Diabetes patient. Include: Initial medication recommendations Blood glucose monitoring schedule Lifestyle modifications Follow-up timeline Warning signs requiring immediate attention Write at an 8th-grade reading level, include both English and Spanish translations, and format as a patient-friendly handout with bullet points and clear sections." ### Excellent Prompt Example: "You are a certified diabetes educator preparing materials for a newly diagnosed Type 2 Diabetes patient. [Use the attached document or copied text to generate the following summary] The patient is a year-old with . Create a personalized management plan that includes: Medication Management: Current medications: Administration schedule Side effects to monitor Monitoring Protocol: Target blood glucose ranges Testing schedule Recording methods Lifestyle Modifications: Dietary recommendations Exercise guidelines Sleep hygiene Format the content at an 8th-grade reading level, include culturally appropriate dietary examples, and provide clear action steps for emergency situations. Include spaces for personalized goals and monitoring logs." ## Healthcare AI Prompt Examples in Practice ### Example Set 1: Therapy Session Documentation Poor Prompt Example: "Write therapy notes for my patient." Better Prompt Example: "You are a licensed mental health therapist writing a BIRP note for a 45-minute CBT session. Include: Client's presentation and affect Interventions used today Progress toward treatment goals Safety assessment Plan for next session Write in professional clinical language appropriate for insurance documentation and include standard disclaimers." Excellent Prompt Example: "You are a licensed mental health therapist documenting a cognitive behavioral therapy session. [Use the attached document or copied text to generate the following summary] The client is a year-old in session for . Create a comprehensive BIRP note that includes: Behavioral Observations: Mental Status: Affect/Mood: Risk Assessment: Engagement Level: Interventions: Specific CBT techniques: Skills taught/practiced: Homework reviewed: Resources provided: Response: Progress on treatment goals: Insight/participation level: Symptom changes: Homework compliance: Plan: Next session focus: New homework assigned: Referrals needed: Follow-up schedule: Format the note using current CPT documentation requirements, include appropriate clinical terminology, and add standard confidentiality statements. Flag any safety concerns for immediate supervisor review." ### Example Set 2: Patient Chart Summary Poor Prompt Example: "Summarize this patient's chart" Better Prompt Example: "You are a medical scribe creating a chart summary for a complex patient with multiple chronic conditions. Include: Significant diagnoses Current medications Recent hospitalizations Outstanding tests Care team members Follow-up appointments Format the summary in a clear, hierarchical structure using standard medical terminology." Excellent Prompt Example: "You are a clinical documentation specialist preparing a comprehensive chart summary for care team review. [Use the attached document or copied text to generate the following summary] The patient is a year-old with . Create a structured summary including: Demographics and Background: Insurance: PCP: Preferred Language: Advanced Directives: Active Problems (prioritized): Primary Diagnoses: Current Status: Date of Onset: Treating Providers: Medication Profile: Current Medications: Recent Changes: Allergies/Reactions: Pharmacy: Recent Care Events: Hospitalizations: ED Visits: Procedures: Specialist Consults: Outstanding Items: Pending Tests: Referrals: Care Gaps: Due/Overdue Items: Care Coordination: Team Members: Recent Communications: Care Conference Notes: Social Support: Format the summary with clear headers, highlight critical information, include relevant dates, and flag any immediate action items. Add ICD-10 codes for active conditions and note any quality measure gaps that need addressing." ## Ready to streamline your workflows with AI that understands healthcare? BastionGPT helps medical professionals reduce documentation time, improve accuracy, and generate structured, compliant outputs without adding complexity to their day. By applying the prompt engineering best practices outlined above, you can unlock even greater value from every interaction. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/bias-in-healthcare-ai-a-guide-for-end-users Healthcare Innovation # Recognizing and Mitigating Bias in Healthcare AI: A Guide for End Users Written By: Josh Spencer • May 26, 2024 ## Introduction As generative AI continues to integrate into healthcare, understanding its inherent biases becomes critical for every user. While these biases originate from the data and systems beneath the technology, there are practical steps that healthcare professionals can take to mitigate their impact. This guide explores common types of AI bias and offers actionable advice to ensure that the technology is used responsibly and effectively. ## Common Types of Bias in Generative AI 1. Anchoring Bias When interacting with AI, the language you choose can significantly affect the responses you receive. For example, inputting "remedy" might yield less scientifically valid suggestions than "treatment." Similarly, documenting a patient as “crazy” rather than “exhibited erratic behavior” could lead to less medically appropriate AI responses. Being mindful of the words used in queries can help reduce this form of bias. 2. Generalization Bias AI systems can struggle with rare conditions or underrepresented demographics due to insufficient data. When querying AI about less common diseases or patient groups, it's important to critically evaluate the relevance and accuracy of AI responses, understanding that the system may not have comprehensive data to draw from. 3. Statistical (Stereotyping) Bias AI might produce outputs that inadvertently stereotype groups, such as associating certain health conditions with specific ethnicities. When using AI, actively question such correlations and consider the broader context of the patient's condition and history, rather than relying solely on AI-generated stereotypes. 4. Data (Socioeconomic) Bias Training data for AI often mirrors the existing biases in healthcare data collection, which can disproportionately affect marginalized communities. For example, AI models have been found less likely to recommend necessary medications to patients described as homeless. When AI provides a recommendation, consider whether the decision is representative of what is best for that patient, and adjust your queries or follow-up questions accordingly. 5. Automation Bias Reliance on AI can also lead to automation bias, where users trust AI decisions without sufficient scrutiny. Even if AI provides correct solutions most of the time, the occasional critical error can have dire consequences, particularly in healthcare settings. Always maintain a critical eye towards AI recommendations and cross-check with established medical guidelines and your professional judgment. ## Strategies for End Users to Mitigate AI Bias • Precision in Language: Use specific, clinically appropriate terminology in your queries to guide AI towards more accurate and relevant responses. • Critical Evaluation: Always review AI-generated information critically. Look for signs of generalization or stereotyping that might not apply to a specific patient or situation. • Cross-Verification: Use AI as a tool to supplement, not replace, your expertise. Cross-verify AI suggestions with other medical resources or consult with colleagues to ensure comprehensive patient care. • Patient-Centric Queries: Adapt your questions to reflect individual patient circumstances, particularly when dealing with AI's potential socioeconomic biases. Make sure your AI queries consider the unique context of each patient. • Awareness of Limitations: Educate yourself and remain aware of the potential biases and limitations of the AI tools at your disposal. This knowledge will allow you to better navigate and utilize AI in your daily practice. ## Conclusion While the vast majority of healthcare professionals may not control the foundational aspects of AI models, understanding and mitigating bias at the point of use is crucial. By employing careful language, critically assessing outputs, and maintaining an informed and vigilant approach, healthcare professionals can enhance the utility of AI , making it a safer and more effective tool in their practice. Together, these strategies empower users to counteract biases and leverage AI to improve healthcare outcomes responsibly. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/clinical-ai-scribes-need-true-speaker-recognition Healthcare Innovation # Why Clinical AI Scribes Need True Speaker Recognition Written By: Robert Henehan • December 6, 2025 Every word in clinical documentation matters, and so does who said it. In any patient encounter, who said something is just as important as what was said. Did the patient report the symptom, or did the caregiver? Did the physician recommend the medication, or did the patient ask for it? Many generic ambient AI scribes operate by recording audio and converting it into a single, continuous block of text. They then rely on the Large Language Model (LLM) to infer who is speaking based solely on the context of the words. If a voice says "I have a headache," the AI assumes it is the patient. If a voice asks "How long have you had it?" the AI assumes it is the provider. This contextual guessing creates potential clinical risk and degrades the ability for AI to craft high quality clinical documentation. In the era of Google and ChatGPT, patients frequently use sophisticated medical terminology. If a patient recounts a previous specialist's recommendation by saying "We decided to increase the dosage to 50 milligrams," a generic AI scribe might misattribute this statement to the current physician. This converts a piece of medical history into an incorrect active treatment plan. Without distinct speaker identification, this historical report can become a documentation error that could affect treatment decisions, insurance claims, or legal proceedings. Correcting these misattributions requires the clinician to scour the entire large block of unattributed text. This turns what should be a quick review into a detailed editing session. If an error is missed, it risks becoming part of the legal medical record. ## How It Works: Acoustic Fingerprinting and Diarization To understand the BastionGPT difference, it helps to look at the technology beneath the surface. Most automated note generation platforms treat the transcript as a flat document. They feed this text into the AI model and ask it to add labels like "Doctor" and "Patient" after the fact. This method often struggles when the conversation becomes complex, rapid-fire, involves multiple speakers, or has interruptions. ### The BastionGPT Approach BastionGPT's AI Scribe is designed to avoid relying on the LLM to guess the speaker based on text alone. Instead, the platform utilizes specialized speaker identification technology driven by these core components: ### Acoustic Fingerprinting & Speaker Embeddings The system creates a unique digital ID, or speaker embedding, for each participant’s voice. By analyzing the distinct acoustic fingerprint of each speaker, BastionGPT can track "Speaker 1" versus "Speaker 2" throughout the session. . ### Voice Activity Detection (VAD) Before identifying who is speaking, the system must identify that someone is speaking. Advanced VAD technology filters out non-speech audio such as the hum of HVAC systems, typing sounds, or background noise. This is especially valuable for in-person sessions where ambient recording picks up environmental sounds that would otherwise confuse speaker detection. ### Speaker Diarization This is the processing engine that combines VAD and acoustic fingerprints to map "who spoke when." Diarization partitions the audio stream into distinct segments so that the transcript accurately reflects the back-and-forth flow of conversation even when speakers interrupt one another or speak in short bursts. ### Ambient Optimization The identification tool is optimized for ambient recordings where devices may be stationary while speakers move around the room. This helps maintain accuracy even when audio volume varies between the clinician near the microphone and a patient across the room. ### Speed vs. Accuracy Quality takes time. Some generic AI scribes prioritize instant output. They stream text to the screen as fast as possible by skipping complex audio processing. BastionGPT prioritizes clinical accuracy. Our system takes the time to map speaker embeddings and filter background noise based on clues from the full length of the session. This processing may add time to the generation process compared to basic tools, but it results in a safer, more reliable medical transcript. The transcripts can take between 5 and 60 minutes to be generated but you can process multiple sessions at once without requiring the prior file to complete before you start a new session. ## Clinical Use Case: Solving the "Cocktail Party Problem" In audiology, the difficulty of focusing on a single voice in a noisy or crowded room is known as the "Cocktail Party Problem." In healthcare, this manifests in complex clinical environments where multiple voices contribute to the care plan. BastionGPT’s diarization engine is built to handle these multi-source dialogues effectively. ### Family and Couples Therapy (Triad Encounters) Appointments in pediatrics, geriatrics, and couples therapy rarely happen in isolation. They involve third parties such as parents, caregivers, or partners whose input must be distinguished from the patient’s. A generic AI scribe might attribute a symptom description to the patient when it was actually the caregiver reporting it. BastionGPT distinguishes between the provider, the patient, and the collateral source, providing a solution for the documentation to reflect who provided the history. This supports higher accuracy in family therapy notes and geriatric assessments. ### Group Therapy Mental health workflows often require summarizing sessions with multiple participants, such as group therapy. In a group setting, a flat transcript is difficult to use for tracking individual progress because contextual guessing often fails when multiple patients discuss similar themes. BastionGPT supports multi-speaker recognition for up to ten distinct speakers. This allows facilitators to generate summaries that better attribute insights to specific individuals rather than the group as a whole. ### Clinical Rounds Complex hospital workflows involve interdisciplinary clinical rounds where multiple providers discuss a patient's care. Attributing input to the specific specialist (e.g., the pharmacist vs. the physician) is critical for the care plan to reflect the specific expertise of each team member. Voice recognition capabilities help create care plans that more accurately reflect the contributions of each team member, streamlining care coordination and supporting accountability. ### Psychosocial and Learning Assessments Lengthy evaluations in educational and clinical psychology often integrate data from multiple sources including students, teachers, and parents. Accurate speaker attribution is essential to allow observations to be assigned to the correct source, which is critical for school assessments and learning disability evaluations. Distinctly separating the clinician's observations from the parent’s developmental history, helps to prevent the diagnostic error of conflating sources (e.g., attributing a clinician's report of "sensory overwhelm" to the parent) making the generation of the legal assessment more accurate when reflecting on the source of every observation. ### Speech-Language Pathology (SLP) Sessions In therapy sessions where a clinician models a sound or phrase and the client repeats it, attribution is critical for measuring success. A standard transcript often merges the two voices, making it impossible to distinguish the prompt from the performance. You cannot measure accuracy if the record does not reflect who pronounced the word correctly. BastionGPT uses diarization to separate the therapist’s cue (e.g., "Say 'Rabbit'") from the client's attempt (e.g., "Wabbit"), which supports more accurate progress tracking within the Subjective and Objective data fields. ### Palliative Care and Family Goals of Care Sensitive end-of-life discussions often involve a provider, a social worker, a patient, and multiple distinct family members deciding on a course of action. Documenting consensus is vital; if a daughter agrees to hospice but a son objects, a generic note stating "Family agreed to hospice" is legally dangerous and clinically inaccurate. BastionGPT helps distinguish between specific family members, allowing the note to reflect disparate opinions (e.g., Speaker 3 expressed concern about pain, while Speaker 4 asked about feeding tubes), helping the medical record reflect the true complexity of the decision. ### Independent Medical Exams (IMEs) and Workers' Comp Independent Medical Exams (IMEs) and workers’ compensation evaluations depend on clear, structured documentation. Providers need an easy way to capture what was asked, what was answered, and how to translate that into an accurate report. BastionGPT helps by generating organized notes that help distinguish the examiner’s questions from the claimant’s responses. This reduces ambiguity during report writing and makes it easier to draft summaries, narratives, and findings. ## The Efficiency Argument: Editing vs. Reviewing The primary goal of an AI scribe is to save time. However, the hidden cost of low-quality AI is the time spent editing. Clinicians must always review the entire generated note to ensure accuracy. However, when a provider receives a note where speaker attribution is scrambled, the cognitive load of reviewing that note increases. The clinician must remember the conversation to verify who said what. This effectively re-does the work the AI was intended to handle. If the AI attributes the patient's refusal of medications to the doctor, the clinician has to delete and rewrite that section to avoid a documentation error. BastionGPT’s approach minimizes these issues. When speaker labels are more accurate from the start, the review process becomes faster and less prone to error. You are simply verifying clinical nuance and not having to heavily edit a low quality AI guess at who said what. This difference reduces the risk that a misattributed statement slips into the permanent record. Furthermore, clean data inputs lead to cleaner data outputs. Whether you need a SOAP note, clinical assessment, or to capture the details of a group therapy session, more accurate speaker embeddings help ensure that the "Subjective" section actually contains the patient's report, and the "Plan" section contains the provider's instructions. This structural integrity is valuable for automated coding and billing compliance. ## What This Means for Clinicians Accurate speaker attribution transforms the review process. Instead of scanning an entire transcript for misattributed statements, clinicians can have more trust that the doctor's recommendations appear under the doctor's label and patient-reported symptoms stay with the patient. This reduces review time and lowers the risk of documentation errors reaching the legal medical record. For a closer look at how this approach stacks up against a leading ambient dictation platform, see how BastionGPT compares with DAX Copilot . ## Conclusion: Beyond Convenience In the world of healthcare AI, accuracy is a priority. As models continue to advance, the ability to generate text will become a commodity. The true differentiator for clinical tools will be the ability to structure and attribute that data correctly within a HIPAA compliant AI framework. BastionGPT provides access to the top models, including the latest GPT, Claude, and Gemini models. Our investment in true multi-speaker recognition and VAD technology reflects a commitment to clinical integrity using these powerful tools. By addressing the "wall of text" problem, BastionGPT provides a solution designed to be robust enough for the complexities of real-world medicine. Whether managing a busy school assessment, a group therapy session, or a complex geriatric consult, providers need an AI partner that is built to understand who is speaking. BastionGPT's approach to speaker recognition reflects a broader design principle: clinical AI should reduce work, not create more of it. By handling speaker identification via voice prints rather than asking an AI model to guess, the platform delivers transcripts that clinicians can review efficiently and have more trust in the output of the content they generate. If you are interested, most organizations are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Frequently Asked Questions (FAQ) ### What is the difference between an AI scribe and standard transcription? Standard transcription converts speech to text word-for-word. An AI scribe goes further by analyzing that text to generate drafts of structured clinical documentation, such as SOAP notes or progress reports, and extracting key medical data. BastionGPT combines both, offering more accurate transcription with advanced automated note generation. ### Is ChatGPT HIPAA compliant for creating patient notes? The standard version of ChatGPT is not HIPAA compliant because your data is often used to train the AI models and it does not include a Business Associate Agreement (BAA). BastionGPT provides a HIPAA compliant ChatGPT solution by accessing these models through a secure, private environment where data is never used for training, data is never given to OpenAI, and a BAA is included with every subscription. Click here to learn more. ### Does BastionGPT support multi-speaker recognition? Yes. Unlike many generic AI scribes that rely on context to guess speakers, BastionGPT utilizes specialized speaker identification (diarization) technology designed to distinguish up to ten speakers. This supports accurate attribution for complex encounters like family therapy or clinical rounds. ### Can BastionGPT generate different note formats like SOAP, DAP, or BIRP? Yes. BastionGPT is highly flexible and can be prompted to generate notes in any preferred clinical format, including SOAP, DAP, BIRP, and allows for custom prompting to create documentation from your transcriptions. ### Which AI models does BastionGPT use? BastionGPT integrates the most advanced and capable AI models available, including the latest OpenAI GPT, Anthropic Claude, and Google Gemini models. The platform automatically routes tasks to the best model for the job or allows users to select their preferred model within a secure, HIPAA compliant infrastructure. ### How does BastionGPT handle data privacy? BastionGPT is designed exclusively for healthcare professionals, so all plans include data security features that meet or exceed healthcare compliance regulations like HIPAA. Your data is never shared with third-party AI providers (like OpenAI or Google) for training or marketing purposes. BastionGPT automatically includes a BAA for all of our customers to support full compliance with HIPAA, PIPEDA, and APP regulations. ### Why do Zoom or Teams transcripts know who is speaking, but AI scribes often struggle with this? Platforms like Zoom, Google Meet, and Microsoft Teams can label speakers accurately because each participant is connected through a distinct device and user account. The system knows which audio stream belongs to which person before transcription even begins. In clinical settings, however, participants commonly speak into the same microphone in the same room. This makes speaker identification much harder and forces AI scribes to rely on audio patterns instead of device-based identity. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/comparison-of-word-limits-in-generative-ai Healthcare Innovation # Comparison of Word Limits in Generative AI Written By: Josh Spencer • October 30, 2023 Medical professionals spend hours each day drafting creating clinical documentation. With BastionGPT , entire meeting transcripts and large patient histories can be processed in seconds and converted into drafts of various documents, such as SOAP notes, treatment plans, summaries for patients, reports to insurance, and many more. Try our BastionGPT today to discover how hundreds of medical professionals are shaving hours out of their day by leveraging the power of BastionGPT. For a detailed look at how many pages BastionGPT can process in a single chat, see BastionGPT's document limits explained . Understanding the limitations of AI assistants and AI scribes can be confusing unless you are an AI expert. It requires understanding AI terms like “context lengths” and “tokens”. To help healthcare professionals understand different AI limitations, without taking a course on artificial intelligence, we have made some generalizations and conversions of data. ## What are Tokens in Simple Terms? Different AI models are able to handle different amounts of data by using units called tokens. A "token" usually refers to a chunk of text that the AI reads as one unit. Think of it as a piece of a puzzle. The puzzle, in this case, is a sentence or a paragraph that you want the AI to understand and process. This chunk could be as small as a single letter or as long as a word. For example, in the sentence "Dr. Smith treats patients," each word and punctuation mark counts as a token: "Dr", ".", "Smith", "treats", "patients". ## Real-World Token and Word Data for Healthcare We studied real-world healthcare chats to give you a sense of how tokens and words typically measure up in medical settings. Here are the averages we found: 5.5 characters per word 3.6 characters per token 1.5 tokens per word This data was compiled on September 20, 2023. Please note that these figures could change, and we'll do our best to update this information if that happens. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/consent-documents-healthcare-ai Healthcare Innovation # AI Scribe and Healthcare AI Consent Forms: Free Templates Written By: Robert Henehan • August 16, 2026 Last reviewed August 16, 2026. This page is general information, not legal advice; confirm your own situation with counsel or your compliance lead. If you use BastionGPT to draft notes, record and transcribe visits, or prepare letters and summaries, you may need to tell patients about it, and in a growing number of jurisdictions you need their written consent before AI records a session. This page gives you free consent form templates for the United States, Canada and Australia in mental health and medical versions, a plain-language summary of when consent or disclosure is required, and a workflow for documenting it. ## When patient consent is required for AI: the short answer It depends on where you practise and on how you use AI, and the rules tightened in 2025 and 2026: United States. HIPAA does not require patient consent to use an AI vendor that operates under a Business Associate Agreement for treatment or operations, so drafting and summarizing with BastionGPT sits in the same category as your EHR or a transcription service. Two other layers do apply: recording-consent statutes in the all-party-consent states, which are triggered the moment an AI scribe records a conversation, and a new set of state laws (listed below) that require written, revocable consent before AI records or transcribes a therapy session, or require you to disclose AI use to patients. Canada. PIPEDA and provincial health-privacy statutes such as Ontario’s PHIPA rest on knowledgeable consent, and regulators have said an AI scribe needs it. The CMPA advises obtaining consent before recording any clinical encounter and documenting the discussion, the CPSO’s 2025 advice says physicians must obtain consent before recording conversations with AI, and Ontario’s Information and Privacy Commissioner (January 2026) calls express consent recorded in the chart best practice. Australia. Ahpra’s guidance expects practitioners to tell patients when AI is used and to obtain informed consent before patient information goes into an AI tool or a consultation is recorded, and state and territory surveillance-devices laws can make recording a private conversation without consent an offence. Medical defence organisations recommend a brief verbal confirmation at each consultation, noted in the record. The practical rule: if you only type or dictate your own notes into BastionGPT, notice (a sentence in your privacy notice or intake paperwork) is usually enough outside the disclosure states. If BastionGPT will record or transcribe the visit, get express consent first, in writing where your state or regulator requires it, and document it in the record. ## Download the consent form templates (DOCX and PDF) Each one is two pages, written in plain language for patients, and covers: what BastionGPT does and does not do; audio recording and transcription as a separate tick-box choice; how information is protected (encryption, no model training, in-country hosting, clinician review of every draft); benefits and limits; the right to decline or withdraw without affecting care; the access, correction and complaint routes for the region; a signature block with a parent, guardian or substitute decision-maker line and a verbal-consent option; and a shaded box of practice notes (which state, provincial and regulator rules apply) to delete before use. Edit the DOCX; use the PDF to preview or print. ### Mental health consent forms United States: editable DOCX · PDF preview Canada: editable DOCX · PDF preview Australia: editable DOCX · PDF preview Generic (any jurisdiction): editable DOCX · PDF preview ### Medical consent forms United States: editable DOCX · PDF preview Canada: editable DOCX · PDF preview Australia: editable DOCX · PDF preview Generic (any jurisdiction): editable DOCX · PDF preview Free to use and adapt, no signup. Fill in every bracketed item, delete the uses you do not offer, and have your counsel or compliance lead review the final version. The mental health versions say “client” and “session”; the medical versions say “patient” and “visit”. ## US state laws that require consent or disclosure for AI documentation These are the laws that bind a clinician who uses AI to record, transcribe or draft documentation, current to August 16, 2026. Our 50-state guide to healthcare AI laws covers the whole landscape; the entries below are the ones that change a consent workflow. Colorado (HB 26-1195, effective August 12, 2026). Psychotherapy licensees need clear, written, revocable consent before AI records or transcribes a session, must review AI-generated recommendations, and may not deny care to a client who refuses. Illinois (WOPR Act, effective August 1, 2025). AI may record or transcribe a therapy session only after the client is informed in writing and gives explicit, revocable written consent. Maine (LD 2082, effective July 29, 2026). Written notice and written, revocable consent before AI is used in a recorded or transcribed therapy session; AI is limited to administrative and supplementary support under the licensee’s responsibility. Rhode Island (two laws signed June 22, 2026). Any provider who uses AI to document visits must notify patients and review the AI documentation for accuracy (R.I. Gen. Laws ch. 23-106); mental health providers need written consent before AI is used in recorded or transcribed sessions (H 7349/S 2197). Louisiana (Act 649, effective August 1, 2026). Tell patients before the appointment or treatment begins if AI recording or transcription software will be used. Texas (SB 1188, effective September 1, 2025, and TRAIGA, effective January 1, 2026). Practitioners must disclose AI use to patients and review AI-created records; TRAIGA adds a general duty to disclose AI use in health care services no later than the first date of service. Utah (SB 226, effective May 7, 2025). Anyone in a licensed occupation must prominently disclose, verbally at the start of a spoken interaction or in writing before a written one, when a patient is interacting with generative AI in a high-risk interaction, which includes collecting health data or giving medical or mental health advice. Nevada (AB 406, effective July 1, 2025). Mental and behavioral health providers may use AI for notes and billing only if they independently review the AI output; AI may not deliver care directly. Arizona (Board of Behavioral Health Examiners rule, effective January 1, 2027). Documented informed consent before any service that uses AI to provide, record or document clinical services. California (AB 3030, effective January 1, 2025). AI-generated patient communications about clinical information need a disclaimer unless a licensed provider reviews them. California is also an all-party recording-consent state, and class actions filed in 2026 against health systems allege that ambient AI scribes recorded patients without meaningful notice or consent under the state’s privacy and medical-confidentiality laws. Recording consent applies everywhere. Independently of AI-specific law, an AI scribe that records audio is subject to state recording statutes. All-party consent (or notice) states include California, Florida, Illinois, Maryland, Massachusetts, Montana, New Hampshire, Oregon, Pennsylvania and Washington, with mixed rules in a few others. The safe habit in every state is to say what the scribe does before it starts and to record the patient’s answer in the chart. ## Canada and Australia: what regulators expect ### Canada PIPEDA and provincial health-privacy statutes are built on knowledgeable consent, and there is no provision that lets an AI scribe collect, use or disclose personal health information without it. Ontario’s IPC ( AI Scribes: Key Considerations for the Health Sector , January 2026) says express consent recorded in the health record is best practice, that consent must be knowledgeable (the patient knows a scribe is in use and why, and can withhold consent), that patients who decline must receive the same care, and that companions in the room are being recorded too. The CMPA’s AI Scribes FAQ (revised December 2025) advises obtaining consent before any recording, explaining the purpose and the privacy and accuracy risks, and documenting the discussion; the CPSO’s 2025 Advice to the Profession on AI says to obtain consent before recording conversations and to review every AI-generated note. Federal reform is pending: Bill C-36 (introduced June 15, 2026) would replace PIPEDA and is scheduled for second reading in the fall. Our Canada templates reflect all of this and leave a marked space for your provincial law. See also our guide to PIPEDA compliance for Canadian users . ### Australia The Privacy Act 1988 and the Australian Privacy Principles govern the collection of health information, and Ahpra’s guidance ( Meeting your professional obligations when using Artificial Intelligence in healthcare ) expects practitioners to understand the tool, tell patients when AI is used, obtain informed consent before patient data is entered or a consultation recorded, and apply their own judgement to every output. State and territory surveillance-devices laws are why recording consent matters most: in several jurisdictions recording a private conversation without consent is an offence. The practical routine recommended by medical defence organisations is a short explanation at the first visit, a verbal confirmation at each consultation, and a note in the record; if the patient declines, the consultation proceeds without the scribe. Our Australia templates include the OAIC’s contact details and a space for your state or territory health-records act. See also our guide to the Australian Privacy Principles . ## Two ways to handle consent: general notice or a product-specific form Practices take one of two approaches, and many use both. 1. Notice in your privacy documents. Where the law requires only disclosure, or nothing at all for typed and dictated use, a sentence in your Notice of Privacy Practices, privacy policy or intake pack that names AI-assisted documentation and your vendors’ obligations may be enough. Sample wording you can adapt: United States: “Business associates and AI-assisted documentation. We use contracted service providers, including secure AI documentation software, to help prepare clinical notes and correspondence. These business associates are required by federal law (HIPAA) to safeguard your health information, may use it only to perform the work we have asked them to do, and do not use it to train AI models.” Canada: “Third-party service providers. We use contracted service providers, including secure AI documentation software hosted in Canada, to help prepare clinical notes and correspondence. These providers are bound by written agreements and by law to protect your personal health information and to use it only for the purposes for which we engaged them.” Australia: “Contracted service providers. We engage contracted service providers, including secure AI documentation software hosted in Australia, to help prepare clinical notes and correspondence. These providers must protect your personal information under the Privacy Act 1988 and the Australian Privacy Principles and may use it only to carry out the services they are contracted to provide.” 2. A signed, product-specific consent form. Where AI will record or transcribe, where your state or regulator requires written consent, or where you simply want the conversation on the record, use the templates above. They name BastionGPT, separate recording from drafting, and give patients a clear way to say no. ## How to obtain and document consent: a workflow that holds up Before the first use. Give or read the form (or the notice) before BastionGPT touches the patient’s information. Texas requires disclosure no later than the first date of service; Louisiana requires it before the appointment when recording is involved. Say it out loud, briefly. A one-sentence script is enough: “I use a secure AI tool called BastionGPT to help me write up your visit. If you agree, it will record our conversation, produce a draft note that I review and correct, and delete the recording after 30 days. You can say no or ask me to stop at any time and it will not change your care.” Record the answer. Scan the signed form to the chart, or add a dated line noting verbal consent and who obtained it. Utah requires the disclosure to be prominent; Rhode Island, Nevada and Texas require you to review the AI note, so note that step in your workflow too. Handle the room. If a family member, carer or interpreter is present, tell them the visit is being recorded. Offer the alternative. If the patient declines, document by hand or by dictation without recording, and note the refusal. Re-consent when something changes. A new use (for example you start recording), a minor turning 18, a change of substitute decision-maker, or a yearly review at minimum. In Australia, confirm verbally at each consultation. ## What the templates say about BastionGPT Patients tend to ask the same four things, and the forms answer them in plain language: A Business Associate Agreement is included on every BastionGPT plan; Canadian and Australian customers are covered by written data-protection terms that name PIPEDA, PHIPA and the Privacy Act. Data is encrypted in transit and at rest and hosted in the United States, Canada or Australia to match the practice’s billing region. Patient information is never used to train AI models, is never sold, and is never used for advertising or marketing. Recordings and transcripts follow a retention window the practice sets (30 days by default), and the clinician reviews and approves every draft before it becomes part of the record. The detail behind those statements is on our security and trust page and in our guide to what makes an AI HIPAA compliant . ## Customizing the forms The templates are a starting point. Add your practice name and contacts, keep only the uses you offer, set the retention period you have chosen in BastionGPT, translate the form where your patients need it, and have counsel or your compliance lead review the result against your state, provincial or territorial rules. If you would like us to look at your organization’s existing consent forms or contracts and suggest updates, contact us ; we offer that as an hourly engagement. Quick questions are always free, and our support article on consent form templates stays current with the templates on this page. ## Related templates and guides Informed consent for treatment template Telehealth consent form template Consent for treatment of a minor Notice of Privacy Practices acknowledgment 42 CFR Part 2 consent form Psychotherapy notes authorization State laws on healthcare AI (50-state guide) BastionGPT AI medical scribe ## Frequently asked questions ### Do I need patient consent to use an AI scribe? In most cases, yes, and it should be obtained before recording starts. In the United States, recording-consent laws in the all-party-consent states and AI-specific laws in Colorado, Illinois, Maine, Rhode Island, Louisiana, Texas and Utah require consent or disclosure; in Canada, the CMPA, the CPSO and Ontario’s IPC expect express consent documented in the record; in Australia, Ahpra expects informed consent before a consultation is recorded. Even where no statute applies, consent is the standard of practice for ambient recording. ### Does HIPAA require patient consent to use AI for documentation? No. HIPAA permits a covered entity to use a business associate for treatment and health care operations without patient authorization, and BastionGPT signs a Business Associate Agreement on every plan. Consent requirements come from state recording statutes, state AI laws, professional ethics codes and, outside the US, from privacy laws such as PIPEDA and the Australian Privacy Act. ### Which US states require written consent or disclosure for AI documentation? As of August 2026: Colorado, Illinois, Maine and Rhode Island require written, revocable consent before AI records or transcribes a psychotherapy session; Rhode Island also requires patient notification whenever AI documents a visit; Texas and Louisiana require disclosure of AI use, and Utah requires prominent disclosure when a patient interacts with generative AI; Nevada, Rhode Island and Texas require the clinician to review AI-generated notes; Arizona behavioral-health licensees must document informed consent from January 1, 2027. Recording-consent statutes apply on top of these in the all-party-consent states. ### Is verbal consent enough, or do I need a signed form? Verbal consent noted in the chart is acceptable in many settings and is the routine recommended by Australian and Canadian medical defence organisations, but written consent is required by law in Colorado, Illinois, Maine and Rhode Island for AI recording or transcription of therapy sessions, and a signed form is the easiest way to prove consent later. Our templates support both: a signature block and a verbal-consent line. ### Do I need consent if I only type or dictate my own notes into BastionGPT? Usually not as a separate signed form. Typed and dictated use is processing by a contracted vendor under a BAA or equivalent data-protection terms, which health-privacy laws generally permit with appropriate notice and safeguards. Disclosure states such as Texas and Louisiana, Utah’s rule for patient-facing generative AI, and any organizational or payer contract that requires you to name processors still apply, so a sentence in your privacy notice is the minimum. ### What should an AI consent form include? What the tool is and what it will do; whether the visit is recorded and transcribed, as a separate choice; how information is protected (encryption, no model training, hosting location, clinician review); benefits and limits; the right to decline or withdraw without affecting care; access, correction and complaint routes; retention of recordings; and a signature block that covers guardians or substitute decision-makers and verbal consent. ### What happens if a patient declines or withdraws consent? You document the visit another way, by hand or by dictation without recording, and note the refusal or withdrawal in the record. Care must not change; Colorado’s law and Ontario’s IPC guidance say so expressly, and every template on this page tells the patient the same thing. ### Do I need to get consent at every visit? A signed form covers ongoing use until it is withdrawn or your use changes, but you should confirm briefly whenever recording starts and re-consent when you add a new use, when a minor becomes an adult, when a substitute decision-maker changes, and at least yearly. In Australia, medical defence organisations recommend a verbal confirmation at each consultation. ### Is patient information used to train BastionGPT’s AI models? No. BastionGPT never uses customer inputs, documents, recordings or transcripts to train AI models, and does not sell or use them for marketing. Data is encrypted in transit and at rest and hosted in the United States, Canada or Australia to match the practice’s billing region. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/dental-ai-use-cases Healthcare Innovation # AI For Dental Practices: 29 Prompts Dentists Can Use Written By: Robert Henehan • August 4, 2025 Dentists and their teams spend a significant portion of every workday on work that has nothing to do with patient care. The administrative bottleneck in dentistry isn't just inconvenient, it has direct business consequences. Notes that take longer to write mean fewer patients seen. Insurance letters that pile up mean slower reimbursement. Front desk staff drafting the same emails manually means less time on scheduling and patient experience. As of 2026, 35% of dental practices have already implemented AI and the gap between early adopters and those still doing everything manually is widening. When AI handles the drafting burden, the effects compound quickly: Dentists who spend less time on after-hours charting see more patients without extending hours. Billing teams that generate claim narratives in seconds rather than minutes submit cleaner claims faster and win more appeals. Front desk coordinators who use AI for routine communications have more capacity for the high-value interactions that improve patient retention. Whether you're a dentist dictating clinical notes, a hygienist summarizing patient histories, a biller drafting an appeal letter, or a practice manager writing staff communications, BastionGPT works across every role in the practice. BastionGPT is a HIPAA-compliant AI assistant designed for dental practices and healthcare, which means it operates under a Business Associate Agreement and keeps patient data protected by design unlike general-purpose tools like ChatGPT. Read on for 30 examples of how to use AI in dentistry in clinical documentation, patient communication, billing, and practice management, each with a sample prompt your team can use right away using BastionGPT. ## How Dental Practices Use AI With These 29 Prompts ### Clinical Documentation & Treatment Notes Keeping clear, detailed treatment notes is vital and yet writing them can eat up hours. This section highlights how BastionGPT streamlines clinical documentation, acting like a dental charting AI that transforms conversations and exam findings into well-structured notes. Dentists can record an exam and have BastionGPT produce a formatted note or summary, significantly reducing after-hours charting. Whether it’s a routine check-up or complex procedure, the AI assists in identifying potential omissions and inconsistencies , and even suggests improvements or additions, helping clinicians to create comprehensive notes. From periodontal exams to orthodontic visits, BastionGPT helps clinicians maintain thorough documentation with minimal effort. #### 1. Transcribe & Draft Notes Use BastionGPT’s medical scribe capabilities to convert voice dictations or rough notes into polished clinical documentation. For example, after an exam, the AI can generate a complete SOAP note (Subjective, Objective, Assessment, Plan), capturing chief complaint, findings (e.g. “caries on #14 D”), treatment provided, and follow-up steps. This saves dentists from typing lengthy narratives and ensures consistency in format. You can simply click the SOAP note tab to summarize or use the following prompt to convert your notes or other documentation. Ex. You are a dentist completing a clinical note. Review the exam findings and generate a SOAP note including the chief complaint, observed conditions, diagnosis, treatment performed, and plan for follow-up. #### 2. Perio Chart Summaries Periodontal exams generate extensive data. BastionGPT can digest full perio chart data (probing depths, bleeding points, recession) and generate an easy-to-read perio chart summary. It highlights areas of concern – e.g. “Generalized 4-5mm pockets with bleeding on probing in posterior sextants, suggestive of moderate periodontitis – recommend SRP.” This summary can be included in the patient’s record or used to explain periodontal status to the patient, eliminating manual transcription. Ex. You are a dental hygienist documenting a periodontal exam. Analyze the probing depths, bleeding, and mobility to create a concise clinical summary with treatment recommendations. #### 3. Aligner Progress Notes For orthodontic cases, BastionGPT helps track and document progress at each aligner check-in. Simply provide details like current aligner number, any tooth movement observations, and patient compliance notes. The AI assistant drafts a clear aligner progress note e.g. “Week 12: Patient has advanced to aligner 5 of 30. Mild spacing remains between upper incisors; attachment on #7 intact. Patient reports wearing aligners ~22 hrs/day. Continue as planned; next check at aligner 8.” This helps each orthodontic visit be thoroughly documented, aiding continuity of care. Ex. You are an orthodontic assistant documenting an aligner check-in. Draft a progress note detailing the current aligner number, observed tooth movement, and patient compliance, including the next check-in details. #### 4. Complete Procedure Narratives When a procedure is finished, BastionGPT can generate the detailed narrative for the dentist to review. For example, after a complex extraction or implant placement, the dentist can provide key details via the AI Scribe or a recording (tooth, anesthesia, complications, materials used) and have the AI draft the full operative note. This can include pre-op and post-op conditions, steps taken (“#32 extracted in fragments, irrigation and suture placed…”), and post-op instructions given. These AI-drafted narratives help chart entries be comprehensive and properly formatted for future reference or audits. Ex. You are a dentist finishing a complex procedure. Draft a detailed operative note including pre-op and post-op conditions, steps taken, materials used, and post-op instructions given. ### Front Desk & Patient Coordination Workflows Efficient patient coordination keeps a practice thriving. Front-desk staff handle appointment scheduling, reminders, intake forms, and more; tasks BastionGPT assists with through natural language generation. By using well-crafted dental AI prompts, even non-clinical staff can leverage the AI for polished communications and streamlined workflows. Imagine significantly reducing the time spent drafting emails or making reminder calls: BastionGPT can produce friendly, professional messages in seconds. From managing recalls to coordinating referrals, the AI supports having patients stay informed and engaged, all while helping you to maintain a consistent, courteous tone that reflects your practice’s brand. #### 5. Personalized Appointment Reminders Draft reminder emails or texts for upcoming appointments. BastionGPT can integrate patient name, date/time, and any prep notes (e.g., “arrive 10 minutes early, remember to bring insurance card”), outputting clear, warm reminder messages. This prevents no-shows and enhances patient experience, saving considerable staff time over manual drafting. Ex. You are a front desk coordinator preparing an appointment reminder. Use the patient’s name, appointment time, and visit type to write a courteous and informative message. #### 6. Missed Appointment Follow-Ups When a patient cancels or no-shows, the AI assistant generates a polite follow-up offering to reschedule. For example: “We’re sorry we missed you today. Please let us know if you’d like to rebook your dental visit and we’ll be happy to accommodate a new date. Your oral health is important to us!” Such timely communications keep patients on track with care and demonstrate responsiveness. Ex. You are a scheduler following up on a missed appointment. Draft a message offering to reschedule and reinforcing the importance of care continuity. #### 7. Streamlined Patient Intake Forms Create and manage intake paperwork with AI assistance. BastionGPT can help draft comprehensive patient intake forms, or condense a completed form into a summary for the dental team. For instance, after a new patient fills out their history, staff can have the AI summarize key points (allergies, chief complaint, insurance info) in a single paragraph. This gives clinicians a quick overview at a glance, and can provide visibility for important details from lengthy forms that may be overlooked. Ex. You are a receptionist reviewing a completed intake form. Extract allergies, medications, and concerns to generate a summary for clinical staff to review before the appointment. #### 8. Quick Email Responses & FAQs Front desk teams frequently answer recurring questions (e.g., “Do you accept XYZ insurance?”, “What are your hours?”, “How to handle a dental emergency?”). The AI generates clear, adaptable responses that staff can personalize. It can also translate messages for non-English-speaking patients on the fly. By drafting these emails or website FAQ content, BastionGPT helps patients get accurate information quickly, freeing up staff for in-person interactions. Ex. You are a front desk team member responding to a common patient inquiry. Draft a clear and professional email response addressing the frequently asked question, and write the message at a 3rd-grade reading level for non-native English speakers. #### 9. Referral & Specialist Coordination When sending a patient to a specialist, the front desk can use AI to draft referral cover letters or transfer documents. Provide BastionGPT basic case information (patient name, reason for referral, pertinent history, attached x-ray reports) and it will produce a professional letter to the specialist. This letter can outline the case (e.g., “Dr. Jones, we are referring Mr. Smith for evaluation of a possible endodontic retreatment on #15…”) and include contact details for follow-up. This simplifies providing continuity of care between providers. Ex. You are a front desk coordinator preparing a specialist referral. Summarize the patient's contact info, treatment needs, and attach supporting documentation in a formatted handoff note. ### Billing, Claims & Insurance Dental billing and insurance claims are notoriously detail-heavy, where small errors or missing information can lead to denied claims or delayed payments. BastionGPT acts as a dental billing automation assistant by evaluating forms and narratives for completeness and suggesting appropriate coding. Instead of staff laboring over claim narratives or cross-referencing coding manuals, the AI for dental insurance can generate CDT narratives and even suggest relevant codes based on chart notes. It can also interpret dense insurance policy documents for you, or draft appeal letters when needed. By automating these tedious steps, BastionGPT speeds up reimbursements and significantly reduces administrative overhead, all while maintaining accuracy and compliance. #### 10. Automated Claim Narratives Instantly generate clear, thorough narratives to accompany insurance claims. For procedures that require justification (like a crown for a cracked tooth or periodontal scaling), BastionGPT can draft the explanation including clinical findings and why the procedure was necessary. It helps CDT narratives include all relevant details such as patient symptoms, diagnostics (x-rays or findings), treatment done, and prognosis and they can be phrased in insurer-friendly language. These AI-crafted narratives align with payer requirements to help you make sure nothing vital is omitted. You are a billing specialist preparing an insurance claim. Generate a clear and thorough narrative justifying the necessity of the procedure, including clinical findings, diagnostics, and prognosis, in insurer-friendly language. #### 11. Accurate Coding & Documentation Reduce coding errors by having the AI double-check and suggest codes. For example, if a dentist writes a note about “deep distal caries on #18 treated with a composite filling,” BastionGPT might suggest CDT code D2391 (one-surface posterior composite) if not already noted. It can also help the narrative include necessary keywords for that code (like “resin restoration” and tooth surfaces). By cross-verifying documentation with coding, the AI supports preventing claim rejections due to documentation deficiencies. Ex. You are a billing assistant reviewing chart notes for coding accuracy. Suggest the appropriate CDT codes based on the documented clinical findings and ensure the narrative includes necessary keywords for compliance. #### 12. Insurance Benefits Summaries BastionGPT can parse through insurance plan details or benefits breakdowns and summarize them in plain language. If a patient provides a 10-page dental insurance policy, the AI can outline key points: deductible, annual maximum, covered percentages for preventive/basic/major, and any waiting periods or frequencies (e.g., “2 cleanings per year”). This saves your billing coordinator from deciphering fine print and provides quick answers when patients ask, “What will my insurance cover for this procedure?” Ex. You are a billing coordinator explaining a patient's insurance benefits. Summarize the key points from the patient's insurance policy, including deductibles, annual maximums, and covered percentages for different service types. #### 13. Denial Appeal Letters In cases of denied claims or preauthorization refusals, the AI can draft professional appeal letters. Provide the denial reason along with patient information, and BastionGPT will formulate a response citing relevant facts and attachments. For instance: “We are writing to appeal the denial of D4341 (periodontal scaling) for patient X. Enclosed are chart notes and x-ray evidence of bone loss indicating this treatment was medically necessary. Please reconsider coverage in light of this documentation.” Such letters, generated in seconds, help practices fight for rightful reimbursements without spending hours on writing. Ex. You are a billing manager appealing a denied insurance claim. Draft a professional appeal letter citing the denial reason, patient information, relevant facts, and evidence to reconsider coverage. #### 14. Patient Billing Explanations For patient-facing bills, BastionGPT can create easy-to-understand summaries. If a patient is confused about their invoice (insurance adjustments, co-pays, non-covered services), the AI can translate it into layman’s terms: “Your recent visit total was $200. Insurance paid $150. The remaining $50 is your co-pay for the filling on tooth #3 per your plan’s 80% coverage for basic services.” This clarity improves patient trust and reduces calls to clarify bills. Ex. You are a billing coordinator explaining a patient's invoice. Create an easy-to-understand summary of the patient's bill, clarifying insurance adjustments, co-pays, and non-covered services in layman’s terms. ### Diagnostic Imaging & Scan Summaries Dental practices generate extensive clinical imagery, from routine X-rays and intraoral photos to 3D scans for implants or aligners. Interpreting and documenting these findings is time-consuming. A prime frontier for generative AI in dentistry is assisting with image-based workflows. While BastionGPT isn’t a diagnostic device, and doesn’t process images directly, it can summarize radiologists’ reports to produce descriptive summaries for notes or treatment plans. This means quicker write-ups of X-ray findings and clearer patient explanations. For instance, instead of manually noting every visible restoration or issue on a full-mouth series, a dentist could get a first draft from the AI. #### 15. Radiograph Findings to Notes After taking radiographs (bitewings, PA, or panoramic), use BastionGPT to generate a summary of key findings for the record. For example, feed the AI assistant a list of observations or even the radiology report, and it can output: “Bitewing exam: Caries noted on #15D, #14M; existing restorations on #2, 3, 18; no periapical radiolucencies; bone levels appear normal in posterior regions.” This provides a standardized radiographic exam note without the dentist typing it line by line. It can also serve as a patient-friendly imaging explainer, translating dental jargon into plain language when needed. Ex. You are a dentist documenting radiographic findings. Summarize key observations from bitewing, PA, or panoramic X-ray reports to create a standardized note for the patient's record along with a separate patient-friendly explanation. #### 16. Patient Consultation Visual Descriptions As an educational tool, the AI helps create explanations for imaging findings that patients often struggle to understand. If a patient’s panoramic X-ray reveals an impacted wisdom tooth, BastionGPT can craft a short explanation: “Your lower wisdom tooth is trapped under the gum (impacted) and pushing against the next tooth. This can cause pain or infection. We’ll likely need to extract it to prevent issues.” By leveraging the AI Assistant, dentists better communicate imaging findings, improving patient trust and treatment compliance. Ex. You are a dentist explaining imaging findings to a patient. Draft a short, personalized, plain-language explanation targeted to a 6 th -grade reading level of the finding from this imaging report and related documentation to improve patient understanding and compliance. ### Lab & Prosthetic Workflow Behind every great dental restoration is often a great dental lab. Effective communication with dental labs (or in-house milling units) is crucial to get quality crowns, dentures, aligners, and more. BastionGPT streamlines the lab & prosthetic workflow by helping draft precise lab prescriptions, case communications, and even checking for completeness. Instead of writing out every lab slip by hand, dentists can let the AI take a first pass for them and provide a second set of eyes to avoid missing things like shade, material, or due date. It can also summarize lab reports or technical jargon when labs send back case notes. The result is a smoother hand-off between clinic and lab, fewer errors or remakes, and time saved on both ends. #### 17. Detailed Lab Prescriptions Generate complete lab orders for crowns, bridges, dentures, orthodontic appliances, etc. Dentists provide key parameters (patient info, tooth #, restoration type, shade, special instructions), and BastionGPT drafts a professional lab prescription. For example: “Please fabricate a Zirconia crown for tooth #19. Shade: A2. Margin: shoulder prep. Include interproximal contacts; ensure occlusion with opposing #14. Patient has a low smile line – prioritize lingual metal for strength if needed.” The AI’s draft covers essential points, acting as a checklist to prevent overlooking items on the lab form. Ex. You are a dentist preparing a lab order for a restoration. Draft a complete lab prescription for a crown, bridge, or denture, including patient info, tooth number, restoration type, shade, and any special instructions or requirements. #### 18. Lab Case Summaries When sending complex cases (like a full-mouth rehab or implant case) to the lab, dentists can include a cover letter summary generated by AI. It can outline the case background (“Patient had failing dentition, all teeth extracted; we’re fabricating an upper denture and lower implant overdenture…”) and what the lab needs to do for each item. Dentists can review the summary to provide lab technicians the “big picture” without sifting through lengthy clinical notes. It’s especially useful if multiple appliances or sequential steps are involved, providing the lab a clear treatment plan and expectations. Ex. You are a dentist sending a complex case to a dental lab. Generate a concise cover letter summary outlining the patient's background, the overall treatment plan, and specific lab instructions for each appliance or sequential step. #### 19. Quality Assurance Checks BastionGPT acts as a second set of eyes on your lab forms or lab communications. For instance, after drafting a lab order, you can ask the AI to verify if all critical info is included. It might flag, “No shade specified for the crown” or “Due date not indicated.” This helps catch omissions before the case is sent out. Similarly, if a lab returns a case for clarification, the AI can parse their message and highlight the questions you may need to answer (e.g. “Lab asks: confirm if bridge design is 3-unit vs 4-unit”). This reduces back-and-forth. Ex. You are a dental assistant reviewing a lab form. Verify that all critical information, such as shade and due date, is included, or parse a message from the lab to highlight questions needing your answer. ### Document Review, SOPs & Team Training Running a dental practice involves more than treating patients; it encompasses protocols, policies, and continuous team training. Here, BastionGPT shines as an administrative ally. The AI assistant drafts and polishes Standard Operating Procedures (SOPs), employee handbooks, or compliance policies tailored to your clinic. Have a 30-page OSHA guideline to implement? The AI summarizes it into actionable steps for your team. Need to train staff on new software or procedures? BastionGPT generates outlines, quizzes, or even role-play scenarios. By handling the heavy lifting of document creation and review, BastionGPT frees managers to focus on implementation, supporting consistency and a professional tone across all internal documents. #### 20. Drafting SOPs and Policies Create clear step-by-step SOPs for anything from sterilization protocols to front-desk phone scripts. Provide BastionGPT with a rough outline or key points, and it generates a structured document for you to review. For instance, for infection control: it can outline sections on hand hygiene, instrument sterilization (autoclave steps), surface disinfection, waste disposal, etc., complete with bullet points or numbering. Ex. You are a practice manager creating a new office policy. Generate a structured Standard Operating Procedure (SOP) for a specific protocol (e.g., sterilization) with step-by-step instructions, bullet points, or numbering. #### 21. Summarizing Guidelines & Manuals When new regulations or product manuals emerge, the AI condenses them for quick understanding. If the CDC updates its dental clinic infection guidelines, BastionGPT can summarize the update into a one-page “cheat sheet” for your staff, highlighting new or changed elements. Similarly, for a new X-ray unit, the AI extracts key operating instructions and safety warnings from a long user manual into a handy reference. This saves time and helps staff learn only what they need without wading through technical jargon. Ex. You are a practice manager reviewing new regulations or equipment manuals. Condense the information into a one-page "cheat sheet" for staff, highlighting key updates, operating instructions, or safety warnings. #### 22. Document Proofreading & Consistency Use BastionGPT to review existing documents for clarity and consistency. For example, if your practice manual has evolved over years, feeding sections through the AI helps identify duplicate or contradictory information and even corrects grammar. It might catch that one policy says “scrubs are navy blue” while another says “scrubs are black” and flag the discrepancy. By proofreading and suggesting edits, the AI acts as a virtual editor, giving your internal documents a professional polish. Ex. You are an office administrator reviewing practice documents. Proofread for clarity, consistency, and grammar, identifying duplicate or contradictory information within the text. #### 23. Onboarding & Job Aids Quickly generate onboarding packets or job aids with AI assistance for new team members. For instance, create a “Dental Assistant Quick-Start Guide” outlining daily tasks, important contact info, software login instructions, etc. BastionGPT helps the guide become more comprehensive and easy to understand, even tailoring the tone for a new hire. These standardized documents accelerate training and provide new staff with quick references for faster ramp-up. Ex. You are a practice manager onboarding a new team member. Create a comprehensive "Quick-Start Guide" or job aid detailing daily tasks, important contacts, software login instructions, and other essential information for a new hire. ### Compliance & Audit Readiness Compliance is non-negotiable in healthcare, and dentistry has its share of regulations, from HIPAA to OSHA to dental board standards. Failing an audit can be costly. BastionGPT helps practices stay audit-ready by acting as a compliance assistant. Engineered as a HIPAA-compliant AI platform, BastionGPT provides a secure environment for processing patient data. The AI scans documents to spot missing elements (like unsigned consent forms or incomplete charts) and helps you fix them proactively. It also assists in anonymizing patient information when needed. Essentially, BastionGPT provides an extra layer of reassurance that your documentation and processes meet required standards, guiding you in corrections before an auditor arrives. #### 24. PHI De-Identification If you need to share patient records or cases outside the practice BastionGPT can help you anonymize them. By prompting the AI with a document, you can ask it to remove or replace all Protected Health Information (PHI) – names, birthdates, addresses, etc. The result is a sanitized document (e.g. “John Doe” becomes “Patient A,” dates adjusted or blanked) that you can safely use. This is invaluable for creating training materials or consulting on cases while staying within HIPAA rules. Read this article to learn more about how to anonymize patient data with BastionGPT. Ex. You are a compliance officer preparing patient records for external sharing. Anonymize the document by removing or replacing all Protected Health Information (PHI) to create a sanitized version for training or consultation. #### 25. Chart Audit Preparation Before an insurance or regulatory audit, use the AI to audit your own charts. For example, feed it a batch of 10 patient records and ask it to check for required documentation. BastionGPT might flag that 2 of them are missing a signed treatment consent, or that one hasn’t had a HIPAA acknowledgment form updated in 6 years. It can rapidly scan for the presence (or absence) of key items. By aiding you in filling these gaps, your team can correct them proactivley, vastly improving audit outcomes. Ex. You are a compliance officer preparing for a chart audit. Review a batch of patient records and identify any missing required documentation, such as unsigned consent forms or un-updated HIPAA acknowledgment forms. #### 26. Billing Compliance Checks Similar to chart audits, the AI can review billing records for potential coding or compliance issues. It can look at a day’s insurance claims and highlight anomalies, like “Procedure D1110 (adult prophy) billed twice within 3 months for the same patient, check if one should be perio maintenance (D4910).” Or if a note doesn’t justify a billed code, it could point that out too. This kind of AI-driven compliance scrub helps prevent fraud or billing errors that could trigger penalties. Ex. You are a billing auditor reviewing insurance claims. Highlight any anomalies or potential coding and compliance issues, such as duplicate billing for the same patient or codes not justified by notes. #### 27. Incident Documentation & Analysis In case of a compliance incident (e.g., an email with PHI accidentally sent unencrypted), BastionGPT assists in documenting what happened and drafting the necessary reports or patient notifications. It maintains a professional tone and helps ensure all key information is included (what occurred, when, who was affected, remedial actions taken). Well-documented incident reports are often a compliance requirement, and the AI helps avoid missing critical elements while drafting under stress. Ex. You are a compliance officer documenting a security incident. Draft a professional incident report, ensuring it includes all key information about what occurred, when, who was affected, and the remedial actions taken. ### Patient Communication & Education Educating patients and fostering clear communication are essential for satisfaction and treatment acceptance. BastionGPT assists by creating polished, easy-to-understand educational content tailored to each patient. Dentistry often involves jargon and anxiety; AI for patient communication helps translate “doctor-speak” into friendly guidance. Want to provide aftercare instructions for a procedure in the patient’s preferred language? Done. Need a quick summary of what a root canal entails, written at a 6th-grade reading level? The AI has it covered. It also generates content like blog posts, newsletters, or social media tips, enabling your practice to engage and educate your community without extensive writing. By leveraging AI, dentists empower their patients to stay informed, comfortable, and confident in their care. #### 28. Personalized Aftercare Instructions Generate custom post-op or post-treatment instructions for patients. Instead of handing out a generic sheet, BastionGPT can incorporate specific details: “After your deep cleaning today, John, focus on gentle brushing in the lower right area where your gums are tender. Rinse with warm saltwater tonight. It’s normal to have mild soreness for a day or two. Call us if you experience any sharp pain or bleeding that won’t stop.” By personalizing and simplifying instructions, patients are more likely to follow them which can lead to better outcomes and fewer panicked phone calls. Ex. You are a dentist providing post-treatment instructions. Generate custom aftercare instructions for a patient, incorporating specific details about their procedure and personalized advice for recovery. #### 29. Condition & Treatment Explainers Use the AI assistant to create plain-language explanations for diagnoses or procedures. If you diagnose early-stage periodontal disease, you can prompt BastionGPT to produce a one-page explainer for the patient: what it is, how it’s treated, and how to prevent progression. These can include helpful analogies or visuals (if you choose to add images later). Similarly, for a procedure like an implant, the AI can write a step-by-step overview (“First, we place a titanium post…”) so the patient understands the journey ahead. Providing these handouts or emails reinforces what you discussed and helps patients remember and share accurate info with their family. Ex. You are a dentist explaining a diagnosis or procedure to a patient. Create a plain-language explanation (e.g., a one-page explainer for periodontal disease or a step-by-step overview for an implant) to enhance patient understanding and recall. ## Conclusion Dental practices embracing AI are experiencing significant gains: saving time, reducing errors, and boosting both patient and staff satisfaction. By automating tedious documentation and enhancing every workflow from the front desk to the operatory, BastionGPT empowers your team to focus on what truly matters, delivering exceptional patient care. It’s a professional, approachable assistant working 24/7 in the background, consistently supporting your dental team. Ready to transform your dentistry practice with AI-driven efficiency? Click here to start your 7-day free trial. Experience firsthand how this cutting-edge tool can elevate your dental practice – from streamlined records to happier patients. Let BastionGPT handle the busy work while you shine in your role as a caregiver. If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/dme-generative-ai-use-cases Healthcare Innovation # How DME Companies Are Using AI for Operational Efficiency Written By: Robert Henehan • April 13, 2025 Durable Medical Equipment (DME) companies are increasingly using Generative AI to increase efficiency and improve the quality of their work. However, generic AI tools like public ChatGPT are not HIPAA-compliant and cannot be used with PHI. This is where BastionGPT comes in. BastionGPT is a HIPAA-compliant, private version of ChatGPT and many other leading edge AI models that are engineered for healthcare settings. It runs in a secure, isolated environment that meets strict regulatory requirements, allowing DME companies to leverage generative AI on sensitive patient and billing data safely. In the sections below, we explore practical use cases for generative AI assistants like BastionGPT across DME operations. Each example highlights how AI can enhance efficiency, reduce costs, and support compliance in daily workflows. ## Streamlining Documentation and Recordkeeping DME staff spend considerable time on documentation – from patient intake forms and physician order transcriptions to compliance reports. Generative AI can assist in creating accurate documentation, freeing employees from tedious typing. For instance: Automated Transcription & Note-Taking: BastionGPT can listen to voice dictations or phone calls and convert them into written notes or visit summaries. This reduces manual data entry for staff. Our natural language processing (NLP) models now achieve high accuracy in transcribing medical terminology. This means a delivery technician or clinician can dictate equipment setup notes, and BastionGPT will generate a structured record with minimal errors. Drafting Medical Necessity Letters & Forms: DME companies often need to compile letters of medical necessity, prior authorization requests, or other insurance forms based on patient records. BastionGPT can quickly generate a first draft of these documents using the patient’s PHI and the relevant insurance criteria, which staff can then review. This not only saves time but also helps include all required details. Generative models can analyze medical records and suggest appropriate codes or phrasing for such documentation, reducing the chance of omissions. Summarizing and Organizing Records: When dealing with lengthy physician notes or hospital discharge summaries that justify a DME item, BastionGPT can summarize the key points (diagnoses, functional limitations, etc.) for internal use. It can highlight the information from an EHR that supports equipment qualification, making it easier for staff to build a compliant case file. By identifying missing information or inconsistencies in documentation, AI tools help prevent avoidable errors before they cause issues. This means staff might get a prompt if a physician’s order is missing a required element or if a rental agreement lacks a required field, for example. This helps speed up paperwork while minimizing human errors that could lead to billing mistakes or compliance gaps. ## Enhancing Internal Communication and Knowledge Sharing BastionGPT can also improve day-to-day communication and information access within a DME company. Many DME teams juggle communications between departments (sales, warehouse, billing, etc.), and have extensive internal knowledge bases (policy manuals, product catalogs, insurance guidelines). AI assistants can make these interactions and knowledge retrieval more efficient: Email and Memo Drafting: Routine internal communications – such as writing emails, drafting policy updates, or composing meeting minutes – can be offloaded to an AI for a first draft. For example, a manager could ask BastionGPT to “Draft an email to the respiratory therapy team about the new oxygen concentrator maintenance protocol” . The AI will produce a clear, professional email in seconds, which the manager can tweak and send. This ensures consistency in tone and frees up time for more critical thinking tasks. Meeting Summaries and Action Items: Integrating BastionGPT into meetings (virtual or in-person) allows it to transcribe and summarize key discussion points. After an operations meeting or a compliance review session, BastionGPT could generate a summary highlighting decisions made and action items assigned. Team members who couldn’t attend get a concise brief, helping to ensure nothing falls through the cracks. Internal Knowledge Base Q&A: BastionGPT can be provided your company’s documents and SOPs. Instead of sifting through dense binders or SharePoint folders, employees can ask natural language questions and get instant answers. Picture this: A staff member needs to know the Medicare requirements for a power wheelchair rental. Rather than search multiple PDFs, they query the AI assistant. Because it can be loaded with coverage criteria and past approval examples, it can instantly respond with the Medicare requirements and key considerations. In this way, our healthcare AI turns the labyrinth of internal knowledge into a quick conversational lookup tool. This not only saves time (studies show employees often spend significant portions of their day searching for information) but also ensures staff use the most up-to-date and consistent information. Over time, this kind of AI-driven knowledge sharing boosts productivity and confidence – new hires ramp up faster and even veteran employees benefit from instant cross-checks (e.g. asking “What’s our company policy on emergency equipment replacement?” and getting an answer citing the relevant policy section). ## Billing and Claims Management Automation Billing and insurance claims processing are core operations for DME providers – and ones rife with complexity and paperwork. BastionGPT, especially when combined with a DME’s billing data and payer rules, can greatly streamline the revenue cycle. BastionGPT can serve as a tireless billing assistant, handling everything from suggesting possible code assignment to drafting appeal letters: AI in Revenue Cycle Management: One immediate use case is AI-assisted coding and claim preparation. BastionGPT can assist with DME billing, translating medical orders and documentation into billing codes (HCPCS, ICD-10, etc.) and helping to ensure each claim has all required info. BastionGPT’s powerful AI models excel at pattern recognition in text and can suggest appropriate billing codes from descriptions. In practice, this means after reviewing a doctor’s order and patient condition, the AI might propose the correct HCPCS code for a hospital bed rental along with any necessary modifiers. This helps reduce coding errors and ensure accurate billing. Such consistency is crucial, as even minor coding mistakes can lead to denials or delayed reimbursement. Generative AI can also “scrub” claims for errors or omissions before submission. It will analyze the claim details and documentation, flagging missing elements (e.g. lack of a chart note to support the diagnosis) or likely mistakes, so staff can fix them proactively. This kind of AI-driven claim validation within BastionGPT increases first-pass acceptance rates. Appeals and Prior Authorization Management: Beyond initial submission, BastionGPT proves invaluable when claims are denied or require additional info. Drafting appeal letters and prior authorization requests is a natural language task that BastionGPT can handle quickly. Rather than writing an appeal from scratch, a billing specialist can ask the AI to generate a letter to the insurer explaining why a particular claim (e.g. a custom wheelchair) meets coverage criteria, referencing the patient’s medical history and the payer’s policy. Indeed, early adopters report that generative AI is already being used to produce appeal letters for claim denials. This not only saves time, but often results in well-structured, thorough letters that cite the correct rules – increasing the chance of a successful appeal. Similarly, for prior authorizations, the AI can assemble the requisite info (diagnoses, physician notes, etc.) into a concise narrative justifying the equipment need, which staff can review then submit to insurers. Patient Billing Communications: Generative AI can assist with patient follow up. For example, if a patient has a co-pay or needs to be billed after insurance, BastionGPT can draft a clear, empathetic billing statement or payment plan letter for the staff to review and send. All told, integrating generative AI into billing and claims workflows can significantly reduce administrative burden, lower denial rates, and speed up reimbursement, which is a major competitive advantage for DME executives. ## Ensuring Compliance and Reducing Risk Compliance in the DME industry spans HIPAA privacy, billing regulations (Medicare, Medicaid, private payer rules), and even anti-fraud laws. BastionGPT can act as an assistant to help DME companies stay compliant and avoid costly pitfalls. Importantly, because BastionGPT is built with healthcare data security in mind, it can be trusted to handle compliance-related data without leaking sensitive information to OpenAI (ChatGPT) or similar AI companies for their AI training. PHI De-identification and Anonymization: BastionGPT’s healthcare generative AI can be tasked with removing or anonymizing PHI in documents when needed. Suppose the compliance team wants to use a real case as a training example – they could ask the AI to generate a version of the document with all names and other HIPAA-identifiers changed, redacted or removed. BastionGPT, having the ability to process PHI, can do this internally while ensuring no sensitive data is exposed externally. Regulatory Update Summarization and Policy Generation: DME providers must keep up with frequent changes in healthcare regulations (e.g. CMS updates to DME fee schedules, documentation requirements, or new state laws). Generative AI can help by quickly summarizing lengthy regulatory documents into key takeaways. A compliance officer could feed in a new 30-page Medicare bulletin and get back a one-page summary of what changed and what actions the company needs to take. This allows faster response and ensures nothing important is missed in interpretation. The AI can even draft updated internal policies or checklists based on that new regulation. For instance, if Medicare introduces a new prior authorization rule for power wheelchairs, BastionGPT could draft an internal memo or policy document outlining the steps staff must follow, referencing the official guidelines. The compliance team can then fine-tune this draft, saving considerable writing time. Audit Preparation and Documentation Consistency: BastionGPT also supports audit preparation and documentation consistency. Since DME companies are subject to audits from payers and accreditation bodies, BastionGPT could review a batch of patient files and generate a report on whether each file has the required documents (prescription, face-to-face exam note, delivery proof, etc.). By comparing against a predefined checklist, the AI can quickly highlight files that are possibly non-compliant or missing elements, allowing staff to correct them before an external audit occurs. This proactive approach helps avoid penalties. ## Supporting Staff Training and Onboarding Training new staff and keeping existing employees updated is an ongoing effort in any DME company. Generative AI can lighten this load by providing on-demand training assistance, personalized learning content, and even simulated scenarios. By integrating generative AI into training and development, DME companies can reduce the costs and time of training, ensure staff are more uniformly knowledgeable, and foster a culture of continuous learning. New hires get up to speed faster, and existing staff stay sharp on the latest procedures – all of which translates to better operational performance and compliance. AI Mentoring for New Hires: One practical application is using an AI assistant as a 24/7 mentor for new hires (being careful given all generative AI technologies are capable of making errors). During onboarding, a new employee can ask the AI questions anytime, such as “How do I process a Medicare claim for a wheelchair?” or “What are the steps for equipment delivery setup?” using saved prompts that contain the company’s procedures and manuals, can provide step-by-step guidance instantly. This reduces the need for senior staff to constantly shadow or answer repetitive questions, allowing new employees to learn at their own pace with company information and industry best practices. It also ensures consistency in training — everyone gets information from authoritative sources such as company policy and SOPs. Customized Training Materials and Scenarios: BastionGPT can also create training materials quickly. Instead of writing slides or documents from scratch, trainers can leverage our AI to generate drafts. For example, an internal educator could prompt the AI: “Create a summary of safety guidelines for operating home oxygen equipment for our training handbook” . The AI would produce a well-structured summary covering key points (fire safety, storing tanks, checking alarms, etc.), which the educator can then refine. If your DME company needs to train staff on a new software or process, our health AI can generate example datasets or practice exercises. Notably, BastionGPT can help produce realistic yet anonymized patient scenarios for training purposes. Using generative techniques, it’s possible to create synthetic patient profiles and documents that mimic real cases without exposing actual PHI. Trainees could practice tasks (like verifying an order meets coverage criteria or responding to a customer inquiry) using these lifelike scenarios. This hands-on practice is far more engaging and effective than abstract examples, and because the data is fake but realistic, there are no privacy concerns. Personalized Learning and Interactive Education: BastionGPT can adapt to different learning needs, effectively enabling personalized education. If a staff member is struggling with, say, understanding complex insurance rules, they could have the AI explain it in simpler terms or even quiz them in a friendly manner. BastionGPT could simulate a Q&A where it acts as a “tester,” asking the trainee how they’d handle certain billing situations and then giving feedback or corrections. Such interactive learning reinforces knowledge and builds confidence. Generative AI like BastionGPT is frequently being used to simulate training scenarios in healthcare that prepare staff for real-world situations – for DME, this might include handling an angry customer call or navigating a sudden recall of equipment, all via role-play with our artificial intelligence. Continuous Education and Policy Updates: Policies and best practices evolve; AI can keep track of updates and proactively educate the team. For instance, if a new insurance requirement comes out, the AI could automatically produce a brief e-learning module or an internal newsletter article explaining the change and even quiz staff on it. It essentially helps in creating a self-updating training program. ## Conclusion From the back office to the warehouse, BastionGPT is poised to transform your DME operations. The use cases highlighted – documentation drafting, internal communications, billing automation, compliance monitoring, and staff training – demonstrate that many relatively simple, routine workflows can quickly be improved with AI assistance. Importantly, these enhancements directly contribute to the metrics that matter for operations managers and executives: greater efficiency, lower error rates, reduced costs, and stronger compliance. By offloading tedious tasks to BastionGPT, employees can focus on higher-value activities such as patient service improvement, business growth, and proactive quality control. With BastionGPT, DME organizations can confidently integrate AI into workflows that involve PHI/PII, knowing that patient data remains protected. This opens the door to leveraging AI in ways that were not possible with consumer chatbot subscriptions like ChatGPT, Gemini or Claude. Including your organization’s specific data and guidelines in saved prompts will make its outputs more accurate and contextually relevant. Having humans in the loop (for oversight and final approvals) ensures that AI remains an aid, not a sole decision-maker. Generative AI has now moved beyond hype to offer tangible, practical benefits for DME companies. By embracing these use cases, and many more that will be discovered by your use, DME providers can streamline their operations while maintaining the high standards of accuracy and compliance that the healthcare environment demands. The result is a leaner, more responsive organization that can ultimately deliver better service to patients and partners. For DME executives planning the next steps in digital transformation, deploying BastionGPT’s HIPAA-compliant AI assistant is a strategic move – one that can reinforce everything from day-to-day productivity to long-term competitive advantage in the evolving healthcare landscape. Most organizations are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/exploring-the-digital-battlefield-of-healthcare-cybercrime Healthcare Innovation # Founder Podcast: Exploring the Digital Battlefield of Healthcare Cybercrime • September 2, 2023 ## Healthcare under siege, exploring the digital battlefield of healthcare cybercrime In this episode, host Bob Kaiser speaks with Josh Spencer, healthcare CISO and founder of FortaTech Security, to uncover the dark underbelly of the multi-billion-dollar healthcare cybercrime industry and reveal how healthcare organizations are leveraging the latest technologies to defend themselves. ## Podcast Captions (Auto-Generated): [Bob Kaiser 00:28] Welcome to the Business of Healthcare podcast from the Center for Healthcare Leadership and Management at the Naveen Jindal School of Management here at the University of Texas at Dallas. We bring together business leaders and other forward thinkers to discuss how best to meet the challenges of a rapidly changing, increasingly complex healthcare ecosystem. I'm Doctor Bob Kaiser, Director of the Masters Program and Healthcare Leadership and Management. Be sure to subscribe on Apple Podcast, Spotify or your favorite podcasting app to ensure you don't miss any of the future episodes. You can also join us online at businessofhealthcarepodcast.com. Today on the Business of Healthcare Podcast, we're going to explore the digital battlefield of healthcare cyber crime. We have a lot of practical guidance to share today. And also a deeper insight into how artificial intelligence is being deployed in the healthcare cybersecurity space. Our guide today is Josh Spencer, Senior Partner and Chief Information Officer at FortaTech Security. Josh is no stranger to the complicated world of healthcare. Prior to his current role, he served as Chief Technology Officer at UT Southwestern Medical Center, guiding massive infrastructure resiliency or improvements. Data center migration and implementation of transformative new technologies. Josh, welcome to the Business of Healthcare podcast. [Joshua Spencer 01:57] Yeah, thanks for having me, Bob. It's a real pleasure to give you and your listeners a peek into the dark underbelly of the healthcare cybercrime world. And I'm really excited to share some of the powerful weapons that we use to keep our healthcare data safe. [Bob Kaiser 02:12] You know, we have many points of interest to provide these your practical awareness and guidance, but we also want to go deeper today, Josh, into how artificial intelligence plays a role and how it's going to fit into the future. So let's start with a little bit on your background as a Certified Ethical Hacker, My guys, that name itself sounds pretty impressive. And your role as the Chief Information Security Officer. You know, protecting the information and assets and technology for a large Medical Center doesn't sound like a very easy job to me. Tell us more about that. [Joshua Spencer 02:46] Yeah, it's certainly not an easy job, but it is exciting. Over the last 13 years, I've been helping keep healthcare organizations secure, and not a single day has gone by that I didn't see or learn something new. As a certified actual hacker, I was brought in to look at healthcare organizations through the eyes of an outside attacker. I would use the same software, same tools that the hackers use, and scan a company's network looking for any signs of weakness. And when I did find a weak spot, weak spot, and the armor of their network, I would then gain access to that system and all of its data, and then continue to move all throughout the network, gaining access to more and more systems along the way. And that was it. Was a fun exercise, but critically important at the end of the day because most companies, they're going to overestimate their defenses and underestimate the capabilities of the attackers. So an ethical hacker like myself, once they've broken in, we'll work with that company to understand and fix the holes in their security, with my ultimate goal being to make it as difficult as possible for that next hacker to get in and do what I just did. [Bob Kaiser 03:53] So there's such a thing as a Certified Ethical Hacker where you have to be qualified. Tell us a little bit more about that. That's a little bit intriguing to me. I didn't know such a position or training existed. [Joshua Spencer 04:05] Yeah, absolutely. So it's a program that makes sure you have not only the the technical knowledge, since there's plenty of information out there about how to how to do that, but also that you have an understanding of the more obligations of what you're doing. As an ethical hacker, you are going to be getting access to a full treasure trove of information that you weren't intended to have. So you need to know what to do with that information, and then also how to properly communicate with with your clients to understand what to do with that information temporarily. How to rapidly put a bandaid to keep that information from getting compromised and then to work with them to make some longterm fixes to fix not only that particular issue, but make sure similar issues don't get introduced in the future? [Bob Kaiser 04:52] Okay. So we're talking about cyber crime here. I've got a few few general type questions. So let's kind of start with the criminals. Since this is cyber crime, you know, who are they? Why do they do it? And and what exactly do they do? So who are these cyber criminals? What? What are they like? Yeah, absolutely. [Joshua Spencer 05:10] So when I think about cyber criminals, they're really going to come from a whole range of very diverse backgrounds spread out all over the world. But that being said, you know, when we look at it, there's really three common attacker profiles. So the first one is going to be the government sponsored hacker groups. We commonly refer to them as Advanced persistent threats for a PT's. Now these are going to be teams of elite hackers. They're going to be directed by their governments to steal secrets. So think of military strategies or even from a healthcare standpoint, medical research that's giving their countries an edge in that industry. These attack groups are also going to be profit motivated and they're going to be netting their governments billions of dollars every year in revenue to help fund the governments, or at least some of the politicians that are employing them. The UN actually, they released a estimate a few years ago saying that N Korea's cyber army was responsible for 50% of North Korea's income. Their attacks have drained billions of dollars from numerous bank accounts across the world. So they have a a very successful program. But yeah, even though these governments that and their attacker groups get a lot of media attention, really the most common of the three attacker profiles are going to be the organized crime syndicates. So these are large organizations with hundreds or even thousands of employees and they operate much like any other company that you think of. They've got HR departments, paid time off, new hire orientations, they're they're most likely going to operate out of countries that aren't very friendly to the US and and European governments. But despite all of the complaints that they're getting from our governments, they're still not being shut down. That's because, you know, the, the police in those countries recognize that, hey, they're employing thousands of people in that community, providing a really great source of tax revenue. And in many cases, there's also bribery schemes in place to keep the authorities from from sweeping in. And then the last category of attacker is the one that's really started to grow the most rapidly just over the past year. We call these the lone wolf, so they can be anyone from a disgruntled IT employee that's fallen on hard times. It could be a high school kid. They just watch some videos on how to hack from YouTube and then the the latest generation of a I has really given these lone attackers an incredible ability to to rapidly develop new attacks and techniques that would have been a year ago impossible for just one person to do. So we're we're pretty quickly adjusting our defensive measures make sure we're we're properly defending against these new threats that are a I enabled. [Bob Kaiser 07:49] While it's like walking down the dark, dark alley at night, you don't know what to expect or who's going to come at you. So I can see, Josh, where financial data would be key of importance. Why do they want the healthcare data? Why is that so important? [Joshua Spencer 08:04] Yeah, answer to that is pretty simple. And it's money. Lots and lots of money. And really, at a scale that I imagine most people couldn't even conceive of, most of the data that's public out there indicates the overall global cybercrime revenue is teetering around $1 trillion per year. So give you an idea of scale, think of the combined revenues of Facebook, Apple, Microsoft and Google. So it's a lot of money. And Healthcare is definitely a prime target for cyber criminals. So much of that revenue is being stolen out of the pockets of healthcare companies that just haven't prepared themselves properly. Some industries, they've got the luxury of being able to easily correct stolen data. So think of a credit card company, they can detect fraud and have a new card in the mail same day. Now in healthcare though we can't generate a new date of birth or a new medical history for one of our patients. So in many cases a cyber criminal can profit off of that stolen data for years. And then Healthcare is also unique and and how concentrated and comprehensive our sensitive data is. So a criminal breaking into a medical record system, they've got everything they need to commit just about any type of identity theft that they can think of. You've got name, date of birth, Social Security number, home address, a lot of very powerful information. Now, I found most people in healthcare, they've they've got a kind of a vague understanding that healthcare data can be illegally monetized. But it really isn't common for folks to understand exactly how that happens. And I find when people know just how easy it is to illegally profit off of that data, they do a much better job of protecting the data. Many times that the burden that comes with being forced to use a longer password or accepting that two factor authentication prompt, it really doesn't seem as onerous as it once did once we realized that a password can protect thousands of lives from identity theft or even worse. So of course today I'm not going to give you a step by step guide to selling healthcare data, but I don't want the FBI my doorstep. But what I can share is an overview of some of the most common methods. You know, I think most people have at least heard of the dark web, but for those who don't, you know the dark web, it's a criminal's paradise. It's a hidden part of the Internet, really hard for authorities to shut down and require some special software to access. But in the dark web, these underground cybercrime markets flourish and all you need is a little bit of Bitcoin. And you can buy anything from prescription drugs to stolen medical ID cards and even passwords to medical record systems. And all these services they they look and feel just like you're buying anything else. Online you can see reviews of the sellers pick your choice of shipping speed, in some cases even ordering free samples. So let's yeah, let's say a hacker. They sent out 10,000 phishing emails to various health systems, maybe 1000 of those recipients, they fell victim, and maybe a hundred of those aren't using two factor security. Well, the hacker can then sell those 100 passwords and the going rate as of today is about $4.00 a password. And so a day's worth of effort just netted them $400.00 on the dark web. Now, someone who's specialized underneath theft, but they can visit that dark web market, buy up a few of those passwords, and they can log into the medical records system and begin collecting patient information. They can either sell that directly on the dark web, or they can, say, use that information to make a fake IDE. They can pick up some narcotic prescriptions and those patient names and do all kinds of really scary things. And then? [Bob Kaiser 11:38] It sounds like an Amazon.com for criminals. [Joshua Spencer 11:41] I mean, my goodness, absolutely it is. It is frightening to see this type of activity just thriving there on the dark web. [Bob Kaiser 11:48] Well it makes sense what the way you explained that why they want like you said it's it's it's monetarily to their advantage to get this healthcare data. So when you stop and think about these hackers getting this data. But you mentioned a couple things. You said you know people can send in phishing, that's that's emails, I guess. What is that a popular way? What are the ways do they go about getting into these healthcare systems and getting all that data? [Joshua Spencer 12:14] Yeah, absolutely. That's a very popular method. And many times when companies are hacked, they're often going to go to the press and they're going to say we were involved and advanced cyberattack and make it seem incredibly sophisticated and possible to defend against. And that's what they're the PR part of that company wants you to hear. But really in reality, what these companies don't want you to know is that they likely fell victim to a foundational security issue. So thank you, Fred in accounting answered a phishing e-mail. Nancy and HR, she kept canceling her security updates until her computer was infected. And unfortunately we we really have to worry about more than just the Fred Nancy's. We have to worry about vulnerabilities not just from our own employees and systems, but all the vulnerabilities that exist with all of our vendors. And in healthcare we have perhaps more than any other. So we might remember some of the names like Humana and Anthem that that hit some of the headlines due to our breach from their vendors. But it is overall just incredibly common in healthcare. If if you were to take the list of the top ten healthcare breaches that were reported last year, most of those, so over 50% of those were due to third parties. So it's incredibly important to understand who has your data, who has access to your systems, and how the data and that access is being secured. [Bob Kaiser 13:33] Do these hackers do these criminals? Do they typically pick out a specific company and go after them? Or is it a a wide, wide cast of their line just to get anybody who bites on it, so to speak? What? What do you see happening there? [Joshua Spencer 13:47] Yeah, so it's going to be a little bit of both. Quite a few attack campaigns are targeted at a single company. Often times this is going to be due to certain news stories that hit the Internet that put the company in the limelight, especially if that news could imply that the company's become much more valuable suddenly, or perhaps is less defended. But for the most part, hackers are going to cast that wide net. They might collect a list of every healthcare company e-mail address they can find on the Internet. Or they might set up their scanning programs to crawl through the Internet looking for any sign or vulnerability, and then prey on whatever weaknesses that they find. [Bob Kaiser 14:22] Is there another motive other than financial gain you hear about people doing what I'd say ruinous activity they'll attack somebody's you know company or website just to put them out of business. I'm not. [Joshua Spencer 14:36] I'm not sure what that terminology is when they they overrun override the the web access for those those sites, the financial fraud, medical identity theft is one of the biggest motivators but it's not the only motivator. So in the case of these you know state sponsored attacks or advanced persistent threats, many times that's to gain a foothold or negotiating leverage with the other governments. So we've seen some of these foreign hacker groups intentionally Target Healthcare infrastructure and try to dig its roots into deep into our our nation's healthcare infrastructure to gain some leverage in those negotiations that sort of things go South that they know that they could cripple some of our our healthcare infrastructure. The other big component there is some people will attack healthcare organizations just for the ideology or because they think it's fun. So there's been a couple cases on the news where I believe there's a Children's Hospital that that treated a allegedly treated a patient poorly. They then launched A denial of service campaign that shut down that healthcare companies network and it wasn't until a bunch of media backlash it that that person relented and gave access back to the hospital. [Bob Kaiser 15:49] That reminds me recently here in our hometown of the city was they had some type of attack on their systems which basically potentially could lose all types of legal files and in court documents and things like that. And it's like what when you get into a situation like that and you're asked to pay ransom, what's the, what's the best practice? What's, what's the advice that you give people Who do you call and what do you do? [Joshua Spencer 16:14] Yeah. So you certainly want to be prepared with an action plan. The time to put together a plan isn't once you have this, this major event. So definitely need to understand, you know, work with a security expert to understand what are the most common threats out there? What are the best practices for responding to it For something like a ransomware attack? If you have an insurance policy, does that cover the payment of of some of these ransoms? If you have disaster recovery plans, making sure that they fit all these scenarios that you can be up and running if bad things happen. [Bob Kaiser 16:46] Well, I know it doesn't have to happen to a big organization. I've got some friends who've said, you know, they've they've had their home computers, ransomware, they've had their Instagram files hijacked and they can't get their their information back and things like that. If it happens to an individual, because a lot of our audience are individuals as well, they're not. They're not what I'd call the chief information security officers, but what would an individual do? Would they call their local Police Department or they did just? Frustrated and give up what's what's what's the best advice for them, Josh? [Joshua Spencer 17:17] Yeah, I'd say they should certainly reach out to their local Police Department. But they can also coordinate with some cybersecurity companies that specialize in in working directly with with consumers. Because the way that someone responds to an individual attack is going to be very different from the the way they respond to a large corporate attack. So you take a look at the healthcare scenario, your average ransomware attack is going to be $4.8 million. Whereas when we see some of these attacks on on individuals, those are going to be closer to around the range of $300.00 per computer. So the $300.00 is much more likely for an individual to to be able to pay. And there certainly are a number of moral considerations as well that you you might need that data. But for every person who who pays that ransom, they're now funding the next round of of ransom attacks that that are going to target even more people and potentially destroy even more lives. So that that's a call that I would definitely recommend you make in consultation with your your local authorities and a cybersecurity expert. Based on unique facts of the case and what's been stolen, this episode is brought. To you by the Center for Healthcare Leadership. And management, the definitive resource for healthcare management education in North Texas. The center is based in the Naveen. [Bob Kaiser 18:37] Jindal School of Management at the University of Texas of Dallas. [Joshua Spencer 18:40] It plays a unique role in training the next generation of healthcare leaders to meet. Local, regional and national demands the Jindal School uses its. Strengths in Accounting, Administration, Finance, marketing. And information systems to educate highly qualified personnel for healthcare administration and executive leadership positions. The center is home to seven healthcare leadership and management programs, including undergraduate and graduate programs, as well as executive programs for physicians and working professionals. For more information, visit us online at Jindal dot UT Dallas dot. [Bob Kaiser 19:15] Edu/ healthcare. Is ransomware becoming more prevalent? Is is that, if that was a product, is that product selling, is that a big growth area? [Joshua Spencer 19:25] Yeah, so it ransomware really started to come popularized right around 2013 and and I would say that's definitely a permanent fixture in the world of cybercrime. It it peaked right around last year and has been fairly steady. But unfortunately healthcare customers are they're one of the most common and lucrative targets. You know this is because Healthcare is one of the most likely markets to pay ransom. So we we take a look at a report that I BM released earlier this year. You know they're showing healthcare as probably the the largest market impacted. Part of the highest rates of highest demand for some of these ransom payments and even at that $4.8 million figure that I I mentioned most analysts and myself included, you know we we find that that number is actually likely to be much higher. But another factor that's really going to impact healthcare so heavily with ransomware is the quality of our healthcare delivery is so closely tied to use of our electronic systems, we have to worry about potentially fatal outcomes if our access to healthcare data isn't there. So factors like you think delayed care and ability to access urgent information like blood type allergies, medication lists, these are huge factors that are going to motivate a healthcare provider to pay a ransom pretty quickly and as quietly as they can. And when we talk about fatalities this this is no longer a theoretical risk. Poneman Institute, they they performed a study last year across 600 healthcare facilities and they found mortality rates increase in facilities that were experiencing A ransomware attack. And then we we we have these large macroeconomic numbers but we can actually zoom in and look at specific cases. Now a case where emergency department, they had to turn away an ambulance during a ransomware attack and that patient died during the reroute to the hospital and that likely would not have been the case if that emergency room was open and operating. And then we have another case in Alabama where a critical piece of modern equipment was down due to ransomware infection and didn't detect an umbilical cord wrapped around an infant's neck until it was too late. And and that's, you know, those stories are exactly why me and and so many others, we choose careers in healthcare, cybersecurity. You know, it's not just about protecting A business's profits. It's really a moral duty. I'm sure we have some of the highest standards of our data protection because it's really, it's our patient's lives that hang in the balance. You know, every bite of stolen data could could not just turn a patient's life upside down. It could actually end it. So definitely to defend against these attacks, when we look at how do we defend against them, you know one of the most critical elements there is what we call an air gap that that goes between the healthcare network and your data backup. So think about you know, storing a copy of your important data on devices that are totally disconnected from electronic system and stored physically in a vault somewhere. So yeah, unless the criminals going to physically break in your vault, your data is back up protected no matter what that hacker has access to electronically. Or there's also a number of additional technologies that can help prevent ransomware detect it, shut it down before it spreads across your network. So you should definitely pair it with a variety of controls to help keep your organization protected or even from a personal standpoint, making sure your computer is up to date with the latest security software and patches and and things like that. [Bob Kaiser 22:55] So I I have a nice little four terabyte solid-state disk drive that I use to back up my personal information. I can have an air gap there. You know, I can. I can stick that over there into a safety deposit box or into a different room not connected to the Internet. But my goodness, what would you do for a large Medical Center where you've got just a tremendous amount? I can't even use the word to describe how much data there is. How can you create an environment like that with an air gap? That seems like impossible. [Joshua Spencer 23:28] It it certainly is a herculean effort with some of these systems storing data and in every corner of the network. It definitely takes a concerted program to look through your organization's data. You're not going to back it all up in a way that might be preventable from some of these ranks more attacks, but you can at least identify these are the most critical pieces of information that we're going to make. The additional investments to make sure that they're secured and backed up and and air gaps. [Bob Kaiser 23:56] At the beginning of our conversation, I promised the audience that we would maybe take a little bit deeper look into the technology pool, looking a little bit more and tur, we we've learned a whole lot here about a lot of the issues and a lot of the mitigation things we can do to help maybe avoid some of this. There's got to be a bigger, better solution in the future that we don't know about yet. So is there? Can you give us a high level overview? I've heard of quantum computing. Can you explain what that is and how that might impact cybersecurity in the future? [Joshua Spencer 24:29] To get a better idea of what quantum computing is and really why it's so impactful to cybersecurity, just just imagine you're in a massive library. It's got thousands of books, and you need to find one specific sentence within one specific book. So the way a computer today is going to do that, it's going to be a very diligent reader. It's going to pick up each book one by one and look for that sentence. It's going to take a very long time, even if your computer's a very fast reader. Now, quantum computer, though, it's going to use some incredibly weird and unique physics that allows it to read through each book at the same time. So it's going to find your sentence much, much faster. So the problem we're facing today is, is the way we encrypt data is by performing a complicated set of math problems that only people with the key are supposed to be able to figure out quickly. Well, these these quantum computers, they can now solve in seconds what would have taken a normal computer decades to solve without the key. Now, thankfully, the the quantum computers that exist at this moment, they're they're very expensive and very fragile and only accessible to large governments and research labs. Although I'd anticipate the next three to five years are going to bring the public access to powerful quantum computers. Computers are going to be capable of breaking encryption that we've depended on for decades in just a matter of seconds. So yeah, if if the data you're trying to protect is things like credit card numbers, really not a huge concern right now because by the time this quantum computing technology gets an attacker's hands, yeah, it's that data is not going to be valuable, it's going to be expired. But in healthcare, we are challenged by the longevity of our data, Social Security number, a date of birth, a black male medical condition that's still going to be just as valuable in five years as it is today. So we're seeing cyber criminals scooping up just large volumes of encrypted data from public networks, with the plan to harvest that data years later using these quantum computers. [Bob Kaiser 26:34] Wow. So when we look at this whole world of quantum computing, it sounded pretty devastating that the bad guys are going to have some tools against us. [Joshua Spencer 26:43] Is there any positive news about the good guys using this? Yeah, absolutely. It's not all bad news with quantum computing. There are some encryption methods already in use today that are what we call quantum resistant. And so we're already seeing security departments implementing those measures. We're getting rid of the easy to break encryption methods. We're also seeing some really powerful research into quantum computing that provides a level of security that is theoretically uncrackable and it gives that by exploiting some very unique properties in quantum physics. So I'd imagine within the next few years we're going to see those new quantum networks and quantum protocols start to pop up that provide a much greater security and privacy than any other technology that we've had in the past. [Bob Kaiser 27:29] Wow. That kind of opens the door for a topic that we hear a whole lot about today. Everywhere you go, you hear more about artificial intelligence. A I Could you give us kind of a high level overview of what that really means? What does artificial intelligence really means and and how can that be used for the good guys and the bad guys? [Joshua Spencer 27:49] When it comes to cybersecurity, I tend to be fairly pessimistic when there's huge hype around a particular technology. So I'm thinking Bitcoin, NFPSVR, lots of potential there, but often overhyped. A I though this this feels entirely different to me. It was just a few years ago that a I really started to impress me in its potential. Certainly put a a lot of research and effort building up the the capabilities within myself and my team and really gained an understanding that this is going to be an incredible potential to do great things that most of us thought would come about in the next 5 to 10 years. But what's really shocked expert technologists though is really just within the last 12 months we have surpassed the A I capabilities that we thought would be there in 20-30. So really, just despite all the hype and hysteria that that A I's gotten even more intense over the past year, I still don't think the world, most of the world truly appreciates how much of an impact AI is going to have in our day to day lives. And then of course with with every other technology from our past, both the good guys and the bad guys, they're going to find a way to make use of it. So on on the bad guy front, we are seeing already daily attacks against healthcare organizations using AI. Some of our clients are getting customized emails from scammers that will read through a public profile on, say, LinkedIn and pretend to be a friend on that page. They're going to use all the details on your profile to tailor that scam to to trick you the using the most likely scenarios that could come up with and offer just pennies for e-mail. I don't know if you've seen on some of the news headlines there's been some heart wrenching stories of mothers getting calls from their distraught daughters saying they've been kidnapped and they need to send a ransom payment to this this address. You know, the criminals are they're using a I they're using free voice cloning technology and using, you know, deep fake video generation to create really believable conversations or even video calls to help extort the money. So we we actually counsel our clients both in in a business and personal sense to establish code words with your coworkers and your family to help verify your identity. Because even with the technology today and its infancy it's it's very possible that the next call you get on your phone could sound just like your loved one or your your coworker asking for a transfer of money or or some other suspicious activity. And and while some of the the large A I services that that you see on the news open A I Google Bard they all have some security controls that prevent them from being used for evil. But there's unfortunately many ways the cybersecurity criminals have have found to either get around those controls or they've also built some powerful uncensored A I technologies that have those security controls removed. So again the the silver lining here is is really seeing the progress our cybersecurity community is making on these powerful new A I protectors. It can look through vast amounts of data much more quickly than a human can and identify and shut down some suspicious behavior. So a I is recognizing potential hackers in a network and crafting automated responses, all in less time it would have taken for a human just to even read that first security alert. And I'm sure many of us, myself included, we we're ready for the end of phishing and spam emails. And it's really great to see this latest generation of a I technology is is much more capable of detecting and deleting some of those unwanted emails. And then given some of the the tight budgets that we've seen in in many health coordinations, this this A I technology, even if it's current state. It's allowing, even as a senior junior security analyst, to be perhaps 20 to 50% more effective at their job when they're using a I to much more quickly analyze attacks and progress, or to help build stronger defenses into their networks than what they could have by themselves. So if you haven't checked it out yet, Microsoft Security Copilot, it's not generally available yet. But if you do want to see what types of capabilities and features are right at our doorstep, I definitely recommend checking out their demo. It's some pretty powerful stuff. [Bob Kaiser 32:10] I'll tell you what, it's a pretty, pretty scary world that we're entering right now. And we all need to be better educated. We all need to know more. We all need a guide. The old saying is, if you're going through the Grand Canyon on a Colorado River with your 10 year old child, who's the most important person in the boat? And most people would say it's the child. However, it's the guide. It really is the guide, the one who knows what's ahead and how to deal with it. Josh, thanks for being our guide today. We need guides like you. So in closing, what final words of advice do you have for our audience? [Joshua Spencer 32:50] Yes, I would definitely say cybersecurity is not just a concern for IT people or security people. It's really everyone's responsibility to better understand the threat and the ways that we combat it. And that's really why I'm so thankful that your listeners have taken the time to learn a little bit more about this today. You know, whether you're a physician, a nurse, a student, an executive, really any other healthcare professional, we, we all have a role to play in protecting our healthcare data and we really have to make cybersecurity a priority and not just a cost of doing business, but really an investment into the the trust and the safety of our patients and the community we serve. So definitely to to your viewers, I'd say stay diligent, stay informed, don't hesitate to ask questions about how your information is being protected. [Bob Kaiser 33:40] Well, thank you, Josh. We'll be sure to post some of these resources in the footnotes for the podcast. And it's been a great pleasure to have you join us. We'll have you back again in the future too as the world continues to change. Thank you. Thanks for listening to the Business of Healthcare Podcast. To learn more about the Center for Healthcare Leadership and Management, go to jindal.utdallas.edu/healthcare. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/health-ai-legal-use-cases Healthcare Innovation # How Healthcare Legal Professionals & Lawyers Are Using AI Written By: Robert Henehan • April 22, 2025 Healthcare legal professionals from medical malpractice attorneys to hospital compliance officers are increasingly turning to tools like generative AI to interpret medical information and improve productivity and accuracy in their work. However, generic AI tools like public ChatGPT are not HIPAA-compliant and cannot be used with PHI, which limits their utility in healthcare-related cases. BastionGPT is a HIPAA-compliant, private AI assistant (built on the latest GPT, Claude, and Gemini models) that is engineered specifically for healthcare settings. It runs in a secure, isolated environment that meets strict regulatory requirements, allowing legal teams to leverage AI on sensitive patient data safely. BastionGPT can even handle very large documents (up to 1,000 pages at a time) and perform OCR on scanned PDFs – crucial for parsing medical records and insurance files often encountered in legal matters. In the sections below, we explore practical use cases for generative AI assistants like BastionGPT across healthcare legal workflows . Each example highlights how AI can enhance efficiency, reduce manual workload, and support compliance in daily legal tasks. ## How Lawyers & Legal Teams Can Use AI ### 1. Medical Record Summaries and Case Analysis Reviewing medical records is a cornerstone of healthcare-related legal work, whether for malpractice litigation, personal injury cases, or insurance disputes. These records often span hundreds of pages of complex medical jargon. BastionGPT can drastically speed up this review process, helping lawyers, attorneys and paralegals extract the information they need. #### Summarizing Medical Records and Creating Timelines BastionGPT can sift through large patient charts and produce a concise chronology of care or a summary of key facts. For example, a malpractice attorney could upload a 300-page patient record PDF (even if it’s a scanned document) and ask for a chronological timeline of the patient’s treatment. The AI will extract dates, diagnoses, procedures, and outcomes, then list them in order – giving a clear narrative of the patient’s care history. This saves hours that would otherwise be spent piecing together events from various reports. Attorneys get an immediate grasp of what happened when, which is invaluable for identifying potential negligence or pivotal moments in the case. By highlighting major events (e.g. hospital admissions, surgeries, medication changes) and any unusual gaps in care, the AI ensures no critical event is overlooked during case preparation. #### Identifying Inconsistencies or Gaps Because BastionGPT processes all the content in the records, it can flag when something doesn’t add up. For instance, if one doctor’s note mentions a patient was given Medication X on a certain date but elsewhere the chart lists Medication Y for that day (or notes an allergy to X), the AI can alert the legal team to this discrepancy. Similarly, BastionGPT might notice if a required document is missing – say, a consent form that should accompany a surgical report – and bring that to attention. Catching these inconsistencies or omissions early helps lawyers prepare for opposing arguments and ensures they request any missing documentation well before trial. In essence, BastionGPT acts as a second set of eyes on the medical file, reducing the risk of human oversight in document review. #### Extracting Legally Relevant Medical Findings In cases like personal injury or wrongful death, attorneys need to quickly pinpoint the medical findings that support their legal arguments (e.g. the cause of injury or evidence of malpractice). BastionGPT can be prompted to pull out legally relevant facts from medical records. For example, in a complex neonatal injury case with thousands of NICU notes, an attorney could ask the AI, “Summarize any indications of fetal distress or deviation from standard prenatal care in these records.” The AI will return a focused summary, noting specifics like “At 32 weeks, fetal heart rate decelerations were documented; no follow-up tests were ordered until 36 weeks” , or “Nursing notes on 10/05/2024 indicate the IV dosage was twice the recommended amount.” By distilling the medical jargon into clear points, BastionGPT allows legal professionals to quickly grasp the medical strengths or weaknesses of the case and formulate strategies accordingly. ### 2. Drafting Legal Documents and Case Summaries Legal professionals deal with a large number of documents – from discovery responses and demand letters to case summaries and internal memos. Drafting these from scratch can be time-consuming, especially when they must incorporate complex medical information. BastionGPT serves as a tireless writing assistant, generating well-structured drafts that the legal team can then refine. This not only saves time but also helps ensure consistency and thoroughness in documentation. #### Discovery Responses and Motions When faced with a set of interrogatories or a motion that requires detailed answers based on medical facts, attorneys can use BastionGPT to generate a first draft. Imagine a scenario where a lawyer has a list of questions from opposing counsel (e.g. “Detail the patient’s treatment timeline and any complications following the surgery on March 10.” ). By providing BastionGPT with the relevant medical record excerpts or case notes, the attorney can prompt the AI to draft a comprehensive response. The output might include a point-by-point answer citing the medical facts (dates of surgery, noted complications, follow-up care) in a clear, legally relevant manner. Similarly, for motions or briefs, an attorney could ask BastionGPT to outline arguments based on a set of facts – for instance, a motion for summary judgment in a malpractice case – and the AI will produce a structured draft with sections that the attorney can then fill in with legal citations or adjust for tone. This jump-starts the writing process, turning a daunting blank page into an editable draft in seconds. #### Case Summaries and Client Updates Summarizing a complex case for various audiences is another task suited to AI assistance. Law firm associates often prepare internal case memos or client update letters that distill the latest developments or provide an overview of a case. BastionGPT can generate drafts of these summaries efficiently. For example, after reviewing documents in a medical negligence lawsuit, an attorney could prompt the AI: “Summarize the key facts of this case and the current status in 3-4 paragraphs.” The AI might produce a concise narrative: “Mrs. Doe’s case centers on a surgical sponge left behind during a 2019 operation, leading to infection and a prolonged hospital stay. The key points of negligence identified include… Currently, discovery is nearly complete, and expert testimony from Dr. Smith indicates… The next step is…,” and so on. Such a summary can be used as a basis for a client update letter or a briefing to senior partners. It helps ensure all important details are covered and presented logically. #### Professional Correspondence and Reports Beyond litigation, healthcare legal professionals also handle routine but important communications – like letters to regulatory bodies, compliance reports, or emails to colleagues summarizing findings. BastionGPT can draft these materials quickly. For instance, a compliance officer in a hospital’s legal department might need to report the findings of an internal audit to the executive team. By feeding the AI the audit data or notes, they could ask, “Draft a summary report of our Q4 HIPAA compliance audit findings, including any incidents and corrective actions.” The AI would then generate a well-organized report draft, perhaps bulleting the incidents (e.g. “Two minor PHI disclosure incidents were logged, involving misdirected faxes, which have since been mitigated…” ) and highlighting compliance rates and recommendations. The legal professional can then review this draft, ensure accuracy, and finalize it in a fraction of the time it would take to write from scratch. Likewise, even drafting a letter to an opposing counsel or an insurance company – say, a settlement proposal letter that needs to articulate the medical facts supporting a claim – can be accelerated by providing BastionGPT the key points and letting it formulate a polished, formal letter draft that hits all the necessary notes. ### 3. Ensuring Compliance and Reducing Risk Compliance is a critical concern in healthcare law. Legal professionals must ensure that both their own processes and their clients’ actions adhere to HIPAA, Medicare regulations, state health laws, and more. BastionGPT proves valuable as an assistant that not only operates in a HIPAA-compliant manner itself but also helps legal teams maintain compliance and manage risk in their workflows. Here’s how our healthcare AI can support compliance efforts: #### PHI Redaction and Anonymization Sharing medical documents in legal proceedings or training often requires removing protected health information to maintain privacy. BastionGPT can be tasked with automatically anonymizing or redacting PHI/PII in documents. For example, suppose a lawyer wants to use a real patient’s chart as evidence in court or as a case study in a legal seminar where all patient identifiers should be removed or changed. The attorney can prompt BastionGPT with the document and ask for an anonymized version. The AI will output a copy where names are replaced with initials or generic labels (e.g. “John Doe” becomes “Patient A”), dates of birth and addresses are obscured or removed, and any other identifiers are stripped out. This process takes a fraction of the time of manual redaction and reduces the chance of accidentally leaving in a piece of identifying info. Importantly, because BastionGPT is operating in a secure environment, the original PHI never leaves the compliant platform during this process, ensuring the legal team stays within HIPAA bounds at all times. #### Regulatory Updates and Policy Summaries Healthcare regulations change frequently – whether it’s new CMS rules, updates to Stark Law, OCR guidance on privacy, or state-level healthcare laws. Keeping up with these changes is a constant challenge for compliance attorneys and legal advisors. BastionGPT can help by summarizing dense regulatory documents and even drafting policy updates based on them. For instance, if the government releases a new 30-page update to HIPAA or a revision to Medicare reimbursement policies, a legal professional could feed that text into BastionGPT and ask for a summary of key points and implications relative to their areas of practice. The AI might return a concise briefing: “The new rule expands the definition of telehealth services under Medicare, effective Jan 2025, and requires providers to include X documentation for reimbursement…” etc. Taking it a step further, the attorney could then prompt, “Draft an internal memo to our clinic managers explaining what they need to do to comply with these new rules.” BastionGPT would generate a draft memo or checklist, outlining necessary actions (e.g. “Ensure all telehealth visits have consent documented in EHR per new guidelines.” ). The legal team can then fine-tune this draft. This capability means faster turnaround in responding to regulatory changes – the firm or organization can adapt policies and inform staff promptly, reducing the risk of non-compliance due to oversight or delay. #### Audit Preparation and Documentation Checks Both healthcare providers and insurers face regular audits (for example, HIPAA compliance audits, billing audits by Medicare, or internal quality audits). Legal teams often assist in these audits by reviewing documentation to ensure everything is in order. BastionGPT can dramatically reduce the grunt work in audit prep by reviewing large sets of documents against specified requirements. For instance, a hospital’s legal compliance officer could use BastionGPT to scan patient files to help verify each contains the required forms (admission consent, HIPAA notice acknowledgment, physician orders, etc.). The AI can be asked to generate a report of any files that are possibly missing components or have inconsistencies. As an example, you could prompt BastionGPT with a batch of 10 discharge summaries and related documents, asking it to check each for the presence of a signed patient consent form and a HIPAA privacy notice. The AI will analyze each file and might output a summary like: “48 files have all documents in place; 2 files are missing a HIPAA notice acknowledgment (Patients X and Y).” Armed with this information, the legal team can ensure those two exceptions are scrutinized. Similarly, for an insurance claims audit, BastionGPT could review claim files and flag ones where, say, a required physician certification is potentially absent. By using healthcare AI to perform an initial compliance sweep, organizations can catch and fix issues proactively – before an external auditor does – thereby helping to avoiding penalties, fines, or legal exposure. ### 4. Supporting Insurance Claims and Appeals Attorneys and legal professionals involved in healthcare insurance claims – whether working for providers, patients, or insurers – also benefit from generative AI assistance. These cases involve extensive documentation, policy details, and often the need to argue medical necessity or coverage. BastionGPT can lighten the load by analyzing claim information and drafting well-supported arguments for appeals or defenses. Consider these applications: #### Drafting Appeal Letters for Denied Claims When an insurance company denies a claim for a medical procedure or treatment, legal experts must often prepare a detailed appeal letter that lays out why the decision should be reconsidered. This usually involves referencing the patient’s medical records, the policy’s terms, and medical necessity criteria. BastionGPT can expedite this process significantly. For example, suppose a patient’s claim for a costly cancer treatment was denied by the insurer as “not medically necessary.” An attorney or advocate can provide BastionGPT with the pertinent details – the denial letter, relevant excerpts from the medical records (diagnoses, prior treatments, outcomes), and the insurance policy language – and ask the AI to draft an appeal letter. In seconds, BastionGPT will produce a structured letter that might say: “We are writing to appeal the denial of coverage for [Treatment X] for Patient A. As the medical records indicate, Patient A has undergone standard therapies which were ineffective (see Dr. Smith’s notes, 08/12/2024), and [Treatment X] is recommended as the medically necessary next step. According to [InsuranceCo]’s policy section 4.3, treatments are considered necessary when… In Patient A’s case, all criteria are met: … Therefore, we respectfully request a reconsideration of coverage.” This draft provides a strong starting point, ensuring that no key detail or argument is missing. The legal professional can then refine tone or add any case-specific citations, rather than writing the whole letter from the ground up. The result is a faster, more efficient appeal process, potentially leading to quicker resolutions for patients and providers. #### Summarizing Claims and Policy Details In complex insurance disputes or healthcare reimbursement litigation, there may be voluminous records including claim forms, correspondence, policy manuals, and medical documentation. BastionGPT can help lawyers make sense of this complexity by summarizing and organizing the information. For instance, an attorney handling a healthcare insurance fraud case could use the AI to summarize each claim in question: “Summarize Claim #12345: what was billed, what was paid or denied, and any reasons given.” The AI would parse through claim data and output a clear summary (e.g. “Claim 12345 – billed $X for 10 sessions of Physical Therapy; insurer paid $Y for 5 sessions and denied the rest, citing lack of pre-authorization. Communications: Jan 5 letter requesting additional records, Feb 2 denial letter referencing policy 10.2…” ). Having a quick reference like this for dozens of claims allows the legal team to better spot patterns (maybe all denials were related to a specific code or provider) and develop a strategy. Moreover, if the case involves interpreting dense insurance policy documents (which can be as convoluted as legislation), BastionGPT can be directed to answer questions or summarize sections of the policy relevant to the dispute. For example: “What is the exclusion criteria for ‘experimental treatments’ in this policy?” BastionGPT, having been given the policy text, could respond with a distilled explanation of that clause. This kind of on-demand clarification is incredibly useful during litigation prep or negotiations, ensuring the legal team fully understands the fine print without needing to manually parse every paragraph. ### 5. Enhancing Legal Research and Knowledge Access Beyond case-specific documents, healthcare legal professionals constantly refer to a body of knowledge – statutes, regulations, prior case files, internal guidelines, and more. BastionGPT can serve as an intelligent research assistant when provided with these reference materials, making it easier to retrieve information and get answers quickly during the course of a case or compliance project. Here’s how BastionGPT helps on this front: #### Instant Q&A from Uploaded References BastionGPT can be loaded with your relevant legal texts or past documents (for example, you might upload OCR-scanned PDFs of healthcare regulations, accreditation standards, or even a database of past case summaries from your firm). Once this information is in the system, your team can query the AI in natural language instead of manually searching through binders or PDFs. For instance : an in-house counsel could upload the text of the EMTALA law or a state’s health code, then ask, “What are the reporting requirements under this law if a data breach occurs?” If that information is in the uploaded documents, BastionGPT will provide a possible answer drawn from the text (e.g. summarizing the section that mandates notification to patients and HHS within a certain timeframe). This turns what might be a 30-minute search through dense legalese into a quick answer in seconds. It’s like having a smart legal librarian (though one that can occasionally make mistakes) at your fingertips. During fast-paced litigation or compliance reviews, this capability means you can double-check rules, guidelines, or past case facts immediately. ## Save Time with Generative AI Generative AI has moved beyond hype into truly practical benefits for those in healthcare law . By embracing these use cases (and undoubtedly discovering more of your own), legal professionals have streamlined their workflows while maintaining the high standards of confidentiality and compliance that the healthcare industry demands. The end result is a more agile and productive practice that can deliver better outcomes for clients, whether it’s winning cases, ensuring compliance, or expediting claims. From the law office to the courtroom (and everywhere in between), BastionGPT is poised to transform how healthcare legal professionals work. The use cases highlighted – medical record summarization, document drafting, compliance support, insurance claim analysis, and knowledge retrieval – demonstrate that many routine, time-intensive workflows can be improved or accelerated with AI assistance. Importantly, these enhancements directly contribute to outcomes that matter for legal teams and their clients: greater efficiency in case preparation, lower risk of overlooking crucial facts, reduced turnaround time for documentation, and stronger compliance oversight. By offloading tedious and repetitive tasks to BastionGPT, attorneys and legal staff can focus on higher-value activities such as case strategy, client interaction, negotiation, and substantive legal analysis. ## Get Started with BastionGPT With BastionGPT, law firms and healthcare organizations can quickly integrate AI into workflows that involve PHI/PII, knowing that sensitive data remains protected. This opens the door to leveraging healthcare AI in ways that were not possible with consumer-grade chatbots. Incorporating your organization’s specific data (e.g. hospital protocols, litigation templates) into BastionGPT’s prompts makes its outputs even more tailored and relevant. Of course, maintaining human oversight – with attorneys reviewing and finalizing AI-generated content – ensures that the use of AI remains an aid to professional judgment, not a replacement for it. For legal teams in the medical field looking to stay ahead in an evolving landscape, deploying BastionGPT’s HIPAA-compliant AI assistant and AI scribe is a strategic move – one that can reinforce everything from day-to-day productivity to long-term competitive advantage. Most organizations are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/healthcare-administration-ai-use-cases Healthcare Innovation # How Healthcare Administrators Can Use AI Written By: Robert Henehan • June 25, 2025 Healthcare administrators are increasingly leveraging BastionGPT's generative AI to streamline operations across patient intake, billing, document management , and more. From front desk coordinators and billing specialists to operations managers and finance directors, administrative staff can use tools like BastionGPT’s secure AI assistant to summarize complex documents, draft articulate communications, and extract insights from data. The result is faster paperwork with fewer errors, and this is all achieved while maintaining strict healthcare compliance and privacy. In the sections below, we explore how you can use AI in healthcare administration can apply BastionGPT across key workflows, including patient intake, document processing, billing & reimbursement, internal communications, strategic reporting, and standard operating procedure (SOP) development. Each section highlights practical use cases and includes a best-practice prompt example to illustrate how healthcare admins can use BastionGPT to enhances productivity in a secure, compliant manner. ## 18 Benefits of Using HIPAA-Compliant AI for Healthcare Administration ## Document Review and Processing Healthcare administrators deal with mountains of paperwork, from insurance policies and regulatory guidelines to internal reports and scanned records. BastionGPT serves as a tireless assistant for reviewing, summarizing, and extracting data from these documents: #### Reviewing and Summarizing Long Documents Whether it’s a new payer policy manual, a 100-page vendor contract, or a stack of clinic operational reports, BastionGPT can read and condense long texts into key takeaways. For instance, an operations manager could feed a 50-page Medicare policy update along with their existing policy into the AI and get back a summary of the potentially important changes in coverage criteria and billing rules. This allows administrators to stay informed on crucial details without hours of reading. Ex. You are a healthcare operations manager reviewing a policy update document. Compare this new policy against our current billing procedures and create a summary highlighting coverage criteria changes, new billing rules, and specific operational adjustments we must implement to maintain compliance and maximize reimbursement. #### OCR and Data Extraction from Scans Important documents often arrive as scans or faxes, think physician orders, lab results, or handwritten notes. Because BastionGPT supports OCR, admins can upload scanned PDFs or image files and have the AI extract the text and organize the information. For example, if you receive a faxed referral form, BastionGPT can convert it to text and pull out fields like patient info, referring doctor, and requested service. It can even alert you if something is illegible or missing on the form. This capability is especially useful for processing paper intake forms or insurance cards, turning them into usable digital data. Ex. You are a records administrator processing scanned files. Extract patient information from this batch of uploaded scanned referral forms. Generate a structured list of key data fields (e.g., patient name, referring provider, requested service) and flag any missing or unreadable fields. #### Flagging Missing Information or Compliance Gaps BastionGPT can act as a second set of eyes on documentation to better ensure completeness and compliance. For example, compliance or audit staff might upload medical records to see if each contains all required elements. BastionGPT's Clinical AI could rapidly scan records and report that “2 of the following notes improperly recorded use of restraints” or “1 file is missing a consent form”. Catching these omissions in minutes allows the team to fix issues before audits or billing submissions. Similarly, if a document set has inconsistent data (say, a patient’s name spelled differently in two places), the AI can often flag it for correction. This proactive review reduces the risk of errors that could lead to compliance violations or claim denials. Ex. You are a front desk coordinator reviewing a new patient packet. Summarize the intake forms including medical history, allergies, and insurance information. Highlight any missing signatures or unanswered questions that require follow-up before the visit. ## Billing and Reimbursement Medical billing and revenue cycle management involve complex coding, stringent documentation requirements, and frequent communication with insurers. BastionGPT assists billing specialists and revenue cycle managers by speeding up documentation checks and correspondence in the reimbursement process: #### Assisting with Coding and Claims Preparation Turning clinical documentation into properly coded insurance claims can be labor-intensive. BastionGPT can analyze a provider’s notes or an order form and suggest relevant billing codes (e.g. ICD-10 diagnostic codes, CPT/HCPCS procedure codes) based on the described services. It also helps verify that each claim has all necessary information before submission. For example, checking that the physician’s NPI, patient demographics, and pre-authorization numbers are included on a claim form. By catching omissions or mismatches in advance, the AI reduces denials due to clerical errors. Ex. You are a medical coder preparing a claim. Analyze this provider's clinical notes and suggest relevant ICD-10 and CPT codes. Generate a checklist of required claim details (e.g., NPI, diagnosis code, authorization number) and flag any missing information. #### Summarizing EOBs and Payer Communications Admin staff often must interpret Explanation of Benefits (EOB) statements and insurance letters to reconcile accounts. BastionGPT can read through a stack of EOBs or denial letters and summarize the outcomes in plain language. For instance, it might output: “Claim 12345 – Billed $850 for MRI; Insurance paid $600, patient owes $250; denial reason: deductible not met”. Compiling these summaries across multiple claims helps billing managers quickly spot patterns, such as a particular procedure always being partially paid or a specific denial code recurring, so they can address underlying issues. Ex. You are a billing analyst reviewing payer communications. Summarize these Explanation of Benefits (EOBs) and denial letters into plain language, showing total billed, paid amount, patient responsibility, and denial reasons. Highlight any patterns or recurring denial codes. #### Drafting Appeal Letters for Denied Claims When an insurer denies a claim or underpays on a reimbursement, time is of the essence to file an appeal with a strong argument. BastionGPT significantly speeds up this process by drafting detailed, well-structured appeal letters. A billing specialist can provide the AI with key inputs such as the denial notice, relevant patient record excerpts (e.g. diagnosis and treatment history), and the payer’s policy language and ask for an appeal draft. In seconds, BastionGPT will produce a letter that cites the medical necessity facts and references the insurer’s own policy terms to argue why the service should be covered. The specialist can then review and customize the tone or add any required references before submitting. Using AI in this way helps the billing specialist to check that no important detail is missed in the appeal, and it often leads to quicker reversals since the letter directly addresses the insurer’s criteria. Ex. You are a billing specialist responding to a denied claim. Draft a persuasive appeal letter based on this denial notice and supporting clinical documentation. Clearly explain the medical necessity and reference the relevant payer policy language. ## Internal Communications Daily internal communication is another area where generative AI boosts productivity for healthcare administrators. Writing routine emails, announcements, and summaries can eat up a lot of time, but BastionGPT helps generate these materials quickly in a clear, professional tone: #### Email and Memo Drafting Whether it’s informing staff about a new policy, composing a meeting invitation, or sending out a quick update, BastionGPT can draft the initial message for you. An operations manager might say, “Draft an email to all staff about our updated patient intake protocol and its start date next month.” The AI will produce a polished, properly formatted email in seconds, covering all key points (what the new protocol is, why it’s changing, effective date, and any actions required by staff). The administrator then just reviews, adds any personal touches, and sends it. This provides consistency in communication and saves time on writing from scratch. Ex. You are an operations manager drafting internal announcements. Write a staff-wide email announcing an upcoming change to the patient intake protocol, including what’s changing, the implementation date, and any staff training required. #### Meeting Transcripts and Summaries By leveraging BastionGPT’s transcription abilities, admins can record meetings or calls and get summaries of what was discussed, even if it includes PHI. For example, after a departmental meeting, you can ask the AI to transcribe the audio and highlight the main decisions and action items. The output might be a concise summary: “Team Meeting Summary (May 5) – Topics covered: patient scheduling backlog, new billing software rollout. Decisions: extend clinic hours on Tuesdays; hire 1 temp for billing. Action Items: [Name] to draft schedule; IT to train staff on software by June 1.” This kind of autogenerated recap helps even those who couldn’t attend are quickly informed of important outcomes. It also provides a written reference for everyone, reducing miscommunication. Ex. You are a department head summarizing a meeting. Transcribe and summarize this recorded team meeting, highlighting key discussion points, decisions made, and individual action items. #### Instant Knowledge Retrieval (Internal Q&A) Healthcare organizations have extensive internal knowledge bases – policy manuals, SOP documents, HR guidelines, payer contract terms, etc. Instead of manually searching through SharePoint or binders, staff can use BastionGPT to ask questions in plain language and get answers sourced from those documents. For instance, a finance officer could upload the clinic’s contract with a payer and then ask, “What is the reimbursement rate for a Level 3 office visit under this contract?” The AI, having ingested the contract text, can respond with the specific rate and relevant terms. Similarly, an office manager could query, “According to our SOP, what steps should be taken when a patient requests their medical record?” and BastionGPT will provide the answer with reference to the SOP. This turns internal documentation into a quick Q&A resource, saving employees from digging through PDFs – and because the information comes from the organization’s own vetted documents, it remains accurate and consistent. (It’s worth noting that while BastionGPT can function like an internal chatbot for staff, all guidance is still based on your provided documents, and staff should use judgment in interpreting the AI’s responses.) Ex. You are an administrator retrieving policy info. Based on the uploaded SOP and other documents, answer this staff question: 'What are the steps for processing a patient request for medical records?' Provide a clear, sourced response. ## Strategic Reporting and Analysis High-level reporting and analysis tasks – such as compiling quarterly performance reports, financial analyses, or quality improvement summaries – can also be accelerated with generative AI. Healthcare administrators in operations and finance roles use BastionGPT to turn raw information into structured narratives and insights: #### Drafting Executive Reports and Presentations Senior leaders often need to present complex information (clinical outcomes, financial metrics, operational benchmarks) in a digestible format for stakeholders or board members. BastionGPT can take in multiple source documents or data summaries and produce a cohesive draft report. For example, a hospital CFO could prompt, “Draft a Q3 performance report summarizing key metrics – patient volumes, revenue vs. budget, notable operational changes – and highlight any important trends or issues.” They could then add an example report for the AI Assistant to use as a template. The AI could generate a multi-paragraph executive summary that the CFO then fine-tunes. This approach helps them validate that critical points are covered and saves hours of writing. Ex. You are a hospital CFO creating a quarterly performance report for board presentation. Using the provided data summaries, draft an executive summary covering patient volumes, revenue vs. budget, and operational changes, highlighting key trends and critical issues that require board attention. Here is an example to use as a template. #### Compiling Quality Improvement (QI) Project Results When administrators undertake initiatives like reducing readmission rates or improving patient satisfaction, documenting the outcomes is essential. BastionGPT can help by drafting quality improvement reports that include the project’s objectives, interventions implemented, data collected, and results. For instance, a quality manager could input notes and data from a project (e.g. a 6-month effort to cut medication errors) and ask for a summary report. The AI might produce a structured draft: “Project Goal: reduce medication errors by 50%. Interventions: implemented double-check protocol, staff training… Results: errors decreased 45% over 6 months (from 20 to 11 incidents). Key Findings: most errors were ordering mistakes; after intervention, ordering errors fell significantly…” . Such a draft provides a solid starting point, which the manager can then refine with specific numbers or charts. Ex. You are a quality manager documenting a 6-month medication error reduction initiative. Create a structured QI report including project objectives, interventions implemented, data results, and key findings with specific recommendations for sustaining improvements. #### Summarizing Regulatory and Payer Policy Changes Keeping up with ever-changing healthcare regulations and insurance policies is a constant challenge. BastionGPT can alleviate this by digesting lengthy policy documents and outlining the actionable points. Suppose Medicare releases a new 30-page rule or a major payer updates its provider agreement. An admin can input that text and ask, “Summarize the key changes in this document and how they affect our billing or operations.” The AI will produce a concise summary, for example: “ New Policy Highlights: effective Jan 2025, telehealth visits are reimbursed at parity with in-person (was 80% before); prior auth is now required for MRI after 1st visit; documentation must include X, Y, Z…”. Armed with this summary, the administrator can quickly understand the impact and even have BastionGPT draft a brief internal memo or an FAQ for staff explaining the changes. This ensures the organization responds promptly to external changes while staying compliant. Ex. You are a practice administrator whose primary insurance payer just released updated prior authorization requirements affecting multiple specialties. Analyze this policy document to identify which services now require pre-approval, extract the new submission timelines and required documentation, then draft a staff training outline explaining how these changes impact our referral and scheduling workflows. ## SOP Development and Training Materials Developing Standard Operating Procedures (SOPs) and training content is another workflow where generative AI proves invaluable. Healthcare admins must frequently create or update SOP documents to reflect new regulations, technologies, or process improvements. BastionGPT acts as a smart drafting assistant for these needs: #### Creating and Updating SOP Documents Rather than writing an SOP from scratch, administrators can input an outline or even a rough description of the workflow, and let BastionGPT generate a structured draft. For example, if an operations manager needs a new SOP for patient telehealth check-ins, they could list the steps or key points (e.g. verify patient identity, confirm consent, test video/audio, document visit in EHR) and ask the AI to flesh it out. BastionGPT can produce a formal SOP document with sections, step-by-step instructions, and even notes on compliance where applicable. In one potential scenario, a hospital administrator can use BastionGPT to draft updated infection control protocols (an SOP for staff) – incorporating new sanitation procedures, PPE usage guidelines, and patient isolation steps – which the infection control team then reviewed and approved. This shows how AI can ensure all necessary details are included in the draft, from regulatory requirements to best practices, while saving considerable writing time. Ex. You are a healthcare operations manager developing a new SOP for this process. Using the provided workflow outline or key process steps, create a comprehensive SOP document with numbered procedures, assigned staff responsibilities, timing requirements, compliance considerations, and quality checkpoints to ensure consistent implementation across the organization. #### Ensuring Compliance in Policies When writing policies or SOPs, it’s critical to align with the latest regulations and accreditation standards. Administrators can feed BastionGPT the relevant guidelines (for example, OSHA safety standards or Joint Commission requirements) alongside their draft policy and prompt the AI to check for any missing elements. The AI might suggest, for instance, “ Include a section on annual staff training per OSHA requirement X ” or incorporate required language about patient rights. BastionGPT’s ability to cross-reference multiple documents helps compliance officers produce thorough, compliant procedures, reducing back-and-forth edits. Additionally, because BastionGPT keeps PHI/PII secure, even if your SOP includes some real-world examples or incident data, you can safely include that in the prompt without privacy concerns. Ex. You are a healthcare compliance officer reviewing a draft policy or SOP for regulatory alignment. Using the provided regulatory guidelines and accreditation standards, analyze the draft document to identify missing compliance elements, suggest required language additions, and recommend specific sections that need enhancement to meet current regulatory requirements and industry best practices. #### Developing Training and Onboarding Materials Beyond formal SOP documents, BastionGPT can help create training guides, checklists, and educational content for staff. Trainers might use it to draft a quick-reference guide or a presentation outline based on the SOP. For example, “ Create a summary of our new patient data privacy SOP that can be used in a staff training handout .” The AI would generate a digestible overview (covering key rules like handling of PHI, proper use of email, etc.), which can then be refined and distributed to employees. This supports consistent messaging in training. Moreover, BastionGPT can generate practice scenarios for onboarding – since it’s a generative model, you could ask it to produce fictitious but realistic examples (with no real patient data) for trainees to work through. You could even use real examples and have them modified or anonymized thanks to the HIPAA compliance. For instance, “ Give an example scenario of a front-desk error in patient intake and a correct approach, for training discussion .” The AI might create a short story of a scenario, which training leads can use for role-play or quizzes. This makes training more engaging and practical for new hires, while avoiding exposure of any real patient’s information. Ex. You are a healthcare training coordinator creating educational materials for staff onboarding. Using the provided SOP or policy document, develop [training guide/quick-reference handout/presentation outline] that translates complex procedures into digestible content for new employees, including practice scenarios and key compliance points to ensure consistent understanding and implementation. ## Patient Intake and Coordination Efficient patient intake and care coordination are vital administrative tasks. BastionGPT helps front-desk staff and patient coordinators collect and distribute information more effectively, reducing manual data entry and ensuring nothing falls through the cracks: These intake gains are especially valuable in high-volume settings—see how AI supports urgent care and walk-in clinics where patients arrive without appointments. #### Summarizing Intake Forms & Documents New patient packets often include multi-page questionnaires, consent forms, and insurance cards. Rather than reading these line-by-line, staff can have BastionGPT generate a concise summary of a patient’s intake form (e.g. key demographics, medical history highlights, allergies) and flag any missing information or signatures that need follow-up. This saves time and helps to ensure critical details aren’t overlooked. Ex. You are a medical receptionist processing patient forms. Generate a structured summary from this multi-page intake packet, highlighting key demographic info, allergies, medications, and missing signatures. Include a note if anything needs to be clarified before the visit. #### Coordinating Referrals and Follow-ups Patient coordinators can use generative AI to streamline communications between departments. If a patient needs a referral to a specialist or a follow-up appointment after a hospital discharge, BastionGPT can help draft the necessary communications. For example, it might draft an email to a specialist’s office summarizing the patient’s case and referral reason, or produce a checklist of steps to coordinate a complex care plan. By quickly generating these documents (which the coordinator can tweak as needed), the AI helps the coordinator to all relevant information is communicated clearly and consistently. Ex. You are a care coordinator preparing a referral message. Draft a professional referral email to a specialist’s office that includes the patient’s diagnosis, reason for referral, relevant clinical findings, and any instructions or next steps. Include a checklist of follow-up items needed to complete the referral process. Details to add: [Insert diagnosis, referral reason, clinic name, discharge date, required documentation, and any patient-specific instructions] ## Can Healthcare Admins Use AI and Maintain HIPAA Compliance? Crucially, these benefits listed above are achieved without compromising privacy or control when using a HIPAA-compliant AI assistant like BastionGPT. Because BastionGPT is a HIPAA-compliant solution engineered for healthcare, organizations can integrate AI into PHI/PII-intensive workflows knowing that sensitive data remains protected – a capability not possible with consumer-grade AI chatbots. And by keeping humans in the loop for oversight and final approvals, BastionGPT ensures that AI remains an aid to decision-making, not a replacement for professional judgment. Administrators still review every AI-generated output, maintaining accountability and adherence to the nuances of their facility’s standards. ## Conclusion Generative AI is proving to be a game-changer for healthcare administration. By embracing tools like BastionGPT in workflows ranging from front-desk intake to executive reporting, administrative teams are dramatically improving efficiency, reducing routine workload, and strengthening accuracy and compliance. Tasks that once took hours such as reading dense documents, transcribing notes, and drafting letters can now be completed in minutes with AI assistance, freeing staff to focus on higher-value activities such as improving patient experience, coordinating care, and strategic planning. The productivity gains are tangible: faster turnaround times, lower error rates, and more streamlined processes across the board. Crucially, these benefits are achieved without compromising privacy or control. Because BastionGPT is a HIPAA-compliant solution engineered for healthcare, organizations can integrate AI into PHI/PII-intensive workflows knowing that sensitive data remains protected – a capability not possible with consumer-grade AI chatbots. And by keeping humans in the loop for oversight and final approvals, BastionGPT ensures that AI remains an aid to decision-making, not a replacement for professional judgment. Administrators still review every AI-generated output, maintaining accountability and adherence to the nuances of their facility’s standards. In summary, healthcare administrators use BastionGPT to augment their productivity in a secure, compliant manner. Whether it’s summarizing a 100-page policy, drafting an appeal letter, transcribing a meeting that may include patient information, or writing a new SOP, the generative AI assistant accelerates the task while the human expert guides and validates the results. The end outcome is a leaner, more responsive administrative operation that can adapt quickly to changing needs and regulations which ultimately enable better support for clinicians, patients, and the healthcare organization as a whole. Adopting the HIPAA-compliant AI BastionGPT is a strategic move for any healthcare administration team looking to save time, reduce burnout, and innovate safely in the age of AI. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/healthcare-ai-comparison Healthcare Innovation # Healthcare AI Comparison • January 23, 2024 We have compared Microsoft 365 Copilot , OpenAI Enterprise and BastionGPT as the top healthcare compliant solutions to help our users make more informed decisions. We have not included solutions that utilize lower quality AI than GPT-4, that aren’t yet on the market, or that don’t offer HIPAA compliance. Microsoft 365 Copilot information was taken from its publicly posted feature and pricing information. Pricing of $62 per user per month is based on the $30 monthly Copilot license fee. However, Copilot can only be purchased if the account also has a sufficient license, of which E3 is the lowest priced at $32 per month. The purchase of Copilot requires a minimum purchase of 300 licenses. This information was taken from published information from Microsoft and is accurate as of this publication. OpenAI Enterprise information was taken from its publicly posted information. However, OpenAI does not officially post pricing information as of this article’s publication. As such, we have utilized 3 rd party pricing information posted on the internet, which indicates a $100,000 minimum annual commitment and a common $250/user cost. BastionGPT information was sourced from official feature and pricing information on their website. For related comparisons, see how tools differ in their word limits across generative AI , read why we switched to BastionGPT , and review which is the best AI scribe . ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/healthcare-ai-prompt-library-guide Healthcare Innovation # Healthcare AI Prompts: Building Your Clinical Library Written By: Robert Henehan • April 16, 2025 BastionGPT's Prompt Library feature offers healthcare professionals a powerful way to save and reuse carefully crafted medical and therapy prompts with just a slash, turning occasional AI assistance into a seamless part of your daily workflow. This HIPAA-compliant solution is transforming how medical and mental health professionals interact with healthcare AI, making documentation, therapy sessions, and patient care more efficient than ever. If you are currently using BastionGPT you can click this link to download a set of prompts for your library. ## What is BastionGPT's Healthcare AI Prompt Library? BastionGPT's Prompt Library allows you to save and organize your most effective healthcare AI prompts for instant reuse. Rather than rewriting or copying and pasting the same prompts repeatedly, you can simply type a slash command to access your saved medical prompts, streamlining your workflow and supporting consistency across all your AI-generated clinical content. ## Why Healthcare Professionals Need a Medical Prompt Library Healthcare documentation requires precision, consistency, and compliance. By creating and saving well-structured healthcare AI prompts, you can: Save valuable time on repetitive documentation tasks Support consistency across patient notes and communications Align with medical documentation standards Help reduce errors by using pre-vetted, effective medical prompts Share best practices with colleagues in your practice ## Building Your Customized Healthcare Prompt Library Unlike other AI solutions where prompts must be copied and pasted from external documents, or can be difficult to modify, BastionGPT's Healthcare AI Prompt Library integrates directly into your workflow. Your carefully crafted prompts are immediately available whenever you need them. This seamless integration turns your Healthcare Prompt Library into: A flexible template system for clinical notes and psychological assessments A repository of letter formats for referrals and communications A collection of educational content frameworks for patient guidance A resource for standardizing documentation across your medical practice ## Best Practices for Building Your Prompt Library To maximize the value of your prompt library, consider these strategies: Curate from trusted sources : Adapt prompts from respected clinical journals, professional organizations, and evidence-based practice guidelines Collect effective examples : When you find high-quality prompts online or from colleagues, save them to your library after customizing for your practice Categorize systematically : Organize prompts by type (assessment, treatment, education) and specialty for easy retrieval Continuously improve : Regularly update prompts based on clinical outcomes and feedback Collaborate with peers : Share and exchange effective prompts with trusted colleagues to expand your library By maintaining a curated collection of proven prompts from various sources, you'll create a valuable resource that grows with your practice and reflects current best practices in your field. ## Real-World Healthcare and Psychology Applications ### Medical Documentation Save prompts for different types of notes: Initial assessments and intake forms Follow-up visit documentation Procedure notes and surgical reports Discharge summaries and care plans ### Mental Health Documentation and Therapy Resources Create specialized psychology and therapy prompts for: Psychological evaluations and assessments Therapy session notes (CBT, DBT, ACT frameworks) Treatment planning documentation Progress monitoring reports Client reflection exercises for between-session work Therapeutic homework assignments and worksheets Guided meditation and mindfulness exercise scripts Mood tracking and symptom monitoring templates ### Patient Communications Create templates for: Referral letters to specialists Patient instructions and health education Follow-up emails and appointment reminders Medication explanations and adherence support ### Administrative Tasks Streamline office communications with healthcare AI prompts for: Insurance documentation and medical billing support Staff training materials and protocols Patient education handouts on various conditions Policy documentation and compliance reporting ## HIPAA Compliance: Security You Can Trust Unlike generic AI platforms, BastionGPT is built specifically for healthcare with HIPAA compliance at its core. Your medical prompt library and all patient information remain protected within a secure environment designed to address healthcare's stringent privacy requirements. Getting Started with Your Customized Healthcare Prompt Library Identify repetitive tasks in your clinical workflow that could benefit from saved medical and therapy prompts Craft detailed healthcare prompts following best practices outlined below Create therapeutic exercises for clients to use between sessions Organize your prompt collection by therapeutic approach, client issues, or session type Save effective prompts to your library for future use Access with ease using slash commands or clicks during patient care and therapy sessions Refine over time as you discover what works best for your specific clinical and therapeutic needs For step-by-step instructions, see our support article on how to save prompts in BastionGPT . ## Creating Effective Healthcare and Therapy Prompt Libraries The most effective prompts for healthcare and mental health settings include several key elements that can be organized into comprehensive therapy prompt libraries. Here's how to build robust therapy journal prompts and clinical templates: ### 1. Define the Professional Role Specify the healthcare role perspective the AI should adopt: "You are a psychiatrist conducting an initial assessment..." "You are a therapist documenting cognitive behavioral therapy progress..." "You are a primary care physician creating a comprehensive care plan..." "You are a mental health counselor creating therapy journal prompts for clients..." ### 2. Specify the Purpose Clearly state what you're trying to accomplish: "Create a discharge summary for a patient following orthopedic surgery" "Generate a psychological assessment report following standard protocols" "Draft a therapy progress note using the SOAP format" ### 3. Include Structural Elements Define the format and components needed: SOAP note structure with specific clinical sections Mental health assessment templates with DSM-5 criteria Educational materials with appropriate reading level for patients ### 4. Provide Examples Include samples of your preferred output style: Specific headings and subheadings for clinical documentation Typical phrasing you use in psychological evaluations Preferred spacing and organizational structure for medical reports ### 5. Effective Prompt Frameworks Try these other proven prompt structures as well: RTF Framework : Role (specify who AI should be) – Task (what to do) – Format (how to present) TAG Framework : Task (define the work) – Action (state the process) – Goal (clarify the purpose) BAB Framework : Before (explain the problem) – After (state desired outcome) – Bridge (how to get there) CARE Framework : Context (background information) – Action (what to do) – Result (expected outcome) – Example (sample to follow) ## Why Choose BastionGPT? BastionGPT's Healthcare AI Prompt Library transforms the way medical and mental health professionals interact with AI. By creating a personalized collection of effective medical prompts, therapeutic worksheets, and clinical documentation frameworks, you can save time while supporting consistency and compliance across all your healthcare documentation. Your customized prompt library becomes an invaluable resource that grows with your practice, allowing you to quickly generate tailored content for different patient needs while maintaining your unique professional voice. The investment in creating quality healthcare prompts pays dividends daily, helping you focus more on patient care and less on paperwork. You can begin building your comprehensive healthcare prompt library today and experience how BastionGPT's healthcare-focused AI can support your practice efficiency while maintaining the highest standards of privacy and security. Start your free 7-day trial of BastionGPT If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/healthcare-ai-scribe-documentation-guide Healthcare Innovation # A Guide to AI Scribes in Healthcare Written By: Robert Henehan • December 12, 2024 Healthcare providers dedicate their lives to patient care, yet many find themselves spending more time on documentation than with patients. While digital tools like electronic health records aimed to streamline workflows, they've often added layers of complexity to an already challenging situation. Even with dedicated support staff, healthcare professionals often spending evenings and weekends catching up on documentation instead of recharging or spending time with family. This growing administrative burden has fueled burnout across the industry, prompting a search for solutions that actually give providers their time back. That’s where AI medical scribes come in. Powered by advances in generative AI, these tools assist with real-time documentation during patient encounters, helping clinicians capture accurate notes without sacrificing face-to-face interaction. If you are comparing options, see our breakdown of which is the best AI scribe . As healthcare organizations begin to adopt AI-driven documentation support, an important question emerges: how can this technology improve efficiency while preserving the trust, privacy, and clinical standards that define quality care? ## What is an AI Medical Scribe? An AI medical scribe is a software tool that uses advanced speech recognition and generative AI to automatically document clinical encounters in real time or after a meeting. It listens to the conversation between a provider and patient and converts that interaction into structured medical notes and therapy notes in various formats. Unlike traditional human scribes, AI scribes can quickly organize information like symptoms, diagnoses, treatment plans, and follow-ups. Many tools also allow providers to review, edit, and approve notes before they are finalized, ensuring accuracy and clinical control. The goal is to reduce the administrative load while maintaining clinical judgment. By handling much of the documentation process, AI medical scribes help providers stay focused on patient care while improving efficiency and consistency in recordkeeping. ## Understanding How AI Scribes Works Think of AI Scribes as a collaborative copilot in your documentation process. The system works alongside you during patient encounters, helping to capture clinical information and create drafts of structured notes based on your style of writing and preferred formatting. This approach helps reduce time spent on administrative tasks while maintaining your control over all clinical documentation. When you work with BastionGPT, you'll find it supports various documentation needs: Creating initial drafts of progress notes, SOAP notes and DAP notes Helping structure consultation summaries Supporting the development of referral letters Organizing clinical information in your preferred format Providing analysis of patient clinical information Improving coding and helping with the revenue cycle and billing processes Delivering recommendations for care and treatment plans Many providers report positive outcomes when implementing these tools thoughtfully. For example, primary care practices often find they could redirect 2+ hours per day from documentation to patient care. A multi-provider specialty clinic reported that providers generally spent about 35% less time on documentation tasks after implementing AI support tools. ## How Drafts Become Final Documentation AI Scribes operate as a collaborative partner in your documentation workflow. The system drafts, organizes, and structures clinical information so you can stay focused on the patient in front of you. You stay in control of what enters the medical record. Every AI-generated document passes through a review-and-approve step before it becomes part of the official chart. That step is built into the workflow on purpose. It keeps the provider as the decision-maker on every note, summary, and treatment plan, and it maintains a clear line of clinical accountability. ## Understanding AI Scribe Capabilities and Limitations When implementing AI documentation support, it's important to understand both its potential benefits and inherent limitations. Let's explore how healthcare providers can work effectively with these tools while maintaining high standards of clinical care. ### Common AI Scribe Documentation Challenges AI scribe tools can sometimes face challenges similar to those encountered by human scribes or medical students. Understanding these potential issues helps providers implement appropriate oversight and quality control measures: #### Transcription Variations : Medical terminology presents unique challenges for AI systems. For example, similar-sounding terms like "hypertension" and "hypotension" might occasionally be confused. While our AI models and transcription services have been tailored to healthcare terminology, your clinical expertise helps catch and correct these variations during review. #### Context and Nuance : While AI tools excel at capturing spoken information, they may miss important non-verbal elements of patient encounters. A patient's body language, emotional state, or subtle clinical signs rely on your professional observation and documentation during review. #### Information Accuracy : AI systems sometimes exhibit what we call "hallucinations" - adding information that wasn't part of the original encounter. Your review process helps ensure all documented information accurately reflects the actual patient interaction. ## Putting AI Documentation in Context Current research into traditional medical documentation suggests that manual entry in electronic health records can lead to various documentation challenges. Studies indicate that patients who review their records often identify discrepancies, highlighting the importance of thorough documentation review regardless of the tools used. BastionGPT's tools aim to support improved documentation quality through: Structured formats that promote completeness Consistent organization of clinical information Tools for efficient review and correction Integration with existing clinical workflows ### Why Healthcare Provider Oversight Makes the Difference Your clinical expertise brings irreplaceable elements to the documentation process: Clinical Context : You understand how each piece of information fits into the broader picture of patient care, including implications that might not be obvious to AI systems. Pattern Recognition : Your experience helps you notice subtle connections between symptoms, findings, and patient history that inform clinical decision-making. Comprehensive Care : Your knowledge of the patient's overall health status, family history, and social context helps ensure documentation captures all relevant factors. Quality Control : Your review transforms AI-generated drafts into comprehensive clinical documents that accurately reflect patient encounters. ## Best Practices for Implementing AI Scribes Into Your Workflow Successfully incorporating AI documentation support into your practice typically involves several key steps: ### Developing Your Review Process Consider establishing a consistent approach to document review that fits your workflow: Review notes shortly after patient encounters while details remain fresh Focus particular attention on key clinical elements like medications and treatment plans Increase engagement with patients during encounters while the AI helps with documentation ### Critical Elements for Review When reviewing AI-generated documentation, consider focusing on: Clinical Content: Accuracy of symptoms and findings Medication names and dosages Treatment plans and follow-up instructions Complex medical histories Any unexpected or unusual information Contextual Elements: Patient affect and behavior Non-verbal communications Environmental or social factors Family dynamics and support systems ### Security and Privacy Considerations Healthcare privacy represents a fundamental responsibility. BastionGPT was founded by healthcare and Cybersecurity experts who understand how to protect your information and follow industry best practices. Many competitors claim that their systems use “Pseudonymization” and “Separation of Patient Identifiers,” but those are no longer best practices and create more risk for the accuracy of your data. We understand how to protect the integrity and security of your data without making false, high risk, or misleading claims about security. BastionGPT's security framework includes: Data Protection: Regular security assessments and updates Strict access controls and authentication measures Time-limited data retention policies Clear protocols for data handling Compliance Support: HIPAA compliant security measures International privacy standard compatibility (APP, PIPEDA, FERPA, and more) Regular compliance monitoring and updates Clear audit trails for system interactions ## Looking Ahead: The Evolution of Healthcare Documentation As healthcare continues to evolve, documentation tools will likely continue developing to better support clinical practice. The key lies in finding the right balance between technological assistance and clinical expertise. BastionGPT remains committed to working with healthcare providers to develop tools that enhance clinical practice while maintaining the highest standards of patient care. We believe that by supporting documentation efficiency, we can help providers focus more energy on elevating patient care. The future of healthcare documentation points toward a collaborative partnership between providers and technology. With thoughtfully implemented AI support, healthcare providers can potentially spend less time on administrative tasks and more time on patient care. ## Getting Started with BastionGPT's AI Medical Scribe We understand that incorporating new technology into clinical practice represents a significant decision. Consider these steps as you explore AI documentation support: Start with a trial period in your specific practice setting Begin with straightforward documentation tasks Provide your existing notes and reports as examples so the AI can match your documentation preferences Create clear protocols for document review and approval Regularly assess the impact on your workflow and documentation quality Our team works alongside providers to support successful implementation, offering resources and guidance based on real-world experience across various clinical settings. ## Your Partner in Clinical Documentation Healthcare providers deliver essential care every day, and they deserve tools that support their important work. While AI technology continues to evolve, the fundamental goal remains constant: supporting healthcare providers in delivering excellent patient care. We invite you to explore how AI scribe documentation support might fit into your practice. Whether you're looking to reduce administrative burden, enhance documentation quality, or find more time for patient care, we're here to work with you in developing solutions that support your clinical practice. If you are weighing dedicated scribe vendors, see how BastionGPT compares with Freed . If you’d like to learn more you can start a trial , reach out to [email protected] or schedule some time with our AI experts . ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/healthcare-balancing-risks-and-rewards-with-chatgpt Healthcare Innovation # Healthcare: Balancing Risks and Rewards with ChatGPT Written By: Josh Spencer • July 11, 2023 Artificial Intelligence (AI), particularly AI-powered language models like OpenAI's ChatGPT , have begun to transform healthcare. They enhance efficiency, drive innovation, and streamline patient care. However, as we enter into the exciting realm of ChatGPT in healthcare applications, healthcare organizations must navigate the associated risks, especially in securing patient health information (PHI). HIPAA compliance is critical, and the conversation around ChatGPT HIPAA integration is continually evolving. Six primary strategies exist for healthcare organizations to address the use of ChatGPT and similar tools in their settings, each offering different degrees of risk mitigation. ##### 1. Ignoring the Issue: High Breach Risk Some organizations might opt to ignore the issue and allow users to establish their own strategies. This laissez-faire approach may seem simple and cost-effective, but it leaves healthcare organizations exposed to significant HIPAA-compliant breach risks. Even a single piece of patient info being copied to ChatGPT, without the proper HIPAA contracts in place (such as a HIPAA BAA) will result in breaches, and possibly fines and sanctions. With this approach, it is only a matter of time before patient data is compromised. ##### 2. User Education: Moderate Breach Risk Educating users on the importance of not entering confidential or PHI into generative AI can be a reasonable approach. This strategy relies on continuous training and awareness programs, emphasizing the importance of HIPAA compliance. The risk level is moderated but not eliminated since human errors can still lead to breaches. It is important to keep in mind though that this approach is playing with fire, as one accidental patient identifier will result in HIPAA breaches if the service chosen is not a HIPAA compliant ChatGPT. ##### 3. Filtering Out PHI: Moderate Breach Risk Third-party services that try to remove PHI from your searches could potentially be used. While this approach could provide an extra security layer, it is still largely at a conceptual stage, and many of the services currently purporting to offer this service are error-prone or not yet ready for purchase. As of now, there's a lack of credible players in this space that a healthcare organization would likely trust, making this a less desirable option. This is a dangerous option as an error with the PHI removal could spell HIPAA compliance issues. ##### 4. Self-hosted ChatGPT: Low Breach Risk but Costly Creating a self-hosted AI infrastructure presents another route for healthcare organizations. This strategy offers tighter control over data privacy and a more straightforward generative AI option, substantially reducing breach risk. However, the resources required to develop and maintain this service, both in terms of time and money, can be prohibitive for many organizations. ##### 5. HIPAA Compliant Third-Party Hosted ChatGPT: Low Breach Risk Using secure third-party hosted ChatGPT services, such as HIPAA Compliant BastionGPT , provides a viable alternative. These services are designed to align with HIPAA regulations, reducing the risk of breaches significantly. Although this approach requires trust in the third party’s ability to protect patient information, it can balance the need for advanced AI capabilities with robust data security, without having to spend thousands of hours developing a private solution. ##### 6. Blocking All Generative AI: Low Breach Risk but Impactful to Productivity The final option is to block all generative AI. This approach drastically minimizes the risk of data breaches, but it can severely impact productivity and stifle competitive advantage. In the AI era, completely rejecting technology could be a missed opportunity for healthcare advancements. In many instances, ChatGPT is such a big time saver for physicians and healthcare workers, that they will use personal devices to circumvent company restrictions on ChatGPT access. The journey towards integrating ChatGPT into healthcare settings should be handled with caution, keeping HIPAA compliance and patient safety at the forefront. It's up to individual organizations to carefully assess their capacity, resources, and patient privacy obligations, and to implement an approach that best aligns with their specific needs. To consult with a ChatGPT Healthcare expert, you can contact us . ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/how-bastiongpt-transformed-my-medical-practice Healthcare Innovation # How I've Leveraged BastionGPT to Transform My Clinical and Business Practice Written By: Dr. Adil Manzoor • May 6, 2025 As a physician and entrepreneur, I'm always looking for tools that don't just promise efficiency but actually deliver it. Over the past year, BastionGPT has proven genuinely valuable—not just a novelty, but a meaningful part of how I run my clinical practice, streamline operations, and improve patient care. ## Improved Clinical Decision-Making and Patient Outcomes One of my biggest wins has been integrating BastionGPT into our clinical decision support systems. For example, when creating workflows to manage conditions like congestive heart failure and hypertension, BastionGPT helped distill complex clinical guidelines into clear, actionable insights. The result? Faster, evidence-based decisions at the point of care, enabling our clinical teams to proactively address patient issues before they escalate. ## Streamlined Practice Management BastionGPT significantly improved the way we handle administrative tasks and patient management. Previously, reviewing patient charts for medication refills or ensuring timely follow-ups required tedious manual effort. With BastionGPT, we've automated and simplified these audits, quickly identifying gaps and ensuring patients consistently receive proactive care. This not only improved patient satisfaction but dramatically reduced administrative workload, enabling my team to focus more directly on patient interactions. ## Enhanced Patient Communication and Engagement BastionGPT has also significantly upgraded our patient communication processes. It helps us draft clear, patient-friendly messaging, ensuring patients understand their care plans, preventive screenings, and follow-up instructions. The result has been increased patient compliance, fewer missed appointments, and stronger patient-provider relationships. ## Enhanced Teaching and Resident Training As an educator, BastionGPT allowed me to rapidly generate engaging educational content for medical residents. Whether creating concise summaries of complex pediatric conditions or facilitating interactive training sessions, BastionGPT helps me teach effectively and keep trainees actively engaged. Ultimately, BastionGPT has reshaped the way I practice medicine, teach, and manage day-to-day operations. For physicians, educators, and healthcare leaders, integrating BastionGPT isn't just about staying ahead; it's about fundamentally improving how we work, communicate, and live. Providers in other specialties report the same shift—see how clinicians are transforming their mental health practice with the same approach. If BastionGPT has transformed your practice too, you can share it with colleagues and earn rewards through our affiliate referral program . Dr. Adil Manzoor DO, MBA/MS Board Certified Internal Medicine & Pediatrics Physician & Board Advisor, BastionGPT ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/how-chiropractors-use-generative-ai Healthcare Innovation # AI For Chiropractors: How Chiropractors Can Use AI Written By: Robert Henehan • November 13, 2025 Chiropractors spend an average of one-third of their workday on documentation. After a full schedule of patient visits, that means hours of writing SOAP notes, drafting insurance letters, and preparing narrative reports. It's time that could go toward patients, professional growth, or getting home on time. AI is changing that. Chiropractors are now using secure, healthcare-focused AI tools to draft clinical notes, write workers' comp narratives, generate insurance letters, and create patient education materials in a fraction of the time it takes manually without sacrificing accuracy or compliance. BastionGPT was built specifically for this. Unlike general-purpose AI tools like ChatGPT, it operates under a HIPAA-compliant infrastructure with a Business Associate Agreement included, meaning patient data is protected by design. It can process large files (up to 1,000 pages) and transcribe lengthy sessions (with multi-speaker recognition up to 4 hours) to support thorough, complete documentation. This guide covers the practical ways chiropractic clinics, including those handling personal injury and workers' comp cases, can put BastionGPT to work across clinical documentation, insurance communications, patient education, and clinic administration. For more on organizing, saving, and reusing your own effective clinical prompts, see our prompt library guide and our prompt engineering quick start guide . ## How Chiropractors Can Use AI To Save Time ### 1. Clinical Documentation Clinical documentation is the backbone of chiropractic care, but writing detailed notes for every visit can consume a large portion of the day. BastionGPT assists by transforming dictations, rough notes, or transcripts into well-structured chiropractic notes, allowing providers to document initial exams, daily visits, and re-exams more quickly without sacrificing thoroughness. #### Initial Evaluation Note Drafting Chiropractors perform comprehensive initial evaluations for new patients, documenting history, exam findings (e.g. ROM measurements, orthopedic test results), diagnoses, and treatment plans. BastionGPT helps by turning the chiropractor’s dictated notes or outline into a polished initial evaluation report. For example, in a personal injury case after a car accident, a chiropractor can use AI to draft the initial exam note describing the following: Patient’s complaints (neck pain, headaches) Objective findings (reduced cervical flexion, positive Spurling’s test) Assessment (cervical sprain/strain with segmental dysfunction) Plan (chiropractic adjustments 3x/week, therapeutic exercises, re-exam in 4 weeks). The AI-generated draft gives the clinician a solid starting point, covering all required sections and medicolegal details. ##### Prompt Example: You are a chiropractor documenting a new patient’s initial evaluation. Summarize the patient’s history of injury (e.g. a work accident with low back pain and radiating leg pain), examination findings (posture, range of motion limits, positive orthopedic tests like Straight Leg Raise, neurological exam results), diagnoses, and treatment plan. Organize the note in a clear format (Subjective, Objective, Assessment, Plan) suitable for the patient’s chart and include all relevant details for a complete initial report. #### Daily SOAP Note Generation Chiropractic follow-up visits often require daily SOAP notes to track patient progress and satisfy billing requirements. BastionGPT can generate these SOAP notes by analyzing a chiropractor’s brief voice dictation or bullet-point recap of the visit. For instance, consider a patient in a workers’ comp case returning for a routine adjustment: the chiropractor notes the patient’s subjective report (“improved mobility, 5/10 lumbar pain today”), objective findings (muscle spasm noted in lumbar paraspinals, moderate restriction at L4), assessment (condition improving, tolerating treatment), and plan (continue current care plan, add core stretching exercise). The AI can quickly format this into a consistent SOAP note entry. This saves the chiropractor time on typing and ensures no section is overlooked, while the chiropractor confirms the note’s accuracy and signs off. ##### Prompt Example You are a chiropractor writing a daily SOAP note after a treatment visit. Draft the SOAP note including: the patient’s subjective update on symptoms (e.g. neck stiffness 4/10, better since last visit), objective findings (muscle tone, ROM changes, any notable exam findings today), your assessment of their progress, and the treatment provided with plan (e.g. adjustments delivered, modalities applied, home care reminders, next visit scheduled). Ensure the tone is concise and professional, suitable for the medical record. #### Re-exam / Progress Report Summaries Periodic re-exams or progress reports are crucial in both routine care and cases like personal injury or workers’ comp, where updates are needed for insurers or case managers. BastionGPT assists chiropractors by compiling prior notes, new exam findings, and outcome measures into an organized progress report. Imagine a chiropractor conducting a 6-week re-exam for a patient with a work-related back injury: the AI can help summarize how the patient’s pain levels, range of motion, and functional abilities have changed since the initial exam, list current objective findings (e.g. improved lumbar flexion from 50° to 70°, orthopedics now negative), and detail an updated treatment plan or MMI (maximum medical improvement) status. The resulting draft reads like a professional narrative that highlights progress and remaining deficits, which the chiropractor can refine and send to the case adjuster or include in the patient’s file. ##### Prompt Example You are a chiropractor preparing a progress report re-exam note for a patient who has completed 4 weeks of treatment for a shoulder injury. Summarize the patient’s initial status versus current status – including pain levels, range of motion improvements (e.g. shoulder abduction improved from 90° to 150°), strength changes, and any persistent issues. Include results of re-assessed orthopedic tests (e.g. Neer’s impingement now mild, Speed’s test negative) and provide an updated assessment and plan (continuing care, any new recommendations, or readiness for discharge if appropriate). Write this in a formal report style as you would for an insurance case manager, ensuring all key changes are documented clearly. ### 2. Narrative & Legal Reports Chiropractors handling personal injury and workers’ comp cases often spend hours crafting detailed narrative reports for attorneys and insurance adjusters. AI for personal injury and workers’ comp documentation can offload this writing burden by producing well-structured draft narratives that cover the patient’s history, treatment, and status, which the chiropractor then reviews for accuracy. These med-legal documents are time-consuming but critical for claims, and BastionGPT helps ensure they’re thorough and professionally formatted. #### Workers’ Comp Narrative Report Workers’ compensation cases frequently require a comprehensive narrative report summarizing the injury, treatment provided, objective findings, and the patient’s ability to work. Chiropractors can feed BastionGPT the patient’s records and prior notes as context and get a draft narrative that hits all the important points. For example, after treating a patient injured on the job (say, a lifting injury causing low back pain), a chiropractor might need to write a narrative for the insurance carrier or employer. The AI will review the details from the patient’s file – mechanism of injury, initial exam findings (like antalgic gait, positive Kemp’s test), treatment timeline, referrals (if any), and current functional status – then produce a logical narrative covering causation, diagnosis, treatment, progress, and work capacity. The chiropractor instructs the AI not to copy any source text verbatim but to create a fresh summary, which the doctor then fine-tunes, ensuring it’s accurate and includes any required legal phrasing. ##### Prompt Example You are a chiropractor preparing a workers’ compensation narrative report for a patient’s case. Using the details from the below (prior reports), draft a clear and comprehensive narrative that includes: the patient’s mechanism of injury and history, initial exam findings (ROM deficits, orthopedic/neuro test results), treatment provided (number of visits, types of therapy), the patient’s progress over time, current clinical status, and any work restrictions or MMI status. Organize the report with headings for Background, Findings, Treatment, and Conclusion. Use the attached or documented information to match the style and content of the example. Do not copy text from ; use it only to inform the new summary. #### Personal Injury Narrative Letter In personal injury (PI) cases (like auto accidents), chiropractors often write narrative letters to attorneys summarizing the patient’s course of care for demand packages. BastionGPT can help draft these letters by weaving together the key facts of the case in a persuasive, medically detailed manner. For instance, for a patient injured in a car accident, the AI-generated narrative letter can outline the initial presentation (e.g. whiplash-associated neck pain, headaches), exam findings (spasm, restricted cervical ROM, positive compression tests), the treatment plan executed (chiropractic adjustments, physiotherapy modalities, duration of care), and the patient’s response to treatment including remaining impairments or need for future care. It may also highlight any delays in recovery, complications, or compliance issues, as well as itemize the total number of visits and medical costs incurred, which are often needed by personal injury attorneys. The chiropractor ensures the tone is factual and objective, yet the content supports the patient’s injury claims by thoroughly documenting the impact and necessity of care. ##### Prompt Example You are a chiropractor writing a personal injury narrative report for an auto accident patient. Draft a letter to the patient’s attorney that summarizes the case: describe the collision and initial injuries, detail the findings from the chiropractic initial exam (e.g. cervical and lumbar sprains with muscle spasm, reduced ROM, positive orthopedic tests), outline the treatment provided (number of adjustments, therapies, duration of care), and report the patient’s progress and current status (noting any residual pain or limitations). Present the information in a clear, chronological narrative, using professional language that would be suitable for an insurance adjuster or attorney’s review. ## 3. Insurance & Payer Communications Dealing with insurance payers is a routine part of outpatient chiropractic, especially in PI and workers’ comp. Chiropractors must often justify the medical necessity of care, request prior authorizations, or appeal denied claims. AI can expedite these insurance communications by drafting well-reasoned letters that cite patient facts and guidelines, helping chiropractors make a compelling case while saving time on paperwork. #### Medical Necessity Letter Drafting When an insurer or third-party payer questions the need for continued care or a specific service, chiropractors write medical necessity letters to explain why the treatment is justified. BastionGPT can use the patient’s clinical data and progress notes to generate a clear, factual letter that supports the need for care. For example, if a personal injury protection (PIP) carrier or health insurer requires justification for extending chiropractic visits beyond a typical limit, the AI can draft a letter highlighting the patient’s diagnosis (e.g. lumbar disc injury with radiculopathy), the functional improvements seen so far (perhaps 50% pain reduction, improved lifting capacity), and the evidence-based rationale for further treatment (such as persistent objective findings or standard guidelines recommending a certain course length). The letter would be addressed to the insurance adjuster or utilization reviewer and formatted professionally, often including references to medical necessity criteria. The chiropractor provides relevant details from the patient’s record so the AI includes specific supporting points, then reviews the draft to ensure it aligns with clinical findings and tone. ##### Prompt Example You are a chiropractor drafting a medical necessity letter to an insurance company. Using the attached or copied patient’s information (exam findings, treatment records, and progress notes), write a letter justifying the need for continued chiropractic treatment for this patient. Begin by stating the patient’s diagnoses and initial condition, then detail the improvements achieved to date (with objective measures like increased ROM or ADL function). Explain why additional treatment is medically necessary – for example, citing remaining functional deficits, the treatment plan per guidelines, or risks of stopping care too soon. Use the as a framework for style, tone, and structure. Address the letter to a case reviewer, use a respectful and evidence-based tone, and do not copy any text from directly replace and rephrase with the attached content or additional instead. Content and notes are attached or below. #### Prior Authorization Request Letter Before providing certain treatments or imaging studies, chiropractors often need to obtain prior authorization from insurance (or workers’ comp utilization review). BastionGPT can expedite the writing of these request letters by organizing the clinical justification and patient history into a concise argument. For instance, a chiropractor might seek prior auth for an MRI for a patient with persistent cervical radiculopathy symptoms. The AI can draft a letter that quickly recaps the patient’s injury and symptoms, outlines conservative treatments tried and their outcomes, and explains why the requested MRI is necessary now (e.g. to rule out a herniated disc given neurological deficits). Similarly, for therapy visit extensions, the AI can emphasize the patient’s progress and the need for additional sessions to reach maximal improvement. By articulating the medical necessity in insurer-friendly language (potentially referencing guidelines or “medically necessary” criteria), the AI-generated draft helps ensure all relevant details are included. The chiropractor then personalizes any specifics and ensures the letter meets any payer-specific requirements before submission. ##### Prompt Example You are a chiropractor requesting prior authorization for additional treatment sessions for a patient. Write a letter to the insurance company that briefly summarizes the patient’s condition, the treatment provided so far and patient’s progress, and the justification for more visits. Clearly state the request (number of extra visits or specific service) and include a professional closing. The tone should be factual and persuasive, focusing on why the continued care is medically necessary for the patient’s recovery. #### Claims Denial Appeal Letter If an insurance claim or reimbursement for chiropractic services is denied (for example, deemed “not medically necessary” or denied due to documentation issues) the chiropractor can appeal with a formal appeal letter. BastionGPT assists by analyzing the denial reason and the patient’s records to craft a point-by-point rebuttal. For instance, suppose a workers’ comp insurer denies payment for a set of treatments, claiming they were excessive. The AI can generate a draft appeal letter that references the patient’s condition and progress, counters the denial rationale (e.g. explaining how the frequency of visits was appropriate given acute injury guidelines), and includes any supporting documentation summary (like outcome measures or prior authorization approvals). The letter will be formatted professionally, addressed properly, and can include direct quotes from your supporting documentation when relevant. By covering all the necessary arguments and maintaining a respectful but firm tone, the healthcare AI’s draft can significantly cut down the time the chiropractor spends on writing appeals. The chiropractor reviews the letter, adds any personal touches or additional evidence, and then submits the appeal to the insurer. ##### Prompt Example You are a chiropractor appealing an insurance claim denial for a patient’s treatment. Draft a letter to the insurance company addressing the specific denial. Begin by identifying the patient and claim in question, then clearly explain why the provided treatment was justified: reference the patient’s diagnosis and significant findings, note prior approvals or guidelines that support the care, and highlight the patient’s improvement under your care. Politely refute the insurer’s reasons for denial with facts (for example, “Contrary to the denial, the patient’s documented functional deficits warranted continued care beyond the usual 12 visits, as supported by guidelines"). Maintain a professional tone and request a reconsideration of the claim based on the evidence provided. Use the attached or information below as well as the following guidelines. ### 4. Structured Findings & Measures Objective measurements like range of motion and orthopedic test results are key parts of chiropractic evaluations, especially for showing progress in PI or workers’ comp cases. Generative AI helps chiropractors organize these structured findings into clear tables or summaries. By converting raw exam data into neat reports (e.g. ROM comparison tables or bullet summaries of neuro tests), AI saves time and supports having documentation that is accessible to patients. #### Range of Motion (ROM) Table Generation Chiropractors frequently measure patients’ spinal range of motion in degrees and compare them to normal values. Documenting this in a table format is ideal for clarity, but creating tables manually can be tedious. BastionGPT can take the chiropractor’s recorded ROM measurements and automatically produce a tidy table listing each movement, the patient’s measured angle, the normal expected range, and any deficit or change. For example, after an initial exam for a lumbar spine injury, the chiropractor might have: flexion 45° (normal ~60°), extension 15° (normal ~25°), left lateral bending 10° (normal ~25°), etc. The AI can draft a table that includes Movement, Patient ROM, Normal ROM, and Deficit...and so on, including each plane of motion. This structured output makes it easy for an insurance adjuster or referral partner to see the extent of restriction. Chiropractors use the AI’s table as a starting point, verifying the values and possibly adding any notes (like pain with motion). The result is a professional-looking ROM chart included in the exam note or report without the chiropractor having to manually type it out completely. ##### Prompt Example You are a chiropractor creating a range of motion table for a patient’s spinal examination. Take the patient’s ROM measurements for various movements and generate a table that includes columns for Movement, Patient’s Measured ROM (in degrees), Normal Expected ROM, and the difference/deficit. Ensure the table is neatly formatted and labeled so it can be inserted into a report to clearly show the patient’s limitations compared to normal ranges. The patient’s ROM and Expected ROM are attached or below. #### Orthopedic/Neurological Test Summary During exams, chiropractors perform orthopedic and neurological tests (like Spurling’s, Kemp’s, reflex tests, sensory exams) and need to document the findings succinctly. BastionGPT can help by summarizing a list of test outcomes into a coherent narrative or bullet list. For instance, after a cervical spine exam, a chiropractor might have these raw notes: Spurling’s test: positive on left causing arm pain; Cervical distraction: relieves symptoms; Shoulder depression: negative; Reflexes 2+ and symmetric in upper extremities; Dermatomes intact; Muscle strength 5/5 except slight weakness in left grip. The AI can turn that into a concise summary paragraph: “Orthopedic tests revealed a positive Spurling’s on the left (radiating pain into left arm), while cervical distraction alleviated the symptoms, consistent with nerve root compression. Shoulder depression test was negative. Neurological exam showed normal deep tendon reflexes (2+ bilateral) at the biceps, triceps, and brachioradialis, intact light touch sensation in C5-T1 dermatomes, and muscle strength 5/5 in all major upper extremity groups (with only a slight decrease in left grip strength).” This kind of summary helps to make sure all key positives and negatives are documented in professional language. The chiropractor provides the AI with the list of findings and then reviews the generated summary, making any tweaks to ensure accuracy and clarity. This saves time and yields a well-structured exam findings section for the report. ##### Prompt Example You are a chiropractor summarizing orthopedic and neurological exam findings for a patient’s report. Take the following exam results and compose a concise summary: list any positive orthopedic test findings and their implications (e.g. “Kemp’s test causes left low back pain, indicating facet involvement” or “Straight Leg Raise positive at 40° on right, suggesting nerve tension”), note significant negatives (tests that were normal), and include the neurological exam results (reflexes graded, sensory intact or any deficits, muscle strength grades). Write this in narrative form as one cohesive paragraph suitable for an evaluation report, using proper medical terminology (e.g. “hypoesthesia” for sensory loss if present, etc.) and ensure it clearly distinguishes the positive findings from the normal findings. ## 5. Patient Education & Self-Management Educating patients is an important part of chiropractic care, from explaining their condition to providing home exercise programs (HEPs). Generative AI helps chiropractors produce clear patient education materials quickly. BastionGPT can draft easy-to-understand explanations of diagnoses and step-by-step exercise instructions, which the chiropractor then reviews for accuracy and tailors to the patient’s needs. #### Condition Explainer Handout Chiropractors often need to explain complex medical terms or diagnoses in plain language for patients. BastionGPT can generate a patient-friendly explainer about a condition, saving time while ensuring key points are covered. For example, if a patient is diagnosed with sciatica due to a herniated lumbar disc, the AI can create a handout explaining what sciatica is (pinched nerve pain down the leg), common causes (like disc bulges or injuries), symptoms (shooting pain, numbness, tingling), and typical treatment approaches (rest, chiropractic adjustments, exercises, possibly medical interventions). It can also include analogies to make the concept clear – for instance, comparing a herniated disc pressing on a nerve to “jelly squeezing out of a donut and touching a wire.” The chiropractor can ask the AI to target a certain reading level (say 6th–8th grade) so that the language is appropriate for most patients. After getting the draft, the chiropractor ensures the information is correct (and in line with their clinical approach) and then can print or email this condition explainer to the patient. This empowers patients with knowledge about their condition, which can improve compliance and satisfaction. ##### Prompt Example You are a chiropractor creating a patient education handout to explain a condition in simple terms. Write an explainer based on these session notes and documents. Describe the key terms and symptoms. Briefly outline how chiropractic care and self-care can help, without giving specific treatment advice. Keep the tone reassuring and the reading level around 6th–7th grade so it’s accessible to all patients. #### Home Exercise Program (HEP) Instructions Prescribing exercises for patients to do at home is common in chiropractic (for rehabilitation and prevention), but writing out detailed instructions for each patient can be tedious. BastionGPT steps in by generating Home Exercise Program (HEP) sheets tailored to the patient’s condition. For example, a chiropractor treating a patient with chronic neck and shoulder tension might want to send them home with exercises. The AI can draft instructions for exercises like chin tucks (cervical retraction), upper trapezius stretch, and scapular squeezes, complete with how to perform them, recommended sets/reps, and any precautions (“stop if sharp pain occurs,” etc.). Each exercise is described step-by-step in simple language: e.g., “Chin Tuck: Stand with your back against a wall and look forward. Gently tuck your chin straight back (making a ‘double chin’) and hold for 5 seconds, then relax. Repeat 10 times. This exercise helps strengthen your neck muscles and improve posture.” The AI might also add general advice like warming up before exercises or maintaining good form using specific details from the visit, chart, transcription, or notes. The chiropractor can specify the reading level or tone (educational but encouraging). After generating the HEP draft, the chiropractor reviews it to ensure each exercise is appropriate and then provides it to the patient. This way, patients receive clear, customized exercise guidance to support their recovery. ##### Prompt Example You are a chiropractor creating a Home Exercise Program (HEP) handout for a patient with the attached information. The HEP should include 3 exercises and for each exercise, provide step-by-step instructions on how to perform it correctly, including posture and movement cues, as well as recommended frequency and any safety tips. Write in a friendly, easy-to-follow tone (about an 7th-grade reading level), so the patient can confidently do these exercises at home to support their chiropractic care. #### Clinic Admin & Templates Running a chiropractic clinic involves paperwork beyond clinical notes, from intake and consent forms to routine letters like no-show warnings or discharge summaries. Generative AI helps create and refine these administrative documents, ensuring they are thorough and professional. BastionGPT can draft form language and template letters that clinic managers or chiropractors then tweak to fit their practice, saving considerable time in administrative tasks. #### Intake and Consent Form Drafting New patient intake packets and consent forms are essential for clinics, covering medical history, treatment consent, privacy practices, and more. Instead of writing these forms from scratch, chiropractors or office managers can use BastionGPT to generate intake and consent form templates tailored to chiropractic services. For example, the AI can produce a draft intake form that includes sections for patient demographics, insurance information, chief complaint description, medical history (surgeries, medications, allergies), and lifestyle factors. Similarly, it can draft a treatment consent form explaining the nature of chiropractic adjustments, possible risks (soreness, rare complications), and obtaining the patient’s acknowledgement. The AI helps the language to be clear and patient-friendly yet covers necessary elements. By leveraging BastionGPT’s knowledge of healthcare compliance, the generated forms can incorporate standard clauses (e.g. consent to treat, release of information, financial responsibility). The clinic can then review these drafts, insert any clinic-specific logos, and have a polished set of intake documents ready for use. This approach can significantly reduce the time spent editing generic form templates and help ensure nothing important is omitted. ##### Prompt Example You are a chiropractic clinic manager drafting new patient intake and consent forms for your practice. Create a comprehensive intake form that gathers patient information (personal and insurance details, medical history, current complaints) and a separate treatment consent form. The consent form should explain chiropractic care at a 7th-grade reading level, outline potential risks or side effects of adjustments, mention privacy protections and include statements of consent for treatment and financial responsibility. Use clear, polite language that patients can understand and include spaces for the patient’s name, signature, and date on each form. Make sure all necessary legal and regulatory points are covered in the text of the forms. #### No-Show and Discharge Letter Templates Maintaining professionalism in patient communications is important, even for tough messages like addressing no-shows or discharging a patient from care. BastionGPT helps chiropractors draft template letters for these scenarios. For instance, if a patient has missed multiple appointments without notice, the AI can write a courteous yet clear no-show policy letter. It would remind the patient of the clinic’s attendance policy (e.g. requiring 24-hour notice for cancellations), express concern for their well-being, and encourage them to reach out to continue care or reschedule, all while noting any fees or consequences per clinic policy in a professional manner. On the other hand, when a patient’s treatment is completed or if the clinic needs to formally discharge a patient (perhaps due to non-compliance or referral to another provider), BastionGPT can draft a discharge letter. This letter can summarize the care provided, the reason for discharge (e.g. treatment goals met, or need for care beyond clinic’s scope), and next steps (like home care advice or referral information), and thank the patient for the opportunity to participate in their care. By starting from AI-generated templates, the clinic better ensures these communications have the right tone that is compassionate but clear, and include all necessary information. Staff can then adjust any specific details (names, dates, policy specifics) before sending them out, thereby standardizing communication and saving time. This can be done manually or by requesting it from BastionGPT. ##### Prompt Example You are a chiropractor writing a no-show policy reminder letter to a patient who has missed several appointments without notice. Draft a polite, professional letter that addresses the patient by name, notes the recent missed appointments, reiterates the clinic’s cancellation/no-show policy (for example, requiring 24-hour notice and any applicable fees or scheduling restrictions for no-shows), and encourages the patient to contact the office to continue their care. The tone should be understanding but firm about the importance of keeping appointments for effective treatment. Close the letter warmly, expressing that the clinic is committed to the patient’s health and hopes to see them back. ## Integrating AI into Your Chiropractic Practice Generative AI tools like BastionGPT provide chiropractors with a secure, time-saving way to handle documentation and communication tasks while keeping the clinician in control of the final content. To explore how this chiropractic AI assistant can benefit your practice, you can start your 7-day free trial. Experience firsthand how this cutting-edge tool can elevate your practice from streamlined documentation to targeted patient education. Let BastionGPT handle the busy work while you shine in your role as a caregiver. If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Frequently Asked Questions (FAQ) ### How can AI help chiropractors with notes and workers’ comp reports? BastionGPT can draft and format routine chiropractic notes and detailed workers’ comp narrative reports by analyzing your session notes and exam findings. This saves time by producing a thorough first draft for each note or report, which the chiropractor then reviews and edits to ensure accuracy and completeness. ### How much do HIPAA compliant AI assistants cost? BastionGPT only costs $20-45 per user per month and offers a 7-day free trial for new users, so clinics can evaluate its benefits before subscribing. ### How is AI for chiropractors used? AI for chiropractors is used to streamline documentation and administrative tasks in a chiropractic clinic. BastionGPT helps with writing evaluation notes, daily treatment notes, re-exam reports, insurance letters, and even patient education materials all while keeping patient data private and complying with in-scope healthcare privacy regulations. ### Can AI help with personal injury chiropractic documentation? Yes. BastionGPT's healthcare AI can draft comprehensive personal injury narrative reports and chart notes for chiropractors. For example, BastionGPT can summarize the patient’s treatment course, injuries, and progress into a clear narrative suitable for personal injury claims, which the chiropractor can then adjust and use for attorneys or insurance adjusters to review. ### How do chiropractors ensure HIPAA compliance when using AI for documentation? Chiropractors in the US must use HIPAA-compliant AI solutions when dealing with PHI. BastionGPT runs in a secure environment that adheres to HIPAA requirements, does not train the AI models on customer data, and includes a Business Associate Agreement (BAA) standard with all plans. ### Can AI transcribe chiropractic sessions for note-taking? Yes. AI tools like BastionGPT include a medical AI Scribe that can securely transcribe chiropractic session audio or dictations and then help generate notes from it. BastionGPT’s transcription can distinguish multiple speakers (e.g. doctor and patient), offers unlimited monthly sessions with each session lasting for up to 4 hours of conversation. Transcripts can then be used to generate drafts, standardized clinical notes, or custom notes that match your writing style and preferred note formatting. ### What chiropractic documents can generative AI review and draft? Users can upload long reports, charts, and legal documents, up to 300,000 words, using Word, PDF, TXT, PowerPoint, and Excel file types. The AI can summarize, extract key data, and draft outputs such as narratives, letters, or handouts, which can then be reviewed and finalized. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/what-makes-an-ai-hipaa-compliant Healthcare Innovation # What Makes an AI HIPAA Compliant? Written By: Josh Spencer • July 3, 2026 An AI tool is HIPAA compliant when it handles protected health information (PHI) under a signed Business Associate Agreement (BAA), protects that data with required technical safeguards like encryption and access controls, and restricts how the vendor can use the data, including a commitment that PHI is never used to train AI models. Compliance is a property of the whole arrangement (the contract, the security, and how your team actually uses the tool), not a feature of the AI model itself. That is the short answer. The rest of this guide explains each requirement, shows you how to evaluate any AI vendor, and covers the mistakes that turn a helpful tool into a reportable breach. ## Why this question matters right now AI has moved from experiment to everyday practice. According to the American Medical Association's 2026 Physician Survey on Augmented Intelligence , 81% of physicians now use AI in a professional context, up from 38% in 2023. The same survey found that data privacy assurances rank among the top requirements physicians cite for adopting AI further. The stakes of getting privacy wrong are just as concrete: 2025 was the worst year on record for large healthcare data breaches by count. The HIPAA Journal reports 772 breaches affecting 500 or more individuals were reported to the HHS Office for Civil Rights (OCR) for 2025, exposing the PHI of roughly 139.7 million people. Healthcare breaches are the most expensive of any industry. IBM's Cost of a Data Breach Report 2025 puts the average healthcare breach at $7.42 million per incident. Penalties scale fast. Under the HHS penalty schedule updated in January 2026 , civil monetary penalties for HIPAA violations range from $145 to $73,011 per violation in the lower tiers, and up to $2,190,294 per violation for willful neglect that goes uncorrected. OCR is also continuing its risk analysis enforcement initiative, with the agency confirming an expansion into risk management in 2026. Every clinician pasting text into a chatbot, and every practice evaluating an AI scribe, is making a compliance decision. This guide gives you the framework to make it well. ## A quick refresher: PHI, covered entities, and business associates Three HIPAA concepts do most of the work in any AI compliance question: Protected health information (PHI). Health information that identifies a patient or could reasonably be used to identify one. HIPAA's Privacy Rule lists 18 categories of identifiers, from the obvious (names, Social Security numbers, medical record numbers) to the easy to miss (dates of service, geographic detail smaller than a state, device identifiers, full-face photos). Covered entities. Healthcare providers, health plans, and clearinghouses. If you treat patients or bill insurance, this is you. Business associates. Any vendor that creates, receives, maintains, or transmits PHI on your behalf. An AI service that processes your patient notes fits this definition. Under 45 CFR 164.502(e) , you may only share PHI with a business associate after obtaining satisfactory assurances in the form of a signed BAA. That last point is where most consumer AI tools fail immediately. Sending PHI to a vendor without a BAA is an impermissible disclosure, which means the violation occurs the moment the data leaves your control, even if nothing ever leaks. ## The four requirements of a HIPAA compliant AI tool ### 1. A signed Business Associate Agreement The BAA is the legal foundation. It binds the vendor to HIPAA's rules: safeguarding PHI, reporting breaches to you within defined timeframes, using the data only to provide the contracted service, and returning or destroying PHI when the relationship ends. Two practical notes healthcare buyers often miss: A BAA covers specific products and features, not the company. OpenAI, for example, offers BAAs for its Zero Data Retention API endpoints and for sales-managed ChatGPT Enterprise accounts, but consumer ChatGPT plans (Free, Plus, Pro, Team) are not eligible. The tier your staff actually uses determines whether PHI is covered. The BAA governs the vendor's side, not yours. You remain responsible for risk analysis, workforce training, minimum-necessary use, and access configuration. A signed BAA with a vendor whose tool your staff misuse is not a defense. A useful rule of thumb from compliance teams: no BAA, no PHI. ### 2. Technical safeguards that meet the Security Rule HIPAA's Security Rule requires administrative, physical, and technical safeguards for electronic PHI. For an AI platform, look for specifics, not adjectives: Encryption in transit and at rest. The current baseline is TLS for data in transit and AES-256 for data at rest. Access controls. Unique user accounts, role-based permissions, and multi-factor authentication so only authorized staff reach PHI. Audit logging. A record of who accessed what and when, available to you for review. Data isolation. Your organization's data segregated from other customers' data. A vendor that cannot name its encryption standard or produce an audit report has not done the work. ### 3. Restrictions on how your data is used Under HIPAA, a business associate may only use PHI to deliver the contracted service. Many general-purpose AI companies built their business models on the opposite assumption: user inputs improve the product. Consumer AI tools commonly retain conversations and may use them for model training unless the user opts out. For healthcare use, the vendor's terms and BAA should state plainly that: Your prompts, documents, and audio are never used to train AI models Your data is not shared with third-party AI providers beyond what the BAA covers Data is retained only as long as needed , with defined deletion timelines you control If a vendor's terms mention using your inputs to "improve our services" without clear carve-outs for PHI, treat that as disqualifying. ### 4. Features scoped for healthcare, with admin control Even a BAA-covered platform can leak PHI through side doors. Integrated features often route data to services outside the BAA's scope. Microsoft 365 Copilot illustrates the risk : it can be covered under Microsoft's enterprise BAA, but Copilot experiences that route queries through consumer Bing services fall outside that coverage, and a prompt containing PHI that triggers a web lookup can leave the protected environment. A healthcare-ready AI platform limits or disables risky capabilities by default: no social sharing, no uncontrolled third-party plugins, no silent hand-offs to non-covered endpoints. It also gives administrators the controls to enforce that posture across the organization. ## Why de-identification alone is not enough A common workaround is to strip the patient's name before pasting a note into a consumer chatbot. This rarely holds up. HIPAA's Safe Harbor method ( 45 CFR 164.514(b) ) requires removing all 18 identifier categories, including dates, locations smaller than a state, and any other unique identifying detail. Clinical notes are dense with exactly this material: a visit date, a rare diagnosis, a hometown mentioned in passing. Any one of them can make the record identifiable again, and regulators treat data as PHI whenever a reasonable basis exists to believe it can identify a patient. The alternative path, expert determination, requires a qualified statistician to formally certify that re-identification risk is very small. Ad-hoc redaction by a busy clinician is neither. The practical takeaway: assume patient data is PHI unless it has been de-identified through a formal, documented process. Manual redaction before pasting into a non-covered AI tool does not change the compliance analysis. For a deeper look at why tokenization and anonymization strategies fall short with generative AI, see HIPAA risks with tokenization and anonymization . ## Are the popular AI tools HIPAA compliant? The honest answer for most consumer AI products is no, with narrow enterprise exceptions. As of early 2026: ChatGPT. Consumer plans (Free, Plus, Pro, Team) are not eligible for a BAA, so any PHI entered into them is an impermissible disclosure. OpenAI does offer BAAs for its Zero Data Retention API and for sales-managed ChatGPT Enterprise, Edu, and healthcare-specific offerings, which require enterprise procurement and careful configuration. The version most individual clinicians use is the version that is not covered. Full guide: Is ChatGPT HIPAA Compliant? Microsoft Copilot. Can be covered under Microsoft's enterprise agreements for commercial tenants, but coverage does not extend to personal Microsoft accounts, and some Copilot surfaces fall outside the BAA. Configuration and account hygiene decide the outcome. Full guide: Is Microsoft Copilot HIPAA Compliant? For a plan-by-plan comparison, see BastionGPT vs Microsoft Copilot . Google Gemini. Google's BAA can cover Gemini within paid Workspace tiers; consumer Gemini accounts are not covered. Full guide: Is Google Gemini HIPAA Compliant? Claude. Anthropic's BAA does not cover Claude Free, Pro, Max, or Team plans; HIPAA-ready use is available only through its sales-assisted Enterprise tier, and some features remain outside BAA coverage even there. Full guide: Is Claude HIPAA Compliant? Healthcare teams weighing the enterprise route can also see BastionGPT vs Claude Team & Enterprise . The pattern across all four: compliance is possible only on enterprise tiers, under a signed BAA, with features configured correctly, and it never extends to the free or personal versions staff reach for by default. That gap between what the organization licenses and what individuals actually use is where violations happen. A December 2025 Wolters Kluwer survey of more than 500 healthcare workers found that 17% admitted to using unauthorized AI tools at work. ## Policy alone will not protect you Many organizations respond to AI risk with a memo: do not enter patient information into chatbots. A policy is necessary, and it is not sufficient. A rushed clinician saving twenty minutes on a discharge summary can violate it with one paste, often without realizing the note still contains identifiers. Effective programs pair policy with environment: Network and access controls. Some health systems block unapproved AI tools on work networks entirely, or interpose a warning screen before access. Data loss prevention (DLP). DLP tooling can detect PHI patterns (names, MRNs, SSNs) being entered into web forms and warn or block before submission. Training that names the risk. Annual HIPAA training should cover AI specifically: what counts as PHI in a prompt, why "I removed the name" is not de-identification, and which tools are approved. An approved alternative. This is the control that makes the others workable. Staff turn to unapproved AI because it solves a real problem: documentation burden. Blocking tools without replacing them pushes usage into the shadows. Providing a compliant platform that is as capable as the consumer tools channels demand somewhere safe. ## Red flags: how to spot a non-compliant AI tool When evaluating any AI product for healthcare use, treat these as disqualifying until proven otherwise: No BAA offered , or a BAA available only on request for a premium tier your team will not buy No mention of HIPAA or healthcare anywhere in the vendor's documentation Terms that permit training on your data or sharing inputs for product improvement No named security specifics : encryption standards, audit logging, access controls, or third-party attestations Consumer-oriented features enabled by default: social sharing, open web search, public link sharing, third-party plugin marketplaces No healthcare footprint : zero healthcare customers, case studies, or compliance documentation When in doubt, involve your compliance or legal team before any PHI is involved. Assume a tool is not compliant until documentation and a signed agreement prove otherwise. ## A buyer's checklist for HIPAA compliant AI Use these questions in any vendor evaluation: Will you sign a BAA, and does it cover every plan and feature we intend to use? Is our data ever used to train AI models or shared with third-party AI providers? What encryption do you use in transit and at rest, and can you provide current security attestations? How long is our data retained, who controls deletion, and where is it stored geographically? Which features are excluded from BAA coverage, and can administrators disable them? How do you support our own obligations: audit logs we can review, access controls we can configure, breach notification timelines in writing? A credible healthcare AI vendor answers all six in writing without hesitation. If you are evaluating AI scribes specifically, see how BastionGPT compares with Abridge , Suki , Nabla , DAX Copilot , and Freed on HIPAA readiness, BAA coverage, and pricing. ## How BastionGPT approaches HIPAA compliance BastionGPT was built for exactly this problem: giving clinicians the capability of advanced AI inside a platform designed for healthcare's rules. BAA included on every plan. Every BastionGPT plan includes a Business Associate Agreement by default, including individual plans, so a solo therapist gets the same contractual protection as a hospital system. Zero data exposure. Your prompts, documents, and audio are never shared with OpenAI or other third-party AI providers for training. You retain full ownership of your data, and chats and transcripts are wiped after 30 days by default, with user control to retain longer or delete sooner. Security you can verify. Data is encrypted with AES-256 at rest and HTTPS/TLS in transit, hosted in HITRUST and ISO 27001-certified data centers, with regional hosting in the USA, Canada, and Australia. BastionGPT also supports PIPEDA, PIPA, and PHIPA (Canada) and the APP (Australia) for teams outside the US. Scoped for healthcare work. BastionGPT combines a clinical AI assistant for drafting SOAP, DAP, BIRP, and progress notes with unlimited secure AI Scribe transcription and document analysis, working alongside EMRs like Epic, Cerner, and TherapyNotes. BastionGPT assists with documentation; it does not replace professional judgment, and clinicians review all AI-generated output. If your team is weighing AI tools against the checklist above, see how BastionGPT compares or start a trial with a BAA in place from day one. ## Frequently asked questions ### What does "HIPAA compliant AI" actually mean? It means the AI service handles PHI under a signed BAA, meets the Security Rule's safeguards (encryption, access controls, audit logging), and restricts data use to providing the service. No AI model is compliant on its own; compliance describes the contractual and technical arrangement around the tool and how your organization uses it. ### Is ChatGPT HIPAA compliant? The consumer versions of ChatGPT (Free, Plus, Pro, and Team) are not HIPAA compliant because OpenAI does not offer a BAA for them. OpenAI does offer BAAs for its Zero Data Retention API and sales-managed enterprise offerings, which require enterprise contracts and configuration. Entering PHI into consumer ChatGPT is an impermissible disclosure under HIPAA. Full guide: Is ChatGPT HIPAA Compliant? ### Does a BAA alone make an AI tool HIPAA compliant? No. The BAA is necessary but covers only the vendor's obligations. Your organization still must conduct risk analysis, train staff, configure access controls, and ensure the covered product tier and features are the ones actually in use. ### Can I use AI with de-identified patient data? Only if the data is de-identified under HIPAA's standards: the Safe Harbor method (removing all 18 identifier categories) or expert determination. Manually deleting a patient's name before pasting a note into a chatbot does not meet either standard, because remaining details like dates, locations, and unique clinical facts can still identify the patient. ### Is it a HIPAA violation if the AI company never leaks the data? Yes, it can be. Disclosing PHI to a vendor without a BAA is itself an impermissible disclosure. The violation occurs when the data leaves your control, regardless of whether any downstream harm follows. ### What penalties apply to HIPAA violations involving AI tools? The same tiered civil penalties as any HIPAA violation. Under the schedule HHS updated in January 2026, penalties range from $145 to $73,011 per violation in the lower culpability tiers and reach $2,190,294 per violation for uncorrected willful neglect, alongside potential state attorney general actions and breach notification costs. A growing patchwork of state laws on healthcare AI can add further obligations on top of HIPAA. ## The bottom line What makes an AI HIPAA compliant is not the model's name or its capabilities. It is the safeguards around it: a BAA that covers the product you actually use, security controls you can verify, contractual limits on data use, and an organization that backs its policies with technical enforcement and a compliant alternative staff want to use. Verify those four things before any patient data touches an AI tool, and you can bring modern AI into patient care with confidence. Privacy is one dimension of responsible AI; clinicians should also weigh other risks such as bias in healthcare AI . ## Sources American Medical Association, 2026 Physician Survey on Augmented Intelligence (March 2026) The HIPAA Journal, Largest Healthcare Data Breaches of 2025 The HIPAA Journal, Healthcare Data Breach Statistics The HIPAA Journal, HIPAA Violation Fines (penalty schedule updated January 28, 2026) IBM, Cost of a Data Breach Report 2025 (healthcare industry average) OpenAI Help Center, How can I get a Business Associate Agreement (BAA) with OpenAI OpenAI, Enterprise Privacy HHS Office for Civil Rights, Guidance on De-identification of PHI (Safe Harbor, 45 CFR 164.514(b)) HHS, Business Associate Contracts (45 CFR 164.502(e), 164.504(e)) Wolters Kluwer, Survey Finds Broad Presence of Unsanctioned AI Tools in Hospitals and Health Systems (January 2026) ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/is-chatgpt-hipaa-compliant-what-healthcare-professionals-must-know Healthcare Innovation # Is ChatGPT HIPAA Compliant? Written By: Josh Spencer • March 26, 2025 Healthcare and mental health professionals are increasingly using AI tools like ChatGPT for tasks such as note-taking, report writing, and patient communication. However, using ChatGPT in a clinical setting can pose serious HIPAA compliance risks and open you up to sanctions, fines and lawsuits. As AI tools become more accessible, it's crucial for healthcare providers, clinics, and administrators to understand how these technologies interact with HIPAA regulations. Even if PHI isn't intended to be used with ChatGPT, even accidental use with PHI is a HIPAA breach. For the full checklist of requirements, see our guide to what makes an AI tool HIPAA compliant . This article breaks down why standard ChatGPT can fail HIPAA requirements, explores potential compliant options, and helps professionals understand how to access and use ChatGPT AI models with patient data in a secure and compliant manner. ## Is ChatGPT HIPAA Compliant? Standard ChatGPT versions (Free, Plus, Pro, Team) are not HIPAA compliant and should not be used with PHI due to the lack of a BAA and data usage policies. OpenAI's API and Enterprise/Edu solutions could potentially be used compliantly, but require obtaining a conditional BAA, involve significant cost (often 6+ figures annually for Enterprise), face functional limitations, and require ongoing configuration and oversight by the healthcare organization. The ultimate responsibility for compliant use rests with the healthcare provider, including robust training, policies, and monitoring. ## ChatGPT Compliance Risks OpenAI offers several readily accessible versions of ChatGPT (Free, Plus, Pro, and Team). While powerful, using these standard tiers with PHI poses significant HIPAA compliance risks: ### Default Data Usage For ChatGPT Free, Plus, and Pro, OpenAI's default policy allows submitted content to be used for training its AI models. While users can opt-out, this default practice conflicts with HIPAA's principle of using PHI only for permitted purposes (like treatment or the specific service requested). Using patient data to train a general AI model without proper patient consent is almost always not a HIPAA-permissible use. By contrast, our support center explains how BastionGPT avoids training on your data . ### Lack of BAA This is the most critical issue. OpenAI explicitly states it does not offer a BAA for its standard ChatGPT services, including Free, Plus, Pro, and even the Team plan. While Team might have SOC 2 compliance, this doesn't fulfill the BAA requirement under HIPAA. ### The Inherent HIPAA Violation Inputting any PHI into these standard ChatGPT versions constitutes a HIPAA violation primarily because there is no BAA in place between the healthcare provider (Covered Entity) and OpenAI (acting as a Business Associate). Without a BAA, OpenAI has not contractually agreed to HIPAA's specific rules for protecting PHI. ### Data Retention Even if chat history is off or training opted out, OpenAI may retain conversation data for up to 30 days for monitoring purposes. Storing potential PHI without a BAA, even temporarily, raises compliance issues. ### Risk of Accidental Exposure Staff might inadvertently paste PHI into these tools due to lack of training or simple error, leading to unauthorized disclosure. Given these factors, standard ChatGPT versions are unsuitable for any tasks involving identifiable patient information. ## What Open AI Solutions Are HIPAA Compliant? OpenAI offers its API platform and premium ChatGPT Enterprise/Edu solutions that provide additional compliance and security features. These present potential pathways toward HIPAA compliance, but with significant caveats: BAA Availability (Conditional): OpenAI is willing to consider signing BAAs for its API, ChatGPT Enterprise, and ChatGPT Edu services. However, this is not automatic. Organizations typically need to contact OpenAI sales or a dedicated email address, undergo a review process, and approval is not guaranteed. Eligibility often requires specific account types or minimum commitments that, as of this writing, are limited to organizations with over $108,000 of annual spend. API BAA Limitations (Zero Data Retention - ZDR): A crucial point for the API is that OpenAI's BAA only covers API endpoints eligible for Zero Data Retention (ZDR). ZDR means OpenAI doesn't store request/response data long-term. Many advanced API features, including Assistants, Threads, Files API, Image Generation, and image inputs to chat, are not ZDR-eligible and thus fall outside the scope of the BAA for PHI use. This significantly limits the API's functionality for complex healthcare tasks involving PHI. Enterprise/Edu Features: These tiers offer enhanced security (SOC 2 Type 2, encryption, SSO), administrative controls, and data submitted is not used for training by default. These are valuable but still require a signed BAA for HIPAA compliance when handling PHI. The Cost Factor: Achieving HIPAA compliance via OpenAI involves substantial cost. API costs are usage-based and can add up quickly. ChatGPT Enterprise reportedly requires minimums like 150 users at ~$60/user/month. This price point makes HIPAA-eligible ChatGPT inaccessible for most small-to-medium practices, clinics, and individual professionals. Additionally, maintaining a secure and compliant configuration within these services requires ongoing technical and administrative maintenance. ## Practical Hurdles and Persistent Compliance Gaps with ChatGPT Even when using the API or Enterprise tiers with a BAA, significant challenges remain: Complexity and Configuration: Correctly configuring API calls (using only ZDR endpoints for PHI), managing security, and understanding the BAA's limitations requires technical expertise. The responsibility for compliant implementation lies heavily with the user organization. Feature Limitations: The exclusion of many advanced API features from ZDR eligibility restricts sophisticated healthcare workflows involving PHI (e.g., analyzing uploaded patient documents via the Files API or interpreting medical images). User Error Risk: Employees can still cause violations by accidentally using non-compliant tools/endpoints, misusing features like web search or research (which could leak PHI in search queries if enabled), or inadequately de-identifying data. Accuracy Concerns: While not a direct HIPAA violation, the potential for AI "hallucinations" (generating incorrect information) poses clinical risks if outputs aren't rigorously reviewed by qualified professionals. Services like ChatGPT make efforts to reduce hallucinations, but those are broad in nature and aren't specialized in reducing healthcare hallucinations. ## ChatGPT vs. BastionGPT: HIPAA Compliance Comparison The complexities, costs, and limitations of adopting a general-purpose tool like ChatGPT for healthcare highlight the need for specialized solutions designed with security and compliance at their core. BastionGPT is the leading solution for healthcare professionals, which utilizes OpenAI-licensed models. Developed by healthcare security and compliance experts, BastionGPT ‘s framework is inherently designed for HIPAA compliance. Key differences include: Compliance by Design: Security and compliance (HIPAA, PIPEDA, APP) are foundational, not add-ons. It assumes all input might be regulated data. Automatic BAA: A HIPAA BAA is included automatically for all subscription plans, removing a significant barrier for healthcare users. Guaranteed Data Confidentiality: BastionGPT contractually ensures user prompts, documents, and AI outputs are never shared or sold to underlying AI providers (like OpenAI) for training. Data remains confidential within our secure platform. Secure Infrastructure: All features use only HIPAA-secure hosting with robust encryption, access controls, and regular security testing. Risk Mitigation by Design: Features known to pose PHI exposure risks, like integrated web search, are intentionally omitted because secure, HIPAA-compliant options from search providers are currently unavailable. This "compliance-by-design" philosophy aims to minimize the operational burden and inherent risks associated with retrofitting compliance onto general-purpose tools. You can learn more about BastionGPT's Business Associate Agreement in our support center. ## Alternative HIPAA Compliant Solution to ChatGPT Evaluating solutions like BastionGPT , which still provide essentially the same cutting-edge AI models as ChatGPT but are explicitly architected for healthcare security and compliance from the ground up, offers a more powerful and often more accessible pathway. By prioritizing purpose-built, compliant tools, healthcare can adopt the power of AI responsibly, ensuring that innovation enhances care without compromising patient trust or regulatory obligations. For a wider view, read about balancing the risks and rewards of ChatGPT in healthcare . BastionGPT provides the technical expertise to support use of generative AI LLMs like those which power ChatGPT while staying in compliance with HIPAA. We do the hard work for you so you can experience the benefits of using a tool like ChatGPT in the healthcare space. Explore our HIPAA-compliant medical ChatGPT built for healthcare professionals. Our support center also covers why clinicians can't use consumer ChatGPT in more detail. Most organizations are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting Disclaimer: This article provides general information about HIPAA compliance and AI tools based on publicly available information as of April 20th, 2026. It does not constitute legal advice. Healthcare organizations should consult with qualified legal counsel and compliance experts to ensure their specific use of any technology meets HIPAA requirements and other applicable regulations. AI provider policies and features are subject to change. ## FAQs ### Is ChatGPT HIPAA compliant? Standard ChatGPT versions (Free, Plus, Pro, and Business) are not HIPAA compliant. OpenAI does not offer a BAA for these tiers, and its default policy allows submitted content to be used for model training. Entering any identifiable patient information into these versions is considered a HIPAA violation. ### Can healthcare providers use ChatGPT securely? OpenAI's API and ChatGPT Enterprise/Edu solutions offer potential pathways toward compliance, but with significant caveats. Enterprise requires a minimum of approximately 150 users at roughly $60/user/month, and API BAAs only cover endpoints eligible for Zero Data Retention, which excludes many advanced features. Most small and mid-size practices find these options inaccessible. ### Does OpenAI sign a BAA for ChatGPT? OpenAI is willing to consider signing BAAs for its API, ChatGPT Enterprise, ChatGPT Edu, and ChatGPT for Healthcare services, but this is not automatic. Enterprise requires a 150-seat minimum and annual contract, putting the cost floor around $108,000/year. Standard ChatGPT plans (Free, Plus, Pro, Business) do not include a BAA. ### What are HIPAA-compliant alternatives to ChatGPT? Purpose-built healthcare AI platforms like BastionGPT provide access to the same advanced AI models (including OpenAI's latest GPT models) within a framework designed for HIPAA compliance from the ground up. BastionGPT includes a BAA automatically on all subscription plans and contractually commits that user data is never shared with AI providers for training. Plans start at $20/user/month with a 7-day free trial. ### How does BastionGPT support HIPAA compliance? BastionGPT's security and compliance framework (HIPAA, PIPEDA, APP) is foundational, and the platform assumes all input might be regulated data. All features use HIPAA-secure hosting with encryption, access controls, and regular security testing. Data is encrypted in transit (TLS) and at rest (AES-256), hosted on Microsoft Azure, and wiped after 30 days by default. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/is-claude-hipaa-compliant Healthcare Innovation # Is Claude HIPAA Compliant? (2026 Guide) Written By: Robert Henehan • April 22, 2026 Artificial intelligence is changing how healthcare professionals document care, communicate with patients, and manage administrative workflows. Anthropic's Claude, one of the most capable generative AI models available, is gaining traction among clinicians, therapists, and healthcare administrators for tasks like drafting SOAP notes, summarizing patient records, and composing referral letters. But healthcare data carries some of the strictest regulatory protections in the United States. Protected Health Information (PHI) is governed by HIPAA, and providers face fines, sanctions, and reputational harm if they mishandle it. That raises a critical question for any healthcare professional considering Claude: ## Is Claude HIPAA compliant? ### The short answer: Claude is not HIPAA compliant by default. Anthropic offers a path to HIPAA-ready use through its Enterprise product, but only on that tier, with a seat minimum and configuration requirements. For most healthcare organizations, whether solo practices or large health systems, using standard Claude plans with patient data creates serious compliance risk. This article breaks down the compliance landscape for Claude in healthcare, outlines the risks, explains what Anthropic does and does not cover, and presents alternatives designed specifically for clinical environments. Claude is one case of a general framework. For the requirements any AI tool must meet before it touches patient data, including the BAA, technical safeguards, and data-use restrictions, see our guide to what makes an AI HIPAA compliant . ## Claude's Standard Plans: Understanding the Compliance Risks Anthropic offers several plans intended for individual or team use: Free, Pro ($20/month), Max ($100 to $200/month), and Team ($20/seat/month billed annually, or $25 monthly, for teams of 2 to 150 seats). These plans give users access to powerful AI capabilities, but none of them include a BAA, and none are appropriate for use with PHI. Here is why: ### 1. No BAA on consumer or standard plans Anthropic's own privacy documentation is explicit that its Business Associate Agreement (BAA) covers Claude Enterprise and the first-party Claude API in HIPAA-ready configurations, and does not cover Claude Free, Pro, Max, or Team plans. Without a BAA, any use of PHI with these plans is a HIPAA violation by definition. ### 2. Data may be used for model training In September 2025, Anthropic introduced an opt-in toggle allowing consumer plan users (Free, Pro, Max) to share their conversations for AI model training. Users who opt in are subject to a data retention period of up to five years, a dramatic increase from the previous 30-day default. Even users who opt out still have data retained for up to 30 days on backend systems. ### 3. No healthcare-specific safeguards Claude's standard plans are general-purpose AI tools. They lack the infrastructure, access controls, audit logging, and data isolation that HIPAA's Security Rule requires when handling electronic PHI. ### 4. Risk of accidental PHI exposure Staff may assume that paying for a Pro, Teams, or Max subscription makes Claude "safe" for clinical use. A therapist drafting session notes, a physician summarizing a patient encounter, or a billing coordinator reviewing a claim could inadvertently expose PHI without realizing the compliance implications. Given these factors, Claude's consumer and standard plans should not be used for any task involving identifiable patient information. ## Claude's Data Training Policy: A Closer Look Anthropic's consumer privacy policy deserves special attention from healthcare professionals. Under Anthropic's current Consumer Terms, users of Claude Free, Pro, Teams, and Max can opt in to allow their conversations to be used for model training. Users who opt in are subject to a data retention period of up to five years, far longer than the 30-day default. Users who opt out continue with the 30-day retention window, though data may still be held on backend systems for that period. The concern for healthcare is twofold: Individual employees may opt in without organizational awareness. A clinician using a personal Claude account might enable the training toggle without understanding that their conversation data, potentially including PHI, could be retained for years in Anthropic's training pipeline. This creates a compliance gap that is difficult to detect and monitor. De-identification does not eliminate risk. HIPAA's Safe Harbor method for de-identification requires removal of 18 specific identifiers. Anthropic states it uses "a combination of tools and automated processes to filter or obfuscate sensitive data," but this is not equivalent to a formal HIPAA de-identification process. Healthcare organizations cannot rely on Anthropic's internal processes to meet their own compliance obligations. These data training policies do not apply to commercial products (Claude for Work, Enterprise), but the risk lies in the gap between what an organization intends and what individual users actually do. ## Can Claude Be Used in a HIPAA-Compliant Way? Anthropic does offer a pathway toward HIPAA-ready use, but it comes with real barriers for smaller healthcare organizations. ### The HIPAA-ready offering requires an Enterprise plan Anthropic's HIPAA-ready configuration is available for Enterprise plans only, and its help center is explicit that this covers both the self-serve and the sales-assisted route. Self-serve Enterprise carries a 20-seat minimum at $20 per seat per month billed annually; sales-assisted Enterprise carries a 50-seat minimum and custom pricing. Either way, usage is billed separately at API rates on top of seat fees. At the published $20 seat rate, a 20-seat HIPAA-ready contract runs $4,800 per year in seat fees before a single token of usage. The Primary Owner enables HIPAA-ready configuration and accepts the BAA directly in organization settings, so the self-serve path involves no sales or legal cycle, but the 20-seat floor still puts it out of reach for solo practitioners, small group practices, and many mid-size clinics that would be paying for seats they cannot fill. ### Some features are excluded from compliance coverage Even on a HIPAA-ready plan, Anthropic publishes a list of features that stay outside the scope of the BAA. Cowork is not an eligible service in any configuration. Claude Code is covered only in local mode with zero data retention on qualified accounts, while its web, remote-desktop, review, and security surfaces are excluded. Beta features such as Claude Design and Claude for Office are excluded, as are several API surfaces including Batch, Files, Skills, Code Execution, Computer Use, and Web Fetch. Connectors, Enterprise Search, and Claude in Chrome are covered except for data sent to third parties. Organizations must understand which features are and are not covered, monitor changes as Anthropic releases updates, and train staff accordingly. ### Configuration and monitoring fall on the organization Signing up for an Enterprise plan does not make Claude compliant by itself. Administrators must configure access controls, disable risky features, maintain audit logs, and monitor usage on an ongoing basis. As Anthropic releases new features, each one must be evaluated for compliance impact before it is enabled. ### Cloud alternatives add more complexity Claude is also available through AWS Bedrock, Google Cloud Vertex AI, and Microsoft Azure, each with their own BAAs covering the infrastructure layer. These paths are technically viable but require cloud engineering expertise, add infrastructure costs, and place full configuration responsibility on the organization. ## Practical Challenges of Using Claude in Healthcare Even for organizations that secure a HIPAA-ready Enterprise plan, day-to-day operational challenges remain. ### User confusion Clinicians may not understand why they can use Claude on their work computer but not on a personal device, or why certain features are unavailable. This creates friction and increases the risk of accidental violations. ### Content filtering gaps General-purpose AI platforms sometimes block discussions of sensitive clinical topics like trauma, abuse, substance use, or anatomical details. Healthcare professionals need an AI tool that can handle these subjects without refusal or content warnings that disrupt clinical workflows. ### Accuracy risks AI-generated clinical content must be reviewed by qualified professionals. Outputs involving medication dosages, diagnostic criteria, or treatment protocols require careful verification. Example: A behavioral health practice signs up for Claude's Enterprise plan and configures it for HIPAA-ready use. A new therapist joins the team and, not understanding the distinction, begins using Claude Pro on a personal account to draft therapy notes . Because the personal account has no BAA, every session note entered constitutes a potential HIPAA violation, even though the organization's Enterprise account is properly configured. ## BastionGPT: A Healthcare-First Alternative For organizations that want access to advanced AI without the seat minimum and configuration burden that Claude's HIPAA-ready pathway requires, BastionGPT is a purpose-built solution designed from the ground up for healthcare. Automatic BAA on all plans. Every BastionGPT subscription includes a HIPAA BAA in the terms of use. There is no sales review, no approval process, and no minimum user requirement. A solo practitioner on the $20/month Professional plan receives the same BAA coverage as a 500-person health system. Compliance by design. BastionGPT was developed by healthcare and cybersecurity professionals specifically for HIPAA, PIPEDA, and APP compliance. The platform assumes all input may contain regulated data and treats it accordingly. Proven compliance track record. BastionGPT has offered HIPAA-compliant AI with a BAA since April 2023. Anthropic's HIPAA-ready Enterprise offering launched in late 2025. Data is never used for training. User prompts, documents, and AI outputs are never shared with underlying AI providers for training. There is no opt-in toggle, no 5-year retention pipeline, and no ambiguity. Data is securely wiped after 30 days by default, and users can delete sooner. Multi-model access with built-in security. BastionGPT integrates multiple leading AI models within a single secure interface configured for medical use cases, giving users the power of advanced generative AI without managing separate vendor relationships or compliance configurations. You can review which models power BastionGPT (including Claude) in the BastionGPT support center. Healthcare-specific workflows and AI Scribe. BastionGPT includes tools built for clinical documentation (SOAP, DAP, BIRP notes, treatment plans, referral letters, prior authorizations, and more), a Saved Prompts library for standardizing workflows, and HIPAA-compliant audio transcription with multi-speaker recognition for up to 10 speakers. Healthcare-appropriate content filtering. BastionGPT allows discussions of sensitive clinical topics (trauma, abuse, substance use, anatomical details) that standard AI platforms may block, a critical capability for mental health professionals and clinicians working with complex cases. Accessible pricing. Professional plans start at $20/user/month. Professional Plus is $45/user/month. All plans include a 7-day free trial and a 45-day satisfaction guarantee. Alexis Arceo, CEO, Expedited Reports: "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." David Lopis, Director, Psychology Squared: "BastionGPT has a lot more capabilities with data protection, so clinicians can more freely use AI." ## Comparison: BastionGPT vs. Claude Feature BastionGPT Claude HIPAA Compliance BAA included automatically on all plans. Built from the ground up for healthcare privacy requirements. BAA available on Claude Enterprise only, self-serve from 20 seats or sales-assisted from 50. Free, Pro, Max, and Team are not covered, and a published list of features stays outside the BAA even on the HIPAA-ready plan. Healthcare Specialization Purpose-built for clinical documentation with templates and workflows designed for healthcare providers across specialties. General-purpose AI assistant with strong reasoning capabilities but limited healthcare-specific features. Data Privacy Data is never used for training. Securely wiped after 30 days by default. Consumer plans allow opt-in data training with up to 5-year retention. Enterprise data is not used for training. Document Processing & Transcription Handles documents up to 1,000+ pages, roughly 700,000+ words (Professional Plus). Includes unlimited HIPAA-compliant transcription with multi-speaker recognition for up to 10 speakers and six output formats on all plans. Capable of processing long documents, but transcription and clinical note formatting require custom configuration. Pricing & Accessibility Starts at $20/user/month with BAA included. No minimum seats. Available to any organization regardless of size. Consumer plans start at $20/month but lack HIPAA compliance. HIPAA-ready Enterprise requires a 20-seat minimum on the self-serve tier (50 on sales-assisted), an annual commitment, and usage billed separately at API rates. Minimum roughly $4,800/yr (20 seats × $20 × 12) before usage. Security Infrastructure Microsoft Azure with HIPAA-compliant BAA. AES-256 encryption at rest, TLS in transit. A+ SecurityScorecard rating. Regular vulnerability scans, penetration testing, and code reviews. Enterprise tier offers strong security controls. Also available through AWS Bedrock, Google Vertex AI, and Microsoft Azure with respective cloud BAAs. For a plan-level breakdown of Claude Team and Enterprise against BastionGPT, including pricing, seat minimums, and BAA coverage, see BastionGPT vs Claude Team & Enterprise . ## Conclusion Claude is a capable AI assistant, but it is not HIPAA compliant by default. Standard plans (Free, Pro, Max, Team) should not be used with PHI. Anthropic's HIPAA-ready Enterprise offering requires an Enterprise plan with a 20-seat minimum on the self-serve tier and 50 on the sales-assisted tier, annual contracts, usage billed at API rates, and ongoing configuration and monitoring by the healthcare organization. Most users can sign up and start using BastionGPT in as little as 10 minutes. There are no setup costs, a 7-day free trial, and no fixed commitments. Whether you need secure transcription, support drafting SOAP notes and treatment plans, or help with prior authorizations and insurance appeals, BastionGPT is designed to deliver fast, reliable, HIPAA-compliant results. Begin your journey with a 7-day free trial of BastionGPT . If you have questions or want to connect: Email: [email protected] Phone: +1 (214) 619-8696 Schedule a Chat: Book a Meeting ## Frequently Asked Questions (FAQ) ### Is Claude HIPAA compliant? Claude is not HIPAA compliant by default. Anthropic's BAA does not cover Claude Free, Pro, Max, or Team plans. HIPAA-ready use requires a Claude Enterprise plan, available self-serve from a 20-seat minimum or sales-assisted from 50 seats, with the Primary Owner enabling HIPAA-ready configuration and accepting the BAA in organization settings. ### Does Anthropic sign a BAA for Claude? Yes, for Claude Enterprise customers and for the first-party Claude API in HIPAA-ready configurations. On Enterprise the Primary Owner accepts the BAA directly in organization settings, with no sales or legal cycle required. The BAA is not available for Claude Free, Pro, Max, or Team plans, and a published list of features stays outside its scope even after HIPAA is enabled. ### Can therapists use Claude for session notes? Therapists and mental health professionals should not use Claude's consumer plans (Free, Pro, Max) for session notes or any documentation containing PHI. These plans lack a BAA, and data entered could be retained or used to improve AI models if training is enabled. HIPAA-compliant alternatives like BastionGPT are designed specifically for this use case. ### What happens if I use Claude with patient data on a standard plan? Using PHI with any Claude plan that lacks a BAA constitutes a potential HIPAA violation. This can result in regulatory penalties, breach notification requirements, and reputational damage. HIPAA fines range from thousands to millions of dollars depending on the severity and nature of the violation. ### Does Claude use my data for AI training? On consumer plans (Free, Pro, Teams, Max), Anthropic offers an opt-in setting that allows conversations to be used for model training, with data retained for up to five years. Users who opt out retain the 30-day default retention. Enterprise plans are excluded from this policy. Deleted conversations are not used for training under any plan. ### Is there a HIPAA-compliant alternative to Claude for healthcare? BastionGPT is designed specifically for healthcare, with HIPAA compliance built into every plan. It includes an automatic BAA, HIPAA-compliant transcription , clinical documentation templates, and access to multiple leading AI models within a secure environment. Plans start at $20/user/month with a 7-day free trial. ### Can I use Claude through AWS or Google Cloud for HIPAA compliance? Claude is available through AWS Bedrock, Google Cloud Vertex AI, and Microsoft Azure, each of which offers HIPAA-eligible infrastructure. This creates a technically viable path, but the organization bears full responsibility for compliant configuration, access controls, audit logging, and ongoing monitoring. This approach requires technical expertise and adds infrastructure costs. ### What are the risks of using Claude with PHI? PHI exposure through non-compliant plans, unauthorized data retention, potential use of data for AI model training, regulatory fines, breach notification requirements, and erosion of patient trust. Disclaimer: This article provides general information about HIPAA compliance and AI tools based on publicly available information as of August 2026. It does not constitute legal advice. Healthcare organizations should consult with qualified legal counsel and compliance experts to confirm their specific use of any technology meets HIPAA requirements and other applicable regulations. AI provider policies and features are subject to change. ### Sources Plans & Pricing, Anthropic: https://claude.com/pricing Business Associate Agreements for Commercial Customers, Anthropic Privacy Center: https://privacy.claude.com/en/articles/8114513-business-associate-agreements-baa-for-commercial-customers HIPAA-ready Enterprise plans, Anthropic Help Center: https://support.claude.com/en/articles/13296973-hipaa-ready-enterprise-plans What is the Claude Enterprise plan, Anthropic Help Center: https://support.claude.com/en/articles/9797531-what-is-the-claude-enterprise-plan Updates to Consumer Terms and Privacy Policy, Anthropic: https://www.anthropic.com/news/updates-to-our-consumer-terms ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/is-google-gemini-hipaa-compliant Healthcare Innovation # Is Google Gemini HIPAA Compliant? Written By: Robert Henehan • August 28, 2025 Artificial intelligence is transforming healthcare. From drafting SOAP notes to summarizing lab results, AI tools can reduce the administrative burden that clinicians, therapists, and healthcare administrators face daily. Google Gemini, Google’s generative AI model, is being used across industries for tasks like content creation and productivity. In healthcare, it attempts to help professionals with "HIPAA compliant" access to the Google Workspace. But with healthcare, the stakes are higher and the workflows are more complex. Protected Health Information (PHI) is among the most tightly regulated types of data in the United States. Under HIPAA, providers face fines, sanctions, and reputational damage if they mishandle PHI. That leads to the key question: ## Is Google Gemini HIPAA compliant? The answer is complex. Gemini can be used in HIPAA-compliant ways, but only in certain enterprise environments, with signed contracts and detailed configuration. For most healthcare organizations, whether large or small, that path introduces risk, expense, and complexity. This article unpacks the compliance requirements, the risks, and the alternatives, providing healthcare professionals with a clear understanding of what Gemini means for HIPAA. Before weighing any single tool, it helps to know what makes an AI HIPAA compliant in the first place: a signed BAA, verified technical safeguards, and strict limits on how your data is used. ### When Can Gemini Be Used in a HIPAA-Compliant Way? Gemini is not HIPAA compliant across all settings. Consumer Gemini, accessed through personal Google accounts, is not HIPAA compliant. Any PHI entered into those tools risks exposure. Gemini integrated into enterprise Google Workspace can be HIPAA compliant, but only when covered under a signed Business Associate Agreement (BAA) and properly configured. This is the company's responsibility when setting up Gemini, and not Google's. Experimental features and consumer-facing add-ons often remain outside HIPAA coverage, so reading the fine print to avoid breaches is essential. This means Gemini is never “HIPAA safe” by default. It requires contractual, business process, and technical safeguards. Example: Suppose you are using Gemini to help draft a prior authorization using your session notes. Google will often pass information given to Gemini to the Google web search engine to improve its results. This Google web search is not appropriate for HIPAA regulated data like PHI and PII, as outlined in the fine print of their product terms. If your organization failed to understand that they needed to go into the settings to remove the ability to web search, or to inform their employees that Gemini should not be used for PHI, then the practice exposes themselves to great compliance risk. ## Google Gemini HIPAA Risks Healthcare organizations should begin by recognizing the inherent risks of using Gemini without safeguards. Consumer versions are never compliant: The Gemini app and website operate under Google’s consumer terms of service, which do not include HIPAA protections. Any PHI entered is exposed to non-compliant logging, storage, or potential use in training AI models. Default data handling policies create exposure: In consumer Gemini, prompts and responses can be accessible by individuals in ways that go against common organizational requirements and retained for reasons beyond what are permitted by HIPAA and best practices for retention of PHI and PII. Accidental usage is common: Staff may assume “Google products” are safe and paste PHI into Gemini or other tools in their Google products without realizing the repercussions. Legal and financial consequences are severe: HIPAA fines range from thousands to millions of dollars, and breaches erode patient trust. Example: Imagine a therapist accesses Gemini and didn't notice that Google logged them into their personal Google account instead of their work account, a frequent occurrence for users who use both work and personal Google accounts. If the therapist provided PHI in that session via dictation or document upload, that information would not be protected under the standards required of HIPAA and other common healthcare regulation, which is often considered a breach that could constitute regulatory penalties. ### Gemini BAA: Do You Need a Business Associate Agreement? A Business Associate Agreement (BAA) is essential when using any third-party service to process PHI. Under HIPAA, a Business Associate is any entity that handles PHI on behalf of a covered entity. The BAA establishes legal obligations for safeguarding PHI. Google does provide BAAs for Google Workspace, including Gemini functionality. This means organizations that use Workspace with Gemini under an executed BAA can process PHI, provided they also follow other HIPAA security practices. Without a BAA: Using Gemini with PHI is a violation, no matter the safeguards. With a BAA: Some versions of Gemini can be compliant, but coverage is limited to specific services and requires configuration and education. ### What Configuration Is Needed for HIPAA Compliance with Gemini? Configuration is the critical step where many organizations fail. Having a BAA does not make Gemini compliant by itself. Administrators must configure the environment to enforce HIPAA standards. Key configurations include: Disable web search grounding. Gemini may suggest retrying a query with Google Search. If PHI is included, this exposes data outside the compliant environment. Block third-party extensions. Only Workspace-native Gemini features are covered under the BAA. Third-party integrations often fall outside compliance. Protect metadata. PHI must not appear in document titles, filenames, or log data. Metadata is often overlooked but can create exposure when placed in many Google systems. Enable information rights management (IRM). IRM restricts who can view, download, or share PHI-containing documents. Example: If a clinician schedules a session with a patient and wants it to be transcribed securely, but Gemini applies the patient name in the details like meeting attendees or subject line, that data could be sent to unauthorized 3rd parties via the use of a default Zoom plugin. ### What Are HIPAA Requirements for Google Gemini? Even with a signed BAA, organizations must still meet HIPAA’s operational requirements to use Gemini compliantly. These include: Use only HIPAA-covered services: Only Gemini in Workspace, under the BAA, can handle PHI. Consumer or experimental versions cannot. Configure security controls: Features must be restricted to prevent leakage or storage of PHI outside approved boundaries. Implement access control: Only authorized clinicians or staff should be able to use Gemini with PHI. User training: Staff must understand the difference between compliant and non-compliant environments, and even how to ensure data stay safe within compliant environments. Audit and monitor use: Organizations must maintain logs and review them regularly for inappropriate usage. Example: A behavioral health clinic may use Gemini in Workspace to help summarize intake assessments. The clinic must configure Workspace settings to prevent PHI from being exported to non-approved locations and must train clinicians never to use personal Gemini apps with patient details. ## Practical Challenges of Using Gemini in Healthcare While technically possible, using Gemini in a HIPAA-compliant way is challenging. Configuration complexity: IT administrators must constantly monitor Gemini features as they evolve, disabling or restricting risky ones. User confusion: Clinicians may not understand why they cannot use Gemini on personal devices. This creates friction and increases risk of accidental violations. Audit burden: Regular reviews of logs and usage patterns take time and resources. Enterprise focus: Gemini HIPAA compliance is designed around large enterprise customers with security and compliance staff to maintain an acceptable level of diligence. Smaller organizations often find the requirements too heavy. Persistent risk: Even in well-configured environments, the risk of error remains. Example: A hospital might successfully configure Gemini under Workspace, but if even one clinician copies PHI into a non-covered Gemini app on their phone or to an insecure plugin, this opens the door for severe compliance risks. Managing this risk across hundreds or thousands of users is a major challenge. ## BastionGPT: A Healthcare-First Alternative For organizations that want Gemini-level AI capabilities without a minimal compliance burden, BastionGPT is a purpose-built solution that combines advanced AI models from OpenAI and Google, but with healthcare-first safeguards. You can read more about how BastionGPT licenses Google's models in our support center. Automatic BAA: Every subscription includes a HIPAA BAA in our terms of use. Compliance by design: BastionGPT was created specifically for HIPAA, PIPEDA , and APP compliance. Features that expose PHI to breaches are disabled by default. Safe data handling: PHI is never used for training or shared with third parties. Healthcare-specific workflows: Includes SOAP notes, treatment plans, referral letters, prior authorizations, claims review, and patient education templates. AI Transcription: Automatically transcribes sessions and generates clinical notes in a HIPAA-compliant manner. AI Assistant: Helps clinicians review claims, draft prior authorizations, and provide structured feedback while staying within compliance boundaries. Multiple model access: BastionGPT gives users access to Gemini’s AI models, alongside GPT and Claude models, so you get the power of Google Gemini with greater flexibility, choice, and built-in security. Accessible for all organizations: Pricing starts at $20 per month, making it available to both large health systems and independent providers. Example: A multi-specialty practice uses BastionGPT to transcribe therapy sessions, generate treatment plans, review claims, and prepare insurance prior authorizations. Because compliance and clinical workflows are built in, the practice does not need to routinely monitor new features for configuration updates, manage complex enterprise licensing, or constantly monitor logs for signs of unapproved uses. They gain the flexibility of using Gemini, OpenAI, and Claude models in a single secure environment tailored for healthcare. ## Comparison: BastionGPT vs Google Gemini HIPAA Compliance BastionGPT: Straightforward compliance with BAA included in all plans. Built from the ground up for healthcare privacy requirements. Google Gemini: HIPAA compliance possible only with enterprise BAA, complex configuration, limited products and continuous oversight. Healthcare Specialization BastionGPT: Purpose-built for clinical documentation with templates and workflows for healthcare providers. Google Gemini: General-purpose AI with limited healthcare-specific features. Requires users to build custom workflows. Document Processing and Transcription BastionGPT: Handles large documents, transcribes sessions, and integrates clinical terminology. Google Gemini: Capable of summarizing text but lacks healthcare-specific tuning and focus. Pricing and Accessibility BastionGPT: Starts at $20 per month, available to any organization regardless of size. Google Gemini: Tied to enterprise licensing and pricing. Requires IT and compliance teams to manage safely. AI Technology BastionGPT: Uses multiple models (GPT, Claude, and Gemini) optimized for healthcare. Google Gemini: Models are optimized for general productivity, often not capable of discussing sensitive healthcare topics like violence and abuse. Healthcare Features BastionGPT: Includes AI Transcription and Assistant tools for SOAP notes, treatment plans, prior authorizations, claims review, patient education, and so much more. Google Gemini: Limited built-in healthcare features. Requires manual customization to approximate clinical workflows. ## Conclusion Google Gemini is powerful, but it is not inherently HIPAA compliant. With a signed BAA, strict configuration, and ongoing monitoring, it can be used in healthcare. However, the complexity, quality drop, and compliance risk are significant for any organization, whether small or enterprise-scale. Most users can sign up and start using BastionGPT in as little as 10 minutes. There are no setup costs, a 7-day free trial, and no fixed commitments . Whether you need secure transcription, help drafting SOAP notes and treatment plans, or support with prior authorizations and claims, BastionGPT is designed to deliver fast, reliable, HIPAA-compliant results. Begin your journey with a 7-day Free trial of BastionGPT . If you have questions or want to connect: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Frequently Asked Questions (FAQ) ### Is Google Gemini HIPAA compliant? Yes, but only when used in enterprise Workspace with a signed BAA and configured for HIPAA. Consumer Gemini is not compliant. ### Does Google sign a BAA for Gemini? Yes, but only for Workspace Gemini in enterprise settings, not for consumer tools. ### What are HIPAA requirements for Google Gemini? Signed BAA, use of only covered features, strict configuration, access controls, training, and audits. ### What configuration is needed for HIPAA compliance with Gemini? Disable risky features like web search, block non-compliant third-party extensions, enforce required security monitoring processes. ### What are the risks of using Gemini with PHI? PHI leakage, unauthorized storage, legal fines, reputational harm, and patient trust erosion when not configured properly. ### Is there a safer alternative to Gemini for healthcare? Yes. BastionGPT is designed specifically for healthcare, with built-in HIPAA compliance, transcription, and AI Assistant capabilities. Disclaimer: This article provides general information about HIPAA compliance and AI tools as of September 2025 based upon publicly available information. It does not constitute legal advice. Healthcare organizations should consult with legal counsel and compliance experts to confirm requirements. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/is-otter-ai-hipaa-compliant Healthcare Innovation # Is Otter.ai HIPAA Compliant? Written By: Robert Henehan • September 19, 2026 Meeting transcription tools have quietly become some of the most widely used software in professional life. Otter.ai records calls, produces a searchable transcript, and writes a summary, and can join calendar meetings automatically and can join calendar meetings automatically. Which is exactly how it ends up in a healthcare setting. A therapist who uses Otter for supervision calls wonders whether it could capture a session. A practice manager who relies on it for staff meetings considers it for intake calls. The tool is already installed, already familiar, and already producing the kind of transcript that would save real time on documentation. Protected Health Information carries some of the strictest regulatory protections in the United States. Before anyone records a patient encounter with any tool, one question has to be answered first. ## Is Otter.ai HIPAA compliant? ### The short answer: yes it might be on Enterprise, no on every other plan Otter.ai supports HIPAA compliance , but only on its Enterprise plan, only as an add-on, and only once a Business Associate Agreement has been signed. Otter's own Help Center states it plainly: "HIPAA compliance is only available for the Enterprise plan. Reach out to your account manager to start the Business Associate Agreement (BAA) process." Otter's pricing page lists "HIPAA compliance (add-on)" as an Enterprise line item, with pricing available from Otter. That produces a clean split: Basic (free), Pro, and Business plans: no BAA is available, so these plans should not be used with PHI. Under the HIPAA Privacy Rule, disclosing PHI to a vendor acting as your business associate without a signed BAA is not a permitted disclosure. Enterprise plan with the HIPAA add-on and an executed BAA: Otter's supported path for healthcare organizations handling PHI. HIPAA business-associate status depends on the vendor’s role; the BAA documents the required obligations. So the practical question for most people searching this is not whether Otter can ever be HIPAA compliant. It is whether the version of Otter they have is, and for a therapist in solo practice on Pro or a five-person clinic on Business, the answer is no. This article covers what Otter's published policies actually say, what the Enterprise path requires, why a recent federal court ruling matters for anyone recording patient conversations, and what to use instead. Otter is one case of a general framework. For the requirements any AI tool must meet before it touches patient data, including the BAA, technical safeguards, and data-use restrictions, see our guide to what makes an AI HIPAA compliant . ## Otter.ai's Plans: Where the Compliance Line Falls Otter offers four tiers. The capability differences between them are real, and so is the compliance difference. Otter.ai plans Plan Price (Sept 2026) Transcription Conversation history BAA available Basic Price (Sept 2026) Free Transcription 300 min/user/month, 30 min per conversation Conversation history 25 most recent BAA available No Pro Price (Sept 2026) $16.99/user/month, or $8.33 billed annually Transcription 1,200 min/user/month, up to 90 min per meeting Conversation history Unlimited BAA available No Business Price (Sept 2026) $30/user/month, or $19.99 billed annually Transcription Unlimited meetings/in-app recordings, up to 4 hours; imports have separate limits Conversation history Unlimited BAA available No Enterprise Price (Sept 2026) Custom, demo required Transcription Unlimited meetings/in-app recordings, up to 4 hours; imports have separate limits Conversation history Unlimited BAA available Yes, as an add-on ### 1. The BAA is available through Enterprise sales A BAA is the specific contract HIPAA requires between a covered entity and a vendor that creates, receives, maintains, or transmits PHI on its behalf. It is not optional, and no amount of encryption substitutes for it. Otter does offer one. Getting it means moving to Enterprise, adding the HIPAA add-on, and working through an account executive. Otter directs customers to sales for the BAA and pricing. This review found no public Otter BAA template. A practice should request the agreement to review its terms, including indemnification, breach notification timelines, and subcontractor provisions. For an organization with counsel on retainer, that is a normal procurement step. For a smaller practice, it means evaluating a compliance decision while requesting the contract that governs it. ### 2. Enterprise security controls carry their own minimums Otter's Enterprise tier includes SSO, SCIM provisioning, domain capture, activity logs, and custom data retention. Those are the controls Otter's own HIPAA guidance tells customers to configure. Two of them carry a license minimum. Otter's pricing comparison lists a "minimum 100-user license required" for both Single Sign-On and SCIM user provisioning. This is a license minimum for SSO and SCIM; 2FA is available across plans, and HIPAA does not specifically require SSO or SCIM. ### 3. Data retention is not configurable below Enterprise Custom data retention appears only in the Enterprise column of Otter's plan comparison. Below it, scheduled retention is unavailable: Basic provides access to the 25 most recent conversations and archives older ones, and Pro and Business list conversation history as unlimited, likely meaning recordings and transcripts persist until someone deletes them by hand. Neither shape is a retention policy. A free-tier cap that limits access to older conversations is not a data lifecycle rule, and an unlimited archive that only shrinks when a staff member remembers to clear it is the opposite problem. HIPAA-conscious organizations need a retention policy they can implement and document. Otter's HIPAA guidance asks healthcare customers to "configure transcript and recording retention in accordance with their HIPAA-compliant data lifecycle policies," and points them to an account executive to set it up. That control is part of the Enterprise package. ### 4. Compliance work sits with the customer Otter's HIPAA article is unusually candid about how much configuration falls on the healthcare organization. It asks customers to control when the Notetaker joins meetings, review calendar integrations and auto-join settings, disable public and link-based transcript sharing, enforce 2FA and SSO, remove access for departing staff promptly, ideally within 24 hours, review usage logs and dashboards, and set retention policies. That is a real compliance program, not a checkbox. A signed BAA is the entry requirement, not the finish line, and an organization that signs one and changes nothing about its configuration has not done the work. ### 5. The tool is built for meetings, not encounters Otter is a general-purpose meeting assistant. It offers summaries and custom templates that can request formats such as SOAP, DAP, or BIRP. Even on a fully compliant Enterprise deployment, customized output still requires review against clinical documentation requirements. ### 6. Accidental exposure is the realistic failure mode The risk here is rarely a deliberate decision. It is a provider who already uses Otter for internal calls leaving it running for a telehealth session, or the Notetaker joining a video visit because it joins every calendar event. Otter names this risk itself, telling customers they are "responsible for controlling when and how PHI is introduced into the Otter environment," including managing the Notetaker's auto-join behavior. A tool designed to be always on requires deliberate work to keep it off. ### 7. Sensitive clinical content needs careful handling Clinical encounters can involve trauma, abuse, substance use, self-harm, and anatomical detail. Healthcare teams should evaluate whether an AI tool preserves clinically relevant details and produces usable documentation when these topics arise. ## What Otter Does Publish About Security Giving Otter credit where it is due matters, because anyone evaluating this needs the real picture rather than a sales pitch. Otter maintains a public Trust Center at trust.otter.ai listing HIPAA, SOC 2, GDPR, CCPA, and VPAT, with a HIPAA report, SOC 2 report, penetration test report, architecture diagram, HECVAT self-assessment, and information security policy available on request. Otter is SOC 2 Type II certified. Its plan comparison lists TLS encryption and AES-256 encryption for real-time content across tiers, with two-factor authentication available on every plan. Those are legitimate security credentials. They are also a separate question from the one this article is about. ## Why "Secure" Does Not Mean "HIPAA Compliant" This distinction causes more confusion than any other in this category, so it is worth stating plainly. A vendor can encrypt data in transit and at rest, restrict employee access, hold a SOC 2 Type II attestation, publish a trust center, and describe its platform as secure and private, and still not be usable with PHI on the plan you happen to be on. Security certifications describe how a company protects data. A BAA is a contract that assigns legal responsibility for PHI and obligates the vendor to specific safeguards, breach reporting, and subcontractor terms. Both safeguards and the required BAA matter under HIPAA. With Otter, the BAA is tied to a specific plan tier and a specific add-on. ## The Data Training Question Otter's privacy policy describes using customer content to "improve and monitor the Services, including training our proprietary AI technology on de-identified audio recordings and on transcriptions." The policy notes that transcriptions may contain Personal Information. Otter's enterprise marketing page says: "Keep your organization's data out of AI model training. Contact your account manager to get started." Otter's Enterprise admin guide says workspaces are excluded from training by default; opting in requires an account manager. For anyone on Basic, Pro, or Business, the reasonable reading is that eligible recorded content may be used to improve Otter's models; Otter says imported customer data, such as Google Workspace documents, is excluded. ## A Federal Court Ruling Worth Knowing About On August 13, 2026, Judge Eumi K. Lee of the U.S. District Court for the Northern District of California ruled on Otter.ai's motion to dismiss in In re Otter.AI Privacy Litigation , No. 25-cv-06911-EKL. The court granted the motion in part and denied it in part. Claims under the federal Electronic Communications Privacy Act, California's Invasion of Privacy Act, Illinois' Biometric Information Privacy Act, plus unjust enrichment and California's Unfair Competition Law, were allowed to proceed. Claims under the CFAA, CDAFA, and the Washington Privacy Act were dismissed with leave to amend, as were several plaintiffs' privacy claims. Three parts of the reasoning matter for healthcare specifically. The court's analysis turned on what the vendor does with the recording. Otter argued its Notetaker is an invited participant acting as the host's own recording tool. The court held that Otter is plausibly a third-party eavesdropper under CIPA Section 631 based on allegations that it independently collects, retains, and uses conversations for its own commercial purposes, including model training. A vendor that hands the transcript back and keeps nothing is in a different position than one that retains and reuses the content. One medical-call allegation survived the pleading stage. One plaintiff's confidential-communication claim survived specifically because the plaintiff alleged a call with a medical professional involving deeply personal medical information. Other plaintiffs who described their calls only as "private" or "sensitive" had those claims dismissed as too vague. The ruling turned on the specificity and sensitivity of the alleged medical communication. The biometric-collection allegations survived the pleading stage. The court found that Otter's speaker-tagging and speaker-identification features plausibly involve capturing voiceprints, which are biometric identifiers under Illinois' BIPA. Two caveats matter, and they are not small ones. This was a ruling on a motion to dismiss, which tests whether allegations are plausible enough to proceed, not whether they are true. Otter contested the claims. This order made no determination of liability. Court order last verified: September 17, 2026. This summary reflects the August 13, 2026 ruling. Still, the practical takeaway for a covered entity is clear enough. Recording consent is a live legal question in this category right now, and the question worth asking any vendor is whether it retains and reuses your conversation content to train its own models. ## Practical Challenges of Using Otter.ai in Healthcare ### Consent obligations sit on top of everything else Recording a patient encounter raises consent requirements independent of HIPAA. All-party consent states, including California, Illinois, Pennsylvania, Florida, Massachusetts, Washington, and Maryland, require all-party consent for specified private or confidential communications, with scope and exceptions varying by state. Check applicable state rules for the intended behavioral-health workflow. Practitioners retain their consent obligations; vendors may also have obligations. A signed BAA does not satisfy a state recording statute, and every tool, including a purpose-built healthcare one, requires a lawful recording workflow. Obtaining and documenting consent in every state is a recommended practice. ### Meeting bots behave in ways healthcare settings cannot absorb Assistants that auto-join calendar events may join a telehealth appointment the same way they join a team standup. A visible bot in a therapy session changes the session. Otter provides workspace controls to disable Notetaker and manage auto-join, but they have to be configured deliberately, and Enterprise administrators can lock auto-join settings across the workspace. Otter also offers botless recording through its desktop app. ### Output requires rework A meeting summary is not a progress note. Custom templates can request clinical fields, including SOAP structure, session time, risk documentation, and treatment plan linkage; each output needs review against applicable requirements. The transcription saves keystrokes and leaves the documentation work in place. ### Accuracy still requires review Any AI-generated healthcare content requires review by a qualified professional before it enters the record. That is true of every tool in this category, including purpose-built ones. Example. A group practice uses Otter Business for weekly staff meetings, which is a reasonable use with no PHI involved. A newly hired therapist , seeing the tool in the shared workspace, enables it during a telehealth intake to save time on the note. Business does not carry a BAA. The recording, the transcript, and the summary now sit with a vendor the practice has no HIPAA agreement with, and the practice has a potential breach to assess. ## BastionGPT: A Healthcare-First Alternative For healthcare professionals who want the time savings a transcription tool promises without a plan-tier gate in front of the compliance paperwork, BastionGPT is built for healthcare from the ground up. It is used by 10,000+ health organizations and is a Microsoft Partner with Qualys-secured infrastructure. A BAA is included with all plans. Every subscription includes a HIPAA Business Associate Agreement in the terms of use, including the free trial. There is no sales review, no add-on to purchase, and no seat minimum. A solo practitioner on the $20 per month Professional plan receives the included BAA, and signed copies are available via DocuSign on request. Customer Data is not used to train, fine-tune, evaluate, enhance, or improve any AI model. The current Terms state this as a contractual commitment. Transcript retention is configurable, with a 30-day default. Chats remain until deletion or 30 days of account inactivity; users can delete sooner. An AI Scribe built for patient encounters. Unlimited transcription on all plans, subject to fair use, with multi-speaker recognition for up to 10 speakers and recordings up to 240 minutes. Record live or upload audio from in-person visits, telehealth, or phone consults. It works with Zoom, Microsoft Teams, and Google Meet without placing a visible bot in the room, and it supports ambient listening or dictation. Output in the formats documentation actually requires. Available outputs include Transcription, Summary, SOAP Note, Detailed SOAP Note, DAP Note, and Custom Note tabs. The scribe also generates BIRP, H&P, and progress notes, plus custom templates that match your own formatting. Healthcare-appropriate content handling. BastionGPT is designed to support legitimate clinical discussion of sensitive topics, including trauma, abuse, substance use, and anatomical details. More than a scribe. Beyond transcription, BastionGPT drafts referral letters, discharge summaries, and intake reports, analyzes uploaded documents and clinical images, and supports prior authorizations and insurance appeals. It runs multiple leading AI models in one interface, configured for healthcare use, and the model lineup is kept current as new versions are released. Works with what you already use. BastionGPT works alongside 50+ EMR and EHR systems, including Epic, Cerner, athenahealth, eClinicalWorks, SimplePractice, and TherapyNotes, through a copy or upload workflow with nothing for IT to install. Security built for regulated data. Cloud infrastructure including Microsoft Azure, with provider-level ISO 27001 certifications, AES-256 encryption at rest and TLS 1.2 or higher in transit, and third-party penetration testing. Alexis Arceo, CEO, Expedited Reports: "The HIPAA compliance is a huge time saver because I do not have to take out identifying information." Catherine Maxted, RN, Nurse Coordinator: "People compliment the quality and clarity of my documentation all the time now. I can't imagine working without Bastion!" ## Comparison: BastionGPT vs. Otter.ai Feature BastionGPT Otter.ai BAA availability Included with all plans, including the free trial. No seat minimum, no add-on purchase, no sales review. Signed copies via DocuSign on request. Available on the Enterprise plan as an add-on, executed through an account executive. Not available on Basic, Pro, or Business. This review found no public agreement. Healthcare specialization Purpose-built for healthcare documentation, with note formats and workflows designed for clinical work. General-purpose meeting assistant. Custom templates can request healthcare note formats. Data and AI training Customer Data is excluded from AI-model training, fine-tuning, evaluation, enhancement, and improvement. Privacy policy describes training Otter's proprietary AI on de-identified recordings and transcriptions that may contain personal information. Enterprise workspaces are excluded by default. Otter says imported customer data, such as Google Workspace documents, is excluded from training. Data retention control Configurable transcript retention, with a 30-day default; chats remain until deletion or 30 days of account inactivity. Users can delete sooner. Custom data retention listed as an Enterprise feature, configured through an account executive. Basic provides access to the 25 most recent conversations and archives older ones; Pro and Business are unlimited with manual deletion. Identity controls SSO available on Ultra. 2FA across plans. SSO and SCIM listed with a minimum 100-user license required. 2FA across plans. Recording length Up to 240 minutes per recording, unlimited transcription on all plans, subject to fair use. Up to 4 hours per meeting on Business and Enterprise; 90 minutes on Pro and 30 minutes on Basic. 300 monthly minutes on Basic, 1,200 on Pro, unlimited meetings/in-app recordings above; imports have separate limits. Transcription output Transcription, Summary, SOAP, Detailed SOAP, DAP, and Custom Note output options. Up to 10 speakers. Meeting transcript, summary, action items, and custom templates. Meeting platform behavior Works with Zoom, Microsoft Teams, and Google Meet without placing a visible bot in the room. Otter Notetaker joins Zoom, Microsoft Teams, and Google Meet as a participant; auto-join is admin-configurable. Otter also offers botless desktop recording. Pricing and accessibility From $20 per user per month with the BAA included. 7-day free trial, 45-day refund policy, subject to published terms. Free to $30 per user per month for self-serve tiers. Enterprise pricing is quote-based and HIPAA add-on pricing is available from Otter. Security infrastructure Microsoft Azure and Google cloud services, provider-level ISO 27001 certifications, AES-256 at rest, TLS 1.2 or higher in transit, third-party penetration testing. SOC 2 Type II certified, AWS infrastructure, TLS and AES-256 encryption, public trust center with HIPAA, SOC 2, and pentest reports on request. ## Conclusion Otter.ai is a capable meeting transcription tool with real security credentials, and it does offer a path to HIPAA compliance. That path runs through the Enterprise plan, a HIPAA add-on, a signed Business Associate Agreement obtained through sales, and a configuration checklist that Otter hands back to the customer. For a health system with a procurement process and an IT team, that is a workable path worth evaluating on its merits. For the therapist in solo practice, the four-person group, or the practice manager who already has Otter open in another tab, it requires an Enterprise arrangement. On Basic, Pro, and Business there is no BAA, which means those plans should not touch a patient encounter. The gap BastionGPT is built to close is exactly that one. A BAA is included with all plans, including the free trial, and healthcare note formats come standard. Most users can sign up and start using BastionGPT through self-service signup. There are no setup costs, a 7-day free trial, and no fixed commitments. Whether you need secure transcription, help drafting SOAP or DAP notes, or support with prior authorizations and insurance appeals, BastionGPT is designed to support HIPAA-compliant workflows. ## Begin with a 7-day free trial of BastionGPT. BastionGPT is designed to assist healthcare professionals with documentation and clinical workflows. It does not replace professional judgment, clinical expertise, or human oversight. All AI-generated output should be reviewed by qualified professionals before use in clinical settings. If you have questions or want to connect: Email: [email protected] Phone: +1 (214) 619-8696 Schedule a Chat: Book a Meeting ## Frequently Asked Questions (FAQ) ### Is Otter.ai HIPAA compliant? Otter.ai supports HIPAA compliance on its Enterprise plan only, as an add-on, and only after a Business Associate Agreement is signed. Otter's Help Center states that "HIPAA compliance is only available for the Enterprise plan." The Basic, Pro, and Business plans do not carry a BAA, so they should not be used to record, transcribe, or summarize patient encounters. ### Does Otter.ai sign a BAA? Yes, for Enterprise customers. Otter's Help Center directs customers to contact their account executive to start the BAA process. This review found no public BAA template or self-serve signature flow. Otter publishes no BAA offering for the free, Pro, or Business tiers. ### How much does Otter.ai HIPAA compliance cost? Otter does not publish a price. HIPAA compliance is listed on the pricing page as an Enterprise add-on, and Enterprise itself is quote-based and requires scheduling a demo. Expect a sales conversation rather than a checkout page. ### Can therapists use Otter.ai for session notes? Not on Basic, Pro, or Business, because no BAA is available on those plans. A practice on Enterprise with the HIPAA add-on and a signed BAA can, provided it also completes Otter's configuration requirements and obtains patient consent for recording. Custom templates can request SOAP, DAP, or BIRP structure; professional review remains necessary. ### Is Otter.ai secure even where it is not HIPAA compliant? Those are different questions. Otter is SOC 2 Type II certified, publishes a trust center, and encrypts data in transit and at rest. Security describes how a vendor protects data. HIPAA compliance requires a specific contract, the BAA, that assigns legal responsibility for PHI. A tool can be well secured and still be unusable with patient information on the plan you are on. ### How long does Otter.ai keep my recordings? It depends on the plan. Scheduled retention is an Enterprise feature. Basic provides access to the 25 most recent conversations and archives older ones. Pro and Business list conversation history as unlimited, so recordings and transcripts persist until someone deletes them manually. Custom data retention is an Enterprise feature configured through an account executive, which is also where Otter's HIPAA guidance points healthcare customers. ### What happens if I already recorded a patient session with Otter.ai? If a BAA was required and the account had none in place, disclosing PHI to that vendor was not a permitted disclosure. The appropriate next steps are to stop the practice, determine what was recorded and where it is stored, preserve the incident facts and coordinate mitigation with your compliance officer, apply HIPAA’s breach presumption, exceptions, and documented risk assessment to determine notification obligations, and consult counsel or your compliance officer. ### Does Otter.ai use my recordings to train AI models? Otter's privacy policy describes training its proprietary AI technology on de-identified audio recordings and transcriptions that may contain personal information. Its enterprise marketing page says: "Keep your organization's data out of AI model training. Contact your account manager to get started." The Enterprise admin guide states that exclusion is the default; opting in requires an account manager. Otter says imported customer data, such as Google Workspace documents, is excluded from training. ### What is the Otter.ai lawsuit about, and does it affect healthcare users? In In re Otter.AI Privacy Litigation (N.D. Cal., No. 25-cv-06911-EKL), plaintiffs allege Otter's Notetaker recorded meetings without all participants' consent and that Otter retained the recordings to train its models. On August 13, 2026, the court allowed ECPA, CIPA, and BIPA claims to proceed while dismissing others. One plaintiff's claim survived based on sufficiently specific allegations about a confidential medical call. Otter contested the claims; this order made no determination of liability. The practical lesson is to ask any vendor whether it retains and reuses conversation content for model training. ### Is there a HIPAA-compliant alternative to Otter.ai for healthcare professionals? BastionGPT includes a BAA with all plans, including the free trial, with no seat minimum and no add-on to purchase. It provides unlimited HIPAA-compliant transcription, subject to fair use, with multi-speaker recognition for up to 10 speakers and recordings up to 240 minutes. Available outputs include Transcription, Summary, SOAP, Detailed SOAP, DAP, and Custom Note outputs. Plans start at $20 per user per month. ### Do I need patient consent to record a session, even with a compliant tool? Recording consent requirements are separate from HIPAA. Practitioners retain their obligations; vendors may also have obligations. All-party consent states, including California, Illinois, Pennsylvania, Florida, Massachusetts, Washington, and Maryland, require all-party consent for specified private or confidential communications, with scope and exceptions varying by state. Check applicable state rules for the intended behavioral-health workflow. As a recommended practice, obtain and document consent in every state. Disclaimer: This article provides general information about HIPAA compliance and AI tools based on publicly available information as of September 2026. It does not constitute legal advice. Healthcare organizations should consult qualified legal counsel and compliance experts to confirm their specific use of any technology meets HIPAA requirements and other applicable regulations. AI provider policies and features are subject to change. ### Sources Otter.ai Help Center, "HIPAA | Otter.ai": https://help.otter.ai/hc/en-us/articles/33975072019991-HIPAA-Otter-ai (Enterprise-only HIPAA availability, BAA process, customer configuration requirements, SOC 2 Type II reference) Otter.ai Trust Center: https://trust.otter.ai/ (HIPAA, SOC 2, GDPR, CCPA, VPAT listings and available reports) Otter.ai Pricing: https://otter.ai/pricing (plan tiers and prices, "HIPAA compliance (add-on)" line item, transcription minutes and per-meeting caps, "25 most recent" conversation history on Basic, 100-user minimum for SSO and SCIM, custom data retention as an Enterprise feature) Otter.ai Enterprise: https://otter.ai/enterprise (AI training exclusion statement, SSO/SCIM, audit logging) Otter.ai Privacy Policy: https://otter.ai/privacy-policy (training proprietary AI on de-identified recordings and transcriptions) In re Otter.AI Privacy Litigation , No. 25-cv-06911-EKL, 2026 WL 2351237 (N.D. Cal. Aug. 13, 2026) 45 CFR § 164.502(e): https://www.ecfr.gov/current/title-45/section-164.502 (business associate disclosures) 45 CFR § 164.504(e): https://www.ecfr.gov/current/title-45/section-164.504 (business associate contract requirements) HHS, Business Associate Contracts guidance: https://www.hhs.gov/hipaa/for-professionals/covered-entities/sample-business-associate-agreement-provisions/index.html ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/microsoft-copilot-hipaa-compliant Healthcare Innovation # Is Microsoft Copilot HIPAA Compliant? AI Healthcare Compliance Guide Written By: Josh Spencer • March 24, 2025 Microsoft Copilot shows great potential in general business settings, and does offer a HIPAA BAA for some components of their Copilot service. However, Copilot currently falls short in many important security and operational aspects when used with Protected Health Information (PHI) or Personally Identifiable Information (PII). Many providers unknowingly compromise their patient's data by believing Microsoft Copilot is always safe for protected health information (PHI), not realizing that the fine print leaves their data exposed to breach and violations of HIPAA when used in common scenarios. One such example is the Copilot service occasionally passes data to their Bing service, which is not secure for PHI and exempted from the HIPAA BAA. New to the rules? Start with what makes an AI HIPAA compliant : the four requirements any tool must meet before it touches PHI. Large enterprises should be vigilant and ensure these settings are blocked; carefully and continuously reviewing all future service updates to ensure continued compliance. However, this is inefficient and not practical for all but the largest organizations. ## Microsoft Copilot's Healthcare Compliance Challenges When evaluating Microsoft Copilot for healthcare use, several important limitations emerge: Complex Compliance Requirements : While Microsoft Copilot can technically be HIPAA compliant, implementation requires meeting specific conditions including a qualifying Microsoft 365 base license (Business Basic, Standard or Premium, Office 365 E1/E3/E5, or Microsoft 365 E3/E5), configuring additional security settings, and understanding complex data handling requirements High Cost: $360/user/year ($30/user/month on an annual commitment) for Microsoft 365 Copilot, or $252/user/year ($21/month, discounted to $18 under 2026 promotional pricing) for Microsoft 365 Copilot Business in organizations up to 300 users, on top of your existing M365 licenses Limited Healthcare Specialization : Designed primarily for general business use rather than healthcare-specific workflows Restrictive Implementation : Requires Microsoft 365 ecosystem adoption and specialized IT knowledge These Microsoft Copilot healthcare limitations create significant hurdles for healthcare organizations, particularly smaller practices without dedicated IT resources or budget for enterprise-level software investments. ### HIPAA Compliant (with proper configuration and an enterprise BAA): Microsoft 365 Copilot (Only in enterprise M365 environments and only with specific features enabled) Security Copilot (Built specifically for enterprise security teams) Dynamics 365 Copilot (Only select features support compliance) Power Platform Copilot (Only select features support compliance) ### NOT HIPAA Compliant: Consumer Microsoft Copilot on the web (Public-facing AI without BAA coverage) Copilot mobile app (Consumer-grade experience, in most cases, not covered by BAA) Microsoft 365 Premium, which replaced Copilot Pro for new customers in October 2025 (Personal subscription, no BAA, not designed for protected health information) Copilot in Windows (Not covered by BAA, even in managed environments) Be careful which Copilot you're using for your patient information In contrast, BastionGPT was purpose-built as a healthcare-focused HIPAA-compliant AI solution with simplified compliance and implementation. This fundamental difference makes BastionGPT the superior choice for healthcare providers who need both regulatory compliance and specialized clinical capabilities. For more detail on the safeguards behind this approach, see how BastionGPT secures healthcare data . ## BastionGPT is the Best HIPAA-Compliant AI for Medical Practices Healthcare organizations require secure AI for medical notes that protects patient information while enhancing productivity. BastionGPT delivers comprehensive compliance features that address healthcare's regulatory requirements: Streamlined HIPAA-compliant AI infrastructure with BAAs included in all plans Will never introduce capabilities outside of those covered by the HIPAA BAA PIPEDA compliance for Canadian healthcare providers Australian Privacy Principles (APP) compliance Global data centers (US, Canada, Australia) for data sovereignty Data is encrypted and never used to train AI models 30-day secure data retention policies No minimum purchase requirements or enterprise licensing complexity These robust security measures make BastionGPT the ideal AI with a BAA for healthcare applications, especially for practices seeking simplified compliance without enterprise-level IT resources. ## How BastionGPT Compares to Copilot for Clinical Notes When comparing Copilot vs BastionGPT specifically for clinical documentation AI capabilities, several critical differences emerge: ### HIPAA Compliance BastionGPT : Straightforward compliance with BAA included in all plans; designed specifically for healthcare privacy requirements Microsoft Copilot : Possible but requires complex configuration, specialized IT knowledge, and enterprise-level licensing ### Healthcare Specialization BastionGPT : Purpose-built for clinical documentation with healthcare-specific prompts, templates, and capabilities Microsoft Copilot : General business productivity tool without healthcare-specific features or clinical workflows ### Document Processing BastionGPT : Handles a single document up to 30 pages on Professional, or multiple documents up to 700,000+ words (1,000+ pages) on Professional Plus, with specialized medical terminology recognition Microsoft Copilot : Limited to processing within Office apps, potential gaps in medical terminology understanding ### Pricing & Accessibility BastionGPT : Starting at $20/month with no minimum purchase requirement; accessible for individual practitioners and small practices Microsoft Copilot : $360/user/year (in addition to a required base Microsoft 365 license) with required HIPAA-compliant enterprise configuration ### AI Technology BastionGPT : Leverages multiple AI models (GPT, Claude, and Gemini) optimized for different healthcare documentation tasks Microsoft Copilot : Limited model approach that may not address specialized healthcare documentation needs ### Healthcare Features BastionGPT : Extensive healthcare-specific capabilities including SOAP notes, treatment plans, referral letters, and patient education materials Microsoft Copilot : General productivity focus without specialized clinical documentation features These differences highlight why BastionGPT is the best Copilot alternative for healthcare and mental health professionals seeking to implement AI note-taking within a secure, accessible framework. See the full side-by-side, including pricing and BAA coverage: BastionGPT vs Microsoft Copilot . ## Conclusion: BastionGPT is the Superior AI for Healthcare While Microsoft Copilot can technically be configured for HIPAA compliance in certain enterprise environments, its significant implementation barriers, high minimum investment requirements, and general-purpose design make it less optimal for many healthcare providers, especially smaller practices, individual practitioners, and organizations without extensive IT resources. BastionGPT's purpose-built healthcare features, straightforward compliance infrastructure, and flexible implementation options delivers what healthcare providers truly need: a secure, compliant AI assistant that enhances clinical documentation while protecting patient information and supporting regulatory compliance. For the thousands of healthcare organizations which have implemented BastionGPT's AI for healthcare documentation, it has significantly improved documentation quality, provider efficiency, and ultimately, patient care, all within reach for healthcare providers of any size. Start a free 7-day trial of BastionGPT If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/hipaa-compliant-ai-power-automate Healthcare Innovation # HIPAA-Compliant AI in Power Automate for Healthcare Workflow Automation Written By: Josh Spencer • September 23, 2026 Short answer: Power Automate itself is covered by Microsoft's HIPAA Business Associate Agreement, but the AI service you connect to a flow is not. Any third-party connector that touches protected health information needs its own BAA with your organization. BastionGPT is now a Microsoft-certified Power Platform connector built for exactly this: HIPAA-compliant AI actions for Power Automate, Power Apps, Copilot Studio, and Azure Logic Apps, with a BAA from Bastion Intelligence and no code to write. As of September 2026, it is the only Power Automate connector that provides HIPAA-compliant access to GPT, Claude, and Gemini models. For most practices, Power Automate is the healthcare workflow automation software they already own through Microsoft 365. This guide covers where Microsoft's coverage ends, why the common AI connectors fall outside it, and how to add HIPAA-compliant AI to a healthcare workflow in a few minutes. Microsoft-certified connector Listed on Microsoft Learn BastionGPT is published by Bastion Intelligence in Microsoft's connector directory, which documents every action, input, and output. Certified connectors go through Microsoft's certification review before they are listed. Power Automate: Premium Power Apps: Premium Copilot Studio: Premium Azure Logic Apps: Standard View the BastionGPT listing on Microsoft Learn ## Is Power Automate HIPAA compliant? What Microsoft's BAA covers The Power Automate cloud service is on Microsoft's list of services in scope for its HIPAA BAA , alongside Power Apps, Power BI, and Office 365 services such as Exchange Online, SharePoint Online, OneDrive for Business, Microsoft Teams, and Forms. Copilot Studio is covered under the same BAA . The BAA is available by default through Microsoft's Data Protection Addendum for covered entities and business associates, so most healthcare organizations on Microsoft 365 already have it. Coverage is not the same as compliance. Microsoft says so directly: a BAA supports your compliance program, and your organization is responsible for how flows are built and secured. In practice that means data loss prevention policies, least-privilege connections, and Secure Inputs and Secure Outputs on any step that handles PHI, because Power Automate stores each step's inputs and outputs in run history. ## Are AI connectors in Power Automate HIPAA compliant? Usually not, and this is where healthcare flows go wrong. Microsoft's BAA covers Microsoft services. It stops at the boundary of any connector that sends data to another company. When a flow passes an email body or a document to an outside AI service, that vendor becomes a business associate and needs its own BAA with you, covering the exact service the connector calls. That rules out the easy options. Consumer ChatGPT has no BAA. General-purpose AI APIs offer one only to certain customers and only for named services, and the independent publisher connectors that call them are not published or supported by the AI vendor. Microsoft's own Azure OpenAI connector keeps the AI inside Microsoft's Azure agreement, but it means an Azure subscription, an Azure OpenAI resource and model deployment, prompt design, and ongoing upkeep for your IT team, and unless Microsoft approves modified abuse monitoring for your subscription, flagged prompts and completions can be stored for human review . Microsoft's AI Builder is the closest built-in option, but its models are general purpose, it is metered in AI Builder credits, and Microsoft's HIPAA in-scope list names Power Automate, Power Apps, and Power BI rather than AI Builder, so confirm coverage with Microsoft before PHI reaches it. For a fuller checklist, see what makes an AI tool HIPAA compliant . ## A Microsoft-certified, HIPAA-compliant AI connector for Power Automate BastionGPT is a HIPAA-compliant AI assistant built for healthcare by Bastion Intelligence. The connector, certified by Microsoft, published by Bastion Intelligence, and listed in Microsoft's connector directory , gives your flows five actions that call the same HIPAA-compliant AI API our customers use directly: Ask a question. Send a single question, with optional instructions and an optional document ID, and get back a plain-text answer. This is the simplest way to draft a summary, a reply, or a triage note. Create a chat completion. Send a full multi-turn conversation when you need more control, including your own system message. Upload a document. Upload a PDF, image, or text file (up to 10 MB) and get a document ID you can reference in questions for 30 days. Submit audio for transcription. Send a recording and get a transcript ID back right away. Get a transcript. Retrieve the finished transcript with speaker labels and timestamps. BastionGPT selects a model for each request from OpenAI's GPT, Anthropic's Claude, and Google's Gemini families ( see which models power BastionGPT ), and every request runs under your BAA with Bastion Intelligence. Your content is never used to train AI models, and the assistant is designed for clinical language. There is nothing to deploy: create a connection with your API key and start building. Details on encryption and data handling are on our security page . ### Four ways to add AI to a healthcare flow Approach Code required Who covers the AI step under a BAA Built for healthcare Independent publisher connector to a general AI API Some: prompts and response parsing Only if the AI vendor signs a BAA for that exact service No Microsoft AI Builder No, but metered in AI Builder credits Confirm with Microsoft; not named on its HIPAA in-scope list No Microsoft's Azure OpenAI connector Prompt design, plus an Azure subscription, Azure OpenAI resource, and model deployment to maintain Microsoft, under your Azure agreement No BastionGPT certified connector No Bastion Intelligence Yes ## Examples: AI healthcare workflow automation in Power Automate ### Triage patient email with AI The flow pictured in this post has three steps and took a few minutes to build. When a new email arrives in a shared inbox, the flow starts. BastionGPT's Ask a question action receives the email body with the instruction "Summarize this patient email and flag anything urgent." Send an email delivers the summary to the care team. Add a condition on the answer and urgent messages can route to a different queue. ### Transcribe session recordings from OneDrive or SharePoint Drop a recording into a folder and a flow sends it through Submit audio for transcription , polls Get a transcript in a Do until loop, and files the speaker-labeled transcript next to the audio. Feed the transcript into Ask a question to draft a note in the format your practice uses. This gives you HIPAA-compliant AI transcription inside Power Automate without a separate medical transcription vendor. ### Summarize referral packets and records When a referral PDF lands in SharePoint, Upload a document returns a document ID, and Ask a question pulls out the diagnosis, the referring provider, and the requested service into a list item or a Teams message. Documents stay available to the flow for 30 days, so follow-up questions do not require a second upload. ### Give a Copilot Studio agent a compliant AI tool Copilot Studio agents can call the same actions. A staff-facing agent in Teams can answer questions about an uploaded policy document, or summarize a record a user attaches, with the AI step covered by your BastionGPT BAA rather than left to a general-purpose model. ## How to add HIPAA-compliant AI to a Power Automate flow Request a BastionGPT API key at bastiongpt.com/api-signup . Keys are typically issued the same business day. Open a flow in Power Automate , choose Add an action , and search for BastionGPT. Pick an action and create a connection by pasting your API key. The connection is reused by every BastionGPT step in your environment. Map your inputs and use the outputs. Answer, Document ID, Transcript ID, and Text are available as dynamic content in the steps that follow. Turn on Secure Inputs and Secure Outputs on each BastionGPT step that handles PHI, so patient data is masked in run history. ## What you need A BastionGPT API subscription. Connector calls are billed on your API subscription at the same rates as direct API requests. A Microsoft plan that includes premium connectors. BastionGPT is a Premium connector in Power Automate, Power Apps, and Copilot Studio, and a Standard connector in Azure Logic Apps. ## Frequently asked questions ### Does Microsoft's BAA cover third-party connectors in Power Automate? No. Microsoft's BAA covers in-scope Microsoft services. A connector that sends data to a non-Microsoft service is only covered if that vendor has its own BAA with your organization for the service the connector calls. ### Can I use ChatGPT in Power Automate with patient data? Not through consumer ChatGPT, which offers no BAA. OpenAI signs BAAs only for certain API and enterprise customers and only for the services it names, so check the exact coverage before any PHI reaches a flow. The BastionGPT connector avoids the question: every action runs under Bastion Intelligence's BAA. ### Is Copilot Studio HIPAA compliant? Yes. Microsoft's Copilot Studio compliance page states that Copilot Studio is covered under Microsoft's HIPAA BAA, and the same rule applies to the tools you add to an agent: each connector needs coverage of its own. The BastionGPT connector is available to Copilot Studio agents as a Premium action. For Microsoft 365 Copilot itself, see Is Microsoft Copilot HIPAA compliant? ### Is Zapier HIPAA compliant? What is the alternative for healthcare? No. Zapier states that it is not HIPAA compliant, does not sign a BAA, and cannot be used to automate anything involving PHI. Ask any other automation platform the same question before patient data reaches it. For organizations already on Microsoft 365, Power Automate is the alternative that Microsoft's BAA already covers, and the BastionGPT connector adds the AI step under a BAA of its own. ### Does the BastionGPT connector work in Power Apps and Azure Logic Apps? Yes. Microsoft lists it for Power Automate, Power Apps, and Copilot Studio as a Premium connector, and for Azure Logic Apps as a Standard connector. ### Is the BastionGPT connector available in GCC or GCC High? Not yet. The connector is available in Microsoft's commercial clouds worldwide. It is not available in GCC, GCC High, DoD, or China environments. ### Which AI models does the BastionGPT connector use? BastionGPT routes each request to a model from OpenAI's GPT, Anthropic's Claude, or Google's Gemini families, chosen automatically for the task, and all of them run under your BAA with Bastion Intelligence. As of September 2026, no other Power Automate connector provides HIPAA-compliant access to all three. ### Does BastionGPT use my data to train AI models? No. Customer content sent through the API or the connector is never used to train AI models. Full documentation for the connector, including every input and output, is on Microsoft Learn . If you do not have an API key yet, request access and we will get you set up. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/navigating-healthcare-compliance-with-bastiongpt Healthcare Innovation # How to Stay Compliant Using AI Within Global Regulatory Frameworks Written By: Josh Spencer • October 17, 2024 AI is transforming healthcare, but using it responsibly requires strict adherence to privacy regulations. When sensitive patient data is involved, compliance with regulatory requirements is critical. BastionGPT addresses compliance with global standards for AI and data privacy helping you to adhere to US (HIPAA), Canadian (PIPEDA), Australian (APP) and other countries standards. BastionGPT was founded by cybersecurity experts with privacy and security as a central principle, making us stand out as a reliable partner for healthcare providers. ## Understanding Key Healthcare Privacy Regulations #### HIPAA The United States Health Insurance Portability and Accountability Act (HIPAA) is the key healthcare privacy regulation in the United States. It mandates that all protected health information (PHI) is securely stored, processed, and accessed to support patient confidentiality. HIPAA compliance requires business associates to sign a Business Associate Agreement (BAA) to demonstrate their commitment to safeguarding PHI. Signup with BastionGPT includes a BAA to support compliance with legal requirements. #### PIPEDA The Canadian Personal Information Protection and Electronic Documents Act (PIPEDA) governs data privacy in Canada, ensuring that organizations collect, use, and disclose personal information responsibly. PIPEDA emphasizes the importance of obtaining informed consent from individuals and maintaining high standards of security. #### APP The Australian Privacy Principles (APP) form the core of privacy regulation in Australia, focusing on managing personal information responsibly. It includes requirements around data transparency, consent, cross-border data disclosure, and maintaining data integrity. ## The Challenge of Using AI in Global Healthcare AI introduces unique compliance risks that traditional software does not. Many general-purpose AI tools may: Store or reuse data in ways that are not transparent Use inputs to train models Lack clear data residency controls Operate without healthcare-specific safeguards This creates uncertainty for healthcare providers, especially when dealing with PHI or regulated data. As a result, organizations must carefully evaluate whether an AI solution is designed for compliance. ## What Compliance-Ready AI Should Include To safely use AI in healthcare, organizations should look for platforms that prioritize privacy and security by design. Key requirements include: ### 1. Secure for Sensitive Data Healthcare professionals often deal with sensitive information that requires utmost care. With BastionGPT, there is no need to waste time redacting data or limiting its use and it is specifically designed for sensitive data. Your data is stored separately from other customers, and not used to train AI models, allowing healthcare providers to leverage AI capabilities without compromising patient privacy. ### 2. Compliant Data Handling BastionGPT's strict data policies uphold user data security requirements and your chat data is never sold or used for inappropriate purposes such as for marketing services to patients. This commitment to privacy directly aligns with the requirements of HIPAA, PIPEDA, and APP. Customers in the USA, Canada and Australia are routed to country specific data centers to support data sovereignty and alignment with legal requirements. ### 3. Transparent Privacy Practices For APP compliance, organizations must be transparent about their privacy practices. BastionGPT supports this by keeping its privacy policy up-to-date and easily accessible. Customers retain full rights and control over their data, and BastionGPT only uses customer data to provide services—not for marketing or sales purposes. ### 4. Reliability and Data Residency BastionGPT has data centers around the globe to support compliance with data requirements. Data centers are in Australia, Canada, and the USA. This helps healthcare providers meet the cross-border data requirements of APP and PIPEDA, storing data within specific jurisdictions as needed. ### 5. Principles of Safety and Privacy BastionGPT is committed to safety and privacy through its AI principles: Privacy and Security: Personal information is always kept private and secure, as mandated by HIPAA and APP. Transparency and Caution: BastionGPT openly communicates its limitations to promote cautious use in healthcare settings. Human Oversight: The platform is designed to be used with medical professional oversight, clinical expertise is the key to effective and compliant use of AI. ### 6. A Flexible and Valuable Tool for Healthcare BastionGPT offers more than compliance—it offers value. One subscription provides access to multiple AI models, all tuned for healthcare use cases, reducing errors and enhancing patient care. With features like unlimited conversational queries, reduced mistakes, and support for many different types of content, BastionGPT is an adaptable and robust tool for healthcare providers. ## BastionGPT's Commitment to Healthcare Providers Protection from Unsafe Features : All new features are reviewed by experts before implementation, protecting users from potential misuse or errors. No Coding Required : BastionGPT's conversational interface means healthcare professionals can interact without needing technical skills. Live Support : Healthcare professionals using BastionGPT have direct access to experts via email, chat, or video. No Request Limits : Unless you're a bot, BastionGPT allows unlimited requests, helping you get the most out of your AI tools. ### FortaTech Security: A Backbone for Compliance BastionGPT relies on FortaTech Security to support its compliance infrastructure. FortaTech Security provides robust encryption, ISO27001-certified data centers, and confirms that customer data is never used without explicit instruction. Customers maintain full control over data collection, use, and deletion—aiding in compliance with APP, HIPAA, and PIPEDA. FortaTech Security's position on compliance includes: Limited Data Handling : Customer data is only used to provide services, not for advertising. Customer Control : Clients are responsible for their privacy policies and ensuring compliance, with BastionGPT providing built-in security features to support this. Cross-border Data Transfers : Customers know that data residency requirements are met for their information. ## Get Started with Compliance Ready AI As AI becomes more embedded in healthcare workflows, compliance has to be built into the foundation of the tools you use. Organizations that prioritize privacy, transparency, and data control will be best positioned to adopt AI safely and sustainably. Solutions like BastionGPT reflect this shift toward compliance-first AI, offering healthcared teams a way to leverage advanced capabilities without exposing sensitive data or increasing regulatory risk. The right platform should not only improve efficiency, but also strengthen your compliance posture. Explore how BastionGPT supports secure, compliant AI use in healthcare and see what responsible AI adoption can look like in practice. Start your free trial today. If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/pipeda-compliance-canada Legal · FortaTech Security # Canada Data Protection Agreement For Customers of FortaTech Security, LLC d/b/a Bastion Intelligence who provide personal information subject to PIPEDA and to Canadian provincial and territorial privacy and health information legislation. Effective September 1, 2026 Version 2.0, supersedes November 22, 2024 Entity FortaTech Security, LLC d/b/a Bastion Intelligence Incorporated Into Terms of Use Questions legal@bastionintelligence.com On this page 01 Scope of the Canada DPA 02 Compliance with Canadian Privacy Legislation 03 Obligations of FortaTech Security 04 Individual Rights 05 Data Retention and Deletion 06 Audits and Certifications 07 Applicable Laws and Jurisdictions 08 Liability and Indemnification 09 General Provisions A Schedule A: Regulatory Authorities • Documentation for Your Privacy Assessment This Canada Data Protection Agreement ("Canada DPA") is incorporated into the Terms of Use or any other agreement ("Agreement") between you ("Customer") and FortaTech Security, LLC d/b/a Bastion Intelligence ("FortaTech Security," "we," "us," or "our") when you use our services and provide personal information subject to Canadian privacy laws. It sets out the terms under which FortaTech Security processes personal information subject to the federal Personal Information Protection and Electronic Documents Act ("PIPEDA") and applicable provincial and territorial privacy legislation, including Ontario's Personal Health Information Protection Act, 2004 ("PHIPA"). If there is any conflict between a provision of this Canada DPA and a provision of the Agreement, this Canada DPA controls with respect to the processing of Canadian Personal Information. This Canada DPA applies on every plan. It becomes effective when the Customer accepts the Agreement, and no separate signature, attestation, or in-product designation is required for it to take effect. A countersigned copy is available on request, as described in Section 9.4. ## SECTION 01 Scope of the Canada DPA ### 1.1 Applicability This Canada DPA applies to the collection, use, and disclosure of personal information from individuals in Canada ("Canadian Personal Information") provided by the Customer in connection with their use of FortaTech Security's services (the "Services"). ### 1.2 Relationship of the Parties For the purposes of this Canada DPA, the Customer is the organization accountable for Canadian Personal Information, and FortaTech Security acts as a service provider processing that information solely on behalf of and under the direction of the Customer. Where the Customer is a health information custodian under PHIPA or an equivalent provincial statute, FortaTech Security acts as an agent or information manager of the Customer, as those terms are used in the applicable legislation. ## SECTION 02 Compliance with Canadian Privacy Legislation This section describes how FortaTech Security applies the ten fair information principles in Schedule 1 to PIPEDA. ### 2.1 Accountability We take responsibility for the management and protection of personal information under our control. A designated Privacy Officer oversees compliance with PIPEDA, PHIPA, and other applicable Canadian legislation, and can be reached at legal@bastionintelligence.com . ### 2.2 Identifying Purposes We identify the purposes for which personal information is collected at or before the time of collection. Categories collected include: Contact details, such as name, email address, and phone number. Organization information, such as company and title. Payment information. Data generated through use of the Services. Any other information the Customer provides directly to us. ### 2.3 Consent We obtain informed consent for the collection, use, or disclosure of personal information. Consent may be express or implied depending on the context and the sensitivity of the information. Where personal health information is involved, the Customer is responsible for obtaining and documenting patient consent, and for determining the lawful authority on which it relies, before using the Services to process that information. FortaTech Security publishes consent form templates that Customers may adapt to assist in meeting these requirements. ### 2.4 Limiting Collection We collect only the personal information necessary to provide the Services, and collect it by fair and lawful means. Examples include the information required to create accounts, process payments, and deliver AI-generated documentation or analyses. ### 2.5 Limiting Use, Disclosure, and Retention Personal information is used or disclosed only for the purposes for which it was collected, unless the Customer consents or a use or disclosure is required by law. Customer Content is not sold, and is not used to develop, train, fine-tune, or evaluate models, including the models of any third-party provider. Where personal health information is involved, we apply the retention and secure disposal practices described in Section 5. ### 2.6 Accuracy We make reasonable efforts to keep personal information accurate, complete, and current for the purposes for which it is used. Customers can update account information directly within the platform or by contacting us. ### 2.7 Safeguards We protect personal information with security safeguards appropriate to its sensitivity, including encryption in transit and at rest, access controls based on least privilege, tenant segregation, audit logging, and monitoring. Safeguards are described in Section 3.3 and in our Technical and Organizational Security Measures . ### 2.8 Openness Our privacy practices are published and readily available. Our Privacy Policy describes how we collect, use, and protect information, and our sub-processor list identifies every provider we engage and whether it may process Customer Content. ### 2.9 Individual Access Individuals may request access to their personal information. On request we will confirm the existence, use, and disclosure of that information and provide access to it within a reasonable time. Where access is refused, we will give reasons and inform the individual of their recourse. ### 2.10 Challenging Compliance Individuals and Customers may challenge our compliance with these principles by contacting our Privacy Officer using the details in Section 9.4. We investigate all complaints, and where a complaint is justified we take appropriate corrective measures. Schedule A lists the oversight bodies to which a complaint may be escalated. ## SECTION 03 Obligations of FortaTech Security ### 3.1 Compliance with Applicable Laws We will comply with PIPEDA and applicable provincial and territorial privacy legislation when processing Canadian Personal Information on behalf of the Customer. ### 3.2 Processing Purposes We will process Canadian Personal Information only for the purposes of providing the Services in accordance with the Agreement and this Canada DPA, and only on the Customer's documented instructions. ### 3.3 Security Measures We will implement appropriate technical, organizational, and administrative measures to protect Canadian Personal Information from unauthorized access, use, disclosure, alteration, or destruction. These measures include, but are not limited to: Encryption of data in transit (TLS 1.2 or higher) and at rest (AES-256). Access controls based on the principle of least privilege, granted on a need-to-know basis and limited to operational purposes such as security monitoring, abuse monitoring, and technical troubleshooting. Segregation of Customer data by tenant. Audit logging of access and actions, monitoring, and regular security assessments. ### 3.4 Sub-processors We engage sub-processors to assist in providing the Services. Each sub-processor is bound by contractual obligations providing a comparable level of protection to that required under this Canada DPA, and FortaTech Security remains responsible for their performance. We publish the current list of sub-processors , including the purpose for which each is engaged, the categories of data involved, and whether that provider may process Customer Content. ### 3.5 Data Residency and Cross-Border Transfers Canadian Personal Information belonging to Customers registered with a Canadian billing address is stored and processed in Canada during normal operation of the Services. A limited set of operations may involve access to, or transfer of, Canadian Personal Information outside Canada, including to the United States. These are exceptions rather than routine processing, and they arise from specific circumstances rather than ordinary use of the Services. They include certain support requests where diagnosing or resolving an issue requires personnel or a sub-processor located outside Canada to access the relevant data, and optional sessions that the Customer initiates or agrees to, such as a screen share with our prompt engineering team. Where such an operation is necessary, FortaTech Security conducts it in accordance with PIPEDA and applicable provincial legislation, applies the contractual and security protections described in this Canada DPA, and limits the transfer to the minimum information required for the purpose. ### 3.6 Breach Notification In the event of a breach of security safeguards involving Canadian Personal Information, FortaTech Security will notify the Customer without unreasonable delay after discovery, and will provide sufficient information for the Customer to meet its obligations under PIPEDA and the Breach of Security Safeguards Regulations , including any obligation to report to the Office of the Privacy Commissioner of Canada, to notify affected individuals, and to maintain breach records. Where a breach involves personal health information, we will also support the Customer's notification and reporting obligations under the applicable provincial health privacy statute, including notification to the relevant Commissioner where required. ## SECTION 04 Individual Rights ### 4.1 Assistance with Requests If FortaTech Security receives a request from an individual to exercise a right under PIPEDA or applicable provincial legislation, such as access, correction, or deletion, we will promptly notify the Customer and will not respond directly except to confirm receipt and to direct the individual to the Customer. We will assist the Customer in responding to the request to the extent required by law and the Agreement. ### 4.2 Customer Responsibility The Customer is responsible for responding to individuals exercising their rights under PIPEDA or other applicable Canadian privacy legislation, and for determining what information must be provided. ## SECTION 05 Data Retention and Deletion ### 5.1 Retention Periods FortaTech Security retains Canadian Personal Information only for as long as necessary to fulfil the purposes set out in the Agreement and this Canada DPA, or for such longer period as is required by law. Retention periods vary by feature. The current default retention period for each feature, and the range within which a configurable period may be set, are published in our help centre . ### 5.2 Customer Control Over Retention The Customer controls the retention of its content. The Customer may: delete specific items, or all of its content, at any time from within the Services; set the retention period for transcription anywhere within the published range; and request deletion at any time by contacting us at legal@bastionintelligence.com . Where a retention period is not configurable within the Services, the applicable default is published and the Customer may still request deletion at any time. Following deletion, a copy may persist for a short published period in a secure audit vault, which exists to support security monitoring, abuse monitoring, and required auditing, and is then removed. ### 5.3 Deletion or Return of Data On termination of the Agreement, or on the Customer's request, FortaTech Security will delete or return Canadian Personal Information unless retention is required by applicable law. Personal health information is deleted using industry-standard secure deletion methods in accordance with applicable retention and destruction requirements. ## SECTION 06 Audits and Certifications FortaTech Security maintains records of its data protection practices and will provide the Customer with reasonable documentation or certifications to demonstrate compliance with this Canada DPA. The Customer may request additional information or audits in accordance with the Agreement. ## SECTION 07 Applicable Laws and Jurisdictions This Canada DPA is designed to support compliance with the federal, provincial, and territorial legislation governing the collection, use, and protection of personal information and personal health information across Canada. FortaTech Security's practices are designed with reference to the following legislation. ### Federal Personal Information Protection and Electronic Documents Act , S.C. 2000, c. 5 Breach of Security Safeguards Regulations , SOR/2018-64 Privacy Act , R.S.C. 1985, c. P-21 ### Alberta Protection of Privacy Act (POPA) Access to Information Act (ATIA) Personal Information Protection Act (PIPA) Health Information Act (HIA) ### British Columbia Personal Information Protection Act (PIPA) Freedom of Information and Protection of Privacy Act (FOIPPA) E-Health (Personal Health Information Access and Protection of Privacy) Act and E-Health Regulation ### Manitoba The Freedom of Information and Protection of Privacy Act (FIPPA) The Personal Health Information Act (PHIA) ### New Brunswick Right to Information and Protection of Privacy Act (RTIPPA) Personal Health Information Privacy and Access Act (PHIPAA) ### Newfoundland and Labrador Access to Information and Protection of Privacy Act, 2015 (ATIPPA) Personal Health Information Act (PHIA) ### Northwest Territories Access to Information and Protection of Privacy Act (ATIPP) Health Information Act (HIA) ### Nova Scotia Freedom of Information and Protection of Privacy Act (FOIPOP) Personal Health Information Act (PHIA) Personal Information International Disclosure Protection Act (PIIDPA) ### Nunavut Access to Information and Protection of Privacy Act (ATIPP) ### Ontario Personal Health Information Protection Act, 2004 (PHIPA) and O. Reg. 329/04 Freedom of Information and Protection of Privacy Act (FIPPA) Municipal Freedom of Information and Protection of Privacy Act (MFIPPA) Child, Youth and Family Services Act, 2017 (CYFSA), Part X ### Prince Edward Island Health Information Act (HIA) Freedom of Information and Protection of Privacy Act (FOIPP Act) ### Quebec Act respecting the protection of personal information in the private sector (P-39.1), as amended by Law 25 Act respecting access to documents held by public bodies and the protection of personal information (A-2.1), as amended by Law 25 Act respecting health and social services information (R-22.1) ### Saskatchewan The Freedom of Information and Protection of Privacy Act (FOIP) The Local Authority Freedom of Information and Protection of Privacy Act (LA FOIP) The Health Information Protection Act (HIPA) ### Yukon Access to Information and Protection of Privacy Act (ATIPP) Health Information Privacy and Management Act (HIPMA) Nothing in this section is a representation that the Services are suitable for every regulated use in every jurisdiction. Customers remain responsible for determining whether their intended use of the Services complies with the legislation and professional obligations that apply to them. ## SECTION 08 Liability and Indemnification FortaTech Security's liability arising from or related to this Canada DPA is subject to the limitations and exclusions of liability set out in the Agreement. ## SECTION 09 General Provisions ### 9.1 Governing Law This Canada DPA is governed by the laws of the State of Texas, United States, being the jurisdiction of FortaTech Security's principal place of business. Nothing in this clause limits any right or remedy an individual or a Customer has under PIPEDA or applicable provincial legislation, or the jurisdiction of a Canadian privacy regulator. ### 9.2 Conflicts In the event of any conflict between this Canada DPA and the Agreement, the terms of this Canada DPA prevail with respect to the processing of Canadian Personal Information. ### 9.3 Amendments This Canada DPA may be amended from time to time to reflect changes in applicable law or in the Services. The effective date and version of the current text are shown at the top of this page. ### 9.4 Contact For questions about this Canada DPA, to request a countersigned copy, or to make a request regarding personal information, contact our Privacy Officer: FortaTech Security Legal 11816 Inwood Rd # 3181 Dallas, TX 75244 United States Email: legal@bastionintelligence.com Website: https://bastionintelligence.com ### 9.5 Acknowledgment By using the Services and providing Canadian Personal Information, the Customer acknowledges that it has read, understood, and agrees to the terms of this Canada DPA. ## SCHEDULE A Regulatory Authorities We are committed to resolving complaints promptly. If you believe our response to a privacy concern is inadequate, you may escalate it to the oversight body for your jurisdiction. Each office publishes its own current mailing address, telephone number, and complaint process on the site linked below. Jurisdiction Oversight body Website Federal Office of the Privacy Commissioner of Canada priv.gc.ca Alberta Office of the Information and Privacy Commissioner of Alberta oipc.ab.ca British Columbia Office of the Information and Privacy Commissioner for British Columbia oipc.bc.ca Manitoba Manitoba Ombudsman ombudsman.mb.ca New Brunswick Ombud NB ombudnb.ca Newfoundland and Labrador Office of the Information and Privacy Commissioner oipc.nl.ca Northwest Territories Office of the Information and Privacy Commissioner of the Northwest Territories oipc-nt.ca Nova Scotia Office of the Information and Privacy Commissioner for Nova Scotia oipc.novascotia.ca Nunavut Office of the Information and Privacy Commissioner of Nunavut atipp-nu.ca Ontario Information and Privacy Commissioner of Ontario ipc.on.ca Prince Edward Island Office of the Information and Privacy Commissioner assembly.pe.ca Quebec Commission d'accès à l'information du Québec cai.gouv.qc.ca Saskatchewan Office of the Saskatchewan Information and Privacy Commissioner oipc.sk.ca Yukon Yukon Ombudsman and Information and Privacy Commissioner yukonaccountability.ca Before you escalate Most concerns can be resolved directly. Contact our Privacy Officer at legal@bastionintelligence.com and we will investigate and respond in writing. ## FOR REVIEWERS Documentation for Your Privacy Assessment The agreement ends at Section 9. What follows is a reading guide for privacy reviewers. Each document linked below governs its own subject matter. Document What it covers Where Sub-processor list Every provider we engage, the purpose, the data categories involved, and whether that provider may process Customer Content Terms, section 3.4 Technical and Organizational Security Measures Encryption, access control, tenant segregation, logging, monitoring, and incident response Terms, section 15 Privacy Policy What we collect, how it is used and disclosed, storage locations by region, and retention bastionintelligence.com/privacy Data retention and deletion Default retention period for each feature, what is configurable, and how to delete content support.bastiongpt.com Security overview Architecture, data handling, and answers to the questions that recur in security reviews bastiongpt.com/security Clinical validation Accuracy methodology and published results bastiongpt.com/clinical-validation AI principles How models are selected, evaluated, and governed bastiongpt.com/company/ai-principles HIPAA Business Associate Agreement Relevant where a Canadian organization also handles information subject to United States requirements bastionintelligence.com/baa ### Available on request A countersigned copy of this Canada DPA, as described in Section 9.4. Documentation or certifications evidencing our data protection practices, as described in Section 6. Something not covered here If your assessment asks a question these documents do not answer, write to legal@bastionintelligence.com and tell us which framework you are working through. It helps to know whether you are completing an Alberta PIA, a Quebec section 17 assessment, or the Ontario IPC checklist, because the three ask for different things. --- > Source: https://bastiongpt.com/post/psychiatrist-ai-use-cases Healthcare Innovation # 40+ Ways Psychiatrists & Therapists Use AI - BastionGPT Written By: Robert Henehan • May 9, 2025 Mental health professionals are increasingly leveraging secure AI tools like BastionGPT to enhance the way they deliver care. From faster documentation to more personalized treatment planning, AI is helping therapists work more efficiently without sacrificing quality or compliance. AI-powered assistants can transcribe sessions, generate clinical therapy notes , summarize extensive patient histories, and create tailored, patient-friendly educational materials. All of this is done within a secure, HIPAA-compliant environment designed to protect sensitive information. In this guide, we’ll explore the most impactful ways psychiatrists and psychologists are using AI in mental health care and how these tools are shaping the future. For a broader overview, see these use cases for generative AI in therapy and psychiatry . ## How to Use Artificial Intelligence in Your Mental Health Practice ### Streamlining Mental Health Documentation and Reporting Heavy documentation burdens are a well-known challenge in mental healthcare. This workload not only eats into time that could be spent with patients, but has been shown to contribute to provider burnout. BastionGPT’s AI medical scribe and transcription capabilities directly tackle this problem by automating drafts of clinical documentation while helping to ensure notes are thorough and accurate. Therapists and psychiatrists use BastionGPT to turn audio recordings or loose written notes into polished, well articulated reports and notes, saving hours on writing. #### Session Notes and SOAP Documentation: BastionGPT can generate well-structured therapy session notes (e.g. in SOAP format) after a session. For example, a psychologist can finish a counseling session and ask BastionGPT to produce a progress note summarizing the client’s mood, issues discussed, interventions used, and plan for next time. The AI will output a professional, coherent note that the clinician can quickly review and edit. This helps ensure that no key detail is missed (such as safety concerns or homework given) and that documentation is completed on time which helps with compliance and continuity of care. By offloading much of the time intensive note-writing to the AI, providers eliminate tedious after-hours charting and reduce their documentation backlog. #### Comprehensive Evaluation Reports: Mental health professionals also use BastionGPT to draft evaluation and assessment reports. For instance, a neuropsychologist could input raw data from cognitive tests, interview notes, and observations, and have BastionGPT compose a draft neuropsychological evaluation report. The AI will organize the information into appropriate sections (background history, test results, interpretations, recommendations) written in clear, empathetic language. This drastically speeds up reporting for complex evaluations (such as autism assessments or cognitive exams) that normally take many hours to write. Clinicians can focus on interpreting results and proofreading the final report while BastionGPT handles the narrative writing, ensuring the final report is thorough and easy to understand. View a step-by-step guide to creating a multi-page report. #### Summarizing Large Records and EHRs: BastionGPT’s ability to process very large text inputs (up to approximately 1,000 pages) lets clinicians generate concise overviews of bulky electronic health records or multi-year treatment notes when time is limited. These summaries are meant to complement and never replace a professional’s direct review of the source documentation. They are most helpful when a full read-through would delay care or is impractical before an urgent appointment. By highlighting diagnoses, medication history, pivotal events, and outstanding questions, the AI summary lets the clinician triage what needs a closer look and quickly locate original passages for verification. The provider still confirms any detail that will guide assessment or treatment; the summary shortens the path to those critical sections. Used in this way, BastionGPT accelerates safe, evidence-based decision-making without compromising the depth of clinical judgment or patient care. By streamlining clinical documentation in these ways, BastionGPT frees mental health providers from paperwork drudgery and helps ensure consistent, high-quality records. All documentation generated by the AI is kept secure and private, and clinicians remain in control and they can review, edit, or refine any AI-drafted note before finalizing it. The result of empowering psychiatrists with BastionGPT is faster, more accurate reporting that meets compliance standards and allows providers to reclaim time for patient care. ## Developing Individualized Treatment Plans Crafting an effective treatment plan for a patient often requires synthesizing a lot of information such as diagnoses, patient preferences, evidence-based guidelines, and creative therapeutic approaches. Mental health professionals are using AI to jump-start the treatment planning process and ensure plans are comprehensive. BastionGPT can serve as a knowledgeable assistant, drafting customized treatment plans that clinicians can then tailor and approve. #### Personalized Care Plans: Psychiatrists and therapists use BastionGPT to generate initial treatment plan drafts for complex conditions. For example, imagine a psychiatrist treating a patient with bipolar disorder: they can prompt BastionGPT with the patient’s diagnosis, symptoms, and some background, and ask for a detailed treatment plan. The AI can propose a structured plan that covers multiple facets, e.g. medication recommendations for mood stabilization, psychotherapy modalities (such as CBT or family therapy), lifestyle modifications (sleep schedule, exercise), and safety considerations. This gives the clinician a strong starting template that they can adjust based on the patient’s unique needs. Nothing is finalized without the clinician’s judgment, but the AI’s draft ensures no major component is overlooked and can introduce ideas the clinician might refine (like suggesting a particular therapy technique or support group). #### Dynamic Updates to Plans: Because BastionGPT can quickly review prior session notes and progress, it also helps with adjusting treatment plans over time . As a patient progresses, the clinician can summarize recent changes (for instance, “patient’s anxiety scores have improved over 6 weeks, but sleep issues remain”) and ask the AI to suggest modifications to the care plan. BastionGPT might recommend focusing more on sleep hygiene strategies or adjusting medication in the plan draft. This assists providers in keeping plans current and responsive to patient progress. The clinician remains in control of decisions, but the AI ensures all relevant data is considered when updating goals or interventions. #### Interdisciplinary Care Coordination: In team-based mental health settings, BastionGPT can help integrate input from multiple professionals into one unified plan. If a patient has input from a therapist, psychiatrist, and social worker, the AI can combine notes from all three to produce a single consolidated care plan that covers therapy, medication, and social support aspects. This is especially useful in settings like treatment planning meetings or discharge planning from a psychiatric hospital, where numerous pieces need to come together into one document. By automating the rote aspects of writing treatment plans, BastionGPT allows clinicians to focus on the clinical reasoning, evaluating which recommendations make sense and customizing them for the individual. The result is high-quality, individualized treatment plans created more efficiently. Every plan generated stays within a secure, HIPAA-compliant environment, so sensitive patient details in the plan remain protected. ## Patient Communication, Follow-Up, and Engagement Effective care doesn’t stop when a patient walks out of the therapy room – ongoing communication and follow-up are key parts of mental health treatment. AI assistants like BastionGPT are helping clinicians stay connected with patients between visits by making it easy to send summaries, resources, and reminders in a personalized yet efficient way. #### Personalized Session Follow-Up Emails: Therapists often reach out to clients after sessions with notes or “homework.” BastionGPT makes this process faster by drafting follow-up emails or messages tailored to each patient. For instance, after a counseling session, a therapist can ask BastionGPT to compose an email to the client summarizing the key points discussed (e.g. “Today we explored your anxiety triggers and practiced a breathing technique...”), reinforcing any coping strategies taught, and listing any homework assignments or reading for the week. The AI will generate a professional, empathetic message that the therapist can quickly tweak if needed and send. This ensures patients leave with a clear understanding of their progress and next steps, and it maintains momentum between sessions. What used to take therapists considerable time writing individual emails is now done in seconds, improving patient adherence and understanding without overburdening staff. #### Appointment Reminders and Outreach: BastionGPT can also help draft caring outreach messages. For example, if a patient hasn’t been seen in a while or misses an appointment, the AI can generate a gentle check-in message: “Hello, I noticed we haven’t seen you recently, I hope you’re continuing to connect with your family and set healthy boundries. We’re here if you need to talk or want to reschedule...” The clinician or staff can review and send such a message, maintaining a supportive connection. This personalized touch can encourage patients to stay on track in their care. BastionGPT helps mental health professionals provide consistent, compassionate communication without adding hours of work. By utilizing AI for these follow-up tasks, clinicians ensure patients feel supported between visits, which can improve treatment adherence and outcomes. ## Summarizing Group Therapy Sessions Group therapy is a powerful modality in mental health care, but keeping track of everything in a group session – and documenting it afterward – can be challenging. Therapists are turning to AI assistants like BastionGPT to summarize group therapy sessions for both internal records and to share with participants. By quickly capturing the key themes and insights from a session, BastionGPT helps facilitators reinforce progress and keep everyone on the same page. #### Detailed Group Session Recaps: After a group therapy meeting, a facilitator can provide BastionGPT with either a transcript or their notes from the session and ask for a summary to distribute. The AI will produce a digest that highlights key discussion topics, shared experiences, and any group decisions or homework. For example, in a weekly anxiety management group, the summary might note: “This week, members discussed challenges with exposure exercises, two members shared personal victories in facing fears, and the group agreed to practice a new breathing technique daily. Next session’s focus will be on handling setbacks.” Such a summary can be emailed to participants (with appropriate consent) so they remember what was learned and feel a sense of continuity. It’s also useful for any group members who missed the session – they can catch up easily. BastionGPT’s ability to identify important points in a free-flowing conversation ensures no critical insight is lost when writing up the notes. #### Record-Keeping and Accountability: From a documentation standpoint, having a clear summary of each group session is valuable for the clinic. It provides a record of what interventions were delivered in the group and how participants responded, which can be important for both clinical follow-up and billing. Instead of writing these summaries from scratch, clinicians use BastionGPT to generate group therapy notes much faster. The AI can structure the notes with headings like Attendees, Topics Discussed, Therapeutic Activities, Observations, and Plan for Next Session , ensuring that the documentation is thorough and standardized across sessions. This helps in reviewing progress over time for the group as a whole and demonstrates the therapeutic work being done, which is useful in supervision or audits. #### Multi-Participant Voice Recognition: BastionGPT’s transcription tool supports multi-speaker recognition, distinguishing up to 10 different voices. In a group therapy context, this means if you record the session, the AI can attribute statements to different speakers in the transcript. This feature allows the facilitator to see who said what in the raw transcript, making it easier to attribute insights or follow up with specific individuals as needed when generating the summary. It’s a level of detail that manual note-taking could hardly capture in a live group. By leveraging AI to summarize and document group sessions, therapists ensure that group therapy participants get the maximum benefit from each meeting. The timely summaries keep clients engaged and informed, and the detailed records help clinicians maintain effective and compliant documentation. It’s another example of how AI extends a professional’s capabilities, handling the clerical work of group therapy so the clinician can focus on facilitating and caring for the group. ## Documenting Crises and Flagging Risk In mental health care, crisis situations demand immediate attention and thorough documentation. Whether it’s a suicide risk assessment, a panic attack intervention, or an emergency counseling call, clinicians must both act quickly and record the event in detail for follow-up and legal reasons. BastionGPT assists in efficiently documenting crisis interventions to help capture critical details, and it can even help clinicians identify warning signs in text data that might indicate a patient is at risk. This proactive use of AI can support early intervention and enhance patient safety. #### Crisis Intervention Documentation: After handling a crisis, writing the incident report is often the last thing a clinician wants to do. BastionGPT can ease this burden by drafting a structured crisis report. For example, imagine a crisis counselor on a helpline who de-escalates a caller experiencing severe suicidal ideation. Once the immediate crisis is managed, the counselor can input notes or a transcript of the call into BastionGPT and ask for a comprehensive intervention summary. The AI-generated report will typically include: the presenting problem (e.g. “caller was having active suicidal thoughts with a plan”), the interventions used (“counselor utilized grounding techniques and involved the caller’s family”), the outcome (“caller agreed to a safety plan and emergency services were contacted”), and recommendations for follow-up. By having BastionGPT produce this first draft, the counselor can ensure all vital information is captured promptly while they focus on coordinating next steps. #### Safety Plan Creation: In urgent situations, BastionGPT can also assist in formulating or documenting a safety plan. If a therapist is working with a client in crisis, they might ask the AI to outline a safety plan template (listing warning signs, coping strategies, emergency contacts, etc.) which the therapist can then finalize with the patient. Having this template ready saves time during a tense session and ensures a consistent approach to safety planning. The plan can then be shared with the patient and included in their chart. #### Behavioral Risk Alert Generation: One innovative use of AI in mental healthcare is scanning text (like therapy session transcripts, patient journal entries, or messages) for red flags that suggest elevated risk. BastionGPT can analyze such text data to help identify language that might indicate suicidality, violence risk, or severe deterioration. For instance, if a patient is using a journaling app and the entries are fed through BastionGPT, the AI might flag sentences that reflect hopelessness, explicit mention of suicidal thoughts, or drastic changes in tone. Research shows that NLP techniques can detect signs of suicidal ideation that might otherwise be missed – one study found that AI analysis could spot risk in patients that 58% of the time would have gone undetected without NLP. BastionGPT could alert the clinician with a notice like, “The latest entry contains statements that indicate possible suicidal thinking; please review immediately.” This early warning system does not replace clinician judgment, but it provides an additional safety net by catching subtle cues in large volumes of text that a busy provider might overlook. It essentially acts as a second set of eyes, continuously scanning for crisis signals in a privacy-protected way. By documenting crises thoroughly and flagging risks proactively, mental health professionals can respond more effectively and prevent patients from “falling through the cracks.” BastionGPT’s assistance in this arena means that crisis notes are completed faster (and with high detail) and potential emergencies are identified sooner. All of this is done within a secure platform, so even the most sensitive information about a patient’s crisis stays confidential. In emergencies, every minute saved matters and BastionGPT helps clinicians use their time where it’s needed most while still keeping impeccable records. ## Research and Literature Review Support Mental health professionals must stay informed about the latest research findings and often engage in their own data analysis – whether it’s evaluating therapeutic outcomes or conducting academic studies. Generative AI has become a valuable tool for these tasks, helping clinicians and researchers sort through large amounts of information and extract meaningful insights quickly. BastionGPT, with its ability to process up to 1,000 pages at a time, enables mental health experts to review literature, analyze data, and draft research summaries with greater efficiency. #### Literature Reviews and Evidence Synthesis: Keeping up with the ever-growing body of mental health research can be daunting. Therapists might want to know the newest evidence for treating a certain condition, or a clinic may want to summarize findings on an intervention they’re considering. BastionGPT can dramatically speed up conducting a literature review. For example, a psychologist could provide the AI with several journal articles or studies (by copying text or uploading PDFs) on the impact of mindfulness meditation for depression. The AI can then summarize key findings from each source and highlight common conclusions, perhaps noting that most studies show moderate improvement in depressive symptoms and emphasizing any differences in outcomes. It can also compile a cohesive report that compares and contrasts the studies, saving the clinician hours of reading. This is especially useful for busy practitioners who want evidence-based guidance without spending all night citing articles. #### Qualitative Data Summarization: For those involved in research or quality improvement projects, BastionGPT can help in analyzing qualitative results and drafting reports. Suppose a clinical director is running an internal study on patient outcomes from a new therapy program – they have survey responses, symptom descriptions, and written feedback from participants. By feeding these into BastionGPT, the AI can identify themes and generate a narrative report. It might highlight, for example, that “Participants commonly reported feeling more empowered in managing anxiety after the program, with frequent mentions of improved coping skills and peer support as key factors.” For qualitative data (like patient comments or therapy session transcripts), it can perform a thematic analysis by summarizing common themes and patterns. #### Academic Writing Assistance: For mental health professionals writing articles, grant applications, or conference abstracts, BastionGPT serves as a handy writing aid. You can ask it to outline a literature background on a topic, suggest relevant references (it can’t access external databases on its own, but it can help organize ones you provide), or even proofread and improve the clarity of a draft. Importantly, because BastionGPT is HIPAA-compliant and used in a private environment, any sensitive or unpublished data you include in your prompts remains secure. This is crucial if you’re working with preliminary research results or patient data; you can use the AI to help write about it without compromising data security or privacy. By integrating AI into research and analysis tasks, mental health practitioners and scholars can stay evidence-based without getting bogged down in information overload. BastionGPT effectively acts as a research assistant that never tires of reading and summarizing. It helps ensure that clinical decisions and program development can be informed by the latest knowledge, as the AI quickly brings the most relevant insights to the forefront. Whether it’s for formal research projects or just staying current on best practices, using BastionGPT in this way leads to more informed care. ## Creating Patient Education Materials and Therapeutic Narratives Educating patients and their families is a big part of mental health care. Therapists often provide handouts, write explanatory letters, or use therapeutic stories and metaphors to help clients understand and engage in their mental health treatment. BastionGPT is proving to be extremely helpful in generating patient-friendly educational materials and even creative therapeutic content, saving clinicians time while enhancing the educational experience. Since BastionGPT is compliance-focused, it can produce these materials using real patient information (if needed) without privacy risk, for example, creating a tailored explanation for a specific patient’s situation. #### Accessible Psychoeducational Content: Mental health concepts can be complex or overwhelming for patients. BastionGPT can help professionals create brochures, handouts, or online articles that break down these concepts into clear, layman-friendly language. For instance, a therapist might want a handout on coping with panic attacks for their clients. By prompting BastionGPT (e.g. “Explain panic attacks and how to cope with them in simple terms, and include some do’s and don’ts”), the AI can generate a draft that covers what a panic attack is, why it happens, steps to manage one (breathing, grounding techniques), and misconceptions to avoid. The clinician can then tweak and brand this handout for their practice. In another scenario, a psychiatrist could ask for a plain-language explanation of a new medication to give to a patient, BastionGPT will describe what the medication does, how to take it, common side effects, and safety tips in an easy-to-understand way. This ensures patients are well-informed about their conditions and treatments, which can improve adherence and reduce anxiety. And because it’s generated within a HIPAA-compliant tool, if the educational content needs to reference aspects of the patient’s own case, that can be done safely (for example, “In your case, since you have X, this medication will help by…”). #### Therapeutic Stories and Analogies: Many therapists use storytelling and metaphors as techniques – for example, creating a story about a hero overcoming challenges that parallels the patient’s situation, or using analogies to explain cognitive-behavioral concepts. BastionGPT can serve as a creative partner in developing these therapeutic narratives. A child therapist, for instance, could say, “Create a short story about a brave fox who learns to cope with loud noises, to help a child with sensory sensitivity” . The AI might spin a tale about “Freddie the Fox” who at first is scared of thunder but learns strategies to stay calm, implicitly teaching coping skills using details from their records. The therapist can adjust details to personalize it further for the child. Similarly, for an adult client who struggles with negative thoughts, a therapist might ask BastionGPT for an analogy that explains cognitive distortions, the AI could come up with, say, a comparison to “a funhouse mirror that warps how you see yourself,” which the therapist can use in session. These creative outputs help keep therapy engaging and relatable. BastionGPT basically provides a well of ideas and drafts that therapists can draw from to enrich their sessions and homework assignments. #### Customized Educational Plans: For clients who want to learn more about their condition, clinicians can use BastionGPT to generate a “learning plan” or series of educational emails. For example, a psychologist treating someone with PTSD might use BastionGPT to produce a series of weekly educational snippets: Week 1 about understanding trauma responses, Week 2 about grounding techniques, Week 3 about managing nightmares, etc. The clinician reviews these and sends them to the patient as part of therapy. This automates patient education in a structured way, making sure the patient gets reliable information gradually, which can supplement in-session therapy. In all these cases, BastionGPT amplifies a clinician’s ability to educate and creatively engage patients. The materials generated include evidence-based information and can be reviewed for accuracy (BastionGPT is designed to prioritize credible medical information), so clinicians can trust it as a starting point. By offloading the first draft of educational content to the AI, mental health professionals save time and can deliver high-quality, tailored educational resources that empower their patients. And because everything is done in a secure environment, even patient-specific educational content (like a summary of “John’s Depression Treatment Plan and How It Works”) can be created without violating privacy. It’s a win-win for enhancing patient understanding and engagement. ## Enhancing Teletherapy with Real-Time Transcription Teletherapy has become a staple in mental health services, allowing clinicians to reach patients remotely through video or phone sessions. However, conducting therapy over Zoom or phone presents its own challenges for note-taking and keeping a record. This is where BastionGPT’s real-time transcription and summary capabilities shine: it can listen to teletherapy sessions and transcribe them, acting as a virtual scribe, and then help turn those transcripts into structured notes or summaries. Additionally, BastionGPT supports uploading existing recordings or importing transcripts from other telehealth or transcription tools. For instance, if a provider recorded a session using Zoom or a specialized telehealth platform, they can simply upload that audio file or transcript to BastionGPT and have the system process and summarize it. This flexibility makes it easy to work with both live sessions and pre-recorded material, offering secure integration into various telehealth workflows. #### Live Transcription of Sessions: Using BastionGPT, a therapist can get a transcript of their teletherapy session. For example, during a video session, the therapist can run BastionGPT’s AI scribe in the background; it will capture the conversation of both the therapist and the patient, distinguishing speakers. This is particularly helpful if the therapist wants to review what was said afterward or ensure they didn’t miss any important detail (like a specific phrase the patient used or a list of symptoms described). The key advantage of BastionGPT doing this is that the transcription is happening in a HIPAA-compliant environment, so the content isn’t being sent to an insecure service or violating confidentiality. Everything stays in the secure BastionGPT platform. #### Automated Telehealth Session Summaries: After the session, BastionGPT can take that transcript and generate a summary or clinical note just as with in-person sessions. In telepsychiatry, for example, a psychiatrist could have the AI produce a consultation note that includes the patient’s chief complaints as discussed on the call, mental status exam observations (if any noted verbally), the diagnosis or impressions given, and the treatment plan (e.g. medication changes or follow-up recommendations). The AI can format this into a proper psychiatric progress note or an intake note for a first tele-visit. Providers describe it as having an “AI scribe” in the virtual room, and with just a click a first draft of the note is ready for review, drastically cutting down documentation time. The provider then quickly edits the draft for any nuances the AI might not capture (like nonverbal cues or important contextual info) and signs off. This approach maintains high-quality documentation for telehealth visits without the provider having to type furiously during or after the appointment. #### Consistency Across Modalities: Teletherapy often is blended with in-person care (some sessions remote, some in office). BastionGPT helps ensure documentation consistency across these modalities. No matter how the session is conducted, the notes can have a similar level of detail and structure, because the AI can be used in both scenarios. For example, a clinic might standardize that every session note – in person or telehealth – should include a summary of patient report, interventions, and plan. BastionGPT can be prompted to follow that template every time. This consistency is great for quality of care and for anyone reviewing the charts later (like supervisors or auditors). It shows that even with teletherapy, the clinic’s documentation standards are maintained or even enhanced. Using BastionGPT in teletherapy means therapists don’t have to choose between being present with the client and taking good notes, they can have both. The technology quietly supports in the background, capturing everything and organizing it afterward. Patients benefit because their clinician is more present and attentive (not distracted by note-taking), and clinicians benefit by not having a pile of notes to write at the end of a long day of telehealth appointments. Plus, the entire process – from transcription to note generation – is secure and compliant, so providers can confidently use these tools in telemedicine without privacy worries. Telehealth combined with BastionGPT becomes a powerful way to deliver care efficiently and safely. ## Tracking Progress and Outcomes Over Time Monitoring a patient’s progress over the course of therapy or treatment is critical in mental health care. Clinicians need to discern trends in symptoms, note what interventions have helped, and report outcomes to stakeholders (patients, families, or even insurers). AI assists in this arena by aggregating and summarizing progress data over time, which can be especially helpful for busy providers or clinics dealing with many clients. #### Individual Patient Progress Reports: Therapists often write progress summaries, say every month or at the end of a treatment block, to evaluate how far a patient has come. BastionGPT can generate these progress reports by pulling from all the session notes to date. For example, after 10 sessions with a client, a therapist can paste them and ask the AI to “Summarize this client’s progress in therapy so far” . Using the notes from each session, BastionGPT might produce a report like: “Over the past 3 months, the client has shown significant improvement in reported depression scores (from severe to moderate range), increased use of coping skills (as evidenced by journaling and reaching out to friends), and progress toward therapy goals (e.g. obtained a new job, which was a goal set in week 2). Remaining challenges include occasional panic symptoms, which have reduced in frequency but still occur under high stress. The client’s adherence to homework has been good, completing assignments 80% of the time. Overall, the trajectory is positive, with continued focus on panic management recommended.” A report like this, generated in seconds, provides a coherent narrative of the treatment journey, which the therapist can then review, refine and condense as needed. It can be shared with the patient (to boost morale by showing improvements), used in treatment planning (to decide next steps), or included in medical records. It’s also handy for communicating with other professionals (like a psychiatrist or primary care physician) about the patient’s status. #### Outcome Trend Analysis: Beyond narrative summaries, BastionGPT can help identify quantitative trends in a patient’s data. If a clinician tracks certain metrics (e.g. weekly anxiety rating scales, or frequency of panic attacks per month recorded in notes), the AI can be prompted to analyze those from the documentation. It might say, “The patient’s self-rated anxiety went from 8/10 at intake to an average of 4/10 in the last three sessions, indicating a downward trend.” Or “Notably, the patient’s mood ratings dipped each holiday season according to journal entries, suggesting a seasonal pattern to address.” These insights can alert clinicians to patterns that might not be immediately obvious when looking session by session. By having an AI comb through months of data points and text, therapists better ensure important patterns in progress (improvement or worsening) are flagged and can be addressed in the treatment plan. #### Quality Improvement and Outcome Reporting: At a program or clinic level, aggregated progress tracking is valuable for quality improvement (QI) and demonstrating treatment effectiveness. BastionGPT can assist administrators or clinical directors by summarizing outcomes across many cases. For instance, a clinic might want to report that “Out of 50 patients who completed our 12-week program, 80% showed a clinically significant reduction in anxiety symptoms.” Staff could feed the data or outcome summaries of those 50 cases into BastionGPT and ask for an overall report. The AI could tabulate common outcomes, identify how many improved vs. stayed the same, and even pull common feedback themes from discharge surveys (like if many patients mentioned appreciating the group therapy component). This saves a ton of manual work in drafting annual reports or grant reports. BastionGPT basically acts as a data analyst and report writer combined, turning raw outcome data into human-readable conclusions. The clinic can then review and present these findings, which can guide QI decisions (e.g. knowing which areas of treatment are most effective or where drop-out rates are an issue). Tracking progress is about closing the loop in care – ensuring that the treatment provided is making a difference. By using BastionGPT, mental health professionals get help in measuring and communicating that difference. It leads to more informed adjustments to care, clear evidence of improvement (or signals if something isn’t working), and robust reporting for accountability. And since BastionGPT works within a compliant system, even these summaries and analyses of PHI are done securely. Clinicians maintain control and confidentiality, while benefiting from the AI’s knack for spotting trends and summarizing trajectories that span across time. ## Clinical Supervision and Training Support Mental health professionals not only provide care, but also continuously train and supervise each other. From experienced clinicians supervising interns, to educators training new therapists, to professionals engaging in continuing education – learning and development is an ongoing part of the field. BastionGPT is being used as a versatile tool in these areas as well, documenting supervision discussions, generating training scenarios, and aiding in creating educational content for clinicians. Its ability to absorb large amounts of text and produce organized outputs can lighten the load in mentorship and training contexts, all while keeping sensitive case material private. #### Supervision Session Notes: In clinical supervision, a supervisor meets with a trainee or a group of trainees to discuss cases, review progress, and provide guidance. Instead of scribbling notes, supervisors can use BastionGPT to record and summarize these supervision sessions. For example, after a weekly supervision meeting, the supervisor can list the cases discussed and key feedback given, then have the AI generate a supervision summary. The output might include: Trainee Jane Doe’s Cases: (1) Case A: discussed boundary setting with client; supervisor advised trying XYZ approach. (2) Case B: reviewed treatment plan for teen client, noted need to involve family more... etc., and similarly for Trainee John Doe’s cases. It can also note general topics covered (ethics, self-care, etc.). This summary can be shared with the trainees as a reference (so they don’t have to rely on memory alone) and kept as a record of supervision (which some licensure boards require). It ensures accountability – showing that supervision covered certain competencies – and helps trainees remember and implement the feedback. Using BastionGPT in this way means supervisors can focus on the conversation in the moment, then quickly produce detailed notes afterward with AI assistance, rather than trying to write everything down during the meeting. #### Training Scenario Creation: Developing realistic case scenarios for training purposes is time-consuming. BastionGPT can generate hypothetical patient cases, role-play dialogues, or ethical dilemmas on demand, which educators can use in teaching. For instance, a professor teaching a counseling skills class might ask BastionGPT, “Give me a detailed mock case of a 30-year-old client with social anxiety and a short dialogue that demonstrates cognitive-behavioral therapy techniques.” The AI could produce a scenario complete with background information (job, family, symptom history) and a sample therapist-client dialogue. The instructor can then have students role-play the scenario or analyze it. Need a quick case for a trainee to practice writing a treatment plan? The AI can spit one out. Need an example of a challenging situation (like a client showing dependency or a boundary issue) to discuss in a workshop? BastionGPT can craft it. These AI-generated scenarios are rich and varied, and trainers can always refine details to better fit their teaching points. The big advantage is saving time – educators don’t have to write every mock case from scratch, yet they have an essentially unlimited supply of cases to pick from. #### Educational Content for Clinicians: Beyond patient education, BastionGPT can help create materials for clinician education and professional development. Think of things like quick reference guides, summaries of new therapy techniques, or even quizzes for workshops. For example, a clinical director might want to disseminate a one-page summary of “best practices for teletherapy” to the team – BastionGPT can draft this by summarizing policy documents or articles on the topic. Or a supervisor running a training on a new EHR system could ask the AI to generate a cheat-sheet of how to write notes that meet audit standards, incorporating the clinic’s guidelines. Since BastionGPT can handle long documents, it could also summarize key insights from books or journal articles for a journal club handout. Importantly, if any of these training materials need to reference case material or PHI from real examples, BastionGPT keeps that data secure. For instance, one could feed in a case study from the clinic and have the AI help write discussion questions about it, without fear that this sensitive info is going to an unauthorized cloud service. #### Mock Conversations and Skill Practice: Another creative use in training is to use BastionGPT as a “mock patient” for practicing conversational skills. A trainee could prompt BastionGPT to respond as a patient would in a scenario (since it’s a conversational model). For example: “BastionGPT, act as a teen client who is resistant to therapy and answer my questions with short, skeptical answers.” The trainee can then practice how they’d engage, and the AI will respond accordingly. While not a perfect substitute for a real role-play, it’s a unique way for trainees to practice in a low-stakes environment and get comfortable with various client personalities or situations. All of these interactions stay within the secure platform, so even if a trainee mentions a real case in their prompt, it remains confidential. In essence, BastionGPT not only helps with patient care but also with shaping the next generation of mental health providers and keeping current ones sharp. It reduces the administrative overhead in supervision and training – documentation, scenario prep, material creation – which allows educators and supervisors to focus on teaching and mentorship. The use of AI in this context means more standardized, high-quality training materials and a more engaging learning experience (with the help of creative case simulations). And it all conforms to privacy standards, as BastionGPT can handle real training cases without exposing them, giving training programs confidence that learning can be enhanced by AI without breaching ethics or confidentiality. ## Streamlining Intake, Triage, and Administrative Tasks Managing the administrative side of a mental health practice is another area where AI is making a major impact. Tasks like intake processing, insurance authorization, and compliance audits are necessary but time-intensive and energy-draining. BastionGPT assists clinics by simplifying many of these workflows – from summarizing intake forms to drafting insurance letters – all while keeping sensitive data protected. This can significantly reduce the clerical load on clinicians and administrative staff, speeding up processes, minimizing errors and improving staff retention. #### Automated Intake Form Summaries: When a new client comes to a mental health clinic, they often fill out lengthy intake questionnaires covering their history, symptoms, and demographics. Instead of a clinician manually reading and extracting the key points, BastionGPT can do it in seconds. Staff can input the client’s responses and ask the AI for an intake summary highlighting critical information. For example: “29-year-old female with history of childhood trauma, presenting with symptoms of PTSD and depression; reports nightmares, hypervigilance, and relationship difficulties; no prior therapy; denies current suicidal ideation but has had passive thoughts in past.” The summary could also flag any high-risk answers (like if the client indicated a history of suicide attempts or current substance use). This gives the therapist a quick briefing before the first session with only a fraction of the traditional time, as staff can start with a largely complete draft that they just need to review for errors before providing to the therapist. It’s especially useful in high-volume settings or when intakes are done online before a live appointment. BastionGPT essentially performs a first-pass triage by condensing the form into the essentials, which can also help in assigning the case to the right clinician (e.g. if the summary indicates a substance issue, maybe assign to someone specialized in addictions). All this is done with PHI handled safely inside BastionGPT, so there’s no worry about exposing the client’s private data. #### Triage and Risk Prioritization: In clinics or community mental health centers where there may be waitlists, BastionGPT can assist in prioritizing clients based on need. By analyzing intake data or initial consultation notes, the AI could help categorize cases by complexity or urgency. For instance, it might help generate a triage report that notes: “Client A: moderate depression, low risk, can wait 2 weeks; Client B: active suicidal ideation, high risk, needs immediate contact; Client C: seeking ADHD assessment, moderate priority.” This sort of AI-augmented triage helps support clients who need help most urgently are flagged, and it provides a standardized way to sort through a large number of referrals. This can be lifesaving in scenarios of limited resources. And because BastionGPT has no issues handling sensitive info, it can incorporate all the PHI (like specific answers from an intake) into its analysis without running afoul of privacy laws. #### Prior Authorization Support: Mental health providers frequently have to obtain prior authorizations from insurance for certain treatments or medications – a bureaucratic hassle that involves writing letters or filling forms to justify the medical necessity. BastionGPT can take on much of the writing for these authorizations. A psychiatrist, for example, could prompt the AI with the necessary details oreven the entire patient chart: patient diagnosis, treatment history, what is being requested (say a specific atypical antipsychotic medication), and why it’s needed over alternatives. BastionGPT can then use its knowledge of prior authorization requirements and conventions to draft a formal prior authorization letter that includes all relevant details: “Patient X has tried and failed two other medications, has a history of side effects Y, and this medication is indicated due to... etc., referencing any guidelines or literature if provided. The clinician just polishes and signs it. Because BastionGPT can reference clinical data from the patient’s record, it helps ensure the letter is comprehensive and evidence-based, increasing the chance of approval. And it does it in a fraction of the time it would take to write from scratch. #### Referral Letter Drafting: BastionGPT also helps clinicians draft referral letters when coordinating care with other providers or services. For example, if a therapist needs to refer a client to a psychiatrist for medication evaluation, they can prompt BastionGPT to generate a professional referral letter summarizing the client's presenting problems, treatment history, and reasons for the referral. This allows for smooth transitions of care and consistent communication between providers, without the therapist having to write each letter manually. #### Insurance Appeal Letters: Similarly, if an insurance claim or treatment request is denied, BastionGPT can help write an appeal letter using its knowledge of the insurers conventions and requirements. Using the denial reason and the patient’s information, it can draft a clear argument addressing the insurer’s criteria. For example, “We are appealing the denial of additional therapy sessions for Patient Y. The patient has shown progress but still meets medical necessity for continued weekly therapy, as evidenced by ... (symptoms, risk factors). According to [guideline], terminating therapy at this stage could result in relapse... We request authorization for X more sessions.” Having these letters generated quickly means patients experience fewer interruptions in care due to administrative holdups. It also reduces frustration for clinicians who often find these paperwork tasks draining. #### EHR Data Cleanup and Summaries: Administrative staff or informatics specialists can use BastionGPT to make sense of data in the EHR. For instance, if a clinic wants to check if their EHR entries meet certain standards, they could have the AI review a batch of notes to see if required fields are present in each (like a diagnosis, a signature, etc.) – essentially a documentation quality audit. Or, if transitioning to a new system, BastionGPT might help summarize key points from old records to input into a new system’s format. The AI’s ability to transform and standardize text can ease such migrations or audits significantly. #### Clinical Audit Preparation: Speaking of audits, mental health services are subject to audits by external bodies (for compliance, accreditation, or funding). BastionGPT can act as a pre-audit checker. For example, one could instruct it: “Review these 100 therapy notes and flag any that do not mention a treatment plan or informed consent discussion.” The AI could then output a list of notes that might be missing those elements, allowing the clinic to correct them before an official audit. It can also help gather required data – e.g., “List all instances in these notes where seclusion or restraint was used” for an audit on those interventions. Essentially, BastionGPT can parse through large volumes of documentation to pull out what auditors might ask for, saving compliance teams countless hours. #### Administrative Correspondence and Reports: On a simpler level, BastionGPT helps with day-to-day admin communications too. Need to draft a memo to staff about a new policy? Or write a reply to a patient’s email asking about billing? The AI can produce a quick draft that is polite, clear, and thorough. It’s like having an unlimited amount of admin staff members who can write well on virtually any topic – then the real staff just fine-tune the tone. Over time, this speeds up all those little tasks that otherwise fragment someone’s day. By applying AI to administrative workflows, mental health organizations improve their efficiency and reduce the chance of human error in tedious tasks. BastionGPT allows staff to focus on people rather than paperwork, knowing that forms, letters, and summaries can be generated with just a few clicks. And crucially, all these tasks – which often involve PHI and sensitive details – are handled within a secure, HIPAA-compliant framework. This means clinics can embrace automation without compromising on privacy or accuracy. The end result is a smoother-running operation that ultimately benefits patients (through faster service and fewer administrative roadblocks to care) and relieves clinician stress (by cutting down bureaucratic overload). ## Speed Up Your Documentation Workflow with BastionGPT From solo practitioners to large mental health organizations, BastionGPT is helping transform daily workflows in a safe, compliant, and effective manner. As we’ve explored, mental health professionals are using this HIPAA-compliant AI assistant to streamline documentation, treatment planning, patient communication, group therapy notes, crisis documentation, research, patient education, telehealth notes, progress tracking, supervision, and administrative tasks. In each of these use cases, BastionGPT acts as a tireless helper – producing drafts, summaries, and insights that would otherwise take hours of human effort – while keeping sensitive patient information secure by design. Its compliance-focused architecture (BAA in place, encrypted storage, isolated environment) means providers don’t have to worry about compliance issues when leveraging AI for healthcare . Providers can also review the APA guidance on generative AI to keep their use aligned with professional standards. And its ease of deployment and flexible access (most clinics can get set up in ~10 minutes, with no upfront costs or long-term commitments) make it an accessible solution for teams of any size. BastionGPT augments, not replaces, the clinician, allowing mental health professionals to work more efficiently and focus on the human aspects of care. Notes get done faster, insights come easier, and patients stay engaged – all within a framework that prioritizes security and ethical use. The time and energy saved can be redirected to what matters most: caring for patients and improving mental health outcomes. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus . If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/psychologists-guide-to-using-ai-for-clinical-documentation Healthcare Innovation # A Psychologist's Guide to Using AI for Clinical Documentation Written By: Dr. Anna DiNoto • October 27, 2025 # Can Psychologists Use AI for Clinical Documentation? Yes, and it's transforming how mental health professionals create clinical reports. A few years ago, I started hearing about AI tools in the testing psychologist community on Facebook. As a licensed psychologist working across the lifespan with both clinical evaluations and forensic cases, I was intrigued but cautious. I went down a rabbit hole with the research. I needed to understand how this technology would affect me as a clinician and what the ethics looked like. One requirement was non-negotiable: lack of HIPAA compliance for AI tools was a dealbreaker. All of the research kept pointing to BastionGPT as being HIPAA compliant and at the top of the tier. That's what led me to Bastion. I became an early adopter over two years ago, and I've been able to see the platform continue to grow, which has been wonderful. ## The Challenge: Writing Clear, Compliant Clinical Reports I'd already adopted a philosophy about clinical documentation: if you can't say what you need to say in a sentence, do you really know what you need to say? That supervisor's wisdom had stuck with me from graduate school. Forensic reports present unique challenges because they end up in judges' hands and live online permanently. The question became: how could AI help me communicate clearly and succinctly while capturing everything required by courts, agencies like DSHS, and professional standards? ## How to Use AI for Psychology Reports: Getting Started ### Learning AI Prompts for Clinical Documentation Working with AI for psychological assessments has been like building a relationship. You should be very concrete and specific because AI is hyper-literal. It's almost like speaking to a kid where you're teaching them to do something. If you don't tell it you want bullet points, it won't give you bullet points, even if you expected them. I built my prompting skills through experimentation. I'd open up a new chat and try different words, different adjectives, different ways of describing what I needed. When I found something that worked, I'd save it. Now I have a library of prompts that consistently deliver results. Here's something personal: I'm nice to BastionGPT. I say please and thank you all the time. I feel like it's my friend. It's almost like a personal assistant; it helps me a lot. So I'll say, "Oh, thank you buddy" when it gives me exactly what I need. Maybe that seems silly, and I know it's probably not necessary, but it keeps me from becoming dictatorial in my interactions. Key insight for AI prompting: Examples are incredibly helpful. Showing the AI the formatting you prefer, giving it samples of your style; that makes a huge difference in the output. ## AI Workflows That Transform Clinical Documentation ### Using AI for Psychological Assessment Tables I'm big on aesthetics. My reports end up in judges' hands and online forever, so I want them to look professional and presentable. Tables became my biggest breakthrough with AI for psychology reports . One day I was working with an IQ test table, trying to organize subtests, scores, and percentiles in a meaningful way. I'd sunk an hour into this table, just messing with formatting. I finally turned to BastionGPT and said, "OK, here's what I was thinking the X and Y axis should be. What would be most helpful? How can I present this data in a meaningful way?" The result was exactly what I needed. That's been repeated so many times now. Then I took it further: "In 7th grade language explain what this means so a parent will understand it." The output was brilliant. Patients consistently appreciate the visuals and accessible formatting that breaks up the monotony of just a solid block of twelve-point font, single-spaced text. For more on interpreting assessment scores with AI , BastionGPT's support guide covers how the platform handles psychological evaluation reports. ### Multi-Document Synthesis for Clinical Evaluations One of my most powerful workflows is for patient evaluations using HIPAA-compliant AI. I upload three documents to BastionGPT: The DSM-5-TR criteria My clinical notes from the session and diagnostic criteria Test scores (like the MIGDOS evaluation) My prompt is straightforward: "Tell me how this patient meets the DSM-5-TR criteria for a specific diagnosis. Tell me all the criteria that it would make in a table, so it's easy to read." I've also pre-loaded reference materials, like the uniform American Academy of Clinical Neuropsychology test scores standards I use for scoring. BastionGPT synthesizes all of that prepared data exactly as needed. I can do that work by myself, but if it can synthesize everything after I've done the bulk of the clinical work, I just need it to do that. ### How to Prevent AI Hallucinations in Clinical Reports Here's the key to accuracy: I tell BastionGPT, "Write the page number you found this information on and do not make anything up." I put it on “exact” response mode. I'm almost like a parent with my instructions, very specific and clear. Then it listens. But I always cross-reference and fact-check; is it 99th percentile? Is it 87%? Does this align with what I'm clinically seeing? Generative AI tools like BastionGPT occasionally hallucinates. I've seen it claim a ninety-nine percentile score that was definitely not correct, so proofreading and verification is essential. ## AI Quality Control for Mental Health Professionals ### Maintaining Clinical Standards with AI Bastion doesn't have my license, so I need to stand behind everything that goes out under my name. But the verification process isn't burdensome; it's just good clinical practice. BastionGPT is also a good quality control for me and my team. Grammar checking has been huge. I upload my two- or three-page reports and BastionGPT catches pronoun errors, spots when I've accidentally referenced John's name in Suzy's report because I reused a template. These catches prevent errors before they go out. But I don't just accept AI output as-is. I'll tweak it because I want my reports to read the same way I talk. I'll take what BastionGPT says and refine it because I don't quite talk that way. I'll shave some pieces off. The voice needs to be authentically mine. BastionGPT also helps with analogies. Sometimes I'll ask, "What's a good analogy to explain XYZ to a patient?" It gives me ideas; it doesn't do the job for me, but it helps me capture things in ways my clients need. ## Time Savings: AI Efficiency in Psychology Practice Time savings appear everywhere when using AI for clinical documentation. Instead of typing out three diagnoses with F84.0 and all the Z codes and abbreviations, in alphabetical order, BastionGPT handles that formatting. Clinical reports are distilled down to one-page formats where appropriate. These aren't marginal improvements; they're meaningful time savings that accumulate across every report, every week. The impact? I'm able to take on a lot more patients because BastionGPT has helped me polish my already-streamlined reports. The efficiency gains have been substantial enough that I can serve more people without sacrificing quality. ## Training Your Team to Use AI for Psychology Reports I've given prompts to my staff members. I tell them, "Have Bastion read it and tell you what can I improve? What makes sense? What's inconsistent? What am I missing? What did I misspell?" Always with the caveat: "Don't blindly listen to it. Just see if it gives you a different perspective." Their reports have gotten so much better as a result. I've noticed the improvement consistently. With patients, I'm transparent. It's in our informed consent documentation. They know we use AI to make information more presentable and user-friendly. The feedback has been positive; patients really like the tables and formatting that make complex clinical information accessible. ## AI Ethics in Clinical Psychology: Finding Balance I see the clinical community polarized around AI for mental health professionals. There's one camp that absolutely refuses to engage with it, viewing it with deep suspicion and drawing hard lines against adoption. Then there are people on the opposite extreme who embrace it without any critical oversight, and honestly, many have got their blinders on. I'm concerned when people accept everything without question. I'm firmly in the middle. It's not the be-all and end-all. We've still got to use our brains. We're here to have AI really add to our work and bolster what we're already doing. Balance is the key - be critical and curious when using AI. Blindly copying and pasting is plagiarism. That's not appropriate professional practice. If you just put in a prompt and copy and paste without review, that's a serious problem. But it's also not realistic to completely dismiss this technology. It's here. Whether we like it or not, it's around us, and we've got to adopt it in some way. BastionGPT is another tool in my toolkit to build a better clinician. That's how I see it. The more we can educate ourselves on ways to integrate it that are appropriate, informed, and meaningful, the more helpful it can be. ## Best Practices: AI for Psychologists For colleagues considering AI adoption for clinical documentation, here's what's made the difference for me: ### 1. Connect with Other Practitioners Talk to people who are already using AI tools for psychology reports. Connect with colleagues who are using these tools. Learn from their experience. ### 2. Prioritize HIPAA Compliance Keep a level head. Find that middle ground between rejection and blind acceptance. Always verify HIPAA compliance before uploading any patient information. ### 3. Be Specific with AI Prompts Be hyper-literal and specific. You're going to get frustrated if you haven't told it to do something in bullet points but expected it to. Be as specific as possible. ### 4. Use Examples to Train AI BastionGPT loves examples. Show it how you want things formatted, and it'll match your style. ### 5. Build a Prompt Library Save your prompts. Build your library. When you find instructions that work, preserve them. ### 6. Experiment Safely Have fun with it. Be open to playing with it and experimenting to see what it can and can't do. Open new chats to test different approaches. ### 7. Always Verify AI Output Fact-check important data points. The AI doesn't hold your credentials. But this isn't burdensome; it's the same verification you'd do anyway. ### 8. Never Plagiarize Never blindly copy AI output. Bastion doesn't have my license; I need to make sure I stand behind everything. ## The Future of AI in Clinical Psychology I love BastionGPT. It works for what I need. I've figured out prompts and ways to interact with it, ways to have that back and forth to get exactly what I'm looking for. The technology keeps improving. Over my two-plus years of use, I've watched the iterations and enhancements. It's been genuinely exciting to be part of that evolution. This isn't about replacing clinical judgment. It's about having a tool that saves time on routine tasks, improves presentation quality, catches errors, and frees up mental space for the clinical work that only licensed professionals can provide. For those still on the fence: BastionGPT has made AI integration straightforward. The learning curve is manageable. The benefits are real. And the middle ground, thoughtful adoption with appropriate oversight, is where innovation happens. After all, Bastion doesn't have my license. But it's become an invaluable assistant in helping me fulfill the responsibilities that license demands. ## Frequently Asked Questions (FAQ) ### Is AI safe for psychological reports? Yes, when you use BastionGPT's HIPAA-compliant AI tools and maintain proper oversight. The key is verification: never blindly copy AI output. Always fact-check clinical data, ensure the voice matches your professional style, and stand behind everything that goes out under your name. ### What is HIPAA-compliant AI for mental health? HIPAA compliant AI for mental health refers to artificial intelligence tools that meet Health Insurance Portability and Accountability Act requirements for protecting patient information. This means the platform has proper security, business associate agreements (BAAs), access controls, and privacy measures to protect patient data. BastionGPT is specifically designed to meet these HIPAA compliance standards and best practices for healthcare professionals. ### How do psychologists verify AI-generated content? I use several verification methods: Exact mode instructions : I tell the AI to "write the page number you found this information on and do not make anything up" Cross-referencing : Check percentiles, scores, and clinical data against source documents Clinical alignment : Ensure AI interpretations match what I'm seeing clinically Grammar and consistency checks : Review for pronoun errors, name mix-ups from templates Voice matching : Refine AI output to match my professional voice and communication style ### Can AI replace clinical judgment in psychology? No. AI is a tool to augment clinical work, not replace professional judgment. Bastion doesn't have my license, so I need to stand behind everything in my reports. AI helps with formatting, organization, grammar checking, and synthesizing prepared data, but the clinical interpretation, diagnosis, and treatment planning remain the psychologist's responsibility. ### What are the best AI prompts for psychology reports? Effective AI prompts for clinical documentation should be: Hyper-literal and specific : Tell the AI exactly what format you want (bullet points, tables, etc.) Include examples : Show the AI your preferred style and formatting Set clear boundaries : Instruct it not to make anything up and to cite page numbers or provide a direct quote Use 7th-grade language : When creating patient-facing content, specify reading level Request specific structures : "Create a table showing how this patient meets DSM-5-TR criteria" ### How much time does AI save in clinical practice? The time savings accumulate across every report. Instead of manually formatting diagnoses with ICD-10 codes, creating tables from scratch, or spending hours on document formatting, AI handles these routine tasks. I'm able to take on significantly more patients because BastionGPT helps me polish my already-streamlined reports. The efficiency gains have been substantial enough that I can serve more people without sacrificing quality. ### Do I need to tell patients I use AI for reports? Yes, transparency is important. I include AI use in our informed consent documentation. Patients know we use BastionGPT to make information more presentable and user-friendly. The feedback has been consistently positive; patients appreciate the tables and formatting that makes complex clinical information accessible. ### What's the learning curve for psychologists using AI? The learning curve is manageable. It involves experimentation: trying different prompts, seeing what works, and building a library of successful instructions. Think of it like teaching someone new; you need to be specific about what you want. Most psychologists can start getting useful results within a few sessions of experimentation. The key is being open to playing with it and testing different approaches. ### How do I choose HIPAA-compliant AI tools for my practice? When evaluating AI tools for clinical psychology: Verify HIPAA compliance (this is non-negotiable) Check for Business Associate Agreement (BAA) availability Review security measures (encryption, access controls) Test the platform with non-sensitive information first Connect with other practitioners who use the tool Evaluate customer support and documentation Consider specialty features for psychology and mental health Research consistently pointed to BastionGPT as meeting these standards, which is why I chose it as an early adopter. ## About the Author Dr. Anna DiNoto is a licensed psychologist specializing in neurodiversity and forensic as well as clinical evaluations. She practices in the Pacific Northwest and provides expert testimony in legal proceedings. Dr. DiNoto has been using HIPAA-compliant AI tools in her clinical practice for over two years. Related Resources: BastionGPT Use Cases for Mental Health Professionals Healthcare AI Prompts: Building Your Clinical Library Writing Therapy DAP Notes with AI Begin your journey with a 7-day Free trial of BastionGPT . If you have questions or want to connect: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat with BastionGPT experts: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/school-psychologists-use-ai Healthcare Innovation # How to Use AI as a School Psychologist Written By: Robert Henehan • May 15, 2025 School psychologists are harnessing the power of AI to dramatically cut down on report writing and the overwhelming documentation burden that often leads to late nights and burnout. Imagine reducing the time spent drafting lengthy evaluations, IEP notes, and behavior plans by over 90%, freeing you to focus more on direct student support. This isn't a far-off future; it's happening now with BastionGPT. As a HIPAA and FERPA-compliant AI assistant and transcription tool, BastionGPT has been trained to craft the highest quality psychology reports, writing them in a way which matches your writing style and preferred formatting. It’s a game-changer for schools and report writing teams, offering a secure and compliant way to improve your documentation quality while reclaiming valuable time without compromising student privacy. Learn about how you can integrate AI as a school psychologist. For a broader foundation, start with our psychologist's guide to using AI for clinical documentation . ## How to Use AI to Streamline School Psychologist Workloads ### 1. Streamlining Comprehensive Evaluations and Eligibility Reports One of the most time-consuming tasks for our school psychologists is writing comprehensive psychoeducational evaluation reports for special education eligibility. These include initial evaluations such as: Full and Individual Evaluations (FIIE) Triennial re-evaluations Independent Educational Evaluations (IEEs) Early childhood assessments In-depth evaluations for specific areas like Learning disabilities (SLD) Autism Intellectual disability (ID), Emotional disturbance (ED) Giftedness Dyslexia Traumatic brain injury (TBI) Transition planning #### How BastionGPT Streamlines the Process BastionGPT dramatically accelerates the report writing process for all of these. “Using Bastion, I can draft a high-quality FIIE in under two hours from start to finish. It used to take me a full day,” one veteran school psychologist shared. By offloading the initial drafting to AI, practitioners can save several hours per report and still ensure a thorough result. In fact, many users report that BastionGPT can cut the time required to write most sections of a psychoeducational evaluation by more than half. BastionGPT is especially effective at transforming raw assessment data into polished report narratives. Psychologists can input: Cognitive test scores Rating scale results Interview notes Classroom observations The AI then generates a structured draft organized into the standard sections of an evaluation report, including: Background history Assessment results Interpretation of findings Recommendations This significantly speeds up reporting for complex cases. For example, an autism evaluation that might normally take many hours to write can be drafted in a fraction of the time. The practitioner always remains in control. School psychologists focus on interpreting results and refining conclusions, while BastionGPT assists with the narrative drafting. Because the system is trained on school psychology terminology and assessment frameworks, it can recognize common instruments such as WISC, Woodcock-Johnson scores, BASC, and Vineland ratings and summarize their outcomes accurately. “I was shocked that Bastion knows all of the different rating scales and cognitive tests, and could use just my notes to write up those sections of my reports in seconds,” notes one school psychologist using BastionGPT. It also helps ensure consistency and clarity: it can catch minor grammar issues and suggest alternative wording to make the report more reader-friendly for parents and teachers. The end result is a high-quality evaluation report produced in a fraction of the usual time, without sacrificing accuracy or compliance. Learn more about writing psych evaluations with BastionGPT in our support center. Watch this video to learn how to generate complex, well-formatted multi-page reports in BastionGPT, then edit and reuse the workflow with saved prompts. ### 2. Simplifying Behavior Assessments and Intervention Plans Behavior-focused evaluations like Functional Behavioral Assessments (FBAs), Behavior Intervention Plans (BIPs), and Positive Behavior Support Plans (PBSPs) are another area where AI for school psychologists can save immense time. Typically, these reports require synthesizing observation data, teacher interviews, incident logs, and behavior rating scales to identify the functions of a student’s behavior and recommend interventions. #### How BastionGPT Streamlines the Process BastionGPT streamlines this process by analyzing the raw information and generating organized, insightful summaries. For instance, a school psychologist can feed in notes from a classroom observation, plus teacher and parent comments and rating scale results, and ask BastionGPT to draft the FBA summary. The AI will quickly: Outline the antecedents, behaviors, and consequences observed Highlight patterns (e.g. "off-task behavior increases during unstructured time") Propose likely functions of the behavior Suggest evidence-based intervention strategies or accommodations for the BIP draft Ex.: reinforcement techniques, environmental modifications, or skill-building supports By automating the heavy lifting of writing these behavior reports, BastionGPT frees psychologists to spend more time actually implementing interventions and consulting with staff. What might have been a multi-day writing project can be completed in a much shorter timeframe. BastionGPT turned my observation notes and rating scale data into a solid FBA summary and draft behavior plan within seconds. It’s like having a hidden clone to do all the grunt work that I dread the most,” says one psychologist. This AI-driven efficiency doesn’t come at the cost of quality, either. Because BastionGPT can summarize assessment results, analyze behavior data, and even write recommendations and goals, the resulting FBA/BIP documents are comprehensive and tailored to the student. The school psychologist still reviews and edits the draft as needed, but far less time is spent on the basic write-up. The final product is a thorough, defensible behavior assessment and plan, completed with significantly less stress and turnaround time. ### 3. Supporting Mental Health & Crisis Protocols School psychologists are often on the front lines of student mental health crises, conducting suicide risk assessments, threat assessments, self-injury (SIB) reviews, and substance abuse screenings. These sensitive situations demand immediate action and detailed documentation. BastionGPT helps clinicians document crisis protocols swiftly and thoroughly, so helping to make sure nothing falls through the cracks. #### Suicide Risk Assessments After a critical incident, writing the incident report or risk assessment summary is often daunting when time is of the essence. Using BastionGPT, a psychologist can input their rough notes or even a transcription of an interview with the student and quickly get a structured report that captures all the critical details. For example, an AI-generated draft of a suicide risk assessment might include: Presenting concern – "Student expressed suicidal thoughts during counseling" Assessment steps taken – "Conducted Columbia Suicide Severity Rating Scale; consulted with parent and counselor" Findings – Level of risk, specific warning signs noted, protective factors Actions/recommendations – "Student agreed to a safety plan; will receive daily check-ins; emergency contact protocol established" By having BastionGPT produce this first draft, the psychologist can ensure all vital information is recorded promptly while they focus on coordinating care and next steps. #### Threat Assessments In threat assessment scenarios (e.g. a student makes a concerning violent statement), BastionGPT can similarly assist in documenting the incident, evaluation of the threat, and the intervention plan, including notifications and follow-up measures. The AI’s lack of content filtering is crucial here – unlike a generic AI that might refuse to discuss violence, BastionGPT will objectively summarize the situation and language used, ensuring the documentation is complete. #### Safety Plans & Behavior Contracts Some school-based teams also use BastionGPT to help draft safety plans or behavior contracts for at-risk students. By asking BastionGPT’s powerful AI for a safety plan template, they can get a quick outline (listing warning signs, coping strategies, contact numbers, etc.) that can be customized to the student’s needs, saving valuable time in a crisis. Importantly, all of this is done within a secure, encrypted platform, so even the most sensitive details about a student's crisis remain confidential and compliant with FERPA requirements. School psychologists report that crisis notes can now be completed faster and with high detail, allowing them to respond more effectively in emergencies. In a field where every minute saved in a crisis can mean a safer outcome, BastionGPT helps ensure the highest quality records without slowing down the urgent response. ### 4. Automating Meeting Notes and Plan Documentation Beyond evaluations and assessments, school psychologists are responsible for a variety of meeting and plan documents that eat up time. Individualized Education Program (IEP or ARD) meeting minutes, Section 504 accommodation plans, Prior Written Notices (PWN) to parents, and even consent forms for evaluations are all essential paperwork in the special education process. BastionGPT acts as a tireless administrative assistant for these tasks. #### IEP & ARD Meeting Minutes Instead of writing meeting summaries from scratch, a psychologist can input a rough outline or transcript of an IEP meeting discussion and have the AI generate well-structured minutes that clearly document the attendees, topics discussed, decisions made, and action items. This can be done using the school’s existing meeting recording tool, or using BastionGPT’s built-in AI scribe service which is included in all plans. The draft minutes can be reviewed for accuracy and edited as needed, but much of the clerical work is done automatically. This not only saves time but also helps ensure no required element (like a note about parent questions or dissenting opinions) is missed in the documentation. #### Section 504 Plans For Section 504 plans and similar documents, BastionGPT can help draft the plan by organizing information about the student’s needs and recommended accommodations into the standard format. Simply by providing notes on the student’s disability and successful strategies, a school psychologist can prompt the AI to produce a polished accommodations plan that the team can then tweak and finalize. Similarly, BastionGPT can generate a professional and polished Prior Written Notice letter by outlining the evaluation results or placement changes and the rationale behind decisions in parent friendly language. #### Consent Forms & Referral Letters Even consent forms or referral letters can be drafted with AI assistance. For example, the tool can create a template letter seeking parental consent for an evaluation, ensuring that all legally required elements (purpose of evaluation, areas to be assessed, rights to refuse, etc.) are included. Speeding up the creation of these routine documents means psychologists and special education administrators spend far less time on paperwork boilerplate. Because BastionGPT works with the actual student information securely, there’s no need to strip out names or details as one would if using a public AI tool. The result is faster turnaround on meeting documentation and plan paperwork, which keeps the special education process moving smoothly and compliantly. ### 5. Streamlining Progress Monitoring and Data Analysis Psychologists must write periodic IEP goal progress reports, summarize Multi-Tiered System of Supports (MTSS) data for intervention meetings, and justify services such as Extended School Year (ESY) in terms of data. BastionGPT helps take the drudgery out of progress monitoring paperwork by analyzing raw data and generating clear summaries. #### IEP Goal Progress Reports For instance, you can provide the AI with a student’s IEP goals along with recent performance data (e.g. “Goal: Improve reading fluency to 100 WPM; Data: Jan – 80 WPM, Feb – 85 WPM, Mar – 95 WPM”) and ask for a quarterly progress update. BastionGPT will produce a concise narrative like: “Reading Fluency: Progressing – [Student] increased from 80 to 95 words per minute this quarter, approaching the goal of 100 WPM. Continues to make steady gains with intervention support.” This saves the psychologist from writing essentially the same updates in different words for each student each quarter. The AI’s summaries are also parent-friendly by default, avoiding overly technical jargon. #### MTSS Data Summaries When preparing for an MTSS meeting or an eligibility re-evaluation, BastionGPT can quickly review multiple data points (classroom assessments, intervention logs, teacher comments) and highlight key trends. For example, it might summarize: “MTSS Reading Intervention: Over 12 weeks, the student’s reading comprehension moved from 50% to 70% on weekly probes, but fluency gains plateaued after week 8 when the intervention frequency was reduced.” Such insights help teams make data-driven decisions faster. #### ESY Justification Additionally, if a psychologist needs to justify ESY services in a report, they can ask BastionGPT to draft the section explaining regression and recoupment concerns using the student’s past break-period data. The AI will articulate how the student’s skills significantly regressed over summer and took weeks to recover, supporting the need for ESY – saving the practitioner from trying to craft that argument from scratch. Throughout these tasks, the psychologist remains the expert in the loop: you verify that the data is interpreted correctly and make the final judgment calls. But by letting BastionGPT handle the first pass of number-crunching and summarizing, progress reports and data analyses can be completed with far less manual effort. ### 6. Creating Short-Form Reports and Addenda Not every report a school psychologist writes is a massive evaluation; there are many shorter reports and addenda that still require careful writing. Classroom observation reports, language proficiency impact statements, sensory profile summaries, and other addendum reports often need to be attached to evaluations or used in decision-making. BastionGPT can assist with these shorter-format documents just as effectively as with long ones. Classroom Observation Reports After conducting a 30-minute classroom observation, a psychologist can list out their raw notes — for example: "10:00 AM – student off-task, looking around room; 10:05 – teacher prompts student, student complies for 5 minutes…" The BastionGPT AI will transform the notes into a well organized and coherent narrative, describing the student’s on-task/off-task behavior, interactions, and environment in a professional tone. It might produce a paragraph noting patterns (such as the student needing frequent redirection, or particular times of day when behavior deteriorates) that the psychologist can then incorporate into the evaluation report or share with the team. #### Language Proficiency Impact Statements If an evaluation requires an addendum on how a student’s limited English proficiency might be affecting their test performance, BastionGPT can help draft that language impact statement by drawing on best practices (for instance, noting if testing was done in the student’s native language or if interpreter assistance was needed, and how the results should be interpreted with caution). #### Sensory Profiles & Supplemental Assessments The AI can similarly summarize the findings of a sensory processing profile or a brief supplemental assessment, providing a concise write-up that the psychologist can fine-tune. By handling the initial draft of these addenda, BastionGPT ensures that even small but important reports don’t get overlooked or delayed. School psychologists can knock out these ancillary documents in minutes rather than hours, confident that the output aligns with professional standards and can be easily edited for the final polish. ### 7. Managing Legal and Compliance Documentation When legal or compliance issues arise, documentation needs often become even more intensive. School psychologists may have to compile due process hearing documentation packets, write up Manifestation Determination Review (MDR) reports, or maintain detailed Child Find referral logs. These documents must be comprehensive and error-free, since they can be scrutinized in legal proceedings or audits. BastionGPT proves invaluable in assembling and refining such high-stakes paperwork. #### Due Process Documentation If a due process case is filed regarding a student’s services, for example, the district might need a complete history of that student’s evaluations, supports, and communications. Instead of manually collating and summarizing years of records, a psychologist can provide BastionGPT the relevant documents (e.g., previous eval reports, IEPs, meeting notes, emails) – taking advantage of the AI’s ability to process tens of thousands of words at a time, and have it generate a cogent summary or timeline of events. The AI can highlight key points (dates of evaluations, eligibility decisions, services provided, concerns raised by parents, etc.) with references to the source documents as appropriate. This draft timeline gives the team a strong starting point to work from, ensuring they don’t miss a critical detail when presenting the case. #### Manifestation Determination Reviews (MDR) For manifestation determination meetings (which decide if a student’s misbehavior is linked to their disability), BastionGPT can help structure the report by outlining: The incident details The disability in question Relevant evaluation findings The team's discussion and conclusion on whether the behavior was a manifestation of the disability By inputting the incident report and the student’s diagnostic profile, the psychologist can prompt the AI to draft the MDR report, saving significant time. #### Child Find Logs Likewise, maintaining Child Find logs – records of how the district is identifying and evaluating students who may need services – can be simplified by AI. BastionGPT could be used to regularly summarize outreach efforts, referrals, and their outcomes in a running log, ensuring that when auditors ask for documentation, the information is organized and readily available. All these sensitive use cases are where BastionGPT’s secure, compliant design shines. The AI enables school psychology teams to produce thorough, legally compliant documentation more efficiently, without exposing confidential student data to unsecured systems. In an era of increasing due process cases and compliance pressures, having an AI assistant that can churn through paperwork quickly is a tremendous asset. ## BastionGPT vs Other AI Tools BastionGPT is a HIPAA- and FERPA-compliant generative AI platform designed specifically for sensitive healthcare and education records. Every plan includes: Business Associate Agreement (BAA) Secure data handling Data is never sold or used to train AI models This means school psychologists can confidently use BastionGPT with identifiable student information (PHI/PII) without risking confidentiality or violating laws. For more on how BastionGPT protects student records, see the support article Is BastionGPT FERPA compliant? The platform offers ChatGPT-like capabilities such as conversational Q&A, content generation, summarization but within a protected service. Users get the benefits of the most powerful AI assistance without wasting time on manual de-identification or worrying about sanctions. ### Designed for the Complexity of School Psychology AI for school psychologists must go beyond basic functionality — it must respect privacy laws and assist meaningfully with complex documentation tasks. BastionGPT delivers: Large document processing — handles up to 1,000 pages, ideal for lengthy evaluation reports or years of student records OCR scanning — processes PDFs and images directly No content filtering — unlike free chatbots, it won't omit important details (e.g. discussions of self-harm or violence) because it's specifically tuned for medical and educational content ## Advantages of Integrating BastionGPT AI in School Psychology From individual school psychologists to entire district departments, BastionGPT is helping transform daily workflows in a safe, compliant, and effective manner. As we’ve explored, educators are using this HIPAA and FERPA compliant AI assistant to streamline evaluation reports, behavior assessments, crisis documentation, meeting notes, progress tracking, addenda, and legal paperwork. In each of these scenarios, BastionGPT acts as a tireless assistant, producing drafts, summaries, and insights that would otherwise take hours of manual writing, all while keeping sensitive student information secure by design. Its compliance-focused architecture (with BAA-backed data protections, encryption, and a private environment) means schools don’t have to worry about privacy violations when leveraging AI technology. Just as importantly, BastionGPT augments rather than replaces the psychologist’s expertise. It allows professionals to work more efficiently and focus on the human aspects of their role. Reports get written faster, insights come more easily, and students receive support sooner, all within a framework that prioritizes student privacy and ethical use. The time and energy saved can be redirected to what matters most: supporting students and improving outcomes. ## Start Using BastionGPT Today Most users are able to sign up and begin using BastionGPT within 10 minutes, with no setup costs, 7-day trial and no fixed commitments. Begin your AI journey with BastionGPT today by starting your trial here: https://bastiongpt.com/plus If you have more questions or would like to connect – you can reach out at: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/state-laws-on-healthcare-ai Healthcare Innovation # State Laws on Healthcare AI Written By: Josh Spencer • April 15, 2026 # State Healthcare & AI Compliance Laws Healthcare professionals using AI tools must navigate a growing patchwork of state privacy, data protection, and AI-specific laws. BastionGPT is HIPAA compliant by design, with a Business Associate Agreement (BAA) included on every plan. Learn more about navigating healthcare compliance with BastionGPT . Below is a state-by-state overview of the laws most relevant to healthcare AI assistants and scribes. Note: Federal laws including HIPAA, the HITECH Act, the FTC Act, COPPA, and Section 1557 of the Affordable Care Act apply in every U.S. state and are not repeated per state below. BastionGPT is built to support compliance with these federal requirements. Last reviewed August 16, 2026. Items marked “pending” have not been enacted, and several 2026 sessions (California, New York, New Jersey, Massachusetts, Pennsylvania, Michigan, Ohio) are still open, so this guide is re-checked quarterly. Updated August 2026: 2026 session outcomes added for 27 states; citations to superseded or never-enacted statutes removed. This page is general information, not legal advice. ## If you use AI to document visits, these laws speak to you directly Consent or notice before AI records or transcribes: Colorado (HB 26-1195 — written, revocable consent; in force Aug 12, 2026), Illinois (WOPR Act — written consent), Maine (LD 2082 — written consent; July 29, 2026), Rhode Island (H 7349/S 2197 — written consent; and ch. 23-106 — notify patients that AI documents the visit), Louisiana (Act 649 — tell patients before the visit; Aug 1, 2026), Arizona (Board of Behavioral Health Examiners rule — informed consent; Jan 1, 2027), Utah (SB 226 — prominent disclosure in high-risk generative-AI interactions), Texas (SB 1188 and TRAIGA — disclose AI use). Review what the AI wrote: Nevada (AB 406 — independently review AI-generated notes and billing), Rhode Island (ch. 23-106 — review AI documentation for accuracy), Texas (SB 1188 — review all AI-created records), Indiana (HB 1271 — no AI-prepared claims without human review). AI may not present as, or replace, a licensed clinician: California (AB 489), Delaware (HB 191), Illinois (WOPR), Nevada (AB 406), Missouri (SB 1019, Aug 28, 2026), Tennessee (SB 1580), Vermont (Act 156), Oregon (HB 2748, nursing titles), Nebraska (LB 525, 2027). All-party recording-consent states: California, Florida, Illinois, Maryland, Massachusetts, Montana, New Hampshire, Oregon (in-person), Pennsylvania, Washington — get and document consent from everyone in the room before an ambient scribe records; Connecticut and Nevada apply all-party rules to phone calls. Jump to a state… Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming ## Alabama AL Alabama Personal Data Protection Act (HB 351, eff. May 1, 2027) Data Breach Notification Act (Ala. Code § 8-38-1) SB 63 — AI in health-coverage decisions (eff. Oct 1, 2026) Alabama enacted its first comprehensive privacy law in 2026. The Alabama Personal Data Protection Act (HB 351, signed April 17, 2026) takes effect May 1, 2027 and gives consumers access, correction, deletion, portability, and opt-out rights; HIPAA covered entities, business associates, and protected health information are exempt. The Data Breach Notification Act requires prompt notice when personal data is compromised. SB 63 (also signed April 17, 2026; effective October 1, 2026) regulates how health plans use AI in coverage determinations: AI cannot be the sole basis for a decision and its use must be disclosed. BastionGPT supports Alabama clinicians with HIPAA-compliant AI that never uses customer data to train models. ## Alaska AK APIPA — breach notification (AS 45.48) HIE Confidentiality & Security (AS 18.23.300–.400) Health Care Services Information & Review Organizations (AS 18.23) Telehealth (AS 08.02.130) Alaska has no comprehensive consumer privacy law and no AI-specific statute. The Alaska Personal Information Protection Act (AS 45.48) governs breach notification and disposal of personal information. AS 18.23 protects the confidentiality of peer-review records and sets confidentiality and security standards for the state health information exchange (AS 18.23.300–.400). Alaska's telehealth statute (AS 08.02.130) applies when AI-assisted care is delivered remotely. BastionGPT's encryption and access controls support these confidentiality expectations. ## Arizona AZ Medical Claims & Prior Auth — HB 2175 (eff. July 1, 2026) BBHE rule — informed consent for AI used to record or document services (eff. Jan 1, 2027) Breach Notification (A.R.S. § 18-552) Insurance Utilization Review (A.R.S. Title 20) Arizona has no comprehensive privacy law or general AI statute, but HB 2175 (signed May 12, 2025; effective July 1, 2026) requires a medical director to exercise independent medical judgment on any denial involving medical necessity, so AI cannot be the final decision-maker on claims or prior authorizations. For behavioral-health licensees, the Board of Behavioral Health Examiners ' updated rules require, from January 1, 2027, documented informed consent before any service that uses AI or another “human simulation modality” to provide, record, or document clinical services. Arizona's breach law requires notice to affected individuals and the Attorney General. A 2026 chatbot-disclosure bill (HB 2311) was vetoed on June 19, 2026. BastionGPT reinforces this human-in-the-loop model: it drafts and assists while licensed clinicians retain every clinical decision, and BastionGPT's consent guidance helps practices document the required consent. ## Arkansas AR APIPA — breach notification (Ark. Code § 4-110-101) Children & Teens' Online Privacy Protection Act (Act 952 of 2025, eff. July 1, 2026) Deceptive Trade Practices Act Arkansas has no comprehensive consumer privacy law and no AI-in-healthcare statute. The Personal Information Protection Act (Ark. Code § 4-110-101) requires reasonable security and breach notification, and the Children and Teens' Online Privacy Protection Act (Act 952 of 2025) took effect July 1, 2026. Two 2025 bills on AI in insurer decisions and AI-generated medical records (HB 1297, HB 1816) were withdrawn. BastionGPT operates as a documentation tool under clinician supervision and does not make clinical or coverage decisions. ## California CA CCPA / CPRA CMIA (Cal. Civil Code § 56) AB 3030 — AI patient-communication disclosures SB 1120 — Physicians Make Decisions Act AB 489 — AI may not imply a health-care license SB 942 / AB 853 — CA AI Transparency Act (operative Aug 2, 2026) SB 243 — Companion Chatbots AB 2013 — GenAI training-data transparency SB 53 — Transparency in Frontier AI Act AB 45 — consumer health data & geofencing SB 361 — data-broker amendments SB 1223 — neural data as sensitive PI All-party recording consent (Penal Code § 632) CPPA automated-decision-making regulations (phasing in from 2027) California has the most extensive healthcare AI framework in the country. The CCPA/CPRA and the CMIA establish broad consumer and patient data rights. AB 3030 requires health facilities, clinics, and physician offices to disclose when patient communications about clinical information are generated by AI unless a licensed provider reviews them first. SB 1120 ensures AI cannot independently deny, delay, or modify care based on medical necessity, and AB 489 (effective January 1, 2026) prohibits AI from using terms that imply it holds a healthcare license. SB 243 regulates companion chatbots, the AI Transparency Act became operative on August 2, 2026, and the CPPA's automated-decision-making regulations phase in from 2027. California is also an all-party recording-consent state (Penal Code § 632), so patients must agree before an ambient scribe records a visit. Several 2026 bills on healthcare AI and chatbots are still before the Legislature. BastionGPT supports compliance by operating as a clinician-supervised assistant that never represents itself as a licensed professional, and BastionGPT's consent guidance helps practices meet AB 3030 and recording-consent obligations. ## Colorado CO Colorado Privacy Act (CPA) SB 26-189 — ADMT transparency law (replaces SB 24-205; eff. Jan 1, 2027) HB 26-1195 — Psychotherapy AI restrictions (eff. Aug 12, 2026) HB 26-1139 — AI in utilization review (eff. Jan 1, 2027) Breach Notification (C.R.S. § 6-1-716) Colorado rewrote its AI framework in 2026. SB 26-189 (signed May 14, 2026; effective January 1, 2027) repeals the 2024 Colorado AI Act and replaces it with transparency duties for automated decision-making technology: notice when ADMT influences a consequential decision, an explanation within 30 days of an adverse decision, and a right to human review. HB 26-1195 , effective August 12, 2026, directly governs psychotherapists: AI may be used only for administrative or supplementary tasks under the provider's full responsibility, providers must review any AI-generated recommendations, and clients must give clear, written, revocable consent before AI records or transcribes a session — a refusal cannot affect access to care. HB 26-1139 (effective January 1, 2027) requires individualized, clinician-reviewed decisions when insurers use AI in utilization review. The Colorado Privacy Act adds consumer data rights and profiling opt-outs. BastionGPT is built for the permitted role — a documentation tool under provider supervision — and its consent guidance helps practices document HB 26-1195 consent. ## Connecticut CT CTDPA (Conn. Gen. Stat. § 42-515 et seq.) CTDPA amendments — LLM-training disclosure (SB 1295 / PA 25-113, eff. July 1, 2026) SB 4 / PA 26-64 — data brokers, geolocation, profiling (2026) SB 5 / PA 26-15 — AI companions & employment ADMT (2026) Breach Notification (§ 36a-701b) Safeguards Law (§ 42-471) CUTPA Connecticut keeps layering AI duties onto its privacy law. Since July 1, 2026 the CTDPA , as amended by SB 1295 (Public Act 25-113), requires businesses to disclose whether personal data is used to train large language models, lowers applicability thresholds, and expands sensitive-data and minor protections; SB 4 (Public Act 26-64, May 2026) adds data-broker, geolocation, and profiling rules. SB 5 (Public Act 26-15, the 2026 online-safety and AI act) regulates AI companion chatbots from January 1, 2027 and automated employment decisions from October 1, 2026; it does not add provider-specific healthcare duties. Because BastionGPT never uses customer or patient data to train models, the CTDPA's core disclosure concern does not arise. ## Delaware DE DPDPA (6 Del. C. ch. 12D) HB 191 — AI may not hold or use medical or nursing titles (2026) Breach Notification (6 Del. C. § 12B-101) Online Privacy & Protection Act (6 Del. C. ch. 12C) Delaware's Personal Data Privacy Act stands out because it does not include an entity-level exemption for HIPAA covered entities: PHI is exempt at the data level, but HIPAA-regulated organizations must still comply with the DPDPA for data outside PHI. The law grants consumers rights to access, correct, delete, and opt out of advertising and profiling. In April 2026 Delaware enacted HB 191 , which bars any nonhuman entity — including AI agents — from being licensed as, or presenting itself as, a physician, physician assistant, or nurse. BastionGPT never represents itself as a licensed professional, and its HIPAA-compliant design and data-minimization practices support Delaware clinicians navigating the DPDPA overlap. ## Florida FL FIPA (Fla. Stat. § 501.171) Florida Digital Bill of Rights (§ 501.701) FDUTPA All-party recording consent (§ 934.03) Florida has not enacted an AI-specific healthcare law; the 2026 AI Bill of Rights (SB 482) passed the Senate but died in the House on March 13, 2026, as did bills restricting AI in psychology and counseling. FIPA imposes strict breach notification (a 30-day window, with medical records covered), and the Florida Digital Bill of Rights adds consumer rights for large controllers with a HIPAA covered-entity exemption. Florida is an all-party recording-consent state (§ 934.03), so patients must consent before an ambient scribe records a visit. BastionGPT is a HIPAA-compliant AI assistant operating under licensed provider supervision. ## Georgia GA Personal Identity Protection Act (O.C.G.A. § 10-1-910) Fair Business Practices Act SB 444 — AI in insurance decisions (eff. Jan 1, 2027) SB 540 — conversational-AI disclosure & minor safety (eff. July 1, 2027) Georgia has no comprehensive consumer privacy statute. Its Personal Identity Protection Act requires breach notice “in the most expedient time possible,” and Bland v. Urology of Greater Atlanta (Ga. Ct. App. 2025) recognized a common-law duty to protect personal information. In 2026 Georgia enacted SB 444 (signed May 5; effective January 1, 2027), which bars insurers from issuing adverse determinations on AI alone, and SB 540 (signed May 11; effective July 1, 2027), which requires conversational-AI operators to disclose AI use and protect minors. BastionGPT operates under provider supervision and does not make coverage determinations. ## Hawaii HI Breach Notification (HRS ch. 487N) Personal Information Protection — SSN rules (ch. 487J) Destruction of Personal Information Records (ch. 487R) Unfair/Deceptive Acts (§ 480-2) SB 3001 / Act 248 — AI companion chatbots (2026) Hawaii has no comprehensive consumer privacy law and no AI-in-healthcare statute; a 2026 healthcare-AI disclosure bill (SB 2281) did not pass. Its breach law ( HRS ch. 487N ) requires notice without unreasonable delay, ch. 487J restricts use of Social Security numbers, and ch. 487R governs destruction of personal-information records. In July 2026 Hawaii enacted Act 248 (SB 3001), which requires AI companion chatbots to disclose that users are talking to AI and to follow self-harm protocols. BastionGPT is not a consumer chatbot; it is a HIPAA-compliant documentation tool used by licensed providers. ## Idaho ID Breach Notification (Idaho Code § 28-51-104) Consumer Protection Act Conversational AI Safety Act (SB 1297, eff. July 1, 2027) Idaho has no comprehensive privacy law. In 2026 it enacted SB 1297 , the Conversational AI Safety Act (signed March 31, 2026; effective July 1, 2027), which requires conversational-AI operators to disclose that users are interacting with a machine and to maintain protocols for adult and minor users. Idaho's breach statute (§ 28-51-104) and Consumer Protection Act otherwise govern. BastionGPT is not a consumer-facing chatbot; it is a HIPAA-compliant documentation tool used by licensed providers. ## Illinois IL BIPA (740 ILCS 14) WOPR Act (Public Act 104-0054) PIPA / Breach (815 ILCS 530) Consumer Fraud Act (815 ILCS 505) All-party recording consent (720 ILCS 5/14-2) SB 3114 — human review of AI claim downcoding (2026) Illinois has one of the most active regulatory environments for healthcare AI. The WOPR Act (HB 1806, Public Act 104-0054, effective August 1, 2025) prohibits AI from providing therapy, making therapeutic decisions, or communicating therapeutically with clients, while allowing licensed professionals to use AI for administrative and supplementary support — but recording or transcribing a session with AI requires that the client be informed in writing and give explicit, revocable written consent. BIPA imposes strict consent requirements for biometric identifiers, including voiceprints, with a private right of action, and Illinois is an all-party recording-consent state. In 2026 the Transparency in Downcoding Act (SB 3114) added human-review requirements for payer algorithms. BastionGPT fits within WOPR's permitted categories — drafting notes, transcribing sessions with documented consent, and supporting administrative tasks — without engaging in therapeutic communication. ## Indiana IN INCDPA (Ind. Code art. 24-15, eff. Jan 1, 2026) Breach Notification (IC 24-4.9) HB 1271 — human review of AI in claims and downcoding (eff. July 1, 2026) Deceptive Consumer Sales Act Indiana's Consumer Data Protection Act (effective January 1, 2026) gives residents access, correction, deletion, and portability rights, requires data protection assessments for sensitive data, exempts HIPAA-regulated data, and has a permanent 30-day cure period. The breach statute allows Attorney General penalties of up to $150,000 per deceptive act. HB 1271 (effective July 1, 2026) bars insurers from using AI as the sole basis for downcoding or medical-necessity denials and requires providers to review any AI-prepared claim before submission. BastionGPT drafts under clinician review and processes clinical data under HIPAA safeguards. ## Iowa IA Iowa CDPA (Iowa Code ch. 715D, eff. Jan 1, 2025) Breach Notification (ch. 715C) Consumer Fraud Act (§ 714.16) HF 2635 — AI in prior authorization (eff. July 1, 2026) Iowa's Consumer Data Protection Act (effective January 1, 2025) exempts HIPAA-regulated entities and PHI. HF 2635 , signed May 13, 2026 and effective July 1, 2026, lets utilization-review organizations use AI for initial prior-authorization review but prohibits AI as the sole basis for denying, delaying, or downgrading a medically necessary service. BastionGPT operates within the HIPAA framework as a provider-supervised documentation tool. ## Kansas KS Breach Notification (K.S.A. § 50-7a01) Consumer Protection Act Health Information Technology Act Kansas has no comprehensive consumer privacy law and no AI-specific statute; the 2026 chatbot-safety proposal (HB 2671) died in committee. The breach law requires notice “in the most expedient time possible” and reporting to consumer reporting agencies when more than 1,000 residents are affected. BastionGPT is a clinical documentation tool, not a consumer-facing chatbot. ## Kentucky KY KCDPA (KRS 367.3611–.3629, eff. Jan 1, 2026) Breach Notification (KRS 365.732) Consumer Protection Act Telehealth Laws Kentucky's Consumer Data Protection Act took effect January 1, 2026 and grants access, correction, deletion, portability, and opt-out rights; HIPAA covered entities' PHI and, after the 2025 amendment (HB 473), health-care providers' HIPAA-governed information are exempt. Kentucky has no AI-specific statute. BastionGPT's privacy-first design supports Kentucky providers in meeting these requirements. ## Louisiana LA Louisiana Data Privacy Act (SB 386 / Act 502, eff. Jan 1, 2027) HB 475 / Act 649 — AI transcription disclosure (R.S. 37:22.1, eff. Aug 1, 2026) Breach Notification (La. R.S. 51:3071) Medical Records Privacy (R.S. 40:1165.1) LUTPA Louisiana enacted two relevant laws in 2026. Act 649 (HB 475), effective August 1, 2026, requires healthcare professionals to tell patients before an appointment begins if AI-powered recording or transcription software will be used. The Louisiana Data Privacy Act (SB 386, Act 502) takes effect January 1, 2027 with access, deletion, and opt-out rights and consent for sensitive data. Louisiana's breach law and medical-records privacy statute (R.S. 40:1165.1) continue to apply, and LUTPA reaches deceptive claims about what AI tools can do. BastionGPT's consent guidance helps practices meet the Act 649 disclosure duty. ## Maine ME Notice of Risk to Personal Data Act (10 M.R.S. §§ 1346–1350-B) LD 2082 / P.L. 2025 c. 687 — AI in mental-health services (eff. July 29, 2026) Broadband Internet Privacy Law (35-A M.R.S. § 9301) Unfair Trade Practices Act Maine has no comprehensive consumer privacy law — the Maine Online Data Privacy Act (LD 1822) failed in April 2026 — but its breach-notification statute and broadband privacy law apply. LD 2082 , effective July 29, 2026, restricts AI in mental-health care: only licensed professionals may provide therapy, AI may be used for administrative or supplementary support under the licensee's responsibility, and using AI in a recorded or transcribed session requires written notice and the client's written, revocable consent. BastionGPT's consent guidance helps Maine clinicians document that consent. ## Maryland MD MODPA (Com. Law § 14-4601 et seq., eff. Oct 1, 2025) MPIPA — breach (§ 14-3501) Confidentiality of Medical Records Act (Health-Gen. § 4-301) HB 820 — AI in utilization review (2025) All-party recording consent (Cts. & Jud. Proc. § 10-402) HIE Regulations (COMAR 10.25.18) Maryland's Online Data Privacy Act (effective October 1, 2025) grants access, correction, deletion, and opt-out rights and exempts HIPAA covered entities and PHI. The Confidentiality of Medical Records Act adds state rules on consent, disclosure, and access, and Maryland's wiretap law requires all parties' consent before a conversation is recorded. HB 820 (2025) requires insurers using AI in utilization review to base decisions on the individual patient's clinical record and keep humans responsible for denials. BastionGPT functions as a clinical documentation tool under provider control. ## Massachusetts MA Data Security Regulations (201 CMR 17.00) Breach Notification (M.G.L. c. 93H) Chapter 93A Consumer Protection All-party recording consent (M.G.L. c. 272, § 99) Patients' Rights (M.G.L. c. 111, § 70E) Consumer Data Privacy Act (in conference committee) Massachusetts has one of the strictest data-security regimes — 201 CMR 17.00 requires a written information security program with encryption, access controls, and training — and an all-party recording-consent law (M.G.L. c. 272, § 99) that matters for any AI scribe. A comprehensive Consumer Data Privacy Act passed both chambers in 2026 and is in conference committee; it is not yet law. Chapter 93A applies to deceptive or unfair AI practices. BastionGPT's security architecture and consent guidance support these requirements. ## Michigan MI Identity Theft Protection Act (MCL 445.61) Consumer Protection Act Medical Records Access Act (MCL 333.26261) Michigan's Identity Theft Protection Act requires businesses to notify individuals and the Attorney General following a data breach, and the Consumer Protection Act prohibits unfair and deceptive practices. The Medical Records Access Act governs patient access to records. Michigan has no comprehensive consumer privacy statute and no AI-specific law; bills on insurer AI decisions and chatbots remain pending. BastionGPT operates as a secure AI assistant under clinician supervision. ## Minnesota MN MCDPA (Minn. Stat. ch. 325O, eff. July 31, 2025) MGDPA (§ 13.01) Breach Notification (§ 325E.61) Health Records Act (§ 144.291) Minnesota's Consumer Data Privacy Act (effective July 31, 2025) grants consumers data access, correction, deletion, and opt-out rights and requires data protection assessments for high-risk processing. The state's Health Records Act provides Minnesota-specific rules that apply in addition to HIPAA, making it one of the more protective health privacy frameworks. Minnesota has no AI-in-healthcare statute; bills on AI in psychotherapy and prior authorization remain pending. BastionGPT is purpose-built as a healthcare AI transcriptionist and assistant under clinician control. ## Mississippi MS Breach Notification (§ 75-24-29) Consumer Protection Act Medical Records Confidentiality (§ 41-9-61) Mississippi maintains a leaner privacy framework, with primary protections from the breach notification statute and Consumer Protection Act ; medical records statutes establish patient consent and confidentiality requirements. Mississippi has no comprehensive privacy law and no AI-in-healthcare statute (a 2026 bill to bar AI from mental and behavioral health care, HB 1720, died in committee). BastionGPT keeps Mississippi providers in control of documentation. ## Missouri MO Breach Notification (RSMo § 407.1500) Merchandising Practices Act Health Information Privacy (§ 191.227) SB 1019 — AI may not be represented as a mental-health professional (eff. Aug 28, 2026) Missouri has no comprehensive privacy law; its breach statute and Merchandising Practices Act are the main consumer protections, and § 191.227 governs patient access to records. SB 1019 , effective August 28, 2026, makes it an unlawful practice to advertise or represent AI as a mental-health professional or as able to provide therapy or a diagnosis, with penalties of $10,000 to $20,000 per violation. BastionGPT is a documentation assistant used under clinician oversight and is never presented as a therapist. ## Montana MT MCDPA (MCA § 30-14-2801; SB 297 amendments eff. Oct 1, 2025) Uniform Health Care Information Act (MCA § 50-16-501) Breach Notification (MCA § 30-14-1704) All-party recording notice (MCA § 45-8-213) Genetic Information Privacy Act (MCA § 30-23-101) Montana's Consumer Data Privacy Act gives residents rights to access, correct, delete, and port personal data and to opt out of sales, targeted advertising, and profiling; SB 297 (effective October 1, 2025) lowered the applicability threshold to 25,000 consumers, added duties for minors' data, and removed the cure period. Health information held by providers is separately governed by the Uniform Health Care Information Act (MCA § 50-16-501 et seq.). Montana requires that everyone in a conversation know it is being recorded (MCA § 45-8-213), so patients and anyone else present should be told before an ambient AI scribe records a visit. Montana has no statute specific to AI in clinical care or mental health. BastionGPT operates as a clinician-supervised documentation tool within these privacy and consent requirements. ## Nebraska NE NDPA (Neb. Rev. Stat. § 87-1101, eff. Jan 1, 2025) Conversational AI Safety Act (LB 525, eff. July 1, 2027) LB 77 — AI in prior authorization (2025) Breach Notification (§ 87-801) Consumer Protection Act (§ 59-1601) Nebraska's Data Privacy Act (effective January 1, 2025) gives consumers rights of access, correction, deletion, portability, and opt-out, requires consent before processing sensitive data such as health information, and requires data protection assessments for higher-risk processing. The Conversational AI Safety Act (LB 525, signed April 14, 2026; effective July 1, 2027) prohibits chatbots from presenting themselves as licensed mental-health professionals or holding clinical credentials. LB 77 (2025) bars utilization-review agents from using AI as the sole basis to deny, delay, or modify care and requires disclosure of AI use. Nebraska has no statute specific to AI scribes or AI-generated clinical notes. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Nevada NV AB 406 — AI in mental & behavioral health (NRS ch. 629 & 433, eff. July 1, 2025) Consumer Health Data Privacy (SB 370; NRS § 603A.400) SB 220 online privacy opt-out (NRS § 603A.300) Breach Notification (NRS § 603A.220) Nevada's AB 406 (effective July 1, 2025) bars licensed mental and behavioral health providers from using AI to deliver care directly to a patient, while expressly allowing AI for administrative support such as scheduling, billing, and organizing session notes — provided the provider independently reviews the accuracy of any AI-generated notes or billing records and complies with HIPAA and state records law; the same act prohibits AI systems from being marketed as therapists or counselors, with civil penalties up to $15,000 per violation. Nevada's Consumer Health Data Privacy Law (SB 370) requires consent for collection and sharing of consumer health data outside HIPAA, bars geofencing of health facilities, and requires opt-in authorization for sales of health data. SB 220 gives consumers a right to opt out of the sale of covered information collected online, and NRS 603A.220 sets breach-notification duties. A 2025 bill limiting insurers' use of AI in prior authorization (SB 128) was vetoed. BastionGPT operates as a clinician-supervised documentation tool consistent with AB 406's administrative-support model, with the clinician reviewing every note. ## New Hampshire NH NH Data Privacy Act (RSA ch. 507-H, eff. Jan 1, 2025) All-party recording consent (RSA 570-A:2) Breach Notification (RSA 359-C:19) Consumer Protection Act (RSA 358-A) Insurance Data Security (RSA 420-P) New Hampshire's data privacy law (RSA chapter 507-H, effective January 1, 2025) gives consumers rights to access, correct, delete, and port personal data and to opt out of targeted advertising, sales, and profiling; it requires consent to process sensitive data such as health information and data protection assessments for heightened-risk processing. New Hampshire is an all-party-consent state (RSA 570-A:2), so patients and anyone else present must consent before an ambient AI scribe records a visit. Insurers are covered by the Insurance Data Security Law (RSA 420-P), and breaches by RSA 359-C:19. New Hampshire has no statute specific to AI in clinical care; 2026 bills on AI providing licensed professional services (SB 640) and on AI in utilization review (HB 1406) were both rejected. BastionGPT operates as a clinician-supervised documentation tool within these privacy and consent requirements. ## New Jersey NJ NJDPA (N.J.S.A. § 56:8-166.4 et seq., eff. Jan 15, 2025) Identity Theft Prevention Act / breach notice (N.J.S.A. § 56:8-163) Consumer Fraud Act (N.J.S.A. § 56:8-1) A4070 — limits on patient identity data collected by health facilities (eff. Apr 1, 2027) Bills barring AI advertised as a licensed mental-health professional (A799/S735) New Jersey's Data Privacy Act (effective January 15, 2025) gives consumers rights of access, correction, deletion, portability, and opt-out, requires consent to process sensitive data — expressly including mental or physical health condition, treatment, or diagnosis — and requires data protection assessments for heightened-risk processing. Breach notification is governed by the Identity Theft Prevention Act , and the Consumer Fraud Act underpins Attorney General enforcement. A4070 (signed March 25, 2026; effective April 1, 2027) restricts what identity information health care facilities may collect and disclose about patients. New Jersey has not enacted a law requiring disclosure of AI use in clinical interactions; bills to bar AI from being advertised as a licensed mental-health professional and to have licensing boards set generative-AI rules remain pending. BastionGPT operates as a clinician-supervised documentation tool within these privacy requirements. ## New Mexico NM Breach Notification (NMSA § 57-12C-1) Unfair Practices Act (NMSA § 57-12-1) New Mexico has no comprehensive consumer data privacy law and no statute specific to AI in health care; every AI bill in the 2025 and 2026 sessions — including the Artificial Intelligence Act (HB 60), the Chatbot Safety Act (HB 174), and the AI Transparency Act (HB 28) — was postponed indefinitely. The Data Breach Notification Act requires notice to affected residents and, for larger breaches, the Attorney General, and the Unfair Practices Act supports enforcement against deceptive practices. New Mexico has not adopted the NAIC Insurance Data Security Model Law; the Office of Superintendent of Insurance has addressed cybersecurity only by bulletin. HIPAA therefore remains the primary framework for clinical documentation. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## New York NY SHIELD Act (GBL §§ 899-aa / 899-bb) AI Companion Models (GBL Art. 47, eff. Nov 5, 2025) DFS Cybersecurity Reg. (23 NYCRR 500) NYC Local Law 144 (AEDT) RAISE Act (eff. Jan 1, 2027) Health Information Privacy Act (vetoed Dec 19, 2025; S9269 reintroduced) Kids chatbot safety bill (S9051, passed June 2026, awaiting Governor) New York regulates data security through the SHIELD Act (reasonable safeguards plus breach notification) and, for insurers and other DFS licensees, the Part 500 cybersecurity regulation, whose 2023 amendments were fully phased in by November 2025. General Business Law Article 47 (in force November 5, 2025) requires AI companion operators to disclose that users are not talking to a human — at the start of a session and every three hours — and to maintain self-harm detection and referral protocols. The RAISE Act (signed December 2025; effective January 1, 2027) imposes safety and incident-reporting duties on frontier-model developers, and NYC Local Law 144 requires bias audits of automated employment tools. The New York Health Information Privacy Act passed the Legislature but was vetoed on December 19, 2025 (a revised bill, S9269, is pending), and bills on AI in psychotherapy and AI in utilization review did not pass in 2026; New York currently has no statute requiring clinicians to disclose AI documentation tools. BastionGPT operates as a clinician-supervised documentation tool within New York's data-security requirements. ## North Carolina NC Identity Theft Protection Act (G.S. § 75-60; breach notice § 75-65) UDTPA (G.S. § 75-1.1) Health Information Exchange Act H 565 — AI in Medicaid/commercial insurance decisions S 963 — AI chatbots North Carolina has no comprehensive consumer data privacy statute and no law specific to AI in health care; 2025–26 bills to limit AI in Medicaid and commercial insurance decisions (H 565) and to license and regulate AI chatbots (S 963) remain pending. The Identity Theft Protection Act requires reasonable disposal of personal information and breach notification to affected residents and the Attorney General, and the Unfair and Deceptive Trade Practices Act supports enforcement and private treble-damage suits. The Health Information Exchange Act governs provider participation in the state HIE. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## North Dakota ND Breach Notification (N.D.C.C. § 51-30-01) Consumer Fraud Act (N.D.C.C. § 51-15-01) Insurance Data Security Act (N.D.C.C. ch. 26.1-02.2) North Dakota has no comprehensive consumer data privacy law and no statute governing AI in clinical care; its only AI enactments (2025) require disclosure of AI-generated political communications and clarify that an AI system is not a “person” under state law. Businesses must notify residents under the breach-notification law (N.D.C.C. ch. 51-30), and the Consumer Fraud Act supports enforcement against deceptive practices. Insurance licensees must maintain information-security programs under the Insurance Data Security Act (N.D.C.C. ch. 26.1-02.2). BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Ohio OH Data Protection Act / cybersecurity safe harbor (SB 220; ORC § 1354.01) Breach Notification (ORC § 1349.19) Consumer Sales Practices Act (ORC § 1345.01) HB 525 — AI in therapy services HB 579 / SB 164 — AI use by health insurers Ohio has no comprehensive consumer data privacy law and no statute specific to AI in clinical care; bills on AI in therapy services (HB 525) and on insurers' use of AI (HB 579, SB 164) were pending as of August 2026. The Data Protection Act (SB 220, ORC 1354) gives businesses an affirmative defense to data-breach tort claims if they maintain a written cybersecurity program that reasonably conforms to a recognized framework such as the NIST Cybersecurity Framework or HIPAA's Security Rule. Breaches of unencrypted personal information must be disclosed under ORC 1349.19, and the Consumer Sales Practices Act supports Attorney General enforcement. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Oklahoma OK Oklahoma Consumer Data Privacy Act (SB 546, eff. Jan 1, 2027) Security Breach Notification Act (24 O.S. § 161; SB 626 amendments eff. Jan 1, 2026) Consumer Protection Act Insurance Data Security Act (36 O.S. § 670) Oklahoma enacted a comprehensive consumer data privacy law in 2026 ( SB 546 , signed March 20, 2026; effective January 1, 2027) giving residents access, deletion, correction, and opt-out rights and imposing duties on controllers that process sensitive data such as health information. The Security Breach Notification Act (24 O.S. 161 et seq.), as amended by SB 626 effective January 1, 2026, requires notice to affected residents without unreasonable delay and to the Attorney General within 60 days when more than 500 residents are affected, with a safe harbor for HIPAA-compliant entities. Insurers must maintain information-security programs under the Insurance Data Security Act . Oklahoma has no statute specific to AI in clinical care — a 2026 bill requiring informed consent for AI use by mental-health professionals and other providers (SB 2037) failed — so HIPAA governs AI documentation tools. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Oregon OR OCPA (ORS § 646A.570; HB 2008 amendments eff. Jan 1, 2026) AI Companions Act (SB 1546, eff. Jan 1, 2027) HB 2748 — AI barred from nursing titles (2025) In-person recording notice (ORS § 165.540) Consumer Information Protection Act / breach (ORS § 646A.604) Genetic Privacy (ORS § 192.531) Oregon's Consumer Privacy Act gives consumers rights to access, correct, delete, and port personal data and to opt out of targeted advertising, sales, and profiling, requires consent for sensitive data such as health information and data protection assessments for higher-risk processing; 2025 amendments (HB 2008, effective January 1, 2026) ban selling precise geolocation data and data of consumers under 16. Under ORS 165.540 , an in-person conversation may not be recorded unless all participants are specifically informed, so patients and companions must be told before an ambient AI scribe captures a visit. SB 1546 (effective January 1, 2027) requires AI-companion operators to disclose that users are talking to AI and to maintain self-harm response protocols, with a private right of action, and HB 2748 (2025) bars AI and other nonhuman entities from using nursing titles. Oregon has no statute specific to AI-generated clinical notes; a 2026 bill on insurers' AI downcoding (HB 4054) did not pass. BastionGPT operates as a clinician-supervised documentation tool within these privacy and notice requirements. ## Pennsylvania PA Two-party recording consent (18 Pa.C.S. § 5701 et seq.) Breach of Personal Information Notification Act (73 P.S. § 2301; Act 33 of 2024 amendments) UTPCPL (73 P.S. § 201-1) Insurance Data Security Act (40 Pa.C.S. § 4501) Consumer Data Privacy Act (HB 78, passed House Oct 2025) HB 1925 — AI in health care & insurance Pennsylvania's Wiretap Act (18 Pa.C.S. 5701 et seq.) requires the consent of all parties before a conversation is recorded, so an ambient AI scribe may only capture a visit after the patient — and anyone else present — agrees. Pennsylvania has no comprehensive consumer privacy law: HB 78 , the Consumer Data Privacy Act, passed the House on October 1, 2025 and was still before the Senate as of late June 2026. The Breach of Personal Information Notification Act , as amended by Act 33 of 2024, requires notice to affected residents and, in some cases, free credit monitoring; insurers must also comply with the Insurance Data Security Act (40 Pa.C.S. 4501–4536). Bills on AI in health facilities and insurance decisions (HB 1925) and on AI in mental-health therapy (HB 1993, HB 2100) were pending and unenacted as of August 2026. BastionGPT has published guidance on consent practices and operates as a clinician-supervised documentation tool within these consent and data-security requirements. ## Rhode Island RI Use of AI by Healthcare Providers Notification Act (R.I. Gen. Laws ch. 23-106, eff. June 22, 2026) Oversight of AI Technology in Mental Health Care Act (H 7349 / S 2197, signed June 22, 2026) Data Transparency & Privacy Protection Act (§ 6-48.1, eff. Jan 1, 2026) Health Care Confidentiality Act (§ 5-37.3) AI Companion Models (S 2195, eff. Jan 1, 2027) Identity Theft Protection / breach (§ 11-49.3) Insurance Data Security (§ 27-1-46) Rhode Island now directly regulates clinical AI. The Use of Artificial Intelligence by Healthcare Providers Notification Act (R.I. Gen. Laws ch. 23-106, signed June 22, 2026 and effective immediately) requires providers who use AI to document patient visits to notify patients and to review the AI-generated documentation for accuracy, and the Oversight of Artificial Intelligence Technology in Mental Health Care Act (also signed June 22, 2026) bars AI from making independent therapeutic decisions and requires written consent before AI is used to record or transcribe therapy sessions. The Data Transparency and Privacy Protection Act (effective January 1, 2026) adds consumer data rights and consent for sensitive data, while the Confidentiality of Health Care Communications and Information Act continues to govern health information held by providers. A companion-AI disclosure law (S 2195) takes effect January 1, 2027. BastionGPT operates as a clinician-supervised documentation tool designed for the notify-and-review workflow these laws require, and its consent guidance helps practices document the required notice and consent. ## South Carolina SC Physicians' Patient Records Act (§ 44-115-10) Breach Notification (§ 39-1-90) Insurance Data Security Act (§ 38-99-10) Unfair Trade Practices Act (§ 39-5-10) South Carolina has no comprehensive consumer data privacy law and no statute governing AI in clinical care; a 2026 bill to restrict AI in therapy and psychotherapy (S 788) and chatbot-regulation bills (S 896, H 5138) did not become law. The Physicians' Patient Records Act governs ownership, confidentiality, and release of medical records, and the breach-notification law requires notice to affected residents. Insurers must maintain information-security programs under the Insurance Data Security Act , modeled on the NAIC model law. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## South Dakota SD Breach Notification (SDCL § 22-40-19) Deceptive Trade Practices & Consumer Protection (SDCL ch. 37-24) South Dakota has no comprehensive consumer data privacy law and no statute specific to AI in health care; 2026 bills to restrict AI in therapy (HB 1144), require chatbot disclosures (SB 170), and govern insurers' AI coverage determinations (SB 169) were withdrawn or tabled. The breach-notification law (SDCL 22-40-19 et seq.) requires notice to affected residents, and the Deceptive Trade Practices and Consumer Protection chapter supports Attorney General enforcement. South Dakota has not adopted the NAIC Insurance Data Security Model Law; insurers are subject to Division of Insurance safeguards rules. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Tennessee TN TIPA (Tenn. Code Ann. § 47-18-3201) SB 1580 / Pub. Ch. 647 — AI may not pose as a mental-health professional (eff. July 1, 2026) ELVIS Act (§ 47-25-1101) Breach Notification (§ 47-18-2107) Insurance Data Security Act (§ 56-2-1001) HB 1866 / SB 2010 — Regulate AI in Health Care Act Tennessee's Information Protection Act (TIPA, effective July 1, 2025) gives consumers rights of access, correction, deletion, portability, and opt-out (including profiling), requires consent for sensitive data such as health information, and offers an affirmative defense to businesses that maintain a written privacy program reasonably conforming to the NIST Privacy Framework. Public Chapter 647 (SB 1580, effective July 1, 2026) prohibits anyone who develops or deploys an AI system from representing that it is or can act as a qualified mental health professional, enforceable under the Consumer Protection Act with $5,000-per-violation penalties. The ELVIS Act (2024) bars unauthorized AI replication of a person's voice or likeness. Tennessee has no statute governing AI-generated clinical documentation; a broader Regulate AI in Health Care Act remained pending in 2026. BastionGPT operates as a clinician-supervised documentation tool and does not present itself as a mental health professional. ## Texas TX SB 1188 — AI disclosure & record review by practitioners (Health & Safety Code § 183.005, eff. Sept 1, 2025) TRAIGA (HB 149; Bus. & Com. Code § 552.051 health-care AI disclosure; eff. Jan 1, 2026) TMRPA (Health & Safety § 181.001) TDPSA (Bus. & Com. § 541.001) SB 815 — AI in utilization review (eff. Sept 1, 2025) Breach Notification (Bus. & Com. ch. 521) DTPA Texas now imposes two provider-level AI duties. SB 1188 (effective September 1, 2025) permits health care practitioners to use AI for diagnostic purposes only if they review all AI-created records consistent with Texas Medical Board standards and disclose their use of AI to patients, and requires electronic health records to be stored in the United States from January 1, 2026. The Texas Responsible Artificial Intelligence Governance Act (HB 149, effective January 1, 2026) adds a general duty for anyone providing licensed health care services to disclose, no later than the first date of service, that an AI system is used in relation to the service, alongside prohibitions on manipulative or discriminatory AI and Attorney General enforcement with a 60-day cure period. The Medical Records Privacy Act applies HIPAA-style duties to a broader set of entities, and the Texas Data Privacy and Security Act grants consumer rights and requires consent for sensitive data; SB 815 (2025) bars insurers from relying solely on AI for adverse utilization-review decisions. BastionGPT operates as a clinician-supervised documentation tool consistent with the disclose-and-review workflow these statutes require. ## Utah UT UCPA (Utah Code § 13-61-101) Generative-AI disclosure in regulated occupations (SB 226; § 13-75-103, eff. May 7, 2025) Mental-health chatbot rules (HB 452; § 13-72a-101, eff. May 7, 2025) AI Policy Act (SB 149; § 13-72-101; sunset July 1, 2027) SB 319 — AI in preauthorization (eff. Jan 1, 2027) Breach Notification (§ 13-44-101) Utah's Consumer Privacy Act gives consumers access, deletion, portability, and opt-out rights for targeted advertising and sales, with notice-and-opt-out treatment of sensitive data. Since May 7, 2025, SB 226 (Utah Code ch. 13-75) requires anyone providing services in a state-licensed occupation — including physicians and mental-health professionals — to prominently disclose, verbally at the start of a spoken interaction or in writing before a written one, when the person receiving services is interacting with generative AI in a “high-risk” interaction such as one collecting health data or delivering medical or mental-health advice; a clear self-disclosure by the AI is a safe harbor. HB 452 (also effective May 7, 2025) separately regulates mental-health chatbots, requiring non-human disclosure and barring sale of health inputs and targeted ads. The original AI Policy Act (SB 149) remains in force until July 1, 2027, and SB 319 (effective January 1, 2027) restricts insurers' use of AI in preauthorization. BastionGPT operates as a clinician-supervised documentation tool; clinicians remain responsible for any required disclosure to patients. ## Vermont VT Vermont Data Privacy & Online Surveillance Act (S.71 / Act 145, eff. Jan 1, 2028) H.816 / Act 156 — mental-health services by professionals, not AI (eff. June 17, 2026) H.814 / Act 101 — neurological rights & AI Advisory Council (2026) Data Broker Regulation (9 V.S.A. § 2446; 2026 amendments) Security Breach Notice Act (9 V.S.A. § 2435) Insurance Data Security (8 V.S.A. § 4728) Vermont enacted its first comprehensive privacy law in June 2026: the Vermont Data Privacy and Online Surveillance Act (S.71, Act 145) takes effect January 1, 2028 and will give consumers rights to access, correct, delete, and opt out, require consent for sensitive data such as health information, and require data protection assessments. Act 156 (H.816, effective June 17, 2026) provides that mental health services may not be delivered independently by an AI system — only by mental health professionals — while preserving administrative uses such as scheduling, billing, and transcription that support rather than replace clinical judgment; AI-driven therapeutic decision-making is unprofessional conduct. Act 101 (H.814, 2026) recognizes neurological rights, including neural-data privacy, and extends the state's AI Advisory Council. Vermont's pioneering Data Broker Regulation was tightened in 2026, and breaches are governed by 9 V.S.A. § 2435. BastionGPT operates as a clinician-supervised documentation tool — an administrative-support role consistent with Act 156. ## Virginia VA VCDPA (Va. Code § 59.1-575) Health Records Privacy (Va. Code § 32.1-127.1:03) Breach Notification (§ 18.2-186.6) Insurance Data Security Act (§ 38.2-621) SB 269 — AI in mental-health services (continued to 2027) Virginia's Consumer Data Protection Act was among the first comprehensive state privacy laws, giving consumers access, correction, deletion, portability, and opt-out rights and requiring opt-in consent to process sensitive data such as mental or physical health diagnoses, with data protection assessments for higher-risk processing. Health records held by providers are separately governed by Va. Code § 32.1-127.1:03, breaches by § 18.2-186.6, and insurers by the Insurance Data Security Act . Virginia has no enacted AI-specific health law: the High-Risk AI Developer and Deployer Act (HB 2094) was vetoed on March 24, 2025, and 2026 bills allowing mental-health providers to use AI in therapy with patient consent (SB 269) and requiring insurers to disclose AI use in claims decisions (SB 586) passed the Senate but were continued to the 2027 session. BastionGPT operates as a clinician-supervised documentation tool within Virginia's privacy and health-records requirements. ## Washington WA My Health My Data Act (RCW 19.373) All-party recording consent (RCW § 9.73.030) Uniform Health Care Information Act (RCW 70.02) AI companion chatbots (HB 2225, eff. Jan 1, 2027) SB 5395 — AI in prior authorization (eff. June 11, 2026) Breach Notification (RCW § 19.255) Consumer Protection Act (RCW 19.86) Washington's My Health My Data Act applies to consumer health data outside HIPAA, requires consent to collect or share such data beyond what is necessary for a requested service and separate authorization to sell it, bans geofencing of health facilities, and is enforceable through a private right of action under the Consumer Protection Act. Washington is an all-party-consent state: under RCW 9.73.030 a private conversation may not be recorded without the consent of everyone present, which may be obtained by announcing the recording to all parties (and capturing that announcement in the recording) — a rule that directly applies to ambient AI scribes. Provider-held records are governed by the Uniform Health Care Information Act (RCW 70.02). In 2026 the state enacted HB 2225 (effective January 1, 2027), regulating AI companion chatbots, and SB 5395 (effective June 11, 2026), barring insurers from relying solely on AI to deny or limit care in prior authorization; there is still no statute specific to AI-generated clinical documentation. BastionGPT's consent guidance is especially relevant here, and BastionGPT operates as a clinician-supervised documentation tool within these consent and health-data requirements. ## West Virginia WV Breach Notification (W. Va. Code § 46A-2A-101) Consumer Credit & Protection Act (§ 46A-6-101) West Virginia has no comprehensive consumer data privacy law and no statute governing AI in clinical care; a proposed Consumer Data Protection Act (HB 2987) was never enacted, and a 2026 bill to limit AI in mental-health care to administrative support (HB 4770) did not pass. The state's only AI enactment is HB 3187 (2025), which created a Task Force on Artificial Intelligence. Breaches of unencrypted personal information must be disclosed under W. Va. Code § 46A-2A-101 et seq., and the Consumer Credit and Protection Act supports Attorney General enforcement against unfair or deceptive practices. HIPAA therefore remains the primary framework for AI documentation tools in West Virginia. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Wisconsin WI Patient Health Care Records (Wis. Stat. § 146.81) Mental Health Treatment Records (§ 51.30) Breach Notification (§ 134.98) Deceptive Trade Practices (§ 100.18) Wisconsin Consumer Act Insurance Data Security (§ 601.95) Wisconsin has no comprehensive consumer data privacy law and no statute specific to AI in health care; 2025–26 bills to limit AI-driven prior-authorization denials (AB 1109/SB 1066) and to restrict AI chatbots that simulate relationships with children did not pass. Patient health care records are governed by Wis. Stat. 146.81–146.84, which require informed consent for most disclosures, and mental-health treatment records receive additional protection under Wis. Stat. 51.30 . Breaches must be reported under § 134.98, insurers must maintain information-security programs under §§ 601.95–601.956, and § 100.18 prohibits deceptive representations. BastionGPT operates as a clinician-supervised documentation tool within these confidentiality requirements. ## Wyoming WY Breach Notification (W.S. § 40-12-501) Consumer Protection Act (W.S. § 40-12-101) Wyoming has no comprehensive consumer data privacy law and no statute governing AI in clinical care. The breach-notification law (W.S. 40-12-501 et seq.) requires notice to affected residents when personal identifying information is compromised, and the Consumer Protection Act prohibits deceptive trade practices. Wyoming has not adopted the NAIC insurance data-security model law. HIPAA therefore remains the primary framework for AI documentation tools in Wyoming. BastionGPT operates as a clinician-supervised documentation tool within these requirements. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/therapy-dap-notes-ai Healthcare Innovation # DAP Notes: What They Are and How to Write Them With AI Written By: Robert Henehan • March 4, 2025 Healthcare professionals know that writing clear, comprehensive progress notes is essential but time-consuming. AI scribes can offer a secure solution to streamline the creation of DAP notes while maintaining clinical standards and professional requirements. AI chatbots for healthcare provide the flexibility of ChatGPT with specific focus on medical documentation and are designed to support compliance with security standards. ## What Are DAP Notes? DAP notes are a structured progress note format used across behavioral health, counseling, and medical settings. The acronym stands for Data , Assessment , and Plan , with each section serving a specific purpose in the clinical record: Data captures the objective and subjective information from a session, including client statements, behavioral observations, mood and affect, and topics discussed. Assessment reflects the clinician's professional interpretation of the data, including clinical impressions, progress toward treatment goals, risk factors, and changes from prior sessions. Plan outlines the next steps, including session frequency, interventions, referrals, homework assignments, and any adjustments to the treatment plan. ### Why DAP Notes Matter Accurate, well-structured DAP notes support continuity of care by giving every provider on a treatment team a clear picture of what happened in a session, what it means clinically, and what comes next. They also play a direct role in meeting documentation requirements for insurance reimbursement, audits, and regulatory compliance. Payers and accreditation bodies expect progress notes that connect the presenting problem, the intervention, and the treatment trajectory. DAP notes do this in a format that is concise and easy to review. ### Who Uses DAP Notes DAP notes are widely used by licensed counselors, clinical social workers (LCSWs), psychologists, marriage and family therapists (LMFTs), psychiatric nurse practitioners, rehabilitation specialists, and substance abuse counselors. The format is especially popular in outpatient mental health and behavioral health settings, though it appears across medical disciplines where concise, structured documentation is valued. The Value of DAP Notes in Clinical Practice Many clinicians have adopted DAP notes for their clarity and efficiency. Let's look at what makes this format particularly helpful: Clearer organization than narrative notes More concise than some alternative formats Easier integration of therapeutic observations and interventions Better tracking of treatment progress Streamlined communication with other healthcare providers DAP notes bring together the essential elements of clinical documentation while allowing for your personal documentation style to shine through. ## How to Write a DAP Note (Step-by-Step) Each section of a DAP note builds on the one before it. The Data tells you what happened, the Assessment interprets what it means, and the Plan maps out what to do next. Here's what to include in each step. ### Step 1: Data The Data section is your factual record of the session. Include the client's presenting mood and affect, behavioral observations, direct quotes that capture key moments, topics discussed, interventions used, and any measurable progress indicators (such as PHQ-9 scores, frequency of symptoms, or homework completion rates). Write what you observed and what the client reported, keeping interpretation out of this section. ### Step 2: Assessment The Assessment section is where your clinical reasoning lives. Interpret the data you just recorded: What patterns are emerging? How is the client progressing toward treatment goals compared to previous sessions? Note changes in symptom severity, barriers to progress, therapeutic alliance observations, and any risk factors you identified. This is the section that connects the raw session content to your professional judgment. ### Step 3: Plan The Plan section documents the path forward. Include continued treatment goals, specific interventions you intend to use in upcoming sessions, homework or between-session activities assigned, any referrals or coordination with other providers, medication considerations, crisis planning if applicable, and the schedule for the next appointment. The Plan should flow directly from your Assessment, so each next step has a clear clinical rationale behind it ## Creating A DAP Note Using AI Here's a prompt template you might find helpful when working with an AI scribe: "You are a licensed healthcare professional documenting a client session. Create a comprehensive DAP note that includes: Data Section: Client's presentation (mood, affect, appearance) Direct quotes from the client Behavioral observations Content discussed Progress on previous goals Response to interventions Current symptoms Assessment Section: Clinical insights and interpretations Changes from previous sessions Current stage of treatment Barriers to progress Risk assessment findings Treatment goal progress Therapeutic relationship observations Plan Section: Focus for next session Specific interventions to use Homework assignments Referrals needed Follow-up timeline Crisis planning if needed Treatment plan adjustments Format using professional clinical language while maintaining clarity. Only include information that was included in the visit." ## Adapting to Your DAP Note Documentation Style You've developed your own unique documentation approach over years of practice. Rather than changing to fit an AI tool, you can have the AI adapt to your style. Try sharing some of your existing notes with your AI scribe to create a personalized template: Help me analyze my documentation style and create a customized DAP note template matching my clinical approach. Please review my sample notes and analyze them for the following elements: Format Analysis: Section organization and structure Use of headers and subheaders Formatting preferences and patterns Level of detail in each section Note length and consistency Timestamp and dating conventions Clinical Content Review: Depth of objective observations Assessment comprehensiveness Risk documentation patterns Goal tracking methods Progress measurement approach Clinical terminology usage Treatment planning style Documentation Voice: Professional language patterns Client quote integration Balance of clinical/accessible language Narrative structure Documentation flow Assessment style Planning format After completing the analysis, please: Create a summary of my documentation style Develop a DAP note template matching my approach Include appropriate clinical placeholders Provide best practices for maintaining my established style Sample DAP Notes for Analysis:[Insert your 3 DAP notes here] ## Example DAP Note Using BastionGPT Client: John D. Date of Service: 2/10/2025 Session Duration: 50 minutes Service Type: Individual Therapy Session #8 DATA: John arrived on time for our in-person session, appearing well-groomed in business attire. His GAD-7 score decreased to 14 from 18 three weeks ago. John reports continued difficulty with work-related anxiety, particularly around presentations, stating "I lay awake until 2 AM last night going over my presentation deck in my head." Sleep has improved since implementing evening meditation, now averaging 6 hours per night. He successfully used deep breathing exercises 4 out of 7 days this week during anxiety episodes, noting reduced physical symptoms (chest tightness, racing heart) when using these techniques. John completed 5 daily thought records, showing increased awareness of workplace anxiety triggers. ASSESSMENT: John demonstrates improved insight into anxiety triggers, particularly recognizing perfectionist thinking patterns around work presentations. He shows increased ability to challenge anxious thoughts using evidence-based questioning techniques learned in previous sessions. Sleep improvement indicates positive response to behavioral interventions. Risk assessment completed with no current concerns. Support system remains strong with weekly family contact. Progress toward treatment goals shows moderate improvement, particularly in implementing coping strategies and maintaining work-life boundaries. PLAN: Will meet again next week at 2:00 PM. Assigned homework includes completing thought records for 2 work-related anxiety episodes and practicing progressive muscle relaxation (recording provided via portal). Will continue cognitive restructuring work focusing on perfectionist thoughts and introduce additional relaxation techniques. Will consult with psychiatrist Dr. Smith if sleep concerns persist. ## Best Practices for Using AI for DAP Notes When collaborating with an AI chatbot for note creation, consider these approaches: Review and customize the generated content to match your voice Verify clinical accuracy and add your professional insights Add personal observations that only you could have noticed Save successful prompts that work well for your practice Update templates as your documentation needs evolve ## Documentation Considerations for AI-Assisted Notes As you develop your AI-assisted documentation workflow, you might want to: Maintain your preferred balance of objective and subjective language Ensure your usual risk assessment process is reflected Incorporate your consultation or supervision documentation style Keep your approach to noting changes in presentation Continue your established follow-up planning format ## Getting Started with AI for DAP Notes Consider starting with basic templates and gradually refining them to better suit your specific practice needs. Focus particularly on how you document progress toward treatment goals and changes in client presentation. Note AI tools work best when viewed as collaborative partners in the documentation process. They can help reduce the time spent on administrative tasks while allowing you to maintain your clinical voice and professional standards. ##### Ready to enhance your DAP Note documentation process? Start your free 7-day trial to experience how BastionGPT can support your clinical practice while maintaining the highest standards of security and privacy. Our platform helps you create thorough, professional notes while reducing the time spent on documentation, allowing you to focus more energy on what matters most - your clients' care and progress. ## FAQS ### What does DAP stand for in counseling notes? DAP stands for Data, Assessment, and Plan. It is a structured progress note format used by mental health and behavioral health professionals to document session content, clinical interpretation, and next steps in treatment. ### How long should a DAP note be? Most DAP notes run one to three paragraphs, with roughly one paragraph per section. The goal is to be thorough enough to support clinical reasoning and meet payer requirements while staying concise. If your notes regularly exceed 300 words, you may be drifting into transcript territory rather than documenting what is clinically relevant. ### How do DAP notes improve patient care? DAP notes create a clear, consistent record that links what happened in a session to your clinical interpretation and the treatment path forward. This structure makes it easier to track progress over time, coordinate with other providers, and identify when a treatment plan needs adjustment. ### Can AI write DAP notes securely? AI tools can draft DAP notes from session transcripts or clinical input, but security depends entirely on the platform. A HIPAA-compliant tool like BastionGPT is designed to protect patient data with encryption, a signed BAA, and zero use of PHI for model training. Always verify that any AI tool you use for clinical documentation meets your practice's compliance requirements. ### What's the difference between SOAP and DAP notes? SOAP notes separate client-reported information (Subjective) from clinician-observed data (Objective) into two distinct sections, then follow with Assessment and Plan. DAP notes combine all session data into a single Data section, which makes them faster to write and well suited for behavioral health settings where the subjective/objective distinction is less clinically relevant. ### How do you write a good DAP note? Start by recording the facts of the session (client presentation, direct quotes, interventions used, measurable indicators) in the Data section without interpretation. Then write your clinical assessment of that data, linking your conclusions to specific observations and treatment goals. Close with a Plan that includes concrete next steps, such as interventions, homework, referrals, or scheduling. ### Are DAP notes HIPAA-compliant when using AI? The DAP format itself has no bearing on HIPAA compliance. What matters is the platform handling the data. If you use an AI tool that provides encryption and a BAA, your AI-assisted DAP notes can be part of a HIPAA-compliant workflow. BastionGPT is built to support these requirements on every plan. ### How long are DAP notes? DAP notes are typically one to three short paragraphs, with each section (Data, Assessment, Plan) running a few sentences to a short paragraph. Payers and auditors look for documentation that ties directly to treatment goals, so a focused 200-word note that shows clear clinical reasoning is stronger than a longer note filled with vague descriptions. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/tokenization-anonymization-hipaa-compliance-risks Healthcare Innovation # HIPAA Compliance Risks with Tokenization and Anonymization in Generative AI Written By: Josh Spencer • December 30, 2024 As healthcare organizations increasingly explore the potential of generative AI technologies like ChatGPT, ensuring HIPAA compliance remains a critical concern. While various approaches exist to integrate these powerful tools into healthcare operations, not all methods offer equal protection for sensitive patient information. This article examines the risks associated with tokenization and anonymization strategies in healthcare AI applications such as CompliantChatGPT, and explores more secure alternatives for maintaining HIPAA compliance. ## Understanding Tokenization in Healthcare AI Tokenization is a process where sensitive data elements are replaced with non-sensitive equivalents, known as tokens, that maintain the essential format of the data without exposing the actual protected health information (PHI). While this approach might seem promising at first glance, it carries significant, often unacceptable, risks when applied to healthcare data in generative AI systems. ### The Tokenization Approach Some healthcare AI services employ tokenization as a method to achieve HIPAA compliance when interfacing with general-purpose non-compliant AI models like consumer ChatGPT services . The process typically involves: Receiving protected health information from healthcare providers Automatically replacing sensitive elements with tokens Sending the tokenized data to external non-compliant AI models for processing Re-identifying the information when returning results ### The Hidden Risks of Tokenization Tokenization allows companies to more quickly implement AI services, while costing less money than alternatives. While tokenization can appear to be a cost-effective solution, it introduces several critical vulnerabilities that healthcare organizations need to consider: ### Reliability Concerns Even with a 99.9% success rate in tokenization, the implications of failure are severe. For healthcare operations processing thousands of records daily, a 0.1% failure rate could result in hundreds of HIPAA violations each year. Each failure potentially represents a federally reportable security breach, creating significant legal and regulatory exposure. ### Regulatory Compliance Issues HIPAA regulators have shown increasing scrutiny of tokenization-only approaches. During audits, whether routine or incident-triggered, the use of tokenization as a primary security measure may be deemed insufficient, potentially leading to: Regulatory sanctions Substantial fines Mandatory operational changes Damage to organizational reputation ### Technical Limitations The effectiveness of tokenization depends heavily on the accuracy of the algorithms identifying sensitive information. However, healthcare data often contains complex and contextual information that automated systems might miss, such as: Indirect patient identifiers Contextual medical information that could become identifying when combined Novel or unusual data patterns that tokenization algorithms haven't been trained to recognize ## A More Secure Approach: Isolated Environment Implementation Rather than relying on tokenization, a more robust approach involves operating AI models in isolated, HIPAA-compliant environments. This method typically includes: ### Direct Model Integration By using services which license AI models directly and operating them in a controlled and secure environment, organizations can maintain reliable data security assurances. This approach eliminates the need for tokenization by keeping sensitive data within a certified HIPAA-compliant infrastructure. ### Enhanced Security Features A properly isolated environment includes: Complete separation from external non-compliant AI provider services Comprehensive audit trails Controlled access mechanisms Secure data storage and transmission Regular security assessments and updates ## Making the Right Choice for Your Organization When evaluating AI solutions for healthcare applications, organizations should carefully consider the following factors: ### Risk Assessment Consider the volume of PHI your organization handles and the potential impact of even a small failure rate in data protection measures. The consequences of data breaches extend beyond regulatory penalties to include patient trust and organizational reputation. These stakes are underscored by the growing threat of healthcare cybercrime . ### Long-term Viability While tokenization might offer a quicker or less expensive path to AI implementation, the long-term costs of potential breaches, regulatory actions, and necessary security updates could far exceed the initial savings. ### Regulatory Alignment Ensure your chosen approach aligns with both current HIPAA requirements and anticipated regulatory changes in the rapidly evolving landscape of healthcare AI. ## Conclusion As healthcare organizations continue to adopt generative AI technologies, the choice of implementation method becomes increasingly crucial. While tokenization might appear to be a viable solution for achieving HIPAA compliance, its inherent risks and potential for failure make it a problematic choice for healthcare organizations handling sensitive patient data. Organizations should carefully evaluate their options and consider more robust solutions that provide complete control over their AI environment. By prioritizing security and compliance from the ground up, healthcare providers can better protect patient information while still leveraging the powerful capabilities of generative AI technologies. In healthcare technology, the path of least resistance isn't always the safest route. Investing in proper infrastructure and security measures from the start can prevent costly complications and maintain the trust of both patients and regulators. Unlike tokenization-based services, BastionGPT uses licensed LLMs hosted in HIPAA-compliant environments. By prioritizing security and regulatory compliance, it avoids the pitfalls of anonymization while delivering powerful AI capabilities. Healthcare professionals can confidently use BastionGPT knowing their data never leaves secure infrastructure. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/transforming-mental-health-practice Healthcare Innovation # Transforming Mental Health Practice: How AI Gave Me Back Time to Bake Bread and Better Serve My Clients Written By: Emma Levy • June 3, 2025 ## Starting with AI: From Spreadsheets to Life-Changing Technology I'm a clinical psychologist who transitioned from acute mental health work in hospitals to conducting developmental assessments. Today, I work privately, helping parents understand what's happening with their children whether it's learning difficulties like dyslexia, developmental issues like autism or ADHD, or intellectual disabilities. Essentially, I'm a diagnostician who figures out the answers to "what's going on with this young person?" and connects families to the resources they need. My AI journey began with one of those memorable moments that stick with you forever. At a meeting with colleagues who share office space, someone mentioned that their supervisor had recorded a supervision session, and AI created a summary automatically, no note-taking required. We all just went "What? How?" I remember that exact moment because it seemed impossible. Being naturally curious about new technology, I went straight home and signed up for a free trial of that first AI tool. While it wasn't the most user-friendly program, it planted a seed. Over the summer break, I made a decision that 2025 was going to be the year I streamlined my reports because they simply couldn't take as long as they were taking. The real catalyst came when I was trying to automate my Excel tables for scoring assessments. I had a ChatGPT account but didn't really know what to ask it. When I started building spreadsheets to automate score calculations, I quickly hit ChatGPT's daily question limit. That's when I tried my BastionGPT code and spent two intensive weeks creating automated tables for all my assessment tools, tables that now color-code ranges and calculate percentiles automatically. What struck me most was developing this strange relationship with the AI during those weeks. I was actually thanking BastionGPT because it felt like having another person working alongside me, patiently answering question after question about Excel formulas. I thought I'd just use the 30-day trial for those spreadsheets and cancel, but by the end of that period, I realized I couldn't live without this technology. ## Using AI & BastionGPT: The Most Transformative Impact The most transformative use case has been recording and processing parent interviews. When assessing children for developmental issues, I conduct detailed interviews with parents that can last one to two hours, gathering comprehensive developmental history with incredibly specific questions about play, social interactions, and behavioral patterns. Writing up these parent interviews used to be the absolute worst part of my job. I didn't mind the conversations, but creating a narrative from two hours of detailed discussion was torture. Parents don't speak according to the headings in my reports, they'll mention three different topics in one sentence, and I'd need to extract pieces for different sections. I used to highlight each portion I'd transferred to show I'd dealt with it, and it took hours upon hours. Now, I give BastionGPT my headings and an example of my narrative style. At the end of the session, I press a button, and what used to take me hours is done. It's really weird because what used to be the longest part of my report has become one of the quickest. The transformation went from several hours to maybe 15 minutes of reading and tweaking. BastionGPT stands out because of its clean, straightforward interface. You can just go on and start using it immediately. It's like ChatGPT but purpose-built for healthcare. I've tried many other platforms, and I keep going back to BastionGPT. Some competitors had interfaces that felt like a mess with too much going on, while others were too American-focused with terminology and frameworks that didn't apply to our work in New Zealand. With BastionGPT, I could simply ask for British spellings, and it adapted perfectly. One memorable example involved a student with a complex learning profile affecting mathematics. I used BastionGPT to create a separate handout for the math teacher, explaining why the student's specific cognitive weaknesses would impact mathematical learning more than other subjects. I gave the AI my full report and asked it to create a targeted explanation for a teacher who thought the student was just being lazy. I never would have created a separate, personalized handout before because I was simply too busy. But BastionGPT made it effortless to provide customized support. ## The Impact of AI: Professional and Personal Transformation The personal changes have been remarkable. I joked to my husband recently while baking bread that I used to bake all the time, asking "When did I last produce a loaf of bread?" I told him, "This is AI you can thank for me doing this. I literally have got time on my hands again." BastionGPT has genuinely given me hours of my day back. Before adopting this technology, I was at the point of wondering how long I could keep doing assessment work. A colleague and I were discussing whether we could keep writing these extensive reports, they were becoming overwhelming. Now we joke that we can keep doing this work for the rest of our lives, and we can truly keep doing it now. The technology has also transformed my interactions with parents during assessments. Instead of constantly typing detailed notes while looking at a Word document, I can now look directly at them and have genuine conversations. I can maintain eye contact and be fully present instead of switching between typing and glancing at the clients. It's restored the human component to my practice. ## Challenges & Ethics: Navigating New Territory The biggest learning curve was mastering prompt engineering. I didn't initially understand how to communicate with AI effectively. My advice to others struggling with this is to talk to it like you have an intern standing next to you, explaining the full context. For that math teacher handout, I explained that she didn't believe the student had legitimate difficulties and thought he could just try harder. The key breakthrough came from giving examples in prompts. Most of my effective prompts now include an example of my style, saying "write it like this one below." The AI reads your style and adapts accordingly. Ethically, I've created consent forms that address AI use, and it's remarkable that BastionGPT could reference relevant New Zealand legislation when I asked. No parent has refused consent when I explain that I use AI for transcription and organization, they generally respond that this is what professionals are doing now. The murky ethical area involves determining how much AI assistance is appropriate in report writing. I'm very choosy about which sections I allow AI to help with, maintaining human oversight for clinical interpretations. While AI can accurately organize conversations and even suggest diagnoses with surprising accuracy, I believe the clinical judgment and interpretation should remain distinctly human. ## The Future of AI: Imagining Possibilities Honestly, I couldn't have imagined current AI capabilities a year ago, so predicting the future feels impossible. From a practical standpoint, I'd love AI to automatically extract scores from assessment protocols and populate my tables, eliminating manual data entry errors. I envision AI continuing to handle the administrative burden while preserving the essentially human aspects of psychological practice. The technology should augment our clinical expertise, not replace our professional judgment. ## Advice & Final Thoughts: Don't Delay, Just Try It My advice to peers is simple: don't delay this. It feels like a big, overwhelming technology, but it's not. Sign up for a trial and just ask it one question about a current client, maybe request a handout for parents on a specific topic. That's all it is, nothing more complicated. I tell people I've become like an "AI evangelist". I can't stop talking about it and trying to recruit others because the impact has been so transformative. My colleague jokes that we've become like missionaries for this technology. Try a few different platforms to find what works for you, but don't be intimidated. The key is treating AI like a knowledgeable assistant: provide context, include examples, and be specific about what you want. The one message I want readers to understand is that this technology has genuinely transformed my practice and given me back my professional satisfaction. I was approaching burnout from administrative burden, and now I can focus on what I love about this work, helping families understand their children and connecting them to appropriate support. If you're hesitating about AI adoption, remember that every day you delay is a day you could have been baking bread instead of drowning in paperwork. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/use-cases-for-generative-ai-in-healthcare Healthcare Innovation # Top Use Cases for Generative AI in Healthcare Written By: Josh Spencer • April 24, 2024 BastionGPT has helped thousands of healthcare professionals elevate their patient care and regain time in their day to better care for patients and their own mental well-being. We regularly check in with our customers to better understand their service use. As a part of that, we are happy to share the most valuable and common use cases with our community. As a versatile, healthcare-focused , HIPAA-compliant, and intelligent assistant, BastionGPT offers many applications that can significantly enhance productivity, creativity, and decision-making processes in the healthcare sector. Please note that while BastionGPT has taken steps to reduce the errors that generative AI occasionally produces, all output generated by any generative AI (such as ChatGPT, BastionGPT, Microsoft Copilot, Gemini or Anthropic Claude) must be reviewed by a healthcare professional before being shared directly with patients or being used in any manner which could cause harm. Also, ensure you are only using HIPAA-compliant services when you are entering patient data into your prompts or file uploads. For more information about our ethical AI standards and requirements, please see our AI Principles . ## 1. Healthcare Communication and Summarization ##### Example : A doctor needs to summarize a patient's medical history and current treatment plan for a referral letter. BastionGPT can draft the necessary information into a concise, professional document, ensuring all relevant details are included. ##### Capabilities: • Create clear and concise patient notes. • Summarize medical records and lab results. • Draft referral letters and treatment plans. ##### Detailed Example: A pediatrician uses BastionGPT to draft a letter to a specialist, summarizing a young patient's history of asthma, current medications, recent symptoms, and previous test results. The AI ensures that all critical information is communicated effectively, facilitating better-coordinated care. ## 2. Creative Writing and Content Generation in Healthcare ##### Example: A healthcare marketing team needs engaging content for a new wellness program. BastionGPT can generate creative and compelling marketing content, including program descriptions, social media posts, and blog articles. ##### Capabilities: • Generate creative content such as stories, poems, and marketing copy. • Brainstorm ideas and refine drafts. • Ensure content aligns with the desired tone and style. ##### Detailed Example: A healthcare organization uses BastionGPT to draft a series of blog posts about mental health awareness. The AI generates detailed and engaging descriptions of various mental health topics, complete with tips and resources, helping to attract and engage readers. ## 3. Research and Data Analysis in Healthcare ##### Example: A researcher needs to conduct a literature review on the impact of social media on mental health. BastionGPT can process numerous academic articles and summarize key findings, saving the researcher valuable time. ##### Capabilities: • Synthesize complex information and generate summaries. • Assist in literature reviews and data interpretation. • Create research reports and highlight key insights. ##### Detailed Example: A graduate student uses BastionGPT to summarize findings from multiple studies on the effects of exercise on mental health. The AI compiles the results into a cohesive report, highlighting trends, significant data points, and areas needing further research, aiding the student in writing a comprehensive thesis. ## 4. Customer Support and Interaction in Healthcare ##### Example: A healthcare provider needs to improve its patient communication by providing quick and accurate responses to inquiries. BastionGPT can draft personalized replies to patient emails, resolve common issues, and follow up with additional information. ##### Capabilities: • Generate quick and accurate responses to patient inquiries. • Handle common patient service issues. • Draft follow-up emails and communications. ##### Detailed Example: A medical clinic uses BastionGPT to draft patient inquiries about appointment scheduling, test results, and medication information. The AI generates personalized responses, ensuring that each patient receives timely, friendly and accurate information, enhancing overall patient satisfaction. ## 5. Business and Administrative Tasks in Healthcare ##### Example: A hospital administrator needs to generate a summary of a recent staff meeting and create an action plan. BastionGPT can draft a detailed meeting summary, highlighting key points and assigning tasks to team members. ##### Capabilities: • Draft professional emails and reports. • Generate meeting summaries and action plans. • Automate routine administrative tasks. ##### Detailed Example: A hospital executive uses BastionGPT to draft a quarterly report for stakeholders. The AI compiles data from various departments, summarizes achievements, outlines challenges, and presents future goals, ensuring a comprehensive and professional presentation. ## 6. Legal Document Preparation in Healthcare ##### Example: A legal assistant in a healthcare organization needs to draft a summary of a case for an attorney. BastionGPT can generate a detailed and well-organized case summary. ##### Capabilities: • Draft legal documents and case summaries. • Summarize legal research and findings. • Generate client correspondence and reports. ##### Detailed Example: A paralegal uses BastionGPT to draft a summary of a patient's case, including key facts, legal issues, and relevant case law. The AI ensures that the summary is clear, concise, and professionally written, aiding the attorney in preparing for court. ## 7. Financial Analysis and Reporting in Healthcare ##### Example: A financial analyst in a healthcare organization needs to create a report on quarterly earnings. BastionGPT can draft a comprehensive financial report, including data analysis, trends, and projections. ##### Capabilities: • Generate financial reports and summaries. • Analyze financial data and trends. • Draft investor communications and presentations. ##### Detailed Example: A financial advisor uses BastionGPT to draft a quarterly earnings report for a healthcare organization. The AI analyzes the financial data, identifies key trends, and provides a clear and detailed summary, helping the advisor communicate effectively with stakeholders. ## 8. Event Planning and Management in Healthcare ##### Example: An event planner in a healthcare organization needs to create a detailed plan for an upcoming health fair. BastionGPT can draft an event plan, including the schedule, budget, and vendor information. ##### Capabilities: • Generate event plans and schedules. • Draft vendor communications and contracts. • Summarize event details and logistics. ##### Detailed Example: An event planner uses BastionGPT to draft a comprehensive plan for a health fair. The AI includes a detailed schedule, budget breakdown, and vendor contact information, ensuring that all aspects of the event are well-organized and communicated effectively. ## 9. Patient Education Materials ##### Example: A healthcare provider needs to create educational materials for patients with diabetes. BastionGPT can generate informative and easy-to-understand content, including brochures, handouts, and digital resources. ##### Capabilities: • Generate patient education materials. • Summarize complex medical information in layman's terms. • Create engaging and accessible content. ##### Detailed Example: A diabetes educator uses BastionGPT to draft a series of brochures on managing diabetes. The AI generates clear and concise information on diet, exercise, medication management, and monitoring blood sugar levels, helping patients better understand and manage their condition. ## 10. Telehealth Communication and Documentation ##### Example: A telehealth provider needs to maintain accurate and detailed records of virtual consultations. BastionGPT can draft detailed consultation summaries, including patient concerns, diagnoses, treatment plans, and follow-up recommendations. ##### Capabilities: • Draft consultation summaries and treatment plans. • Maintain accurate and detailed telehealth records. • Generate follow-up recommendations and patient communication. ##### Detailed Example: A telehealth provider uses BastionGPT to draft a summary of a virtual consultation with a patient. The AI includes the patient's reported symptoms, the provider's observations, the diagnosis, and the recommended treatment plan, ensuring that the telehealth records are comprehensive and accurate. ## Conclusion BastionGPT is the most capable and secure generative AI on the market, which transforms how we approach documentation, communication, and creative endeavors in the healthcare sector. Its applications in healthcare communication, creative content generation, research, patient support, administrative tasks, legal documentation, financial analysis, event planning, patient education materials, and telehealth communication highlight its potential to drive efficiency and innovation across multiple domains. Subscribing to BastionGPT will unlock new levels of productivity and creativity, making it an essential tool for healthcare professionals. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/use-cases-for-the-generative-ai-in-therapy-and-psychiatry Healthcare Innovation # Top Use Cases for the Generative AI in Therapy and Psychiatry Written By: Josh Spencer • June 10, 2024 Generative AI technology is rapidly transforming healthcare. As a versatile AI assistant and AI scribe , BastionGPT offers numerous applications that can significantly enhance productivity, creativity, and decision-making processes in the fields of therapy and psychiatry. However, it is essential to note that the output generated by BastionGPT should be reviewed by a qualified therapist, psychiatrist or other qualified professional before being shared directly with clients. Here are some of the top mental health and psychiatry-related use cases for BastionGPT, complete with detailed examples: ## 1. Mental Health Documentation and Reporting ##### Example: A therapist needs to compile a neuropsychological report for a child with autism. BastionGPT can draft a comprehensive report using raw data from assessments, parent interviews, and behavioral observations. This report includes sections like developmental history, current functioning, and recommendations, written in clear, empathetic language. ##### Capabilities: • Generate neuropsychological evaluation reports. • Create SOAP notes for therapy sessions. • Summarize patient histories and progress notes. ##### Detailed Example: A therapist uses BastionGPT to draft a progress note for a session with a client who is dealing with anxiety. The generative AI summarizes the session's focus, their reported symptoms, the interventions used (such as cognitive-behavioral techniques), and the client's response to the therapy, ensuring thorough and professional documentation. ## 2. Individualized Treatment Plans ##### Example: A psychiatrist needs to develop a treatment plan for a patient with bipolar disorder. BastionGPT can draft a detailed treatment plan, including medication management, psychotherapy recommendations, and lifestyle modifications. ##### Capabilities: • Generate individualized treatment plans. • Summarize patient progress and adjust treatments. • Create comprehensive care plans. ##### Detailed Example: A psychiatrist uses BastionGPT to draft a treatment plan for a patient with bipolar disorder. The AI includes medication recommendations, therapy options, and lifestyle changes, ensuring a holistic approach to the patient's care. ## 3. Patient Communication and Follow-Up ##### Example: A therapist needs to send follow-up emails to clients after sessions. BastionGPT can draft personalized follow-up emails, summarizing key points from the session and providing additional resources or homework assignments. ##### Capabilities: • Generate personalized follow-up emails. • Summarize session highlights and recommendations. • Provide additional resources and assignments. ##### Detailed Example: A therapist uses BastionGPT to draft follow-up emails for clients, summarizing the session's key points and including any recommended exercises or resources. This ensures that clients have a clear understanding of their progress and next steps. ## 4. Group Therapy Session Summaries ##### Example: A therapist conducts weekly group therapy sessions and needs to summarize each session for the participants. BastionGPT can draft detailed summaries, highlighting key discussions, shared insights, and collective goals. ##### Capabilities: • Summarize group therapy sessions. • Highlight key discussions and insights. • Document collective goals and progress. ##### Detailed Example: A group therapist uses BastionGPT to draft summaries of each session, including the main topics discussed, shared experiences, and agreed-upon goals. This helps participants stay informed and engaged in their therapeutic journey. ## 5. Crisis Intervention Documentation ##### Example: A crisis counselor needs to document an intervention with a client experiencing a mental health crisis. BastionGPT can draft a detailed report, including the client's presenting issues, interventions used, and safety plan. ##### Capabilities: • Document crisis interventions. • Summarize presenting issues and interventions. • Create safety plans and follow-up recommendations. ##### Detailed Example: A crisis counselor uses BastionGPT to draft a report after intervening with a client experiencing a panic attack. The AI includes the client's symptoms, the de-escalation techniques used, and a safety plan for the client moving forward. ## 6. Research and Data Analysis in Mental Health ##### Example: A researcher needs to conduct a literature review on the impact of cognitive-behavioral therapy (CBT) on depression. BastionGPT can process numerous academic articles and summarize key findings, saving the researcher valuable time. ##### Capabilities: • Synthesize complex information and generate summaries. • Assist in literature reviews and data interpretation. • Create research reports and highlight key insights. ##### Detailed Example: A graduate student uses BastionGPT to summarize findings from multiple studies on the effectiveness of CBT for treating depression. The AI compiles the results into a cohesive report, highlighting trends, significant data points, and areas needing further research, aiding the student in writing a comprehensive thesis. ## 7. Patient Education Materials in Mental Health ##### Example: A therapist needs to create educational materials for patients with anxiety disorders. BastionGPT can generate informative and easy-to-understand content, including brochures, handouts, and digital resources. ##### Capabilities: • Generate patient education materials. • Summarize complex psychological concepts in layman's terms. • Create engaging and accessible content. ##### Detailed Example: A therapist uses BastionGPT to draft a series of brochures on managing anxiety. The AI generates clear and concise information on coping strategies, relaxation techniques, and resources for further support, helping patients better understand and manage their condition. ## 8. Teletherapy Communication and Documentation ##### Example: A teletherapy provider needs to maintain accurate and detailed records of virtual consultations. BastionGPT can draft detailed consultation summaries, including patient concerns, diagnoses, treatment plans, and follow-up recommendations. ##### Capabilities: • Draft consultation summaries and treatment plans. • Maintain accurate and detailed teletherapy records. • Generate follow-up recommendations and patient communication. ##### Detailed Example: A teletherapy provider uses BastionGPT to draft a summary of a virtual consultation with a patient. The AI includes the patient's reported symptoms, the provider's observations, the diagnosis, and the recommended treatment plan, ensuring that the teletherapy records are comprehensive and accurate. ## 9. Progress Tracking and Reporting ##### Example: A therapist needs to track and report a client's progress over several months of treatment. BastionGPT can generate detailed progress reports, highlighting improvements, setbacks, and recommendations for continued care. ##### Capabilities: • Generate progress reports for clients. • Highlight improvements and setbacks. • Provide recommendations for continued care. ##### Detailed Example: A therapist uses BastionGPT to draft a progress report for a client who has been in therapy for six months. The AI summarizes the client's progress, including improvements in mood and coping skills, as well as any challenges faced, and provides recommendations for future sessions. ## 10. Clinical Supervision and Training ##### Example: A clinical supervisor needs to document and summarize supervision sessions with trainee therapists. BastionGPT can draft detailed supervision notes, including trainee progress, discussed cases, and feedback provided. ##### Capabilities: • Document supervision sessions. • Summarize trainee progress and discussed cases. • Provide detailed feedback and recommendations. ##### Detailed Example: A clinical supervisor uses BastionGPT to draft notes from a supervision session with a trainee therapist. The AI includes a summary of the cases discussed, the trainee's progress and challenges, and the supervisor's feedback and recommendations for further development. ## Conclusion BastionGPT is a versatile, HIPAA compliant and intelligent assistant capable of transforming how we approach documentation, communication, and creative endeavors in the fields of therapy and psychiatry. Its applications in mental health documentation, individualized treatment planning, patient communication, group therapy summaries, crisis intervention documentation, research, patient education, teletherapy communication, progress tracking, and clinical supervision highlight its potential to drive efficiency and innovation across multiple domains. Embracing BastionGPT can unlock new levels of productivity and creativity, making it an essential tool for mental health professionals. ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial --- > Source: https://bastiongpt.com/post/why-we-switched-to-bastiongpt Healthcare Innovation # We Tested 5 AI Platforms for Writing Complex Reports: Why We Switched to BastionGPT Written By: Kiley Cuene and Symone Gelay • January 22, 2026 When our practice leadership asked us to find the most robust, HIPAA-compliant, and user-friendly AI platform for our clinical psychology practice, we knew we had our work cut out for us. Our practice creates comprehensive psychological assessment reports that often run 30-80 pages, include multiple standardized tests, and sometimes end up in court. We needed an AI tool that could handle complexity without sacrificing quality or compliance. Our practice evaluated five different platforms over several weeks, using real case data to compare how each one handled the demanding work we do every day. What we discovered surprised us and ultimately changed how our entire practice approaches clinical documentation. ## The Challenge: More Than Just Report Writing Before diving into what we found, it helps to understand what makes complex psychological reports unique: independent education evaluations (IEEs) are a prime example. When parents disagree with their school district about special education services, our reports become legal documents. They go to trial. Attorneys scrutinize every sentence. A single mistake (like mentioning "Shannon" when you're writing about "Billy") can discredit an entire evaluation and cost a family the services their child needs. These reports require us to: Analyze multiple assessment tools (behavioral rating scales, cognitive tests, achievement measures) Integrate data from parents, teachers, and direct observations Explain technical results in narrative form, not just list scores Connect test findings to real-world classroom functioning Provide evidence-based recommendations that hold up under legal scrutiny Meet strict timelines while maintaining forensic-quality standards Record reviews could take significantly longer than the three hours we typically bill for. Analyzing extensive IEPs, cross-referencing services, pulling relevant information, and synthesizing it into a coherent narrative was time-consuming work. Our clinical supervisor emphasized the need for efficiency without sacrificing accuracy. That's when we started looking at AI solutions. ## The Platforms We Tested Our practice evaluated five platforms designed for psychological assessment work: Assessment Assistant, Schoolpsych.ai, Leitner AI, Professionalpsyctools.com, and Sophia Report Writer . For each, we used the same approach: take a completed report we'd already written, input the raw data, and compare outputs. ## What We Found Several platforms used dropdown menus and pre-formatted fields. While this sounds efficient, it created significant limitations. These platforms only accepted specific assessments from predetermined lists. If we needed to analyze major cognitive measures like the WISC-V or WAIS-IV, achievement batteries like the KTEA-3, or behavioral assessments like the BASC-3 and BRIEF-2, we could only proceed if those exact tests were in their database. Any specialized measure we administered? Not supported. The output came in tables or bullet points. We write narrative reports in paragraph form because parents find them clearer and courts expect comprehensive explanations. Converting bulleted data into flowing prose meant starting from scratch anyway. Setup created barriers too. One platform required downloading a browser extension. Another made us enter complete demographic information before we could even test features. Compared to just opening a chat and asking a question, these felt like unnecessary obstacles. A couple of platforms offered chat interfaces, but had critical flaws. During free trials, some capped interactions at 2-5 queries. One system stopped us after two submissions with "You've reached the end of your free trial." Even when the format was right, the content wasn't there. We'd get generic summaries that listed scores without explaining what they meant for classroom functioning. ## Why BastionGPT Stood Out From our first interaction with BastionGPT, we noticed fundamental differences: ### 1. Immediate Usability No downloads. No demographic data entry. No restricted test lists. We uploaded our assessment reports, pasted in IEP sections, or simply typed our prompts. The platform worked with whatever we gave it. ### 2. Depth and Detail BastionGPT generated comprehensive narrative content that matched the complexity we needed. When we asked it to analyze the BASC results, it didn't just spit back scores. It explained what elevated anxiety scores typically indicate, how they might manifest in classroom behavior, and what interventions research supports. The platform understands medical and psychological terminology. When we needed to explain how ADHD connects to academic underachievement, BastionGPT provided evidence-based explanations with sources, something we rarely got from generic AI tools like Gemini (which would essentially say "it affects it because it affects it"). ### 3. Iterative Refinement This became our favorite feature. We could build reports section by section, refining as we went: "Include scores in parentheses" "Write in past tense" "Incorporate the teacher's results into the paragraph I already wrote" "Remove the word 'suggesting' and use 'demonstrating' instead" Each request worked. We could scroll back through our conversation to see what data we'd already provided. We could add information as it came in (like when a teacher submitted rating scales late). The chat format meant we weren't locked into a rigid workflow. For longer evaluations, the 10-plus-page report technique explains how to build full-length reports without losing quality. ### 4. No Artificial Limits Unlike competitors that restricted trial users to a handful of queries, BastionGPT let us fully test every feature we needed. We could process multiple documents, try different prompting strategies, and really understand whether it worked for our use cases. ### 5. Time Savings We Could Measure Record reviews that took 12 hours manually now take 3-4 hours with BastionGPT handling the initial analysis. We still review everything carefully (we have to, especially for cases going to trial), but we're starting from a comprehensive draft rather than a blank page. Report edits dropped from 3-4 hours to about an hour. When our supervising psychologist gives feedback like "this section needs more declarative language" or "you can't use tentative phrasing here," we can tell BastionGPT exactly what needs to change and get refined text in seconds. For our practice, this translates directly to cost savings. We bill record reviews at 50 pages per hour, capped at three hours. When reviews took six hours manually, the practice absorbed that cost. Now we consistently stay within the estimated time, which means better margins without charging families more. ## What About the Other Use Cases? Beyond report writing and record reviews, we use BastionGPT for IEP summarization (even 120-page documents), edit implementation when we receive feedback on drafts, test interpretation guidance, and meeting prep. Some colleagues ask it strategic questions before IEP meetings about how to present findings or address parent concerns. ## The Learning Curve Getting the best results from BastionGPT requires understanding how to prompt effectively. We learned to be specific about format ("Write in paragraph form"), build incrementally rather than asking for an entire report at once, provide examples of our preferred style, and always review critically. We've created a shared document of effective prompts our team uses, like: "Please create a concise summary paragraph using clear and firm language, referring to absolute weaknesses and absolute strengths" or "Incorporate the teacher's results into the paragraph I already wrote, including T scores in parentheses." ## Why Healthcare-Specific AI Matters One thing we didn't fully appreciate until this evaluation: general-purpose AI tools don't understand healthcare contexts the same way. BastionGPT has built-in knowledge about medical conditions, psychological constructs, and evidence-based interventions. It's not just accessing the same information as ChatGPT or Gemini; it's accessing it through a healthcare lens. When we ask about how working memory deficits affect reading comprehension, BastionGPT understands we need research-backed explanations suitable for clinical documentation, not a Reddit summary or general definition. It knows to cite sources, use appropriate terminology, and maintain professional standards. ## Our Recommendation After testing five platforms, BastionGPT was the clear choice for our practice. It's the only tool that combines: Flexibility to handle any assessment or document type Depth to support forensic-quality reporting Speed to save meaningful time on every case Healthcare-specific intelligence that understands clinical contexts No artificial usage barriers The other platforms might work for simpler use cases or practices with different workflows. If you only use three standardized tests and prefer bulleted outputs, a structured platform might suffice. But for comprehensive psychological assessment work, especially when reports might go to trial, we needed something more capable. BastionGPT saves us hours per case, supports our complex reporting needs, and helps us provide the detailed evaluations that families and courts require. For our work, that makes it invaluable. ## What We Tell Other Practices If you're evaluating AI tools for clinical work: Test with real cases: Use actual completed reports to compare outputs Look beyond features lists: A platform that claims to do everything might do nothing well Consider your workflow: Chat interfaces offer more flexibility than structured forms Evaluate depth, not just speed: Fast but shallow outputs don't save time if you have to rewrite everything Check usage limits: Some "free trials" barely let you test basic functionality Think about edge cases: Can it handle your most complex reports, not just average ones? For us, those criteria led to BastionGPT. We love that we found a tool that actually makes our work better, not just faster. ## About the Authors Kiley Cuene and Symone Gelay are doctoral candidates in clinical psychology specializing in psychological assessments. They conduct comprehensive evaluations for children with suspected learning disabilities and developmental differences. Related Resources: BastionGPT Use Cases for Mental Health Professionals Healthcare AI Prompts: Building Your Clinical Library How a Physician Transformed Their Practice with BastionGPT Video: How to Write Complex Reports with BastionGPT Begin your journey with a 7-day Free trial of BastionGPT . If you have questions or want to connect: Email: [email protected] Phone: (214) 619-8696 Schedule a Chat with BastionGPT experts: Book a Meeting ## Try BastionGPT free for 7 days Draft progress notes, letters, and treatment plans in a HIPAA-compliant AI assistant. A signed BAA is included on every plan, and pricing starts at $20 a month. Start your free trial